WO2012015898A1

WO2012015898A1 - Systems and methods for measuring velocity of a particle/fluid mixture

Info

Publication number: WO2012015898A1
Application number: PCT/US2011/045506
Authority: WO
Inventors: Eric J. Markel; Robert O. Hagerty; Harry W. Deckman
Original assignee: Univation Technologies, Llc
Priority date: 2010-07-28
Filing date: 2011-07-27
Publication date: 2012-02-02

Abstract

Systems and methods for estimating a velocity of a particle/fluid mixture are provided. The method can include flowing a particle/fluid mixture through a fluid conveying structure past a sensor. A signal detected by the sensor as particles in the particle/fluid mixture pass the sensor can be measured to provide a measured signal. The measured signal can be manipulated to provide an output. A velocity of the particle/fluid mixture can be determined using the output.

Description

SYSTEMS AND METHODS FOR MEASURING VELOCITY OF A PARTICLE/FLUID

MIXTURE

BACKGROUND

[0001] An important process control parameter for polymerization processes, especially gas phase polymerization, is the superficial velocity of the cycle gas flowing through the reactor. If the superficial velocity of the cycle gas is too low fluidization within the reactor can be lost causing the fluidized bed to collapse and shut down the reactor. Conversely, if the superficial velocity of the cycle gas is too high an excessive amount of catalyst and polymer particles from the fluidized bed are carried out of the reactor and into the cycle fluid line

[0002] The velocity of the cycle fluid within the cycle fluid line is typically measured using a venturi flowmeter. From the cycle gas velocity measurements the superficial velocity of the cycle gas through the reactor can readily be calculated. The venturi flowmeter consists of a smoothly varying constriction in the cycle fluid piping, where the velocity of the cycle fluid is forced to increase, thereby causing the pressure of the cycle fluid flowing therethrough to decrease due to the Bernoulli effect. A differential pressure ("DP") transmitter or other suitable instrument is used to measure the pressure at the point of minimal cross-sectional area or "throat" of the venturi flowmeter relative to the upstream pressure of the cycle fluid. This pressure difference is then used to infer the cycle fluid velocity, typically by employing the Bernoulli equation.

[0003] Venturi flowmeters, however, are especially prone to errors when used in polymerization processes. The errors may be caused by at least one of two types of fouling. The first type of fouling occurs when polymer buildup partially or completely plugs the small-diameter flow passages used to connect the DP transmitter to the venturi flowmeter. The second type of fouling occurs when the inner surfaces of the venturi flowmeter are partially or completely coated with polymer buildup. The result of either or both types of fouling causes incorrect cycle gas velocity measurements that in turn cause errors in calculating the superficial velocity of the cycle gas through the reactor. Incorrect superficial velocity determinations increase the potential for fouling to occur within the reactor.

[0004] There is a need, therefore, for improved systems and methods for estimating a cycle fluid velocity within a polymerization system cycle fluid line and/or the superficial velocity of the cycle gas fluid through the polymerization reactor.

SUMMARY

[0005] Systems and methods for estimating a velocity of a particle/fluid mixture are provided. The method can include flowing a particle/fluid mixture through a fluid conveying structure past a sensor. A signal detected by the sensor as particles in the particle/fluid mixture pass the sensor can be measured to provide a measured signal. The measured signal can be manipulated to provide an output. A velocity of the particle/fluid mixture can be determined using the output.

[0006] Another method for estimating a velocity of a particle/fluid mixture can include flowing a particle/fluid mixture through a fluid conveying structure past a first sensor and a second sensor, wherein the first and second sensors are separated from one another by a predetermined distance (d). At least one signal detected by the first sensor as particles in the particle/fluid mixture pass the first sensor can be measured to produce a first measured signal. At least one signal detected by the second sensor as the particles pass the second sensor can be measured to produce a second measured signal. The first and second measured signals can be manipulated to provide an output. A transit time can be determined from the output. The distance (d) can be divided by the transit time to provide a velocity of the particle/fluid mixture.

[0007] The system for estimating a velocity of a particle/fluid mixture can include a fluid conveying structure, a sensor, an electrometer, and a processor. The fluid conveying structure can have an internal volume for flowing a particle/fluid mixture therethrough. The sensor can be adapted to detect at least one signal generated as particles in the particle/fluid mixture pass the probe without contacting the probe. The electrometer can be communication with the sensor and can be adapted to measure the signal detected by the sensor. The processor can be in communication with the electrometer. The processor can receive the measured signal and manipulate the measured signal to provide an output. The processor can also determine a velocity of the particle/fluid mixture using the output.

[0008] Another system for estimating a velocity of a particle/fluid mixture can include a fluid conveying structure, a first sensor, a second sensor, and a processor. The fluid conveying structure can have an internal volume for flowing a particle/fluid mixture therethrough. The first sensor can be in communication with the internal volume and can be adapted to detect and measure at least one first signal generated as particles in the particle/fluid mixture pass the first sensor. The second sensor can be in communication with the internal volume and can be adapted to detect and measure at least one second signal generated as the particles pass the second sensor, wherein the first and second sensors are located a predetermined distance (d) apart from one another. The processor can be in communication with the first and second sensors and can be adapted receive and manipulate the first and second measured signals to provide an output. The processor can determine from the output a transit time. The processor can divide the distance (d) by the transit time to provide a velocity of the particle/fluid mixture. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figure 1 depicts a schematic of an illustrative velocity measurement system for estimating a velocity of a particle/fluid mixture flowing through a fluid conveying structure.

[0010] Figure 2 depicts a schematic of an illustrative velocity measurement system for estimating the velocity of a particle/fluid mixture flowing through a fluid conveying structure.

[0011] Figure 3 depicts a schematic of an illustrative gas phase polymerization system having the velocity measurement systems depicted in Figures 1 and 2.

[0012] Figure 4 is a graphical depiction of measured entrainment static detected by a static probe during monitoring of a cycle fluid flowing through a gas phase polymerization reactor cycle line over five minutes of operation.

[0013] Figure 5 is a close-up view of an absolute autocorrelation vector calculated from the measured entrainment static depicted in Figure 4 focused on a time lag of zero (0) seconds.

[0014] Figure 6 depicts a graphical depiction of measured entrainment static detected by a static probe and measured acoustic emissions detected by an acoustic emissions sensor during monitoring of a cycle fluid flowing through a gas phase polymerization reactor cycle line over five minutes of operation.

[0015] Figure 7 depicts a graphical depiction of the absolute values of the mean centered measured entrainment static and the measured acoustic emissions depicted in Figure 6.

[0016] Figure 8 depicts a graphical depiction of a cross-correlation sequence calculated from the absolute values of the mean centered measured entrainment static and the measured acoustic emissions depicted in Figure 7.

DETAILED DESCRIPTION

[0017] Figure 1 depicts a schematic of an illustrative velocity measurement system 100 for estimating a velocity of a particle/fluid mixture 105 flowing through a fluid conveying structure 115. The velocity measurement system 100 can include one or more probes or sensors (one is shown) 103, electrometers (one is shown) 140, and processors (one is shown) 150. The sensor 103 can be configured to detect one or more electrical signals or properties from charged particles (two are shown 107, 109) in a particle/fluid mixture 105. For example, a tip 104 of the sensor 103 can be positioned or located within an internal volume or "flow path" 1 17 of a fluid conveying structure 115 such that the electrical signals or properties of the charged particles 107, 109 can be detected by the sensor 103 as the particle/fluid mixture 105 flows through the fluid conveying structure 1 15. As shown in Figure 1, two separate charged particles 107, 109 are depicted with each charged particle 107, 109 following a separate path 108, 110, respectively, through the fluid conveying structure 1 15. The particles 107, 109 can each be separate or discrete particles, as shown, and/or agglomerations or aggregations of multiple particles, not shown. The particles 107, 109 can be suspended, entrained, or otherwise contained in or carried by the fluid of the particle/fluid mixture 105. The fluid in the particle/fluid mixture 105 can be a gas, a liquid, or a combination thereof.

[0018] The sensor 103 can detect the electrical signal of the particles 107, 109 as the charged particles 107, 109 approach and pass and/or approach and contact the sensor 103. For example, in following path 108, the charged particle 107 passes the sensor 103 without contacting with the sensor 103. As the particle 107 approaches the sensor 103 an approaching electrical signal or "leading lobe" can be detected via the sensor tip 104. After the particle 107 passes and advances away from the sensor 103 a leaving electrical signal or "lagging lobe" can be detected via the sensor tip 104. The detected electrical signal via line 106 can be communicated to the electrometer 140 and then to ground. The electrometer 140 can measure or otherwise estimate the electrical signal detected via the sensor 103 as the particle 107 passes the sensor 103. In another example, in following path 1 10, the charged particle 109 comes into direct contact with the sensor 103. The electrical signal of the particle 109 can be transferred to the sensor 103. For example, the charged particle 109 can transfer its charge to the sensor tip 104 and the transferred charge via line 106 can be communicated to the electrometer 140 and then to ground. The electrometer 140 can measure or otherwise estimate the charge transferred from the particle 109 to the sensor 103. In other words, the sensor 103 can detect the one or more electrical signals of the particles 107 that pass the sensor 103 without contacting the sensor 103 and the one or more electrical signals of the particles 109 that approach the sensor 103 and contact the sensor 103. As such, the electrometer can measure or otherwise estimate the electrical signals detected by the sensor 103 as some charged particles 107 pass the sensor without contacting the sensor while other charged particles 109 contact the sensor 103.

[0019] The charged particles 107, 109 can be positively charged or negatively charged. For example, the charged particle 107 can be positively charged and the charged particle 109 can be negatively charged. In another example, the charged particle 107 can be negatively charged and the charged particle 109 can be positively charged. In another example, the charged particles 107, 109 can both be positively charged or negatively charged. The sensor 103 can detect both positive electrical signals and negative electrical signals. For example the sensor 103 can be capable of detecting both positive current and negative current. As such, in at least one example, the sensor 103 can be referred to as a "bipolar" sensor. A suitable and commercially available sensor 103 can be the Electrostatic Monitor Probe (model ESM3400) available from Progression, Inc. [0020] It has been surprisingly and unexpectedly discovered that the one or more electrical signals or properties detected by the sensor 103 can be used to determine or estimate the velocity of the particle/fluid mixture 105 flowing through the fluid conveying structure 1 15. The electrical signal(s) of interest from which the velocity of the particle/fluid mixture can be determined from are the particles 107 that approach and pass the sensor 103 without striking or contacting the sensor tip 104. As such, reducing or eliminating the number or frequency at which the particles 109 strike or contact the sensor tip 104 can be desirable.

