CN103234610A - Weighing method applicable to truck scale - Google Patents
Weighing method applicable to truck scale Download PDFInfo
- Publication number
- CN103234610A CN103234610A CN2013101771825A CN201310177182A CN103234610A CN 103234610 A CN103234610 A CN 103234610A CN 2013101771825 A CN2013101771825 A CN 2013101771825A CN 201310177182 A CN201310177182 A CN 201310177182A CN 103234610 A CN103234610 A CN 103234610A
- Authority
- CN
- China
- Prior art keywords
- weighing
- model
- layer
- training
- formula
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Landscapes
- Control Of Transmission Device (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a weighing method applicable to a truck scale, and the method comprises a weighing sensor, a data acquisition device, a microprocessor and a display which are used, and the following steps that a weighing mathematical model is established, a weighing signal is acquired and on-line weighing is carried out; the step of establishing the weighing mathematical model comprises an ideal weighing model, an actual weighing model and the training methods thereof; the ideal weighing model is a linear function; the actual weighing model is a three-layer BP neural network, wherein a first layer is an input layer, a second layer is a hidden layer and a third layer is an output layer; before on-line weighing, the training of setting sample sizes must be carried out on the ideal weighing model and the actual weighing model; the training is carried out when the microprocessor is connected with an external computer; sample information of set quantity is acquired; the ideal weighing model and a derivative thereof serve as constraint conditions; finally the parameters W, b (1), V and b (2) of the actual weighing model are acquired and saved in the microprocessor; and then the external computer is removed.
Description
Technical field
The present invention relates to a kind of Weighing method that is applicable to truck scale.
Technical background
Truck scale is the important branch of weighing apparatus, is mainly used in the bulk supply tariff metering.Present analog electronic vehicle weighing apparatus occupies the leading position in truck scale market, and it mainly is made up of load-bearing force transmission mechanism (scale body), simulation LOAD CELLS, weighting display instrument three big master units.Truck scale generally has 4 ~ 12 tunnel LOAD CELLS according to the range difference.These sensors are distributed in scale body below symmetrically according to certain topological structure, have constituted a multisensor syste.There is coupling in this multisensor syste, and each road sensor output is interrelated, and relevant with the load loading position.Truck scale is concentrated the output signal of each road LOAD CELLS in the simulation terminal box and is added up, obtain one and the proportional voltage signal of tested quality of loads, after signal condition, A/D conversion, handle the acquisition weighing results by single-chip microcomputer, send demonstration, communication, finish weighing of tested load.Uneven loading error and linearity error are two principal elements that influence truck scale weighing results accuracy.Uneven loading error is because truck scale is subjected to various effect of non-linear, when tested load is on the truck scale loading end diverse location, and the inconsistent and error that produces of weighing results; Linearity error is because the characteristic of each road LOAD CELLS is inconsistent, causes the input and output and nonideal linear relationship of truck scale, thus the weighting error that produces.The uneven loading error of existing truck scale is what to separate with the linearity error compensation process, traditional uneven loading error compensation method is by manually regulating resistor in the truck scale terminal box repeatedly, change the sensor passage gain of every road, the compensation uneven loading error, this method manually-operated is loaded down with trivial details, inefficiency, compensation effect is poor; For this reason, there is the scholar to propose multiple linear regression analysis method (" large-scale weighing machine system partial load digitizing Research on compensation method ", Chen Chang, Wang Xiaoliang, Qin Zijun, Dalian University of Technology's journal, 1994,1), utilize method (" based on the intelligent weighing sensor research of advance data treatment technology ", Zhu Zijian, Nanjing Aero-Space University's PhD dissertation of Solving Linear angular difference correction factor, 2005), but these methods can not solve each sensor output relevance problem of bringing because of topological structure, also do not consider the various non-linear factor influences of truck scale, so compensation effect are relatively poor; Have the scholar adopt neural net method carry out the compensation of truck scale uneven loading error and linearity error (" based on the truck scale error compensation of multi-sensor information fusion ", vast stretch of wooded country army, Teng Zhaosheng, sea late, etc., Chinese journal of scientific instrument, 2009,6; " based on the truck scale error compensation of many RBF neural network ", vast stretch of wooded country army, Teng Zhaosheng, extra large late, etc., Hunan University's journal, 2010,5), though can significantly reduce weighting error, neural network needs a large amount of training samples, workload is big; The main cause that workload is big is that the truck scale range is big, and the standard test weight that needs during test is many, load(ing) point is many, and the information of weighing is obtained difficult.Existing truck scale linearity error compensation is after the uneven loading error compensation is finished, utilize following method to finish: at first to utilize the standard test weight of Different Weight to be carried on the truck scale body successively, obtain target weighing results and actual weighing results, then actual weighing results is doubly taken advantage of correction factor, make it to equal the target weighing results.This method is based on the truck scale input-output on the basis that is linear relationship, differ bigger with actual, so compensation effect is relatively poor.
Summary of the invention
The objective of the invention is to overcome deficiency of the prior art, a kind of new Weighing method that is applicable to truck scale is provided: namely utilize the good nonlinear function ability of approaching of neural network, the actual model of weighing of structure truck scale; Utilize the weigh constraint condition of model constructing neural network of the ideal of truck scale simultaneously, to reduce the required sample of neural metwork training, reduce workload, finish the actual Model Optimization of weighing of truck scale, realize that finally truck scale accurately weighs and error compensation.
