CN102521651A - Bow net contact force prediction method based on NARX neural networks - Google Patents
Bow net contact force prediction method based on NARX neural networks Download PDFInfo
- Publication number
- CN102521651A CN102521651A CN2011104362224A CN201110436222A CN102521651A CN 102521651 A CN102521651 A CN 102521651A CN 2011104362224 A CN2011104362224 A CN 2011104362224A CN 201110436222 A CN201110436222 A CN 201110436222A CN 102521651 A CN102521651 A CN 102521651A
- Authority
- CN
- China
- Prior art keywords
- data
- test
- mrow
- normalization processing
- contact force
- 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.)
- Pending
Links
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 38
- 238000000034 method Methods 0.000 title claims abstract description 32
- 238000012360 testing method Methods 0.000 claims abstract description 55
- 238000010606 normalization Methods 0.000 claims abstract description 45
- 238000012545 processing Methods 0.000 claims abstract description 41
- 238000012549 training Methods 0.000 claims abstract description 11
- 238000004088 simulation Methods 0.000 claims abstract description 6
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 5
- 230000008878 coupling Effects 0.000 claims description 10
- 238000010168 coupling process Methods 0.000 claims description 10
- 238000005859 coupling reaction Methods 0.000 claims description 10
- 238000006073 displacement reaction Methods 0.000 claims description 7
- 238000005094 computer simulation Methods 0.000 claims description 6
- 230000001934 delay Effects 0.000 claims description 3
- 238000012886 linear function Methods 0.000 claims description 3
- 238000005516 engineering process Methods 0.000 abstract description 3
- 230000003137 locomotive effect Effects 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000005299 abrasion Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000010219 correlation analysis Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005265 energy consumption Methods 0.000 description 1
- 230000005284 excitation Effects 0.000 description 1
- 238000012417 linear regression Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 239000000725 suspension Substances 0.000 description 1
- 230000001052 transient effect Effects 0.000 description 1
Images
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Feedback Control In General (AREA)
Abstract
The invention, which belongs to the railway safe operation control technology field, discloses a bow net contact force prediction method based on nonlinear auto-regressive with eXogenous input (NARX) neural networks. The method comprises: test data are obtained by a simulation test; Normalization processing is carried out on the test data; an NARX neural networks prediction model is established; the first set number of data pairs are extracted from the test data that have been processed by the normalization and are used as training samples, and a bayesian regularization algorithm is employed to train the NARX neural networks prediction model; the second set number of data pairs are extracted from the test data that have been processed by the normalization and are used as testing samples; and data in the test samples are used as input data and are input into the trained NARX neural networks prediction model, reverse normalization processing is carried out on an output result and then the processed output result is used as a bow net contract force predicted value. According to the invention, the NARX neural networks model is employed to predict a bow net contact force, so that prediction precision of the bow net contact force is improved.
Description
Technical Field
The invention belongs to the technical field of railway safe operation control, and particularly relates to a pantograph-catenary contact force prediction method based on a NARX neural network.
Background
The development of high-speed railways is a necessary trend of modern construction of railways in China, and electric locomotives become the leading force in high-speed locomotives due to the advantages of large transportation capacity, high speed, low energy consumption, little pollution, low transportation price, safety, reliability and the like. In a high-speed electrified railway system, a pantograph-catenary current receiving system is directly related to the speed of a train, namely, a stable current receiving state needs to be maintained when the train runs at a high speed, namely, a certain contact pressure needs to be kept between a pantograph and a contact line. When the contact pressure is too small, the pantograph is easy to be off-line, namely, the pantograph is separated from the contact line and generates electric arcs; when the contact pressure is overlarge, the lifting amount of the contact line is overlarge, so that the contact line is locally bent, the fatigue damage of the contact line is caused, the abrasion of the contact line is increased, and the bow net accident is caused when the contact pressure is serious. Therefore, the measurement of the bow net contact force has important significance for ensuring the train running safety and the development of high-speed railways in China.
In Yuanhua, Pichong et al, the bow net coupling is realized by establishing a contact pair between bow nets, establishing a rigid suspension contact net coupling model, and carrying out nonlinear transient dynamic analysis on the model to obtain a contact force and a change curve of the vertical displacement of the sliding plate. Xuhaidong, xu min et al established a pantograph-catenary dynamic model based on the railway large-system dynamic theory, and applied the vibration response at the base of the pantograph at the top of the locomotive as the excitation of the pantograph-catenary system to the dynamic model, and discussed the influence of rail coupling vibration on the contact pressure of the pantograph-catenary.