[0021] Any one or more of a number of steps can be taken in order to reduce the number or frequency of particles 109 contacting the sensor tip 104. For example, to reduce the number or frequency of particles 109 striking or contacting the sensor 103, the sensor tip 104 can be flush with an inner surface 1 16 of the fluid conveying structure 115. In another example, the sensor tip 104 can be recessed within the inner surface 116 of the fluid conveying structure 115. In still another example, the positioning of the sensor tip 104 can be adjustable, e.g., can be moved inward and outward relative to the internal volume 117. As such, the position of the sensor tip 104 within the internal volume 117 can be changed or modified. For example, if the particle/fluid mixture 105 includes a large amount of particles 107, 109, e.g., greater than about 0.5 wt%, the sensor tip 104 could be positioned flush with or recessed within the inner surface 116. In another example, if the particle/fluid mixture 105 includes a relatively small amount of particles 107, 109, e.g., less than about 0.1 wt%, based on the total weight of the particle/fluid mixture, the sensor tip 104 could be positioned such that the sensor tip 104 extends into the internal volume 1 17.

[0022] In another example, the sensor tip 104 can be coated with an insulating material to reduce or eliminate undesired charge transfer by particles 109 contacting the sensor tip 104. For example, the sensor tip 104 and/or the sensor 103 can be coated with one or more polymers or polymer containing materials. Illustrative coating materials can include, but are not limited to, polytetrafluoroethylene (PTFE), perfluoroalkoxy (PFA), fluorinated ethylene propylene (FEP), or any combination thereof. The coating material can include XYLAN®, commercially available from Whitford Corporation. The coating can have a thickness ranging from a low of about 10 μιη, about 50 μιη, about 100 μ, or about 150 μιη to a high of about 200 μιη, about 300 μιη, about 500 μιη, or about 1,000 μιη.

[0023] In still another example, the sensor tip 104 can be shielded by positioning a plate, wall, screen, or other shielding structure (not shown) upstream from the sensor tip 104. Shielding the sensor tip 104 with a shielding structure can deflect or block the particles 109 from contacting the sensor tip 104. In other words, shielding the sensor tip 104 with a shielding structure can alter the flow path 110 of the particles 109 such that the particles 109 approach and pass the sensor tip 104 without contacting the sensor tip 104.

[0024] The electrometer 140 can measure or estimate, for example, a current and/or voltage, detected via the sensor 103. The measured current and/or voltage can also be referred to as "entrainment static," which is caused by the charged particles entrained or carried in the fluid. An electrometer 140 that detects a flow of current from the sensor tip 104 to ground can include, but is not limited to, an ammeter, a picoammeter (a high sensitivity ammeter), or a multi-meter. In another example, the electrometer 140 could also detect the current flow indirectly by measuring or estimating a voltage generated as the current flows through a resistor. As such, the sensor 103 can include any sensor, sensor, or other device capable of being monitored via the electrometer 140 to measure, estimate, or otherwise detect one or more electrical signals such as current and/or voltage.

[0025] The electrometer 140 can have a response time of about 0.05 seconds ("sec") or less, or about 0.01 sec or less, or about 0.009 sec or less, or about 0.007 sec or less, or about 0.005 sec or less. For example, the electrometer 140 can have a response time ranging from about 0.0001 sec to about 0.01 sec, or from about 0.001 to about 0.008 sec, or from about 0.003 sec to about 0.006 sec. The electrometer 140 can also include a 4 mA to about 20 mA transmitter.

[0026] The sensor 103 and electrometer 140 can detect and measure the signal(s) at any desired sampling rate or frequency. For example, the sensor 103 and electrometer 140 can detect and measure the electrical signal(s) at a sampling frequency of greater than or equal to about 90 Hz, or greater than or equal to about 100 Hz, or greater than or equal to about 125 Hz, or greater than or equal to about 150 Hz, or greater than or equal to about 200 Hz, or greater than or equal to 200 Hz. In another example, the sensor 103 and electrometer 140 can detect and measure the electrical signals(s) at a sampling frequency of about 100 Hz, about 500 Hz, about 1,000 Hz, about 5,000 Hz, about 10,000 Hz, or more than about 10,000 Hz.

[0027] The signal(s) detected via the sensor 103 and measured via the electrometer 140 can be communicated as "raw" data via line 142 to the processor 150. The processor 150 can manipulate the detected electrical signal(s) or "raw" data received via in line 142 to provide an output or manipulated electrical signal via line 152. The output in line 152 can provide information as to one or more conditions of the particle/fluid mixture 105 within the fluid conveying structure 115.

[0028] The processor 150 can manipulate the electrical signal or "raw" data received via line 142 using any desired process or combination of processes. For example, the signal in line 142 can undergo one or more mathematical operations to produce the output or manipulated electrical signal via line 152. In one example, the processor 150 can manipulate or process the electrical signal received via line 142 from the electrometer 140 using the absolute autocorrelation method. For example, the data communicated via line 142 from the electrometer 140 to the processor 150 can undergo a certain correlation process that correlates the absolute values of the mean centered data. Correlation of the absolute values of the mean centered electrical signal in line 142 can provide a signal-processing tool capable of extracting from the electrical signal in line 142 information that can be used to estimate or determine the velocity of the particle/fluid mixture 105.

[0029] The main or primary features of the absolute autocorrelation of a measured electrical signal, e.g., current, in line 142, and provided as the output via line 152, can include, but are not limited to, an approaching curve or "leading lobe," a leaving curve or "lagging lobe," and a zero- lag peak or peak at zero (0) time lag. The leading and lagging lobes can be indicative of the charge on the particles 107, as the particles 107 approach the sensor 103 and as the particles 107 pass and move away from the sensor 103, respectively. The peak at zero (0) time lag can be indicative of the charge on the particles 109 that strike or contact the sensor 103. Either the leading lobe or the lagging lobe can be used to estimate or determine the velocity of the particle/fluid mixture 105.

[0030] The correlation calculations can be performed using the function "xcorr" in the commercially available Matlab software (available from The MathWorks). Alternatively, the correlation calculations can be performed on a computer or other processing system (e.g., processor 150) programmed in another appropriate manner. To calculate the correlation of vectors x and y (of equal size, n) using the Matlab software, the command "output = xcorr(x,y)" can be executed in the Matlab environment. Autocorrelation of vector x with itself is performed as a special case, using the command "output=xcorr(x)." Other suitable software that can be used to perform the absolute autocorrelation calculations can include, but are not limited to, Labview, Mathematica, and MathCad. We have found a correlation method especially useful for the analysis of data which involves (i) mean centering the data vector x (by subtracting the mean of the vector from each value) to get vector y, (ii) computing the vector z (comprising the absolute values of each datum in vector y), and (iii) calculating the output vector q (using the correlation function q = xcorr(y,z)). This general procedure, for the present purposes, will be referred to as "absolute autocorrelation." The vector q (the absolute autocorrelation vector) has 2n-l terms, with the n^th term comprising the absolute autocorrelated value corresponding to zero time lag. Covariance is a well known parameter related to autocorrelation and cross-correlation. Covariance values rather than cross correlated values can be determined in some embodiments of the invention. Similarly, the Matlab function "detrend' is closely related to the mean centering procedure describe above and can be used in some embodiments of the invention. [0031] The absolute autocorrelated data or output in line 152 can include or indicate an offset time or passing time for the particles 107 that approach and pass the sensor tip 104 without contacting or striking the sensor tip 104. Depending on the average charge of the particles 107 passing the sensor tip 104, the leading lobe has a peak (either positive or negative) that is offset from zero (0) time lag by a period of time and the lagging lobe has a peak that is opposite in sign of the leading lobe and is offset from zero (0) time lag by a period of time that is substantially equal to the offset of the leading lobe. If the leading lobe has a negative peak, the particles 107 passing the sensor tip 104, on average, are negatively charged. If the leading lobe has a positive peak, the particles 107 passing the sensor tip 104, on average, are positively charged. For example, if the leading lobe has a negative peak that is offset from zero (0) time lag by a period of about -0.1 seconds, the lagging lobe would have a positive peak that is offset from zero (0) time lag by a period of about +0.1 seconds. As such, in addition to indicating the charge of the particles 107, the position of the leading and/or lagging lobe peak(s) (positive or negative) relative to zero (0) time lag can be used to estimate or determine the velocity of the particle/fluid mixture 105.

[0032] The particular offset times of the leading and lagging lobes are dependent on the particular velocity of the particle/fluid mixture 105. For example, as the velocity of the particle/fluid mixture 105 increases, the offset times, i.e. the position between the leading lobe's peak and the lagging lobe's peak relative to zero (0) time lag decrease toward the zero (0) time lag. Conversely, as the velocity of the particle/fluid mixture 105 decreases, the offset times, i.e. the position between the leading lobe's peak and the lagging lobes peak relative to zero (0) time lag, increases away from the zero (0) time lag.

[0033] The absolute autocorrelation process can be carried out on the signal(s) detected and measured by the sensor 103 and electrometer 140 acquired over relatively short periods of time, e.g., 10 seconds, with these relatively short periods of time being repeated over a given time window to provide a data set that includes a series of relatively short term values. The absolute autocorrelation of the data set, i.e. the series of signal(s) detected and measured over relatively short periods of time, could then provide a series of leading and lagging lobes. The median value of the leading lobes or the lagging lobes could then be used to provide the offset time.

[0034] The velocity of a particular particle/fluid mixture 105 flowing through the fluid conveying structure 115 can be determined using one of at least two approaches. The first approach can include developing or building a calibration curve that measures a plurality of different velocities of a particle/fluid mixture 105 flowing through a particular fluid conveying structure 1 15 using, for example, a conventional venturi flowmeter, and also estimating or determining the offset time(s) of the leading lobe peak and/or the lagging lobe peak from the raw data in line 142 at the different velocities using, for example, absolute autocorrelation. The calibration curve prepared can then be used to determine the velocity for which subsequent offset times are determined for a given particle/fluid mixture 105.

[0035] The second approach can include determining a proportionality constant (k) that is dependent on the particular geometry of the fluid conveying structure 105 and the particular geometry of the static sensor 103. In using this approach, the velocity of the particle/fluid mixture can be expressed by Equation 1 :

k

v = (Equation 1)

offset time

where v is the velocity of the particle/fluid mixture, k is the proportionality constant (k), and offset time is the time difference between the leading lobe peak or the lagging lobe peak and zero (0) time lag as determined from the absolute autocorrelation.

[0036] The proportionality const determined by using the following equation:

(Equation 2)

where q is the charged particle, ε₀ is the permittivity constant, v is the velocity of the charged particle, and L is the distance from the wall.

[0037] The value of the proportionality constant (k) can determined by numerical integration of the absolution autocorrelation of Equation 2 given the geometry of the a cycle gas piping (3 ft diameter pipe, considered essentially infinite) and an entrainment static probe extending to the center of the cycle gas line piping. Monte-Carlo computations can then be used to extract the value of the proportionality constant (k). There is no known closed-form solution for the integral of the absolute autocorrelation of Equation 2.