Purpose of the present invention is achieved by following technical proposals:
Described Weighing method comprises use LOAD CELLS, data collector, microprocessor and display; Described LOAD CELLS is connected with microprocessor by data collector; Described display is connected with microprocessor; The step of described Weighing method comprises sets up weigh mathematical model, weighing-up wave collection, online weighing; The weigh mathematical model of mathematical model of described foundation comprises ideal weigh model, actual model and their training method of weighing, its step:
1) the described ideal model of weighing is linear function; Being input as of described linear function
NThe data of road LOAD CELLS output
, be output as A (X); Its input-output relational expression is formula (1):
In the formula,
p i Be the gain coefficient of LOAD CELLS, its value is by model training acquisition that ideal is weighed;
2) the described actual model of weighing is three layers of BP neural network, and ground floor is input layer, and the second layer is hidden layer, and the 3rd layer is output layer, and their network structure is as follows respectively:
The neuronal quantity of input layer
NNumber for LOAD CELLS;
The neuronal quantity of hidden layer
M=
, in the formula:
k=1 ~ 10 is correction factor;
LNeuronal quantity for output layer; The hidden layer excitation function adopts the Log-Sigmoid function, i.e. output
Be formula (2):
The neuronal quantity of output layer
LIt is 1; The output layer excitation function adopts linear function; The neural network output of output layer
Be formula (3):
In the formula,
WBe the weight matrix of neural network input layer to hidden layer,
b (1)Be the hidden layer bias vector,
VBe the weight vector of hidden layer to output layer, b
(2)Be the output layer bias,
XBe the neural network input vector,
w Mi Be input layer
iThe road is input to of hidden layer
mIndividual neuronic connection weights,
Be hidden layer
mIndividual neuronic bias,
v m Be hidden layer
mIndividual neuron is to the connection weights of output layer,
x(
i) be input layer
iThe road input;
3) before truck scale drops into online weighing, must carry out the training of set point number to weigh model and the actual model of weighing of ideal, training process carries out under microprocessor and situation that outer computer is connected, be constraint condition with weigh model and derivative thereof of ideal, obtain the actual model parameter of weighing at last
W,
b (1),
VAnd b
(2)Be kept in the microprocessor, withdraw outer computer then; Special-purpose training software is installed in the outer computer;
Step to the actual model training of weighing is as follows:
ⅰ) gather training sample: prepare the standard test weight of some, each standard test weight quality difference is carried in the standard test weight of different quality on the truck scale body at random,
NThe road sensor just has
NIndividual output data,
NIndividual output data and corresponding standard test weight quality constitute one group of training sample and are kept in the outer computer;
ⅱ) structure training objective function, its relational expression is formula (4):
ⅲ) ask the weigh derivative of model of truck scale ideal, its relational expression is formula (5):
ⅳ) ask the ideal weight coefficient of model in the training objective function of weighing, its relational expression is formula (6):
ⅵ) ask respectively
W, b (1) , VAnd b
(2)Increment
,
,
,
, and right
W, b (1) , VAnd b
(2)Upgrade, their relational expression is respectively formula (8), (9):
(9)
In the formula (9),
,
,
,
Be respectively
,
,
,
Value after the renewal,
,
,
,
Be respectively
,
,
,
Value before upgrading;
The training starting condition vii) is set, sets the training of quantity according to formula (6) ~ (9), make the error amount of training generation in setting range, obtain input layer respectively to the weight matrix of hidden layer
W, the hidden layer bias vector
b (1), hidden layer is to the weight vector of output layer
V, output layer bias b
(2)End value, and be kept in the storage element of microprocessor, be called during for online weighing.
In the training process of model that ideal is weighed, at first utilize the training sample of gathering in the actual model training process of weighing, utilize least square method to train then, obtain final model coefficient
p i , its relational expression is formula (10):
In the formula (10),
PServe as reasons
p i Vector, namely
P=[
p 1,
p 2...,
p N ]
T,
NNumber for LOAD CELLS;
X=[
X 1,
X 2...,
X j...,
X K ] be the input sample matrix,
Y=[
y 1,
y 2,
y j ...,
y K ].
Described online weighing should carry out after the actual model training of weighing is qualified, and its step is as follows:
1) will gather
NThe weighing-up wave of road LOAD CELLS is as the input vector of BP neural network ground floor
X
2) with input vector
XWith the actual model parameter of weighing that is kept in the storage unit
W,
b (1),
VAnd b
(2)In the substitution formula (3), try to achieve the output of BP neural network together
Be final weighing results;
3) show final weighing results on the display.
The method that described weighing-up wave is gathered is carried out the data that obtain after signal amplification, filtering and the analog-to-digital conversion process with the output signal of each road LOAD CELLS, as the weigh input vector of model and actual weigh model training and online weighing of ideal
X
Described microprocessor is single-chip microcomputer, dsp processor or other embedded system devices, and has storage unit.
Compared with the prior art the present invention has following advantage: the present invention can realize weighing automatically of truck scale, and carries out the automatic compensation of uneven loading error and linearity error, the accuracy that has improved weighing results greatly simultaneously; Reduce the truck scale required sample size of model training of weighing simultaneously, improved work efficiency.