The study methods of scholars at home and abroad on the pantograph-catenary include field test, semi-physical and semi-virtual (the pantograph is a real object, and the catenary is obtained through computer simulation) test, computer simulation and the like. The field test has extremely high requirements on a measuring method, data processing and the like, and only a few countries such as germany and the like can directly measure the contact force between the pantograph and the contact network. With the advent of high-speed electronic computers, it has become possible to simulate the huge structure of a catenary completely by using a numerical method, and the computer simulation method has become the most common research method.
The method for coupling the contact net with the pantograph is basically divided into two methods: one is the coupling achieved by equal displacement of the bow and the net at the contact point, proposed by wu tianxing, etc., and the other is the coupling achieved by a spring, proposed by zhangweihua, etc. The former does not have to choose the contact stiffness, but cannot be considered offline; in the latter case, the relationship between the dynamic contact force between the pantograph and the displacement of the slide plate is: when the two contact each other, the contact rigidity relationship is kept; once offline, the contact force will always be zero, and there is no longer any connection between the two. Therefore, simulating the bow-net coupling with a spring cannot study the vibration of the bow and net after off-line.
The pantograph can also be simulated to move along a contact line by using a life and death unit technology in engineering design software ANSYS, specifically, a pantograph is established at each node of the contact line, and the corresponding pantograph is activated or killed by the life and death unit technology in the solving process. The biggest defect of the method is that the node speed, the displacement continuity and the transmissibility of the pantograph cannot be considered, so that the calculation result is greatly different from the fact and is very inaccurate.
As can be seen from the above description, the existing bow net contact force prediction methods all have different degrees of defects, so a new bow net contact force prediction method needs to be provided to improve the accuracy of bow net contact force prediction.
Disclosure of Invention
The invention aims to provide a pantograph contact force prediction method based on a NARX neural network, which is used for solving the problem that the precision of the pantograph contact force calculated by a common pantograph contact force prediction method is not high.
In order to achieve the purpose, the invention adopts the technical scheme that the bow net contact force prediction method based on the NARX neural network is characterized by comprising the following steps of:
step 1: obtaining test data through a simulation test; the test data comprises contact line irregularity data and bow net contact force data corresponding to the contact line irregularity data;
step 2: carrying out normalization processing on the test data;
and step 3: establishing a NARX neural network prediction model;
and 4, step 4: extracting a first set number of data pairs from the test data after the normalization processing to be used as training samples; the data pairs refer to contact line irregularity data after normalization processing and pantograph-catenary contact force data after normalization processing corresponding to the contact line irregularity data;
and 5: respectively taking the contact line irregularity data after normalization processing in the training sample and the bow net contact force data after normalization processing corresponding to the contact line irregularity data as input data and output data, and training an NARX neural network prediction model by adopting a Bayesian regularization algorithm;
step 6: extracting a second set number of data pairs from the normalized test data to serve as test samples; the data pairs refer to contact line irregularity data after normalization processing and pantograph-catenary contact force data after normalization processing corresponding to the contact line irregularity data;
and 7: inputting the contact line irregularity data after normalization processing in the test sample as input data into the NARX neural network prediction model trained in the step 5, performing inverse normalization processing on an output result, and taking the output result after the inverse normalization processing as a bow net contact force prediction value.
Specifically, the step of obtaining the test data through the simulation test is to establish a bow net coupling dynamic model, and then utilize MATLAB/Simulink software to carry out dynamic simulation to obtain contact line irregularity data and bow net contact force data corresponding to the contact line irregularity data.
The normalization processing of the test data specifically utilizes a formula
For test data xiCarrying out normalization processing; wherein,
n is the number of test data.
The intermediate layer nodes of the NARX neural network prediction model adopt tan-sigmoid functions, the output layer nodes adopt linear functions, the number of the input layer nodes is 1, the number of the intermediate layer nodes is 15, the number of the output layer nodes is 1, and the input and output delays are 45; wherein the tan-sigmoid function isx is input data of the hidden layer, T is a scaling coefficient, and theta is a displacement coefficient.
The first set number of data pairs is 1300 data pairs.
The second set number of data pairs is 700 data pairs.
The step 7 is followed by a step of evaluating the performance of the NARX neural network prediction model by adopting a root mean square error method, in particular to a formula
Evaluating the performance of the trained NARX neural network prediction model; wherein y (i) is a target value in the test sample, ym(i) And N is the number of data in the test sample.
The invention adopts the NARX neural network model to predict the bow net contact force, thereby improving the prediction precision of the bow net contact force.