[0038] For each integration at a specific velocity, 10,000 random particles were used to generate an ensemble of current waveforms (with data collection at 100 Hz), each with an associated absolute autocorrelation and leading lobe "passing time" or "delay time." The median "passing time" for each assumed velocity was determined in this way to generate a simulation that relates the average velocity (ft/s) to the absolute autocorrelation of the leading lobe. Linear regression was then used for a series of simulations at 0.5 ft/s to 8 ft/s to determine the value of the proportionality constant (k).

[0039] Monte Carlo methods (or Monte Carlo experiments) are a class or type of computational algorithms that use repeated random sampling data to compute their results. Because of their reliance on repeated computation of random or pseudo-random numbers, these methods are most suited to calculation by a computer and tend to be used when it is unfeasible or impossible to compute an exact result with a deterministic algorithm.

[0040] An exemplary process in which the velocity measurement system 100 can be used to determine the velocity of the particle/fluid mixture 105 from the measured electrical signals can be a gas phase polymerization system. The gas phase polymerization system can use one or more metallocene catalysts, for example, to polymerize one or more olefins. From the measured electrical signals or data the relative amounts of entrained solid particles (i.e. catalyst particles and polymer particles) that pass and/or strike the sensor 103 (as well as their average charges) can be extracted from the measured electrical signals. For example, a current can be measured via the electrometer 140 and the processor 150 can provide an output 152 that shows the absolute autocorrelation of the current. As discussed above, the absolute autocorrelation of the current (or voltage) can be used to estimate or determine the velocity of the particle/fluid mixture 105.

[0041] The particles/fluid mixture 105 flowing past the sensor 103 can be at a high speed, e.g. about 15 m/s, and as such, the passing and striking events can occur in a short period of time, e.g. less than 1 second. Additionally, there can be a substantial number of particles 107, 109 interacting with the sensor 103. The absolute autocorrelation methods can provide a useful means to extract an average description of these multiple, fast interactions between the moving charged particles 107, 109 and the sensor 103. Because the interactions occur on such a short time scale, the interactions are observed in the absolute autocorrelation of data in the region corresponding to a fraction of a second lead/lag time.

[0042] As mentioned above, the primary features of the absolute autocorrelation of the measured current in line 142 and provided as the output via line 152 can include a leading lobe, a lagging lobe, and a zero-lag peak or peak at zero time lag. For a typical gas phase polymerization system, the leading lobe can be located at about -0.11 seconds and the lagging lobe can usually be a mirror image of the leading lobe and be located at about +0.11 seconds. For gas phase polymerization using metallocene catalyst, it has been found that the leading lobe is nearly always a minimum-type peak and the lagging lobe is nearly always a positive-type peak. This observation corresponds with typical entrainment static measurements that indicate entrained polymer product usually has a negative charge. As such, the "negative" or minimum- type leading lobe usually indicates negatively charged polymer particles approaching the sensor 103 and the lagging lobe usually indicates the negatively charged polymer particulates leaving or passing away from the sensor 103. Accordingly, in most data reviewed for gas phase polymerization using metallocene catalyst, the leading and lagging lobes are minima and maxima, respectively, indicating the charged particles in the cycle gas are dominated by negatively charged polymer particles.

[0043] The offset time between a peak of either the leading or lagging lobe can be used in Equation 1 (discussed above) to estimate or otherwise determine the velocity of the flowing particle/fluid mixture 105 through the fluid conveying structure 1 15.

[0044] Figure 2 depicts a schematic of another illustrative velocity measurement system 200 for estimating the velocity of a particle/fluid mixture 205 flowing through a fluid conveying structure 215. The velocity measurement system 200 can include two or more sensors (two are shown 220, 225) and processors (one is shown) 230. The first sensor 220 and second sensor 225 can be configured to detect and measure one or more signals generated by particles 207 as the particle/fluid mixture 205 flows through the fluid conveying structure 215 and past the first and second sensors 220, 225, respectively. The particle/fluid mixture 205, particles 207, and fluid conveying structure 215 can be similar to the particle/fluid mixture 105, particles 107, 109, and fluid conveying structure 115, discussed and described above with reference to Figure 1.

[0045] The one or more signals detected and measured via the first sensor 220 can be communicated via line 222 as a first measured signal or first data set to the processor 230. Similarly, the one or more signals detected and measured via the second sensor 225 can be communicated via line 227 as a second measured signal or second data set to the processor 230.

[0046] The first and second sensors 220, 225 can be separated by a distance (d) along the fluid conveying structure 215. The distance (d) can be any desired length. For example, the first and second sensors 115, 120 can be separated by a distance (d) ranging from a low of about 1 cm, about 10 cm, about 1 m, or about 5 m to a high of about 10 m, about 30 m, about 50 m, or about 100 m. The distance (d) can depend, at least in part, on the particular type of fluid conveying structure 1 10.

[0047] The concentration or amount of particles 207 can vary or fluctuate over time as the particle/fluid mixture flows through the fluid conveying structure 215. In other words, the distribution of the particles 207 in the particle/fluid mixture 205 can be non-uniform or variable as the particle/fluid mixture 205 flows through the fluid conveying structure 215.

[0048] As shown, the first sensor 220 can be disposed on an outer wall 218 of the fluid conveying structure 215 and can include a probe or tip 221 that extends into an internal volume or "flow path" 217 of the fluid conveying structure 225. The second sensor 225 can be disposed on the outer wall 218 of the fluid conveying structure 215 and can monitor the flowing particle/fluid mixture 207 from outside the internal volume 217. In another example, the first and second sensors 220, 225 can both include a probe tip or other extension that can be disposed within the internal volume 217. In some embodiments, the probe tip 221 or other extension can be coated or shielded as described above with reference to sensor tip 104 in Figure 1. In another example, the first and second sensors 220, 225 can both monitor the flowing particle/fluid mixture 207 from outside the fluid conveying structure 215.

[0049] Depending on the particular type of sensor, the at least one signal generated as the particles 207 approach and pass the first and second sensors 220, 225 can include, but is not limited to, one or more electrical signals such as current or voltage, acoustic signals, dielectric signals, optical signals, capacitance signals, or any combination thereof. As such, the first and second sensors 220, 225 can be or include a static probe or static sensor, also referred to as an "entrainment" static probe; an acoustic sensor; a dielectric sensor; an optical sensor; or a capacitance sensor.

[0050] As such, the particles 207 can be charged particles and similar to the charged particles 107, 109 discussed and described above with reference to Figure 1. In another example, the particles 207 could be neutral in charge and other properties or signals generated as the particles 207 pass the first and second sensors 220, 225 can be detected and measured by the first and second sensors 220, 225. For example, an optical sensor could use an optical wavelength emitted from a source that could be directed toward the particle/fluid mixture 205, interact with the particles 207, and the interaction could be detected via a detector. The detected interactions could include reflection, transmission, absorbance, and the like. A dielectric sensor could detect the permittivity of the particles 207 as the flow past the dielectric sensor. A capacitance sensor could detect and measure, for example, particle motion, particle composition, and electric fields.

[0051] The first and second sensors 220, 225 can be the same type of sensor. For example, the first sensor 220 and the second sensor 225 can both be static probes or static sensors. In another example, the first sensor 220 and the second sensor 225 can both be acoustic emissions sensors. Alternatively, the first and second sensors 220, 225 can be different types of sensors. For example, the first sensor 220 can be an acoustic emissions sensor and the second senor 225 can be a static probe or static sensor, or vice versa.

[0052] The first sensor 220, the second sensor 225, or both can be configured to detect and measure two or more different signals or different types of signals. For example, the first sensor 220 and/or the second sensor 225 can be configured to measure an electrical signal and an optical signal. In another example, the first sensor 220 and/or the second sensor 225 can be configured to measure an electrical signal and an acoustic signal.

[0053] The processor 230 can manipulate all or a portion of the first and second measured signals introduced via lines 222 and 227 from the first and second sensors 220, 225, respectively. For example, the processor 230 can manipulate the measured signals in lines 222 and 227 using a correlation calculation or process. If the first and/or second measured signals in lines 222, 227 include two or more different types of signals, e.g., acoustic emission signals and optical signals, the processor can select a particular data set, e.g., the acoustic emission signal, for the correlation calculation. The processor 230 can manipulate the measured signals in lines 222 and 227 using cross correlation to produce a cross correlation sequence as the output or manipulated signal via line 231. The output via line 231 can be introduced to a display such as a monitor, an automated control system, another processor, a computer, or the like, or combinations thereof.

[0054] The cross correlation sequence calculations can be performed using the function "xcorr" in the commercially available Matlab software (available from The Math Works). Alternatively, the cross correlation sequence calculations can be performed on the processor 230 programmed in another appropriate manner. To calculate the cross correlation sequence of vectors x and y (of equal size) using the Matlab software, the command "output = xcorr(x,y)" is executed in the Matlab environment. If the number of data points in the measured signals in lines 222, 227 are different, averaging or decimation methods can be used to reduce oversampled data sets to match the time axis of the slowest dataset.

[0055] When the first and second detected and measured signals via lines 222, 227 both include "state variable" data or data that reflects the state of the particle/fluid mixture type signals such as acoustic emission data, the cross correlation calculations can include cross correlating the two detected and measured signals. When one of the first and second detected and measured signals via lines 222, 227 includes "non-state variable" type of data, such as a current, additional steps can be included in the cross correlation calculations. For example, when the first measured signal via line 222 includes a non-state variable type data set (e.g., current) type signal and the second measured signal via line 227 includes a state variable type data set (e.g., acoustic) type signal, the cross correlation calculation can include determining the absolute values of the mean centered non-state variable data set (e.g., the current) to provide an absolute value vector of the mean centered first measured signal. The absolute value vector of the mean centered first measured signal can then be cross correlated with the second measured signal to provide the cross correlation sequence via line 231. In another example, if both the first and second measured signals via lines 222, 227 both include a non-state variable (e.g., current) type signal the absolute values of the mean centered first and second measured signals 222, 227 can first be determined and then cross correlated with one another to provide the cross correlation sequence via line 231.

[0056] The cross correlation sequence via line 231 can be used to determine a velocity of the particle/fluid mixture 205 flowing through the fluid conveying structure 215. The cross correlation sequence via line 231 can provide a transit time. The transit time can be determined, for example, from a graph of the cross correlation sequence. The transit time can be the time shift of a peak of the cross correlation sequence relative to a zero (0) transit time. The distance (d) between the first and second sensors 220, 225 can be divided by the transit time to provide an estimate or determination of the velocity of the particle/fluid mixture 205. The formula for determining the velocity can be expressed by Equation 3 : v = (Equation 3)

transit time

where v is the velocity, d is the distance between the first sensor 220 and the second sensor 225, and the transit time is estimated or determined from the cross correlation sequence.