Description of drawings
Fig. 1 trains the FB(flow block) of embodiment for the present invention.
Fig. 2 is the FB(flow block) of online weighing embodiment of the present invention.
Fig. 3 is neural network embodiment of the present invention, wherein,
f1 is the hidden layer excitation function,
f2 is the output layer excitation function.
Fig. 4 is the signal acquisition circuit theory diagram of one embodiment of the invention.
Fig. 5 is embodiments of the invention 5 truck scale online weighings and error compensation simulation result figure, and wherein (a) is the forward and backward weighing results comparison diagram of compensation, (b) is the forward and backward weighing results graph of errors comparison diagram of compensation.
In Fig. 4: 1-modulate circuit, 2-analog to digital conversion circuit, 3-microprocessor, 4-power module, 5-outer computer, 6-keyboard, 7-display, 8-communication interface.
Embodiment
The invention will be further described below in conjunction with drawings and Examples:
Referring to Fig. 1-5, described Weighing method comprises use LOAD CELLS, data collector, microprocessor 3 and display 7; Described LOAD CELLS is connected with microprocessor by data collector; Described display is connected with microprocessor; The step of described Weighing method comprises sets up weigh mathematical model, weighing-up wave collection, online weighing; The weigh mathematical model of mathematical model of described foundation comprises ideal weigh model, actual model and their training method of weighing, its step;
1) the described ideal model of weighing is linear function; Being input as of described linear function
NThe data of road LOAD CELLS output
, be output as A (X); Its input-output relational expression is formula (1):
In the formula,
p i Be the gain coefficient of LOAD CELLS, its value is by model training acquisition that ideal is weighed;
2) the described actual model of weighing is three layers of BP neural network, and ground floor is input layer, and the second layer is hidden layer, and the 3rd layer is output layer, and their network structure is as follows respectively:
The neuronal quantity of input layer
NNumber for LOAD CELLS;
The neuronal quantity of hidden layer
M=
, in the formula:
k=1 ~ 10 is correction factor;
LNeuronal quantity for output layer; The hidden layer excitation function adopts the Log-Sigmoid function, i.e. output
Be formula (2):
The neuronal quantity of output layer
LIt is 1; The output layer excitation function adopts linear function; The neural network output of output layer
Be formula (3):
In the formula,
WBe the weight matrix of neural network input layer to hidden layer,
b (1)Be the hidden layer bias vector,
VBe the weight vector of hidden layer to output layer, b
(2)Be the output layer bias,
XBe the neural network input vector,
w Mi Be input layer
iThe road is input to of hidden layer
mIndividual neuronic connection weights,
Be hidden layer
mIndividual neuronic bias,
v m Be hidden layer
mIndividual neuron is to the connection weights of output layer,
x(
i) be input layer
iThe road input;
3) before truck scale drops into online weighing, must carry out the training of set point number to weigh model and the actual model of weighing of ideal, training process carries out under microprocessor and situation that outer computer is connected, be constraint condition with weigh model and derivative thereof of ideal, obtain the actual model parameter of weighing at last
W,
b (1),
VAnd b
(2)Be kept in the microprocessor, withdraw outer computer then; Special-purpose training software is installed in the outer computer;
Step to the actual model training of weighing is as follows:
ⅰ) gather training sample: prepare the standard test weight of some, each standard test weight quality difference is carried in the standard test weight of different quality on the truck scale body at random,
NThe road sensor just has
NIndividual output data,
NIndividual output data and corresponding standard test weight quality constitute one group of training sample and are kept in the outer computer;
ⅱ) structure training objective function, its relational expression is formula (4):
ⅲ) ask the weigh derivative of model of truck scale ideal, its relational expression is formula (5):
ⅳ) ask the ideal weight coefficient of model in the training objective function of weighing, its relational expression is formula (6):
ⅵ) ask respectively
W, b (1) , VAnd b
(2)Increment
,
,
,
, and right
W, b (1) , VAnd b
(2)Upgrade, their relational expression is respectively formula (8), (9):
In the formula (9),
,
,
,
Be respectively
,
,
,
Value after the renewal,
,
,
,
Be respectively
,
,
,
Value before upgrading;
The training starting condition vii) is set, carries out the training of set point number according to formula (6) ~ (9), make the error amount of training generation in setting range, obtain input layer respectively to the weight matrix of hidden layer
W, the hidden layer bias vector
b (1), hidden layer is to the weight vector of output layer
V, output layer bias b
(2)End value, and be kept in the storage element of microprocessor, be called during for online weighing.
In the training process of model that ideal is weighed, at first utilize the training sample of gathering in the actual model training process of weighing, utilize least square method to train then, obtain final model coefficient
p i , its relational expression is formula (10):
In the formula (10),
PServe as reasons
p i Vector, namely
P=[
p 1,
p 2...,
p N ]
T,
NNumber for LOAD CELLS;
X=[
X 1,
X 2...,
X j...,
X K ] be the input sample matrix,
Y=[
y 1,
y 2,
y j ...,
y K ].