Drawings
Fig. 1 is a flow chart of a pantograph contact force prediction method based on a NARX neural network;
FIG. 2 is a diagram of a NARX neural network architecture;
FIG. 3 is a graph of bow net contact force test data output versus NARX neural network output;
fig. 4 is a correlation analysis of bow net contact force test data output with NARX neural network output.
Detailed Description
The preferred embodiments will be described in detail below with reference to the accompanying drawings. It should be emphasized that the following description is merely exemplary in nature and is not intended to limit the scope of the invention or its application.
Fig. 1 is a flow chart of a pantograph contact force prediction method based on a NARX neural network. In fig. 1, the method for predicting the bow net contact force based on the NARX neural network includes:
step 1: obtaining test data through a simulation test; wherein the test data comprises contact line irregularity data and bow net contact force data corresponding thereto.
Specifically, establishing a bow-net coupling dynamics model, and then utilizing MATLAB/Simulink software to perform dynamic simulation to obtain contact line irregularity data and bow-net contact force data corresponding to the contact line irregularity data.
Step 2: and (5) carrying out normalization processing on the test data.
The normalization of the test data includes normalization of contact line irregularity data and bow net contact force data corresponding thereto. The normalization is by formula
Normalization processing is carried out on the contact line irregularity data and the bow net contact force data; wherein, 
n is the number of test data, xiContact line irregularity data/bow net contact force data.
And step 3: and establishing a NARX neural network prediction model.
The structure of NARX neural network (Nonlinear Auto-Regressive with eXogenous input neural networks) is shown in FIG. 2. The intermediate layer nodes of the NARX neural network prediction model adopt tan-sigmoid functions, the output layer nodes adopt linear functions, the number of the input layer nodes is 1, the number of the intermediate layer nodes is 15, the number of the output layer nodes is 1, and the input and output delays are 45. Wherein the tan-sigmoid function is
x is input data of the hidden layer, T is a scaling coefficient, and theta is a displacement coefficient.
And 4, step 4: 1300 data pairs are extracted from the test data after the normalization processing to be used as training samples; the data pairs refer to contact line irregularity data after normalization processing and pantograph contact force data after normalization processing corresponding to the contact line irregularity data.
And 5: and respectively taking the contact line irregularity data after normalization processing in the training sample and the bow net contact force data after normalization processing corresponding to the contact line irregularity data as input data and output data, and training the NARX neural network prediction model by adopting a Bayesian regularization algorithm.
Bayesian Regularization (BR) algorithm means that in order to improve the network popularization capability, a special performance parameter composed of output errors, weights and thresholds of each layer is established in the training process, and the weights and the thresholds of the network are adjusted according to Levenberg-Martqualdt optimization theory, so that the parameter is minimized.
Step 6: extracting 700 data pairs from the test data after the normalization processing as test samples; the data pairs refer to contact line irregularity data after normalization processing and pantograph contact force data after normalization processing corresponding to the contact line irregularity data.
And 7: inputting the contact line irregularity data after normalization processing in the test sample as input data into the NARX neural network prediction model trained in the step 5, performing inverse normalization processing on an output result, and taking the output result after the inverse normalization processing as a bow net contact force prediction value.
The performance of the NARX neural network prediction model can be evaluated by adopting a root mean square error method, and particularly, a formula is utilized
Evaluating the performance of the trained NARX neural network prediction model; wherein y (i) is a target value in the test sample, ym(i) And N is the number of data in the test sample.
The smaller the RMSE value, the higher the prediction accuracy of the model, and the closer the predicted value is to the target value. Next, curve fitting is performed on the model output and the target output, so that the approximation degree between the target output value and the model output value can be more intuitively reflected, as shown in fig. 3. Finally, linear regression analysis is performed on the model output and the target output, and the correlation coefficient between the target output value and the model output value can be accurately calculated, as shown in fig. 4, where a is the neural network output data and T is the test output data.
The above description is only for the preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present invention are included in the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (7)
1. A bow net contact force prediction method based on a NARX neural network is characterized by comprising the following steps:
step 1: obtaining test data through a simulation test; the test data comprises contact line irregularity data and bow net contact force data corresponding to the contact line irregularity data;
step 2: carrying out normalization processing on the test data;
and step 3: establishing a NARX neural network prediction model;
and 4, step 4: extracting a first set number of data pairs from the test data after the normalization processing to be used as training samples; the data pairs refer to contact line irregularity data after normalization processing and pantograph-catenary contact force data after normalization processing corresponding to the contact line irregularity data;
and 5: respectively taking the contact line irregularity data after normalization processing in the training sample and the bow net contact force data after normalization processing corresponding to the contact line irregularity data as input data and output data, and training an NARX neural network prediction model by adopting a Bayesian regularization algorithm;
step 6: extracting a second set number of data pairs from the normalized test data to serve as test samples; the data pairs refer to contact line irregularity data after normalization processing and pantograph-catenary contact force data after normalization processing corresponding to the contact line irregularity data;
and 7: inputting the contact line irregularity data after normalization processing in the test sample as input data into the NARX neural network prediction model trained in the step 5, performing inverse normalization processing on an output result, and taking the output result after the inverse normalization processing as a bow net contact force prediction value.