[0057] Continuing with reference to both Figures 1 and 2, the velocity measurement systems 100, 200 can be used to monitor any process or system that includes, produces, uses, potentially could include, potentially could produce, potentially could use, or otherwise contains or could contain a charged particle/fluid mixture 105, 205. Illustrative systems can include, but are not limited to, slurry based polymerization systems, solution based polymerization systems, gas phase polymerization systems, coal gasification, catalytic reforming, catalytic cracking, cement processing, ash or carbon processing operations, and the like.

[0058] Accordingly, the particles 107, 109, and 207 in the particle/fluid mixtures 105, 205 can include polymer particles, catalyst particles, coal, ash, zeolites, and the like. The particle/fluid mixtures 105, 205 can also include a combination of two or more different particles, e.g., a polymerization system or process could include a particle/fluid mixture containing both polymer particles and catalyst particles. The fluid can be in the gaseous phase, liquid phase, or a combination thereof. Illustrative fluids can include, but are not limited to, hydrocarbons, e.g., alkanes and alkenes, liquid water, steam, nitrogen, carbon dioxide, carbon monoxide, hydrogen, oxygen, air, or any combination thereof.

[0059] The size of the particles 107, 109, and/or 207 can vary between different systems or processes and/or during operation of any particular system or process. For example, the particles 107, 109, and 207, depending on the particular process or system, can have a diameter or cross-sectional length ranging from a low of about 0.01 μιη, about 0.1 μιη, about 1 μιη, or about 10 μιη to a high of about 0.1 mm, about 1 mm, or about 5 mm. In another example, a particular process or system can have a particle/fluid mixture 105, 205 that includes two or more different particles and those two or more different particles can have the same average diameter or cross-sectional length or different average diameter or cross-sectional length. In a specific example, in a particle/fluid mixture 105, 205 of a polymerization system that includes both polymer particles and catalyst particles, the polymer particles can have an average diameter or cross-sectional length ranging from a low of about 0.1 mm, about 0.5mm, or about 1 mm to a high of about 2 mm, about 2.5 mm, or about 3 mm and the catalyst particles can have an average diameter or cross-section length ranging from a low of about 5 μιη, about 10 μιη, or about 20 μιη to a high of about 80 μιη, about 100 μιη, or about 125 μιη.

[0060] The particle/fluid mixtures 105, 205 can have a particle concentration ranging from about 0.001 percent by weight (wt%) to about 5 wt%, or from about 0.01 wt% to about 1 wt%, or from about 0.05 wt% to about 0.5 wt %, based on the total weight of the particle/fluid mixture. For example, the particle concentration in the particle/fluid mixtures 105, 205 can range from a low of about 0.01 wt%, about 0.05wt%, about 0.07 wt%, or about 0.1 wt% to a high of about 0.2 wt%, about 0.3 wt%, about 0.4 wt%, or about 0.5 wt%, based on the total weight of the particulate/fluid mixture.

[0061] The velocity of the particle/fluid mixtures 105, 205 flowing through the fluid conveying structures 1 15, 215 can vary depending on the particular process or system. Illustrative velocities for the particle/fluid mixtures 105, 205 can range from a low of about O. lm/s, about 1 m/s, about 5 m/s, about 10 m/s or about 15 m/s to a high of about 20 m/s, about 30 m/s, about 40 m/s, or about 50 m/s. For example, a gas phase polymerization system can have a particle/fluid mixture flowing from a top of a polymerization reactor, through a cycle or recycle line, and to the bottom of the polymerization reactor, with a velocity typically ranging from about 5 m s to about 30 m/s, or from about 10 m/s to about 20 m/s, or from about 12 m/s to about 18 m/s.

[0062] Depending on the particular system or process, the particle/fluid mixtures 105, 205 can be monitored via the velocity measurement systems 100, 200 within a number of different types of fluid conveying structures. Illustrative fluid conveying structures 1 15, 215 can include, but are not limited to, pipes, tubes, hoses, reactors, e.g., polymerization reactors, fluidized catalytic reactors, and the like, ducts, conduits, exhaust or vent stacks, transfer or transportation pipes or pipelines, and the like. For example, the fluid conveying structures 1 15, 215 can be a gas phase polymerization reactor and/or one or more process lines associated with the gas phase polymerization reactor, such as a cycle fluid line, a product recovery line, and/or a vent line.

[0063] The velocity measurement systems 100, 200 can be used to measure the velocity of particle/fluid mixtures 105, 205 that are within and/or cause or create a fouling environment. In other words, the particle/fluid mixture 105, 205 can tend to deposit or cause one or more materials to deposit onto the inner surfaces 116, 216 of the fluid conveying structures 1 15, 215 and/or the sensors, e.g. the sensor 103 and 220 that can be in fluid communication with the particle/fluid mixtures 105, 205, respectively. For example, a polymerization process such as a gas phase polymerization reactor and the cycle line associated therewith can be referred to as having or potentially having a fouling environment or fouling prone environment. In polymerization processes polymer particles can tend to deposit about the inner surfaces 1 16, 216 of the fluid conveying structures 1 15, 215 and/or other surfaces disposed within the internal volumes 1 17, 217, such as the sensor tips 104 and 221. The polymer particles can form random or sporadic deposits and/or coat the inner surfaces forming polymer sheets or other polymer agglomerations. The velocity measurement systems 100, 200 can be operated in fouling environments or fouling prone environments and are unaffected or relatively unaffected by fouling occurring within the internal volumes 117, 217 of the fluid conveying structures 115, 215.

[0064] The internal volume or flow paths 1 17, 217 of the fluid conveying structures 1 15, 215 can have any desired cross-sectional shape or combination of cross-sectional shapes. Illustrative cross-sectional shapes can include, but are not limited to, triangular, circular, rectangular, elliptical, oval, or any other geometrical shape. Protrusions, grooves, recesses, or any other type of surface modifications can be present about all or a portion of the inner surfaces 116, 216 of the fluid conveying structures 1 15, 215. Preferably, the inner surfaces 1 16, 216 are smooth or substantially smooth which can help facilitate a more uniform flow of the particle/fluid mixtures 105, 205 as compared to inner surfaces having protrusions extending therefrom and/or recesses extending therein, for example.

[0065] The cross-sectional shapes of the internal volumes 117, 217 along or over any given length of the fluid conveying structures 1 15, 215 can change or transition from a first cross- sectional shape to a second cross-sectional shape, a first cross-sectional size to a second cross- sectional size, or a combination thereof. For example, a fluid conveying structure having a circular cross-section can change from a first diameter to a second diameter, either smoothly, stepwise, or a combination thereof, over a given length of the fluid conveying structure. Any number of changes or transitions from any number of cross-sectional shapes and/or sizes can be made along the length of the fluid conveying structures 115, 215.

[0066] The velocity measurement systems 100, 200 can be positioned along, about, within, partially within, or otherwise disposed about the fluid conveying structures 115, 215 at any desirable location. For example, the velocity measurement systems 100, 200 can be disposed about a location of the fluid conveying structures 1 15, 215 where a straight or substantially straight flow path for the particle/fluid mixtures 105, 205 is provided. For example, the internal volumes 1 17, 217 of the fluid conveying structures 1 15, 215 can have one or more bends, turns, twists, slopes, changes in cross-sectional shape and/or size, or other non-uniform or differences, but other lengths or sections could have relatively straight or substantially straight lengths or sections. As such, in one or more embodiments it can be preferable to dispose the velocity measurement systems 100, 200 about the relatively straight or substantially straight lengths or sections as opposed to a length or section close to or having one or more bends, turns, twists, slopes, or other non-linear lengths or sections.

[0067] The first sensor 220 of the velocity measurement system 200 can be in communication with the internal volume 217 at a location having a first cross-sectional shape and first area and the second sensor 225 can be in communication with the internal volume 217 at a location having a second cross-sectional shape and second area. The first and second cross-sectional shapes can be the same or different. The first and second cross-sectional areas can be the same or different. Any two or more of the first and second cross-sectional shapes and the first and second areas can be the same or different.

[0068] For example, the first sensor 220 can be in communication with the internal volume 217 at a first location having a circular cross-section with a diameter of about 1.2 m and the second sensor 225 can be in communication with the internal volume 217 at a second location having a circular cross-section with a diameter of about 0.9 m. When the flow path or internal volume 217 varies in cross-sectional shape and/or area or size between the locations of the first sensor 220 and the second sensor 225 various methods can be used to estimate or determine the velocity.

[0069] For example, Equation 3 above can be used to provide an average velocity of the particle/fluid mixture 205 flowing through the fluid conveying structure 215 when the first sensor 220 is in communication with the internal volume that has a first cross-sectional shape and size and the second sensor 225 is in communication with the internal volume that has a second cross-section shape and/or size relative to the first cross-sectional shape and size.

[0070] In another example, a predetermined calibration table can be prepared where a known velocity of the particle/fluid mixture 205 is correlated with the first and second detected and measured signals via lines 222, 227. For example, a venturi flowmeter can be used in conjunction with the velocity measurement system 200 in order to develop or otherwise determine the calibration table or curve. The venturi flowmeter can be located within the internal volume 217 having the first cross-sectional shape and size or within the internal volume 217 having the second cross-sectional shape and/or size. Once the calibration curve has been created the calibration curve can be used to estimate the velocity of the particle/fluid mixture 205 by comparing the detected and measured signals via lines 222, 227 with the calibration curve. The estimated velocity according to the calibration curve would correspond to the velocity of the particle/fluid mixture 205 flowing through the fluid conveying structure 205 where the venturi flowmeter (or other velocity meter) was located during generation of the calibration curve. [0071] For a fluid conveying structure having a circular cross-sectional flow path, the velocity measurement systems 100, 200 can be disposed about the fluid conveying structures 115, 215 such that the flow path has traversed relatively straight or substantially straight length of about 1, about 5, about 10, about 20, about 30, about 40, about 50, about 60, about 70, about 80, about 90, about 100, or more than about 100 times a length of the circular cross-sectional flow path or diameter of the flow path. In other examples, the velocity measurement systems 100, 200 can be disposed about the flow path within the fluid conveying structures 115, 215 at a location where a bend, turn, twist, slope, or other non-linear length or section is located.