Described online weighing should carry out after the actual model training of weighing is qualified, and its step is as follows:
1) will gather
NThe weighing-up wave of road LOAD CELLS is as the input vector of BP neural network ground floor
X
2) with input vector
XWith the actual model parameter of weighing that is kept in the storage unit
W,
b (1),
VAnd b
(2)In the substitution formula (3), try to achieve the output of BP neural network together
Be final weighing results;
3) show final weighing results on the display 7.
The method that described weighing-up wave is gathered is carried out the data that obtain after signal amplification, filtering and the analog-to-digital conversion process with the output signal of each road LOAD CELLS, as the weigh input vector of model and actual weigh model training and online weighing of ideal
X
Described microprocessor is single-chip microcomputer, dsp processor or other embedded system devices, and has storage unit.
Described data collector comprises modulate circuit 1 and analog to digital conversion circuit 2, adopts technique known.
Embodiment 1:
Figure 4 shows that and be applicable to the weigh main structure of part of truck scale of the present invention: the output terminal of each LOAD CELLS connects data collector, and data collector comprises modulate circuit 1 and analog to digital conversion circuit 2; Weighing-up wave carries out handling through amplification, filtering in modulate circuit 1, and carrying out conversion process through analog to digital conversion circuit 2 is digital signal, and digital data transmission is to microprocessor 3; Microprocessor also disposes power module 4, keyboard 6, display 7 and communication interface 8; Power module 4 is power supplies such as modulate circuit 1, analog to digital conversion circuit 2, microprocessor 3.During training, microprocessor 3 connects outer computer 5 by communication interface 8; During online weighing, outer computer 5 is withdrawn.
Embodiment 2: the method for the actual model training of weighing and step.
In the present embodiment, truck scale have 8 tunnel LOAD CELLS (
N=8), range is 40 tons, and the max cap. of every road LOAD CELLS is 20 tons, and the number of divisions is 4000, the calibrating scale division value
eAnd actual graduation value
dBe 10kg; Microprocessor 3 adopts the high-performance single-chip microcomputer MSP430F449 of TI company.Referring to Fig. 1, as follows to the step of the actual model training of weighing:
(1) the structure ideal model of weighing, as the constraint condition of truck scale weighing system: with
NThe weighing-up wave of road LOAD CELLS output
For input, A (X) are output, the structure truck scale ideal model of weighing, as shown in Equation (1);
(2) structure is based on the actual model of weighing of the truck scale of neural network: be input, be output with the truck scale weighing results with No. 8 sensors signal of weighing, construct three layers of BP neural network of one 8 input 1 output, as the actual model of weighing of truck scale, its structure as shown in Figure 3.The hidden layer neuron number of this neural network
MSatisfy
,
k Get 1 ~ 10.
kCan be determined by following method: at first order
k=1, the error of calculating neural network if error meets the demands or minimum, is then determined
kNumerical value; Otherwise,
k Add 1, recomputate the error of neural network, until meeting the demands or the error minimum.By test of many times, present embodiment is determined
k=2, i.e. hidden layer neuron number
M=5.Among Fig. 3,
Be of input layer
iIndividual neuron is to of hidden layer
mNeuronic weights;
,
,
Be respectively hidden layer the 1st, 2,
MIndividual neuronic bias;
For output layer to hidden layer the
mIndividual neuronic weights; b
(2)Be the output layer bias;
,
,
Be respectively hidden layer the 1st, 2,
MIndividual neuronic output;
f 1Be the hidden layer excitation function, it adopts the Log-Sigmoid function; The output layer excitation function adopts linear function, so network output
Represent with formula (3);
(3) objective function of the actual model training of weighing of structure: be constraint condition with weigh model and derivative thereof of truck scale ideal, structure training sample objective function is with formula (4) expression;
(4) load standard test weight, gather weighing-up wave, form the training and testing sample: the standard test weight that utilizes different tonnages such as 0.5 ton, 1 ton, 3 tons, 6 tons, 12 tons, 18 tons, 24 tons, 36 tons, be carried in the diverse location of truck scale body respectively, system is by LOAD CELLS, modulate circuit 1, analog to digital conversion circuit 2 and microprocessor 3, gather 50 group of 8 road load cell signal, obtain 50 groups of samples
, as the formula (11), wherein 30 groups as train samples, and 20 groups are used for the neural network test sample book.Microprocessor 3 is sent to outer computer 5 by communication interface 8 with these samples and preserves, and prepares for the neural network off-line training, and the computing machine here refers to outer computer;
(5) ask the weigh coefficient of model of ideal: utilize formula (10) to ask the weigh coefficient of model of ideal
p i
(6) set the training initial parameter: target square error MSE is 0.0000000001, learning rate
ηBe 0.8, coefficient
μ j Online definite, the hidden neuron number
MBe 5, frequency of training is 10000.
(7) get training sample, neural network training is regulated neural parameter
W,
b (1),
VAnd b
(2): take out training sample in the sample set from be kept at outer computer, and utilize special-purpose training software to train, utilize formula (9) to adjust parameter simultaneously
W,
b (1),
VAnd b
(2)Special-purpose training software adopts the training method shown in formula (4) ~ (10), utilizes the exploitation of MATLAB and Virtual instrument LabVIEW developing platform, also can utilize other programming languages to realize.