2. The method as claimed in claim 1, wherein the obtaining of the test data through the simulation test is specifically that a bow-net coupling dynamics model is established, and then MATLAB/Simulink software is used for performing dynamic simulation to obtain contact line irregularity data and bow-net contact force data corresponding to the contact line irregularity data.
3. The method of claim 1, wherein the normalizing the test data is performed using a formula
For test data xiCarrying out normalization processing; wherein,
n is the number of test data.
4. The method of claim 1, wherein the NARX neural network prediction model employs tan-sigmoid function for intermediate layer nodes, linear function for output layer nodes, number of input layer nodes is 1, number of intermediate layer nodes is 15, number of output layer nodes is 1, input and output delays are all 45; wherein the tan-sigmoid function isx is input data of the hidden layer, T is a scaling coefficient, and theta is a displacement coefficient.
5. The method of claim 1, wherein the first set number of data pairs is 1300 data pairs.
6. The method of claim 1, wherein the second set number of data pairs is 700 data pairs.
7. The method as claimed in claim 1, wherein said step 7 is followed by the step of evaluating the performance of the NARX neural network prediction model using root mean square error, in particular using a formula
Evaluating the performance of the trained NARX neural network prediction model; wherein y (i) is a target value in the test sample, ym(i) And N is the number of data in the test sample.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2011104362224A CN102521651A (en) | 2011-12-22 | 2011-12-22 | Bow net contact force prediction method based on NARX neural networks |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2011104362224A CN102521651A (en) | 2011-12-22 | 2011-12-22 | Bow net contact force prediction method based on NARX neural networks |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102521651A true CN102521651A (en) | 2012-06-27 |
Family
ID=46292562
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2011104362224A Pending CN102521651A (en) | 2011-12-22 | 2011-12-22 | Bow net contact force prediction method based on NARX neural networks |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102521651A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110457822A (en) * | 2019-08-13 | 2019-11-15 | 西南交通大学 | A kind of contact force threshold model construction method for determining bow net contact electric arc and occurring |
| CN111209999A (en) * | 2018-11-21 | 2020-05-29 | 成都唐源电气股份有限公司 | Contact network performance degradation prediction method based on recurrent neural network |
| CN111367173A (en) * | 2020-03-06 | 2020-07-03 | 西南交通大学 | High-speed railway pantograph robust prediction control method based on state estimation |
| CN113028999A (en) * | 2021-02-24 | 2021-06-25 | 河南辉煌科技股份有限公司 | Contact line lift measurement method and system based on convolutional neural network |
| CN113076949A (en) * | 2021-03-31 | 2021-07-06 | 成都唐源电气股份有限公司 | Method and system for quickly positioning parts of contact net |
| CN113267286A (en) * | 2021-07-02 | 2021-08-17 | 中国国家铁路集团有限公司 | Railway bow net contact force identification method and device |
| CN114169223A (en) * | 2021-11-15 | 2022-03-11 | 西南交通大学 | Optimal design method for distribution coordinates of high-speed contact net dropper |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101819407A (en) * | 2010-04-02 | 2010-09-01 | 杭州电子科技大学 | Sewage pump station water level prediction method base on neural network |
| CN101963638A (en) * | 2010-09-07 | 2011-02-02 | 西南交通大学 | Bow net electric contact characteristic testing device |
-
2011
- 2011-12-22 CN CN2011104362224A patent/CN102521651A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101819407A (en) * | 2010-04-02 | 2010-09-01 | 杭州电子科技大学 | Sewage pump station water level prediction method base on neural network |
| CN101963638A (en) * | 2010-09-07 | 2011-02-02 | 西南交通大学 | Bow net electric contact characteristic testing device |
Non-Patent Citations (2)
| Title |
|---|
| XUE-MIAO PANG ET AL: "The Prediction of Derailment Coefficient Using NARX Neural Networks", 《ICTIS 2011》, 2 July 2011 (2011-07-02) * |
| 蔡磊，等: "利用NARX神经网络由IMF与太阳风预测暴时SYM_H指数", 《中国科学》, vol. 40, no. 1, 31 December 2010 (2010-12-31) * |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111209999A (en) * | 2018-11-21 | 2020-05-29 | 成都唐源电气股份有限公司 | Contact network performance degradation prediction method based on recurrent neural network |