[0072] Figure 3 depicts a schematic of an illustrative gas phase polymerization system 300 having the velocity measurement systems 100, 200 depicted in Figures 1 and 2. The polymerization system 200 can include one or more polymerization reactors 301, discharge tanks 355 (only one shown), recycle compressors 370 (only one shown), and heat exchangers 375 (only one shown). The polymerization system 300 can include more than one reactor 301 arranged in series, parallel, or configured independent from the other reactors, each reactor having its own associated discharge tanks 355, recycle compressors 370, and heat exchangers 375, or alternatively, sharing any one or more of the associated discharge tanks 355, recycle compressors 370, and heat exchangers 375. For simplicity and ease of description, embodiments of the invention will be further described in the context of a single reactor train.

[0073] The reactor 301 can include a cylindrical section 303, a transition section 305, and a velocity reduction zone or dome or "top head" 307. The cylindrical section 303 is disposed adjacent the transition section 305. The transition section 305 can expand from a first diameter that corresponds to the diameter of the cylindrical section 303 to a larger diameter adjacent the dome 307. The location or junction at which the cylindrical section 303 connects to the transition section 305 can be referred to as the "neck" or the "reactor neck" 304.

[0074] The cylindrical section 303 can include a reaction zone 312. The reaction zone can be a fluidized reaction bed or fluidized bed. In one or more embodiments, a distributor plate 319 can be disposed within the cylindrical section 303, generally at or toward the end of the cylindrical section that is opposite the end adjacent to the transition section 305. The reaction zone 312 can include a bed of growing polymer particles, formed polymer particles, and catalyst particles fluidized by the continuous flow of polymerizable and modifying gaseous components in the form of make-up feed and recycle fluid through the reaction zone 312.

[0075] One or more cycle fluid lines 315 and vent lines 318 can be in fluid communication with the dome 307 of the reactor 301. A polymer product can be recovered via line 317 from the reactor 301. A reactor feed via line 310 can be introduced to the polymerization system 300 at any location or combination of locations. For example, the reactor feed via line 310 can be introduced to the cylindrical section 303, the transition section 305, the velocity reduction zone 307, to any point within the cycle fluid line 315, or any combination thereof. Preferably, the reactor feed 310 is introduced to the cycle fluid in line 315 before or after the heat exchanger 375. A catalyst feed via line 313 can be introduced to the polymerization system 300 at any point. Preferably the catalyst feed via line 313 is introduced to a fluidized bed 312 within the cylindrical section 303.

[0076] The velocity measurement system 100 and/or 200 can be in communication with the polymerization system 300 at any number of locations. As shown in Figure 3, the velocity measurement systems 100 and 200 are in communication with the cycle line 315 between the reactor 301 and the recycle compressor 370. Other suitable locations the systems 100, 200 can be in communication with the polymerization system 300 can include, but are not limited to, the cylindrical section 303, the transition section 305, and/or the dome 307. For example, the sensor 103 of system 100 and/or the sensors 220, 225 of the system 200 can be in communication with the cylindrical section 303 between an inlet of the cycle line 315 to the reactor 301 and the distributor plate 319 or between the distributor plate 319 and the transition section 305. In another example, the sensor 103 and/or the sensors 220, 225 can be in communication with the cycle line 315 between the heat exchanger 375 and the compressor 370. The sensor 103 and/or sensors 220, 225 can also be in communication with the cycle line 315 between the heat exchanger 375 and the reactor 301.

[0077] Any number of velocity measurement systems 100 and/or 200 can be in communication with the polymerization system 300. For example, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, or more velocity measurement systems 100 and/or 200 can be in communication with the polymerization system 300. If two or more velocity measurement systems 100 and/or 200 are in communication with the polymerization system 300, the two or more systems 100 and/or 200 can be disposed about different locations, the same or similar locations, or a combination of different and similar locations. As shown, the velocity measurement systems 100 and 200 are disposed between compressor 370 and the reactor 301. In another example, the velocity measurement systems 100 and/or 200 could be in communication with the cycle line 315 between the reactor 301 and the heat exchanger 375. In still another example, the velocity measurement systems 100 and/or 200 can be in communication with the reactor 301. The velocity measurement system 100 and/or 200 can be in communication with the cylindrical section 303 above and/or below the distributor plate 319, the transition section 305, or the top head 307.

[0078] The velocity measurement system 100 and/or 200 can be used during predetermined or selected polymerization process periods, continuously, randomly, or any combination thereof. For example, the velocity measurement system 100 and/or 200 can be used to detect one or more electrical signals generated by the passing and/or contacting particles 107, 109 and/or the particles 207 during polymerization reactor start-up, shutdown, idling, stead-state operation, transition periods, and the like. In another example, the velocity measurement system 100 and/or 200 can be continuously or substantially continuously operated during operation of the polymerization system 300.

[0079] As discussed and described above the velocity of the particle fluid mixture flowing through the cycle line 3 15 can be estimated using the velocity measurement system 100 and/or 200 and Equation 1 or 3, respectively. Once the velocity of the cycle fluid has been estimated, the superficial velocity (VSF) of the cycle gas flowing through the reactor 301 can be estimated. The formula for determining the superficial velocity (VSF) of the cycle gas through the reactor 301 can be expressed by Equatio

(Equation 4)

where VSF is the superficial velocity, VFB is the velocity of the cycle fluid flowing through the cycle line, A_CF is the cross-sectional area of the cycle fluid line, and A_FB is the cross-sectional area of the fluidized bed. Accordingly, if the velocity measurement system 100 or 200 were used to estimate or measure the superficial velocity within the reactor 301 , the velocity of the cycle fluid within the cycle cline 3 15 could also be estimated from Equation 4.

[0080] In general, the height to diameter ratio of the cylindrical section 303 can vary in the range of from about 2: 1 to about 5 : 1. The range, of course, can vary to larger or smaller ratios and depends, at least in part, upon the desired production capacity and/or reactor dimensions. The cross-sectional area of the dome 307 is typically within the range of from about 2 to about 3 multiplied by the cross-sectional area of the cylindrical section 303.

[0081] The velocity reduction zone or dome 307 has a larger inner diameter than the cylindrical section 303. As the name suggests, the velocity reduction zone 307 slows the velocity of the gas due to the increased cross-sectional area. This reduction in gas velocity allows particles entrained in the upward moving gas to fall back into the bed, allowing primarily only gas to exit overhead of the reactor 301 through the cycle fluid line 3 15. The cycle fluid recovered via line 3 15 can contain less than about 10% wt, less than about 8% wt, less than about 5% wt, less than about 4% wt, less than about 3% wt, less than about 2% wt, less than about 1% wt, less than about 0.5% wt, or less than about 0.2% wt of the particles entrained in fluidized bed 3 12. In another example, the cycle fluid recovered via line 3 15 can have a particle concentration ranging from a low of about 0.001 wt% to about 5 wt%, from about 0.01 wt% to about 1 wt%, or from about 0.05 wt% to about 0.5 wt %, based on the total weight of the particle/cycle fluid mixture in line 3 15. For example, the particle concentration in the cycle fluid in line 3 15 can range from a low of about 0.001 wt%, about 0.01 wt%, about 0.05wt%, about 0.07 wt%, or about 0.1 wt% to a high of about 0.5 wt%, about 1.5 wt%, about 3 wt%, or about 4 wt%, based on the total weight of the cycle fluid and particles in line 315.

[0082] Suitable gas phase polymerization processes for producing the polymer product via line 317 are described in U.S. Patent Nos. 3,709,853; 4,003,712; 4,011,382; 4,302,566; 4,543,399; 4,588,790; 4,882,400; 5,028,670; 5,352,749; 5,405,922; 5,541,270; 5,627,242; 5,665,818; 5,677,375; 6,255,426; European Patent Nos. EP 0802202; EP 0794200; EP 0649992; EP 0634421. Other suitable polymerization processes that can be used to produce the polymer product can include, but are not limited to, solution, slurry, and high pressure polymerization processes. Examples of solution or slurry polymerization processes are described in U.S. Patent Nos. 4,271,060; 4,613,484; 5,001,205; 5,236,998; and 5,589,555.

[0083] The reactor feed in line 310 can include any polymerizable hydrocarbon of combination of hydrocarbons. For example, the reactor feed can be any olefin monomer including substituted and unsubstituted alkenes having two to 12 carbon atoms, such as ethylene, propylene, 1-butene, 1-pentene, 1-hexene, 1-heptene, 1-octene, 4-methylpent-l-ene, 1-decene, 1- dodecene, 1-hexadecene, and the like. The reactor feed can also include non-hydrocarbon gas(es) such as nitrogen and/or hydrogen. The reactor feed can enter the reactor at multiple and different locations. For example, monomers can be introduced into the fluidized bed in various ways including direct injection through a nozzle (not shown) into the fluidized bed. The polymer product can thus be a homopolymer or a copolymer, including a terpolymer, having one or more other monomeric units. For example, a polyethylene product could include at least one or more other olefin(s) and/or comonomer(s).

[0084] The reactor feed in line 310 can also include the one or more modifying components such as one or more induced condensing agents ("ICAs"). Illustrative ICAs include, but are not limited to, propane, butane, isobutane, pentane, isopentane, hexane, isomers thereof, derivatives thereof, and combinations thereof. The ICAs can be introduced to provide a reactor feed to the reactor having an ICA concentration ranging from a low of about 1 mol%, about 5 mol%, or about 10 mol% to a high of about 25 mol%, about 35 mol%, or about 45 mol%. Typical concentrations of the ICAs can range from about 14 mol%, about 16 mol%, or about 18 mol% to a high of about 20 mol%, about 22 mol%, or about 24 mol%. The reactor feed can include other non-reactive gases such as nitrogen and/or argon. Further details regarding ICAs can be as discussed and described in U.S. Patent Nos. 5,352,749; 5,405,922; 5,436, 304; and 7,122,607; and WO Publication No. 2005/113615(A2). Condensing mode operation, such as disclosed in U.S. Patent Nos. 4,543,399 and 4,588,790 can also be used to assist in heat removal from the fluid bed polymerization reactor. [0085] The catalyst feed in line 313 can include any catalyst or combination of catalysts. Illustrative catalysts can include, but are not limited to, Ziegler-Natta catalysts, chromium-based catalysts, metallocene catalysts and other single-site catalysts including Group 15 -containing catalysts, bimetallic catalysts, and mixed catalysts. The catalyst can also include AICI₃, cobalt, iron, palladium, chromium/chromium oxide or "Phillips" catalysts. Any catalyst can be used alone or in combination with any other catalyst.

[0086] Suitable metallocene catalyst compounds can include, but are not limited to, metallocenes described in U.S. Patent Nos.: 7, 179,876; 7,169,864; 7, 157,531; 7, 129,302; 6,995,109; 6,958,306; 6,884748; 6,689,847; 5,026,798; 5,703, 187; 5,747,406; 6,069,213; 7,244,795; 7,579,415; U.S. Patent Application Publication No. 2007/0055028; and WO Publications WO 97/22635; WO 00/699/22; WO 01/30860; WO 01/30861; WO 02/46246; WO 02/50088; WO 04/022230; WO 04/026921 ; and WO 06/019494.