(8) enter determining program " training is finished ", if finish, obtain the qualified parameter of the actual model of weighing
W,
b (1),
VAnd b
(2)Otherwise, return step and " get training sample " and restart the next round training;
(9) outer computer is downloaded to the actual model parameter of weighing in the microprocessor by communication interface, and is kept in the storage unit, for online weighing is prepared; Withdraw outer computer simultaneously, be i.e. being connected of disconnecting external computing machine and microprocessor, finish training.
According to neural network design theory (" neural network design ", work such as Martin T. Hagan, Dai Kuiyi, China Machine Press, 2005,8), if adopt traditional neural metwork training method (namely do not utilize the ideal of truck scale weigh model), need at least training sample number Num=(
M+ 1) *
N+ (
M+ 1) *
L, in the formula,
MBe the hidden neuron number,
NThe input layer number,
LBe the output layer number.In the present embodiment,
M=5,
N=8,
L=1, so Num=54, namely need 54 groups of training samples at least, otherwise can produce bigger extensive error, thereby the model that causes weighing is unavailable.Yet because the present invention utilized the ideal of truck scale to weigh model as priori, only utilized 30 groups of training samples to meet the demands, the training sample number is far fewer than 54 groups, thereby reduced workload.
Embodiment 3: online weighing.
In the present embodiment, truck scale have 8 tunnel LOAD CELLS (
N=8), range is 40 tons, and the max cap. of every road LOAD CELLS is 20 tons, and the number of divisions is 4000, the calibrating scale division value
eAnd actual graduation value
dBe 10kg; Microprocessor 3 adopts the high-performance single-chip microcomputer MSP430F449 of TI company, utilizes and has trained the qualified actual model of weighing to carry out online weighing among the embodiment 2.Referring to Fig. 2, the online weighing step is as follows:
(1) tested load loads: namely truck is carried on the truck scale body optional position, carries out the weighing-up wave collection after stablizing;
(2) microprocessor collection
NThe weighing-up wave of road LOAD CELLS: 8 tunnel LOAD CELLS output signals are gathered by LOAD CELLS, modulate circuit 1, analog to digital conversion circuit 2 and microprocessor 3 by system
x i (
i=1,2 ..., 8), as the input vector X of the actual model of weighing, i.e. X=[
x 1,
x 2...,
x 8];
(3) call
W,
b (1),
VAnd b
(2): the parameter of from storage unit, calling the actual model of weighing
W,
b (1),
VAnd b
(2)
(4) output of the actual model of weighing of calculating: calculate the output that obtains neural network according to formula (3), this output is the final weighing results of truck scale after the error compensation;
(5) show weighing results at display, and finish this online weighing.
The truck scale that adopts the method for the invention after tested after, obtain weighing and the error compensation effect as shown in Figure 5.Fig. 5 (a) carries out the forward and backward output weighing results of online weighing and error compensation for adopting this method, and Fig. 5 (b) contrasts for adopting the error before and after this method compensation.From figure as can be seen, adopt the truck scale weighing results of the method for the invention accurate, error compensation is effective.
Utilize the concerned countries standard of truck scale calibrating, the truck scale that adopts the method for the invention is carried out on-the-spot test, table 1 is the uneven loading error verification result, and table 2 is the linearity error verification result.The parameter of truck scale: 8 tunnel LOAD CELLS (
N=8), range is that 40 tons, the max cap. of LOAD CELLS are that 20 tons, the number of divisions are 4000, the calibrating scale division value
eAnd actual graduation value
dBe 10kg.
Table 1 uneven loading error verification result
The |
1# | 2# | 3# | 4# | 5# | 6# | 7# | 8# |
Error E(kg) | -7 | 2 | -5 | -6 | -8 | 7 | 4 | -5 |
Permissible error (kg) | ±10 | ±10 | ±10 | ±10 | ±10 | ±10 | ±10 | ±10 |
Table 2 linearity error verification result
Weighing range (t) | 0~0.2 | 0.2~5 | 5~10 | 10~20 | 20~40 |
Measurement error (kg) | +3 | -5 | -8 | -9 | +12 |
Permissible error (kg) | ±5 | ±5 | ±10 | ±10 | ±15 |
In the table 1,1# represents to load zone (being zone, LOAD CELLS position) No. 1, other same meaning arranged.By table 1,2 as can be seen, adopt the truck scale uneven loading error of this method less than the permissible error of concerned countries standard code, less than the permissible error of concerned countries standard code, error compensation is respond well equally for linearity error.
Above embodiment is not as a limitation of the invention.