| CN111209999B (en) * | 2018-11-21 | 2023-04-07 | 成都唐源电气股份有限公司 | Contact network performance degradation prediction method based on recurrent neural network |
| CN110457822A (en) * | 2019-08-13 | 2019-11-15 | 西南交通大学 | A kind of contact force threshold model construction method for determining bow net contact electric arc and occurring |
| CN110457822B (en) * | 2019-08-13 | 2022-04-29 | 西南交通大学 | Contact force threshold value model construction method for judging generation of bow net contact arc |
| CN111367173A (en) * | 2020-03-06 | 2020-07-03 | 西南交通大学 | High-speed railway pantograph robust prediction control method based on state estimation |
| CN111367173B (en) * | 2020-03-06 | 2021-06-25 | 西南交通大学 | High-speed railway pantograph robust prediction control method based on state estimation |
| CN113028999A (en) * | 2021-02-24 | 2021-06-25 | 河南辉煌科技股份有限公司 | Contact line lift measurement method and system based on convolutional neural network |
| CN113076949A (en) * | 2021-03-31 | 2021-07-06 | 成都唐源电气股份有限公司 | Method and system for quickly positioning parts of contact net |
| CN113267286A (en) * | 2021-07-02 | 2021-08-17 | 中国国家铁路集团有限公司 | Railway bow net contact force identification method and device |
| CN113267286B (en) * | 2021-07-02 | 2022-12-13 | 中国国家铁路集团有限公司 | Railway bow net contact force identification method and device |
| CN114169223A (en) * | 2021-11-15 | 2022-03-11 | 西南交通大学 | Optimal design method for distribution coordinates of high-speed contact net dropper |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Song et al. | Wind deflection analysis of railway catenary under crosswind based on nonlinear finite element model and wind tunnel test | |
| CN102521651A (en) | Bow net contact force prediction method based on NARX neural networks | |
| CN111144039B (en) | Train dynamic weighing system and weighing method based on deep learning | |
| Facchinetti et al. | Real-time catenary models for the hardware-in-the-loop simulation of the pantograph–catenary interaction | |
| Yao et al. | A study on the mechanism of vehicle body vibration affecting the dynamic interaction in the pantograph–catenary system | |
| CN103674578B (en) | Bullet train runs the detection method of dynamic performance state | |
| Deng et al. | Correlation model of deflection, vehicle load, and temperature for in‐service bridge using deep learning and structural health monitoring | |
| CN102360454A (en) | Wheel-track force prediction method based on NARX (Nonlinear Auto-regressive with Extra Inputs) neural network | |
| Miyamoto et al. | Online SOC estimation of battery for wireless tramcar | |
| CN102567576B (en) | Method for predicting rate of wheel load reduction | |
| Zhang et al. | Sensitivity analysis and research on optimisation methods of design parameters of high‐speed railway catenary | |
| Song et al. | Assessment of the current collection quality of pantograph–catenary with contact line height variability in electric railways | |
| CN103344395B (en) | A kind of confirmation method of bridge strengthening target bearing capacity and device | |
| CN112989660B (en) | Method for predicting corrosion of pipeline under subway stray current based on partial least square method | |
| Song et al. | Assessment of railway pantograph-catenary interaction performance with realistic pantograph strip imperfection | |
| Yang et al. | Bridge damage identification based on synchronous statistical moment theory of vehicle–bridge interaction | |
| Zhang et al. | Study on a novel field-measured load calibration method for the life assessment of the metro bogie frame | |
| Yin et al. | Reliability analysis of subway vehicles based on the data of operational failures | |
| Zheng et al. | A novel deep learning architecture and its application in dynamic load monitoring of the vehicle system | |
| CN110143217B (en) | Track state measuring method, system and device | |
| CN105741184A (en) | Transformer state evaluation method and apparatus | |
| CN107330264A (en) | A kind of verification method of bridge monitoring data reliability | |
| Wang et al. | Robust optimisation of the streamlined shape of a high-speed train in crosswind conditions | |
| Yang et al. | Braking process modeling and simulation of CRH2 electric multiple unit | |
| Wu et al. | Fast calibration for vibration-based pavement roughness measurement based on model updating of vehicle dynamics |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C12 | Rejection of a patent application after its publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20120627 |