[0087] The "Group 15-containing catalyst" may include Group 3 to Group 12 metal complexes, wherein the metal is 2 to 8 coordinate, the coordinating moiety or moieties including at least two Group 15 atoms, and up to four Group 15 atoms. For example, the Group 15-containing catalyst component can be a complex of a Group 4 metal and from one to four ligands such that the Group 4 metal is at least 2 coordinate, the coordinating moiety or moieties including at least two nitrogens. Representative Group 15-containing compounds are disclosed in WO Publication No. WO 99/01460; European Publication Nos. EP0893454A1; EP 0894005A1 ; U.S. Patent Nos. 5,318,935; 5,889, 128; 6,333,389; and 6,271,325.

[0088] Illustrative Ziegler-Natta catalyst compounds are disclosed in European Patent Nos. EP 0103120; EP 1102503; EP 0231 102; EP 0703246; U.S. Patent Nos. RE 33,683; 4, 1 15,639; 4,077,904; 4,302,565; 4,302,566; 4,482,687; 4,564,605; 4,721,763; 4,879,359; 4,960,741; 5,518,973; 5,525,678; 5,288,933; 5,290,745; 5,093,415; and 6,562,905; and U.S. Patent Application Publication No. 2008/0194780. Examples of such catalysts include those comprising Group 4, 5 or 6 transition metal oxides, alkoxides and halides, or oxides, alkoxides and halide compounds of titanium, zirconium or vanadium; optionally in combination with a magnesium compound, internal and/or external electron donors (alcohols, ethers, siloxanes, etc.), aluminum or boron alkyl and alkyl halides, and inorganic oxide supports.

[0089] Suitable chromium catalysts can include di-substituted chromates, such as Cr0₂(OR)2; where R is triphenylsilane or a tertiary polyalicyclic alkyl. The chromium catalyst system may further include Cr0₃, chromocene, silyl chromate, chromyl chloride (Cr0₂Cl₂), chromium-2- ethyl-hexanoate, chromium acetylacetonate (Cr(AcAc)₃), and the like. Other non-limiting examples of chromium catalysts are described in U.S. Patent No. 6,989,344. [0090] The mixed catalyst can be a bimetallic catalyst composition or a multi-catalyst composition. As used herein, the terms "bimetallic catalyst composition" and "bimetallic catalyst" include any composition, mixture, or system that includes two or more different catalyst components, each having a different metal group. The terms "multi-catalyst composition" and "multi-catalyst" include any composition, mixture, or system that includes two or more different catalyst components regardless of the metals. Therefore, the terms "bimetallic catalyst composition," "bimetallic catalyst," "multi-catalyst composition," and "multi-catalyst" will be collectively referred to herein as a "mixed catalyst" unless specifically noted otherwise. In one example, the mixed catalyst includes at least one metallocene catalyst component and at least one non-metallocene component.

[0091] In some embodiments, an activator may be used with the catalyst compound. As used herein, the term "activator" refers to any compound or combination of compounds, supported or unsupported, which can activate a catalyst compound or component, such as by creating a cationic species of the catalyst component. Illustrative activators include, but are not limited to, aluminoxane (e.g., methylaluminoxane "MAO"), modified aluminoxane (e.g., modified methylaluminoxane "MMAO" and/or tetraisobutyldialuminoxane "TIB AO"), and alkylaluminum compounds, ionizing activators (neutral or ionic) such as tri (n-butyl)ammonium tetrakis(pentafluorophenyl)boron may be also be used, and combinations thereof.

[0092] The catalyst compositions can include a support material or carrier. As used herein, the terms "support" and "carrier" are used interchangeably and are any support material, including a porous support material, for example, talc, inorganic oxides, and inorganic chlorides. The catalyst component(s) and/or activator(s) can be deposited on, contacted with, vaporized with, bonded to, or incorporated within, adsorbed or absorbed in, or on, one or more supports or carriers. Other support materials can include resinous support materials such as polystyrene, functionalized or crosslinked organic supports, such as polystyrene divinyl benzene polyolefins or polymeric compounds, zeolites, clays, or any other organic or inorganic support material and the like, or mixtures thereof.

[0093] Suitable catalyst supports are described in U.S. Patent Nos.: 4,701,432, 4,808,561; 4,912,075; 4,925,821; 4,937,217; 5,008,228; 5,238,892; 5,240,894; 5,332,706; 5,346,925; 5,422,325; 5,466,649; 5,466,766; 5,468,702; 5,529,965; 5,554,704; 5,629,253; 5,639,835; 5,625,015; 5,643,847; 5,665,665; 5,698,487; 5,714,424; 5,723,400; 5,723,402; 5,731,261; 5,759,940; 5,767,032; 5,770,664; and 5,972,510; and WO Publication Nos. WO 95/32995; WO 95/14044; WO 96/06187; WO 97/02297; WO 99/47598; WO 99/48605; and WO 99/50311.

[0094] The cycle fluid via line 315 can be compressed in the compressor 370 and then passed through the heat exchanger 375 where heat can be exchanged between the cycle fluid and a heat transfer medium. For example, during normal operating conditions a cool or cold heat transfer medium via line 371 can be introduced to the heat exchanger 375 where heat can be transferred from the cycle fluid in line 315 to produce a heated heat transfer medium via line 377 and a cooled cycle fluid via line 315. In another example, during idling of the reactor 301 a warm or hot heat transfer medium via line 371 can be introduced to the heat exchanger 375 where heat can be transferred from the heat transfer medium to the cycle fluid in line 315 to produce a cooled heat transfer medium via line 375 and a heated cycle fluid via line 315. The terms "cool heat transfer medium" and "cold heat transfer medium" refer to a heat transfer medium having a temperature less than the fluidized bed 312 within the reactor 301. The terms "warm heat transfer medium" and "hot heat transfer medium" refer to a heat transfer medium having a temperature greater than the fluidized bed 312 within the reactor 301. The heat exchanger 375 can be used to cool the fluidized bed 312 or heat the fluidized bed 312 depending on the particular operating conditions of the polymerization system 300, e.g. start-up, normal operation, and shut down. Illustrative heat transfer mediums can include, but are not limited to, water, air, glycols, or the like. It is also possible to locate the compressor 370 downstream from the heat exchanger 375 or at an intermediate point between several heat exchangers 375.

[0095] After cooling, all or a portion of the cycle fluid in line 315, the cycle fluid can be returned to the reactor 301. The cooled cycle fluid in line 315 can absorb the heat of reaction generated by the polymerization reaction. The heat exchanger 375 can be of any type of heat exchanger. Illustrative heat exchangers can include, but are not limited to, shell and tube, plate and frame, U-tube, and the like. For example, the heat exchanger 375 can be a shell and tube heat exchanger where the cycle fluid via line 315 can be introduced to the tube side and the heat transfer medium can be introduced to the shell side of the heat exchanger 375. If desired, to or more heat exchangers can be employed, in series, parallel, or a combination of series and parallel, to lower or increase the temperature of the cycle fluid in stages.

[0096] Preferably, the cycle gas via line 315 is returned to the reactor 301 and to the fluidized bed 312 through the fluid distributor plate ("plate") 319. The plate 319 can prevent polymer particles from settling out and agglomerating into a solid mass. The plate 319 can also prevent or reduce the accumulation of liquid at the bottom of the reactor 301. The plate 319 can also facilitate transitions between processes which contain liquid in the cycle stream 315 and those which do not and vice versa. Although not shown, the cycle gas via line 315 can be introduced into the reactor 301 through a deflector disposed or located intermediate an end of the reactor 301 and the distributor plate 319. Illustrative deflectors and distributor plates suitable for this purpose are described in U.S. Patent Nos. 4,877,587; 4,933, 149; and 6,627,713. [0097] The catalyst feed via line 313 can be introduced to the fluidized bed 312 within the reactor 301 through one or more injection nozzles (not shown) in fluid communication with line 313. The catalyst feed is preferably introduced as pre-formed particles in one or more liquid carriers (i.e. a catalyst slurry). Suitable liquid carriers can include mineral oil and/or liquid or gaseous hydrocarbons including, but not limited to, butane, isopentane, hexane, heptane octane, or mixtures thereof. A gas that is inert to the catalyst slurry such as, for example, nitrogen or argon can also be used to carry the catalyst slurry into the reactor 301. In one example, the catalyst can be a dry powder. In another example, the catalyst can be dissolved in a liquid carrier and introduced to the reactor 301 as a solution. The catalyst via line 313 can be introduced to the reactor 301 at a rate sufficient to maintain polymerization of the monomer(s) therein.

[0098] The polymer product via line 317 can be discharged from the reactor 301 by opening valve 357 while valves 359, 367 are in a closed position. Product and fluid enter the product discharge tank 355. Valve 357 is closed and the product is allowed to settle in the product discharge tank 355. Valve 359 is then opened permitting fluid to flow via line 361 from the product discharge tank 355 to the reactor 301. In another example, the separated fluid in line 361 can be introduced to the cycle line 315. The separated fluid in line 361 can include unreacted monomer(s), hydrogen, ICA(s), and/or inerts. Valve 359 can then be closed and valve 367 can be opened and any product in the product discharge tank 355 can flow out of the discharge tank via line 368. Valve 367 can then be closed. Although not shown, the polymer product via line 368 can be introduced to a plurality of purge bins or separation units, in series, parallel, or a combination of series and parallel, to further separate gases and/or liquids from the product. The particular timing sequence of the valves 357, 359, 367, can be accomplished by use of conventional programmable controllers which are well known in the art.

[0099] Another product discharge system which can be alternatively employed is that disclosed in U.S. Patent No. 4,621,952. Such a system employs at least one (parallel) pair of tanks comprising a settling tank and a transfer tank arranged in series and having the separated gas phase returned from the top of the settling tank to a point in the reactor near the top of the fluidized bed. Other suitable product discharge systems are described in PCT Publications WO2008/045173 and WO2008/045172.

[00100] The reactor 301 can be equipped with one or more vent lines 318 to allow venting the bed during start up, operation, and/or shut down. The reactor 301 can be free from the use of stirring and/or wall scraping. The cycle line 315 and the elements therein (compressor 370, heat exchanger 375) can be smooth surfaced and devoid of unnecessary obstructions so as not to impede the flow of cycle fluid or entrained particles. [00101] The conditions for polymerization vary depending upon the monomers, catalysts, catalyst systems, and equipment availability. The specific conditions are known or readily derivable by those skilled in the art. For example, the temperatures can be within the range of from about -10°C to about 140°C, often about 15°C to about 120°C, and more often about 70°C to about 110°C. Pressures can be within the range of from about 10 kPag to about 10,000 kPag, such as about 500 kPag to about 5,000 kPag, or about 1,000 kPag to about 2,200 kPag, for example. Additional details of polymerization can be found in U.S. Patent No. 6,627,713.