Claims (5)
1. Weighing method that is applicable to truck scale, described Weighing method comprise and use LOAD CELLS, data collector, microprocessor and display; Described LOAD CELLS is connected with microprocessor by data collector; Described display is connected with microprocessor; The step of described Weighing method comprises sets up weigh mathematical model, weighing-up wave collection, online weighing; It is characterized in that: the weigh mathematical model of mathematical model of described foundation comprises ideal weigh model, actual model and their training method of weighing, its step;
1) the described ideal model of weighing is linear function; Being input as of described linear function
NThe data of road LOAD CELLS output
, be output as A (X); Its input-output relational expression is formula (1):
In the formula,
p i Be the gain coefficient of LOAD CELLS, its value is by model training acquisition that ideal is weighed;
2) the described actual model of weighing is three layers of BP neural network, and ground floor is input layer, and the second layer is hidden layer, and the 3rd layer is output layer, and their network structure is as follows respectively:
The neuronal quantity of input layer
NNumber for LOAD CELLS;
The neuronal quantity of hidden layer
M=
, in the formula:
k=1 ~ 10 is correction factor;
LNeuronal quantity for output layer; The hidden layer excitation function adopts the Log-Sigmoid function, i.e. output
Be formula (2):
The neuronal quantity of output layer
LIt is 1; The output layer excitation function adopts linear function; The neural network output of output layer
Be formula (3):
In the formula,
WBe the weight matrix of neural network input layer to hidden layer,
b (1)Be the hidden layer bias vector,
VBe the weight vector of hidden layer to output layer, b
(2)Be the output layer bias,
XBe the neural network input vector,
w Mi Be input layer
iThe road is input to of hidden layer
mIndividual neuronic connection weights,
Be hidden layer
mIndividual neuronic bias,
v m Be hidden layer
mIndividual neuron is to the connection weights of output layer,
x(
i) be input layer
iThe road input;
3) before truck scale drops into online weighing, must carry out the training of set point number to weigh model and the actual model of weighing of ideal, training process carries out under microprocessor and situation that outer computer is connected, be constraint condition with weigh model and derivative thereof of ideal, obtain the actual model parameter of weighing at last
W,
b (1),
VAnd b
(2)Be kept in the microprocessor, withdraw outer computer then; Special-purpose training software is installed in the outer computer;
Step to the actual model training of weighing is as follows:
ⅰ) gather training sample: prepare the standard test weight of some, each standard test weight quality difference is carried in the standard test weight of different quality on the truck scale body at random,
NThe road sensor just has
NIndividual output data,
NIndividual output data and corresponding standard test weight quality constitute one group of training sample and are kept in the outer computer;
ⅱ) structure training objective function, its relational expression is formula (4):
ⅲ) ask the weigh derivative of model of truck scale ideal, its relational expression is formula (5):
ⅳ) ask the ideal weight coefficient of model in the training objective function of weighing, its relational expression is formula (6):
ⅵ) ask respectively
W, b (1) , VAnd b
(2)Increment
,
,
,
, and right
W, b (1) , VAnd b
(2)Upgrade, their relational expression is respectively formula (8), (9):
In the formula (9),
,
,
,
Be respectively
,
,
,
Value after the renewal,
,
,
,
Be respectively
,
,
,
Value before upgrading;
The training starting condition vii) is set, sets the training of quantity according to formula (6) ~ (9), make the error amount of training generation in setting range, obtain input layer respectively to the weight matrix of hidden layer
W, the hidden layer bias vector
b (1), hidden layer is to the weight vector of output layer
V, output layer bias b
(2)End value, and be kept in the storage element of microprocessor, be called during for the truck scale online weighing;
In the training process of model that ideal is weighed, at first utilize the training sample of gathering in the actual model training process of weighing, utilize least square method to train then, obtain final model coefficient
p i , its relational expression is formula (10):
In the formula (10),
PServe as reasons
p i Vector, namely
P=[
p 1,
p 2...,
p N ]
T,
NNumber for LOAD CELLS;
X=[
X 1,
X 2...,
X j...,
X K ] be the input sample matrix,
Y=[
y 1,
y 2,
y j ...,
y K ].
2. the Weighing method that is applicable to truck scale according to claim 1, it is characterized in that: described online weighing should carry out after the actual model training of weighing is qualified, and its step is as follows:
1) will collect
NThe weighing-up wave of road LOAD CELLS is as the input vector of BP neural network ground floor
X
2) with input vector
XWith the actual model parameter of weighing that is kept in the storage unit
W,
b (1),
VAnd b
(2)In the substitution formula (3), try to achieve the output of BP neural network together
Be final weighing results;
3) show final weighing results on the display.
3. the Weighing method that is applicable to truck scale according to claim 1 and 2, it is characterized in that: the method that described weighing-up wave is gathered, the output signal of each road LOAD CELLS is carried out the data that obtain after signal amplification, filtering and the analog-to-digital conversion process, as the weigh input vector of model and actual weigh model training and online weighing of ideal
X
4. the Weighing method that is applicable to truck scale according to claim 1 and 2, it is characterized in that: described microprocessor is single-chip microcomputer, dsp processor or other embedded system devices, and has storage unit.
5. the Weighing method that is applicable to truck scale according to claim 3, it is characterized in that: described microprocessor is single-chip microcomputer, dsp processor or other embedded system devices, and has storage unit.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310177182.5A CN103234610B (en) | 2013-05-14 | 2013-05-14 | Weighing method applicable to truck scale |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310177182.5A CN103234610B (en) | 2013-05-14 | 2013-05-14 | Weighing method applicable to truck scale |
Publications (2)
Publication Number | Publication Date |
---|---|
CN103234610A true CN103234610A (en) | 2013-08-07 |
CN103234610B CN103234610B (en) | 2015-05-06 |
Family
ID=48882661
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201310177182.5A Expired - Fee Related CN103234610B (en) | 2013-05-14 | 2013-05-14 | Weighing method applicable to truck scale |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN103234610B (en) |
Cited By (16)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105973443A (en) * | 2016-05-17 | 2016-09-28 | 大连理工大学 | Digital truck scale offset load error correction method based on least square method |
CN105973444A (en) * | 2016-06-25 | 2016-09-28 | 湖南师范大学 | Improved automobile scale weighing method |
CN107607182A (en) * | 2017-08-04 | 2018-01-19 | 广西大学 | A kind of truck weighing system and Weighing method |
CN108491404A (en) * | 2018-01-22 | 2018-09-04 | 国电南瑞科技股份有限公司 | A kind of state estimation bad data recognition method based on BP neural network |
CN109377046A (en) * | 2018-10-18 | 2019-02-22 | 上海经达信息科技股份有限公司 | Overload of vehicle method of discrimination, system and device based on BP neural network |
CN109579967A (en) * | 2018-11-27 | 2019-04-05 | 上海交通大学 | Intelligent Dynamic weighing method and system |
CN109668610A (en) * | 2019-01-11 | 2019-04-23 | 东南大学 | The system of vehicle dynamically weighting method and its use based on neural net regression |
CN111664927A (en) * | 2020-05-28 | 2020-09-15 | 首钢京唐钢铁联合有限责任公司 | Method and device for judging metering state of rail weigher |
CN112129386A (en) * | 2019-06-24 | 2020-12-25 | 梅特勒-托利多(常州)测量技术有限公司 | Weighing apparatus and weighing method |
CN112393794A (en) * | 2019-08-18 | 2021-02-23 | 华东理工大学 | Diagnosis and reading correction method for platform scale of four-way weighing sensor when single sensor fault or unbalance loading occurs |
CN112747813A (en) * | 2019-10-29 | 2021-05-04 | 杭州壬辰科技有限公司 | Automatic part weighing and counting system and method |
CN112964345A (en) * | 2021-02-07 | 2021-06-15 | 广东电子工业研究院有限公司 | Freight car weighing system and weighing method thereof |
CN114543952A (en) * | 2022-01-06 | 2022-05-27 | 红云红河烟草(集团)有限责任公司 | Error self-compensation method of electronic belt scale for cigarette production |
DE102020007453A1 (en) | 2020-12-07 | 2022-06-09 | Daimler Truck AG | Method for determining a vehicle mass average and its output in the vehicle |
CN115790804A (en) * | 2023-02-08 | 2023-03-14 | 福建省智能交通信息工程有限公司 | Dynamic truck scale state monitoring method, medium, equipment and device |
CN116481626A (en) * | 2023-06-28 | 2023-07-25 | 深圳市汉德网络科技有限公司 | Vehicle-mounted weighing self-adaptive high-precision calibration method and system |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5228113A (en) * | 1991-06-17 | 1993-07-13 | The United States Of America As Represented By The Administrator Of The National Aeronautics And Space Administration | Accelerated training apparatus for back propagation networks |
US20040148144A1 (en) * | 2003-01-24 | 2004-07-29 | Martin Gregory D. | Parameterizing a steady-state model using derivative constraints |
US20050187643A1 (en) * | 2004-02-19 | 2005-08-25 | Pavilion Technologies, Inc. | Parametric universal nonlinear dynamics approximator and use |
CN102506983A (en) * | 2011-10-31 | 2012-06-20 | 湖南师范大学 | Weighing error automatic compensation method of vehicle scale |
-
2013
- 2013-05-14 CN CN201310177182.5A patent/CN103234610B/en not_active Expired - Fee Related
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5228113A (en) * | 1991-06-17 | 1993-07-13 | The United States Of America As Represented By The Administrator Of The National Aeronautics And Space Administration | Accelerated training apparatus for back propagation networks |
US20040148144A1 (en) * | 2003-01-24 | 2004-07-29 | Martin Gregory D. | Parameterizing a steady-state model using derivative constraints |
US20050187643A1 (en) * | 2004-02-19 | 2005-08-25 | Pavilion Technologies, Inc. | Parametric universal nonlinear dynamics approximator and use |
CN102506983A (en) * | 2011-10-31 | 2012-06-20 | 湖南师范大学 | Weighing error automatic compensation method of vehicle scale |
Non-Patent Citations (3)
Title |
---|
杨仁付: "采用导数约束关系的前向神经网络学习优化方法研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
胡包钢等: "如何增加人工神经元网络的透明度?", 《模式识别与人工智能》 * |
薛福珍: "基于先验知识和神经网络的非线性建模与预测控制", 《系统仿真学报》 * |
Cited By (22)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105973443A (en) * | 2016-05-17 | 2016-09-28 | 大连理工大学 | Digital truck scale offset load error correction method based on least square method |
CN105973443B (en) * | 2016-05-17 | 2018-09-04 | 大连理工大学 | Digital automobile weighing apparatus uneven loading error modification method based on least square method |
CN105973444A (en) * | 2016-06-25 | 2016-09-28 | 湖南师范大学 | Improved automobile scale weighing method |
CN105973444B (en) * | 2016-06-25 | 2018-09-28 | 湖南师范大学 | A kind of improved truck scale weighing method |