[00102] In some embodiments, one or more continuity additives or static control agents can also be introduced to the reactor 301 to prevent agglomeration. Introducing continuity additive(s) can include the addition of negative charge generating chemicals to balance positive voltages or the addition of positive charge generating chemicals to neutralize negative voltage potentials as described in U.S. Patent No. 4,803,251. Antistatic substances can also be added, either continuously or intermittently to prevent or neutralize electrostatic charge generation. The continuity additive and/or antistatic substances, if used, can be introduced with the feed via line 310, the catalyst via line 313, a separate inlet (not shown), or any combination thereof.

[00103] As used herein, the term "continuity additive" refers to a compound or composition that when introduced into a gas phase fluidized bed reactor can influence or drive the static charge (negatively, positively, or to zero) in the fluidized bed. The continuity additive or combination of continuity additives can depend, at least in part, on the nature of the static charge. The particular continuity additive or combination of continuity additives can depend, at least in part, on the particular polymer being produced within the polymerization reactor, the particular spray dried catalyst system or combination of catalyst systems being used, or a combination thereof. Suitable continuity additives and uses thereof can be as discussed and described in European Patent No. 0 229 368; U.S. Patent Nos. 5,283,278; 4,803,251 ; 4,555,370; 4,994,534; and 5,200,477; and WO Publication No. WO2009/0231 11 ; and WOO 1/44322.

[00104] The methods for estimating a velocity of a particle/fluid mixture described herein may comprise, flowing a particle/fluid mixture through a fluid conveying structure past a sensor, measuring a signal detected by the sensor as the particles pass the sensor to provide a measured signal; manipulating the measured signal to provide an output; and determining the velocity of the particle/fluid mixture from the output.

[00105] The particle/fluid mixture may flow through a fluid conveying structure past a first sensor and a second sensor, wherein the first and second sensors are separated from one another by a predetermined distance (d). As the particles in the particle/fluid mixture pass the first sensor, the first sensor can detect at least one signal and measure the signal to produce a first measured signal. As the particles in the particle fluid/mixture pass the second sensor, the second sensor can detect at least one signal and measure the signal to produce a second measured signal. The first and second measured signals can be manipulated to provide an output. A transit time can be determined from the output. The distance (d) can be divided by the transit time to provide a velocity of the particle/fluid mixture.

[00106] When just one sensor, such as a static probe, is used, the manipulation may comprise using the absolute autocorrelation method on the measured signal and determining the absolute autocorrelation vector of the electrical signal. The measured signal may be a current or voltage signal. The absolute autocorrelation of the measured signal may comprise an approaching curve and a leaving curve. The velocity of the particle/fluid mixture may be determined by dividing a proportionality constant (k) by the offset time, where the offset time is the time between a maximum peak or a minimum peak of the approaching curve and a zero lag time.

[00107] When one or more sensors are used, such as a first and second senor, the manipulation of the first and second measured signals may comprise using cross correlation where the output comprises a cross correlation sequence and the transit time is determined from the cross correlation sequence. The first and second measured signals may be of the same type of signal or be different types of signals and may be selected from static signals, acoustic emission signals, dielectric signals, optical signals, and capacitance signals. In one example, the signal detected by the first sensor comprises a static signal and the signal detected by the second sensor comprises an acoustic emission signal. The first and second measured signals may comprise a current or voltage and manipulation of the measured signals may comprise taking the absolute values of the mean centered current or voltage prior to carrying out the cross correlation.

Examples

[00108] To provide a better understanding of the foregoing discussion, the following non-limiting examples are provided. Although the examples are directed to specific embodiments, they are not to be viewed as limiting the invention in any specific respect. All parts, proportions and percentages are by weight unless otherwise indicated.

[00109] Two examples (Examples 1 and 2) were carried out on a commercial scale gas phase polymerization system with Example 1 using the velocity measurement system 100 and Example 2 using the velocity measurement system 200, according to one or more embodiments discussed and described above with reference to Figures 1 and 2, respectively. The gas phase polymerization reactor 301 had a cylindrical section 303, a transition section 305, and a velocity reduction section 307, as described above with reference to Figure 3. The reactor neck was located 13.6 m from the distributor plate 319. The polymerization feed included ethylene and hexene. During normal, steady state operating conditions, e.g., a pressure of about 290 psig and a temperature of about 185°F, about 1 18 klb/hr of an polymer per day were produced in and recovered from the reactor 201. For both Examples 1 and 2, the catalyst feed to the reactor was a metallocene catalyst system.

Example 1 - Estimation of Velocity Using a Single Static Probe

[00110] Data acquired from a gas phase polymerization system operating under steady state conditions using a single static probe is shown in Figures 4 and 5 (Example 1). The data acquired in Example 1 was acquired over a period of five minute period of operation.

[00111] The static probe used for measuring the current ("entrainment static") in Example 1 was a Progression Correflux model 3400 probe with custom "fast," "bipolar" electronics capable of 5 millisecond response. The model 3400 probe was mounted in the cycle line between the heat exchanger and the polymerization reactor. The probe extended about 18 inches into a 36 inch diameter cycle pipe (to the centerline of the pipe). An integral housing located outside the flange contained an electrometer for measuring the entrainment static and a transmitter for transmitting the measured entrainment static to the processor. The electronics were shielded to eliminate outside stray electromagnetic fields from interfering with the probe. The static probe sampled the entrainment static data at 100 Hz for the period of 5 minutes. As such, the number of data points acquired over the 5 minutes was equal to 30,000 data points.

[00112] The measured entrainment static data for Example 1 was introduced to a processor configured to operate the software program Matlab (available from The Math Works). The Matlab function "xcorr" was used to manipulate the measured entrainment static using the absolute autocorrelation methodology for entrainment static data measured via the electrometer.

[00113] Figure 4 is a graphical depiction of measured entrainment static detected by the static probe during monitoring of the cycle fluid flowing through the gas phase polymerization reactor cycle line over five minutes of operation. Figure 5 is a close-up view of the absolute autocorrelation vector calculated from the measured entrainment static depicted in Figure 4 focused on a time lag of zero (0) seconds.

[00114] As shown in Figure 5, the peak 505 of the leading lobe was located at about -0.12 seconds and was a "negative" or minimum type peak. The lagging lobe was essentially a mirror image of the leading lobe (transposed across the x-axis) and the peak 510 of the lagging lobe was located at about +0.12 seconds and was a "positive" or maximum type peak. The peak at zero time lag 515 (while not important for the velocity calculation) was a "negative" or minimum type peak, which was the same as the leading lobe. Since the peak 505 of the leading lobe and the peak 515 at zero time lag were both minima, it can be inferred that the average particle striking the probe was of the same charge sign as the average particle approaching and passing the probe without contacting the probe.

[00115] Equation 1 was used to estimate the velocity of the cycle fluid flowing through the cycle gas line. As such, the proportionality constant (k) for the cycle fluid line had to be determined. The proportionality constant (k) for Example 1 was determined by determining the numerical integration of the absolute autocorrelation of Equation 2, as discussed above. Accordingly, the proportionality constant (k) for the cycle fluid line in Example 1 had a value of 1.3 ft.

[00116] Equation 4 (discussed above) was then used to estimate the superficial gas velocity in the fluidized bed of the polymerization reactor. The cycle fluid line used in Example 1 had a cross- sectional area (ACF) of 7.07 ft², a fluidized bed cross-sectional area (AFB) of 160 ft², and as determined above a cycle fluid velocity (VCF) of 44 ft/sec. Accordingly, the superficial velocity of the cycle gas flowing through the polymerization reactor was estimated to be about 1.9 ft/s.

[00117] The surprising and unexpected discovery of a method for using data detected by a single, sensor or probe for determining the velocity of the cycle fluid and in turn the superficial velocity of the cycle fluid flowing through the polymerization reactor (a fouling environment) provides a more reliable method over current velocity measurement systems. Several improvements could be implemented in order to improve the accuracy of the estimated cycle gas velocity and are worth further explanation.

[00118] One improvement could include increasing the rate at which the signal or data is acquired as compared to that used in Example 1. Example 1 measured the data on the order of about a tenth of a second from the data acquired at 100 Hz. Fundamentally, the accuracy cannot be better than about +/- 10%. Increasing the data acquisition rate to 1,000 Hz or even 10,000 Hz, however, could be made in the design of the electrometer circuitry, transmitter selection, and digital signal processing, and would increase the accuracy to within about +/- 1.0% or about +/- 0.1%, respectively.

[00119] Another improvement could include improving the piping design of the cycle fluid line. Flow meters generally require well-developed flow patterns and often specify that about 30 to 50 pipe diameters of straight section piping precede the flowmeter to ensure good operation. For two-phase flows (such as that in Example 1), downward flow is preferred over horizontal flow to reduce the potential effects of gravity on phase-segregation of the cycle fluid within the cycle line. The particular installation used in Example 1 was only about six pipe diameters downstream of the cycle gas compressor and only about three pipe diameters after a 90-degree bend. For the purposes of a typical gas phase polymerization reactor, a preferred location could be at the bottom of the long straight section of the cycle line coming down from the top of the reactor. Accordingly, optimizing or otherwise improving the particular location of the static probe could improve estimation of the cycle fluid velocity.

[00120] Another improvement could include data averaging. Given the normal fluctuations of the turbulent, two-phase cycle fluid flow, data averaging can be used as a convenient method to improve the accuracy of any one or more measured properties of the flow medium. Example 1 acquired the current signal over a large ensemble of particles (e.g., about 5 minutes) and used the absolute autocorrelation method over all particles to extract the leading-lobe offset time value, which was used to determine the superficial velocity within the reactor. Theoretically, however, the method could use absolute autocorrelation over a short periods (e.g., about every 10 seconds) with those short periods repeating over an extended period of time (e.g., 30 times over 5 minutes) to extract a series of short-term leading-lobe offset time values. The median value of the leading-lobe offset time values could then be used to determine the cycle fluid velocity and could be expressed by Equation 5:

2 3g

V = ^'■ (Equation 5)

Offset Time_leading lobe

where velocity is in units of ft/s and Offset Timei_eadin_g lobe is in seconds. Note, the number 2.38 is a proportionality constant (k) and is valid for 2 inch diameter static probe inserted to a centerline of a 36 inch diameter cycle fluid pipe with well-developed two-phase flow. Calibration constants for other geometries can be calculated from Gausses Law/Absolute Autocorrelation according to the methods discussed and described above or elsewhere herein.