CN107607182A (en) * | 2017-08-04 | 2018-01-19 | 广西大学 | A kind of truck weighing system and Weighing method |
CN108491404A (en) * | 2018-01-22 | 2018-09-04 | 国电南瑞科技股份有限公司 | A kind of state estimation bad data recognition method based on BP neural network |
CN108491404B (en) * | 2018-01-22 | 2022-07-01 | 国电南瑞科技股份有限公司 | State estimation bad data identification method based on BP neural network |
CN109377046A (en) * | 2018-10-18 | 2019-02-22 | 上海经达信息科技股份有限公司 | Overload of vehicle method of discrimination, system and device based on BP neural network |
CN109579967A (en) * | 2018-11-27 | 2019-04-05 | 上海交通大学 | Intelligent Dynamic weighing method and system |
CN109668610A (en) * | 2019-01-11 | 2019-04-23 | 东南大学 | The system of vehicle dynamically weighting method and its use based on neural net regression |
CN112129386A (en) * | 2019-06-24 | 2020-12-25 | 梅特勒-托利多(常州)测量技术有限公司 | Weighing apparatus and weighing method |
CN112393794A (en) * | 2019-08-18 | 2021-02-23 | 华东理工大学 | Diagnosis and reading correction method for platform scale of four-way weighing sensor when single sensor fault or unbalance loading occurs |
CN112393794B (en) * | 2019-08-18 | 2024-01-26 | 华东理工大学 | Diagnosis and reading correction method for single sensor fault or unbalanced load of platform scale |
CN112747813A (en) * | 2019-10-29 | 2021-05-04 | 杭州壬辰科技有限公司 | Automatic part weighing and counting system and method |
CN111664927A (en) * | 2020-05-28 | 2020-09-15 | 首钢京唐钢铁联合有限责任公司 | Method and device for judging metering state of rail weigher |
DE102020007453A1 (en) | 2020-12-07 | 2022-06-09 | Daimler Truck AG | Method for determining a vehicle mass average and its output in the vehicle |
CN112964345A (en) * | 2021-02-07 | 2021-06-15 | 广东电子工业研究院有限公司 | Freight car weighing system and weighing method thereof |
CN114543952A (en) * | 2022-01-06 | 2022-05-27 | 红云红河烟草(集团)有限责任公司 | Error self-compensation method of electronic belt scale for cigarette production |
CN115790804A (en) * | 2023-02-08 | 2023-03-14 | 福建省智能交通信息工程有限公司 | Dynamic truck scale state monitoring method, medium, equipment and device |
CN115790804B (en) * | 2023-02-08 | 2023-04-21 | 福建省智能交通信息工程有限公司 | Dynamic truck scale state monitoring method, medium, equipment and device |
CN116481626A (en) * | 2023-06-28 | 2023-07-25 | 深圳市汉德网络科技有限公司 | Vehicle-mounted weighing self-adaptive high-precision calibration method and system |
CN116481626B (en) * | 2023-06-28 | 2023-08-29 | 深圳市汉德网络科技有限公司 | Vehicle-mounted weighing self-adaptive high-precision calibration method and system |
Also Published As
Publication number | Publication date |
---|---|
CN103234610B (en) | 2015-05-06 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN103234610A (en) | Weighing method applicable to truck scale | |
CN105973444B (en) | A kind of improved truck scale weighing method | |
CN102506983A (en) | Weighing error automatic compensation method of vehicle scale | |
CN100533311C (en) | Embedded type weighing system based on intelligent fuzzy control | |
CN104724638B (en) | The method that forklift weighing device and fork truck are weighed automatically | |
CN205664920U (en) | Modular whole car weighing apparatus weighing system | |
CN106569164A (en) | Method and system for synchronization testing of electric quantity of double-core electric energy meter | |
CN108388962A (en) | A kind of wind power forecasting system and method | |
CN102620939B (en) | Engine torque predicting method and engine torque predicting device | |
CN102759430A (en) | BP (Back Propagation) neural network based high-precision correction and test method for resonance cylinder pressure sensor | |
CN101793621B (en) | Progressive health monitoring method for cable system based on cable force monitoring during support settlement | |
CN105973443A (en) | Digital truck scale offset load error correction method based on least square method | |
CN110633790A (en) | Method and system for measuring residual oil quantity of airplane oil tank based on convolutional neural network | |
CN104568095B (en) | Automobile mounted Weighing method | |
CN105501884A (en) | Remote monitoring and service system for mine capacity | |
CN104102127A (en) | Airborne aerodynamic parameter identification system | |
CN101782473B (en) | Progressive type healthy monitoring method of rope system based on angle monitoring during support settlement | |
CN101782944B (en) | Progressive type healthy monitoring method of rope system based on space coordinate monitoring during support settlement | |
CN201364198Y (en) | Road vehicle dynamic weighing system | |
CN105352571A (en) | Granary weight detection method and device based on index relation estimation | |
CN205483210U (en) | Combined multrirange electronic truck scale of many weighing platforms | |
CN102494732A (en) | Silo material level balance deviation detection method and system | |
CN114152806B (en) | Electric energy sensor with three-way array structure and measurement system and method formed by same | |
CN206649007U (en) | Gas sensor temperature compensation means | |
CN104406671B (en) | Dynamic weighing method based on leakage current model, and dynamic weighing system based on leakage current model |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C14 | Grant of patent or utility model | ||
GR01 | Patent grant | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20150506 Termination date: 20170514 |
|
CF01 | Termination of patent right due to non-payment of annual fee |