Example 2 - Estimation of Velocity Using a Static Probe and an Acoustic Emissions Sensor

[00121] Data acquired from a gas phase polymerization system operating under steady state conditions using two sensors, namely a first static probe and a second acoustic emissions sensor, is shown in Figures 6-8 (Example 2). The static probe was located upstream of the acoustic emissions sensor. The distance between the static probe and the acoustic emissions sensor was about 70 ft.

[00122] The static probe used for measuring the current ("entrainment static") in Example 2 was a Progression Correflux model 3420 that sampled the entrainment static data at 100 Hz for a period of five minutes. As such, the number of data points acquired over the 5 minutes was equal to 30,000 data points. The model 3420 probe was mounted in the cycle line between the heat exchanger and the polymerization reactor. The probe extended about 18 inches into a 36 inch diameter cycle pipe (to the centerline of the pipe). An integral housing located outside the flange contained an electrometer for measuring the entrainment static and a transmitter for transmitting the measured entrainment static to the processor. The electronics were shielded to eliminate outside stray electromagnetic fields from interfering with the probe.

[00123] The second sensor, i.e. the acoustic emissions sensor, used for measuring the acoustic emissions signals in Example 2 was a Granumet XP available from Process Analysis and Automation Ltd. An integral housing contained a transmitter for transmitting the measured acoustic emissions to the processor. The acoustic emission sensor was mounted on an outer wall of the cycle line about 70 ft from the model 3420 static probe. The acoustic emissions sensor sampled the acoustic emission data at 100 Hz for the period of 5 minutes. As such, the number of data points acquired over the 5 minutes was equal to 30,000 data points.

[00124] Figure 6 depicts a graphical depiction of measured entrainment static 605 detected by the static probe and measured acoustic emissions 610 detected by the acoustic emissions sensor during monitoring of the cycle fluid flowing through the gas phase polymerization reactor cycle line over five minutes of operation. The measured acoustic emissions signal is shown on the left y-axis and is in volts (V) and the measured entrainment static is shown on the right y-axis and is in nano-amps (nA). Figure 7 depicts a graphical depiction of the absolute values of the mean centered measured entrainment static 705 and the measured acoustic emissions 610 depicted in Figure 6. Again, the acoustic emissions values 610 are shown on the left y-axis and the absolute values of the mean centered measured entrainment static 705 are shown in the right y-axis. The absolute values of the mean centered static data 705 have a signal line shape very similar to the acoustic signal, but with a time offset due to the transit time required for the entrained solids to move from the static probe to the acoustic emissions sensor.

[00125] Figure 8 depicts a graphical depiction of the cross-correlation sequence 805 calculated from the absolute values of the mean centered measured entrainment static and the measured acoustic emissions shown in Figure 7. The cross-correlation sequence 805 has a peak 810 that is offset from zero time. The transit time 815, i.e. the time the peak 810 is offset from zero time, was about 1.21 seconds. Accordingly, the velocity of the cycle gas within the cycle line was computed using Equation 3 and was 57.8 ft/sec.

[00126] Equation 4 was then used to estimate the superficial gas velocity in the fluidized bed of the polymerization reactor. The cycle fluid line used in Example 2 had a cross-sectional area (ACF) of 7.07 ft², a fluidized bed cross-sectional area (AFB) of 160 ft², and as determined above a cycle fluid velocity (VCF) of 57.8 ft/sec. Accordingly, the superficial velocity of the cycle gas flowing through the polymerization reactor was estimated to be about 2.5 ft/s.

[00127] Examples 1 and 2 provide a reliable method for estimating the both the cycle fluid velocity and the superficial velocity of the cycle gas flowing through the polymerization reactor. Examples 1 and 2 eliminate the reliability issues associated with transitional venturi flowmeters because the static probe, acoustic emissions sensor, and other sensors such as optical, capacitance, or dielectric sensors, unlike the venturi flowmeters, are not susceptible to fouling which introduces errors in the velocity estimations.

[00128] Similar to the velocity measurement system that used a single static probe in Example 1 , one or more improvements or modifications could be made to the setup in Example 2 in which two sensors were used. For example, data sampling rate, data averaging, and/or modifying the cycle fluid piping design and/or the location of the first and second sensors about the cycle fluid line.

[00129] All numerical values are "about" or "approximately" the indicated value, and take into account experimental error and variations that would be expected by a person having ordinary skill in the art.

[00130] Various terms have been defined above. To the extent a term used in a claim is not defined above, it should be given the broadest definition persons in the pertinent art have given that term as reflected in at least one printed publication or issued patent. Furthermore, all patents, test procedures, and other documents cited in this application are fully incorporated by reference to the extent such disclosure is not inconsistent with this application and for all jurisdictions in which such incorporation is permitted.

[00131] While the foregoing is directed to embodiments of the present invention, other and further embodiments of the invention may be devised without departing from the basic scope thereof, and the scope thereof is determined by the claims that follow.

Claims

CLAIMS; What is claimed is:

1. A method for estimating a velocity of a particle/fluid mixture, comprising:

flowing a particle/fluid mixture through a fluid conveying structure past a sensor; measuring a signal detected by the sensor as particles in the particle/fluid mixture pass the sensor to provide a measured signal;

manipulating the measured signal to provide an output; and

determining a velocity of the particle/fluid mixture using the output.

2. The method of claim 1, wherein manipulating the measured signal comprises using the absolute autocorrelation method on the measured signal, and determining the absolute autocorrelation vector of the electrical signal.

3. The method of claim 2, wherein the absolute autocorrelation of the measured signal comprises an approaching curve and a leaving curve.

4. The method of claim 3, wherein a time between a maximum peak or a minimum peak of the approaching curve and a zero time lag is taken as an offset time, and wherein a proportionality constant (k) divided by the offset time equals the velocity of the particle/fluid mixture.

5. The method according to any one of claims 1 to 4, wherein the signal comprises a current or voltage.

6. The method according to any one of claims 1 to 5, wherein the sensor comprises a static probe.

7. The method according to any one of claims 1 to 6, wherein manipulating the measured signal is carried out using a processor.

8. The method according to any one of claims 1 to 7, wherein the signal is measured at a sampling frequency of about 100 Hz or more.

9. The method according to any one of claims 1 to 8, wherein the sensor has a response time of less than about 0.05 seconds

10. A method for estimating a velocity of a particle/fluid mixture, comprising:

flowing a particle/fluid mixture through a fluid conveying structure past a first sensor and a second sensor, wherein the first and second sensors are separated from one another by a predetermined distance (d);

measuring at least one signal detected by the first sensor as particles in the particle/fluid mixture pass the first sensor to produce a first measured signal;

measuring at least one signal detected by the second sensor as the particles pass the second sensor to produce a second measured signal;

manipulating the first and second measured signals to provide an output;

determining from the output a transit time; and

dividing the distance (d) by the transit time to provide a velocity of the particle/fluid mixture.

11. The method of claim 10, wherein manipulating the first and second measured signals comprises using cross correlation, wherein the output comprises a cross correlation sequence, and the transit time is determined from the cross correlation sequence.

12. The method of claim 1 1, wherein at least one of the first measured signal and the second measured signal comprises a current or voltage, wherein manipulating the at least one of the first measured signal, the second measured signal comprising the current or voltage further includes taking the absolute values of the mean centered current or voltage prior to carrying out the cross correlation.

13. The method according to any one of claims 10 to 12, wherein the at least one signal detected by the first sensor and the at least one signal detected by the second sensor are the same type of signal or different types of signals.

14. The method according to any one of claims 10 to 13, wherein the at least one signal detected by the first sensor and the at least one signal detected by the second sensor are selected from the group consisting of a static signal, an acoustic emission signal, a dielectric signal, an optical signal, and a capacitance signal.

15. The method according to any one of claims 10 to 14, wherein the at least one signal detected by the first sensor comprises a static signal and the at least one signal detected by the second sensor comprises an acoustic emission signal.

16. The method according to any one of claims 10 to 15, wherein the first sensor, the second sensor, or both are in contact with the particle/fluid mixture.

17. The method according to any one of claims 10 to 16, wherein the first sensor, the second sensor, or both are disposed on an outside surface of the fluid conveying structure such that the first sensor, the second sensor, or both are not in contact with the particle/fluid mixture.

18. The method according to any one of claims 10 to 17, wherein the first sensor and the second sensor have a response time of less than about 0.05 seconds.

19. The method according to any one of claims 10 to 18, wherein the at least one signal detected by the first sensor and the at least one signal detected by the second sensor are measured at a sampling frequency of about 100 Hz or more.

20. The method according to any one of claims 10 to 19, wherein manipulating the first and second measured signals is carried out using a processor.

21. The method according to any one of claims 1 to 20, wherein a concentration of particles in the particle/fluid mixture ranges from about 0.01 wt% to about 1 wt% based on a total weight of the particles and the fluid.

22. The method according to any one of claims 1 to 21, wherein the particles comprise catalyst particles, polymer particles, or both, and wherein the fluid comprises one or more hydrocarbons.

23. The method according to any one of claims 1 to 22, wherein the flowing mixture is located within a polymerization system.

24. The method according to any one of claims 1 to 23, wherein the flowing mixture is located within a cycle line of a gas phase polymerization reactor.

25. A system for estimating a velocity of a particle/fluid mixture, comprising:

a fluid conveying structure having an internal volume for flowing a particle/fluid mixture therethrough;

a sensor adapted to detect at least one signal generated as particles in the particle/fluid mixture pass the probe without contacting the probe;

an electrometer in communication with the sensor and adapted to measure the signal detected by the sensor; and a processor in communication with the electrometer, wherein the processor receives the measured signal, manipulates the measured signal to provide an output, and determines a velocity of the particle/fluid mixture using the output.

26. The system of claim 25, wherein the sensor comprises a static probe.

27. The system of claim 25 or 26, wherein the sensor is configured to detect both positive and negative charges.

28. The system according to any one of claims 25 to 27, wherein the sensor is at least

partially disposed within the internal volume.

29. A system for estimating a velocity of a particle/fluid mixture, comprising:

a first sensor in communication with the internal volume and adapted to detect and measure at least one first signal generated as particles in the particle/fluid mixture pass the first sensor;

a second sensor in communication with the internal volume and adapted to detect and measure at least one second signal generated as the particles pass the second sensor, wherein the first and second sensors are located a predetermined distance (d) apart from one another; and

a processor in communication with the first and second sensors adapted receive and manipulate the first and second measured signals to provide an output, wherein the processor determines from the output a transit time, and wherein the processor divides the distance (d) by the transit time to provide a velocity of the particle/fluid mixture.

30. The system according to any one of claims 25 to 29, wherein the fluid conveying

structure comprises a polymerization reactor or a cycle line of a polymerization reactor.

31. The system according to any one of claims 25 to 30 wherein the particles comprise catalyst particles, polymer particles, or both, and wherein the fluid comprises one or more hydrocarbons.