Urban rail transit induction current prediction method based on source end and station end synchronous monitoring
By combining source-end and station-end synchronous monitoring with a BP neural network model, the station-end induced current during subway operation can be accurately predicted, solving the problems of abnormal transformer vibration and temperature rise, and optimizing the design of transmission lines.
Patent Information
- Application Number
- CN202411079262.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-07
- Publication Date
- 2024-11-22
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
In existing technologies, the methods for predicting induced current at station terminals during subway operation are not accurate enough, leading to problems such as abnormal vibration and local temperature rise in transformers, and neglecting the influence of induced current in transmission lines.
A method based on synchronous monitoring at the source and station ends is adopted. By acquiring traction current data at the source end, induced current data at the station end, and real-time magnetic induction intensity, the data are decomposed using an empirical mode algorithm, and an induced current prediction model is constructed based on a BP neural network model. The prediction is then made by combining the real-time position of the train and the traction current data.
It enables accurate prediction of induced current at the station end, optimizes the layout of transmission channels, reduces transformer anomalies, and improves power quality.
Smart Images

Figure CN119004973A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of rail transit induced current management, and in particular to a method for predicting induced current in urban rail transit based on source-end and station-end synchronous monitoring. Background Art
[0002] As the main mode of urban rail transit, subways are widely used in major cities due to their environmental protection, high efficiency, convenience and other advantages, and have become the mainstream public transportation mode for the public. Existing subway lines usually use 750V / 1500V DC traction power supply mode. Figure 1 The figure shows the mechanism of the induction current intrusion transformer caused by the subway line. The train obtains traction power through the sliding electrical contact between the pantograph and the contact network, and converts the DC traction power into three-phase AC power through the on-board inverter to drive the on-board AC traction motor. In order to realize the feedback of traction current to the ground traction substation, the train relies on the rolling electrical contact between the wheel pair and the rail to discharge the traction current to the rail, and feeds the traction current back to the ground traction substation through the rail, return line and drainage network. Since the subway train can be regarded as a moving variable current source during dynamic operation, the "transmission line-transformer-grounding loop" constitutes a closed traction current loop. As the train switches between acceleration, uniform speed and deceleration operation modes during operation, the traction current increases during the acceleration of the train, and the traction current is almost constant during the uniform speed of the train. During the deceleration of the train, the train switches to the regenerative braking mode, and the traction current is fed back from the train to the contact network, and will flow back to the ground traction substation along the contact network to realize the feedback of traction power.
[0003] When the train is in dynamic operation, the traction current changes along with the train position, resulting in dynamic changes in the magnetic field generated by the "grid-train-track" traction power supply circuit. The changing magnetic field directly causes low-frequency induced currents to be generated in the transmission lines erected near the subway lines, and invades the windings from the input end of the AC transformer along the transmission lines, causing abnormal vibrations and local temperature rise in the transformer, and directly affecting the quality of electric energy. In the prior art, the focus is usually on the analysis and suppression of stray currents leaking from the subway to the ground through the rails and the return network, and the induced currents generated in the transmission lines during the operation of the subway are not paid attention to. The proportion of station-end induced currents in each route is also often ignored.
[0004] Therefore, in the relevant technology, there is an urgent need for a method that can accurately predict the station-end induced current during subway operation. Summary of the invention
[0005] Based on this, it is necessary to provide an urban rail transit induced current prediction method based on source-end and station-end synchronous monitoring, which can accurately predict the station-end induced current during subway operation in response to the above technical problems.
[0006] In a first aspect, the present application provides a method for predicting induced current in urban rail transit based on source-end station-end synchronous monitoring. The method comprises:
[0007] Acquire source-end traction current data, station-end induced current data, and real-time magnetic induction intensity, wherein the source-end traction current data includes the real-time position of the train and its corresponding traction current;
[0008] Decomposing the source-end traction current data and the station-end induced current data by using an empirical mode algorithm;
[0009] Training an induced current prediction model based on the decomposed source-end traction current data, station-end induced current data, and theoretical induced current data, wherein the theoretical induced current data is calculated according to the real-time magnetic induction intensity, and the induced current prediction model is constructed based on a BP neural network model;
[0010] The real-time traction current data of the source end is measured based on the real-time position of the train, the real-time position of the train and its corresponding real-time traction current data of the source end are input into the trained induced current prediction model, and the real-time induced current of the station end is output.
[0011] Optionally, in one embodiment of the present application, the acquiring source end traction current data includes:
[0012] The real-time position of the train is obtained, the traction current of the train is collected based on the real-time position of the train, and the traction current is reduced in voltage and filtered.
[0013] Optionally, in one embodiment of the present application, the acquiring station-side induced current data includes:
[0014] Collect the induced current and neutral point current at the input end of the transformer at the station end;
[0015] The station-end induced current is determined based on the neutral point current and the induced current.
[0016] Optionally, in one embodiment of the present application, the acquiring of source-end traction current data, station-end induced current data, and real-time magnetic induction intensity includes:
[0017] Correcting the real-time magnetic induction intensity based on the traction current signal;
[0018] The theoretical induced current is determined based on the equivalent loop, the equivalent power transmission loop area and the real-time magnetic induction intensity.
[0019] Optionally, in one embodiment of the present application, the method further includes:
[0020] Based on the source-end traction current data and the station-end induced current data, an empirical mode algorithm is used to determine the source and proportion of the induced current generated on the grid side.
[0021] Optionally, in one embodiment of the present application, the method further includes:
[0022] An error signal is constructed based on the station-end induced current and the theoretical induced current, and interference at the station end is eliminated based on the error signal.
[0023] In the second aspect, the present application also provides an urban rail transit induced current prediction system based on source-end station-end synchronous monitoring, including a vehicle-mounted source-end current and positioning monitoring system, a transformer station-end current monitoring and control system, a transmission line equivalent circuit magnetic induction intensity monitoring system, and a data processing and fitting system, wherein:
[0024] The vehicle-mounted source-end current and positioning monitoring system includes a vehicle-mounted GPS positioning module, a Hall sensor module, a buck circuit module, a signal filtering module, and a communication module, which are used to transmit subway location signals and measure traction current data in real time, and send working signals to the transformer station-end current monitoring and control system, the transmission line equivalent circuit magnetic induction intensity monitoring system, and the data processing and fitting system;
[0025] A transformer station-end current monitoring and control system, including a signal receiving module, a transformer input-end current transformer and a Hall sensor, a multi-channel data acquisition instrument, a transformer input-end current detection device, and a transformer neutral point zero-sequence current detection device, for measuring AC and DC components and neutral point current in real time after receiving a working signal, and determining the measured induced current based on the AC and DC components and the neutral point current;
[0026] The transmission line equivalent circuit magnetic induction intensity monitoring system is installed at the transmission circuit and is used to measure the real-time magnetic induction intensity from the subway in real time after receiving the working signal;
[0027] The data processing and fitting system includes a digital control terminal, a signal receiving module, a signal processing module, a characteristic waveform extraction module and a waveform comparison module. After receiving the working signal, the measured induced current data, traction current data and train running position are processed and fitted to obtain a digital twin model of the induced current about the traction current value and the equivalent loop distance between the train and the transmission line.
[0028] The above-mentioned urban rail transit induced current prediction method based on source-end and station-end synchronous monitoring first obtains source-end traction current data, station-end induced current data, and real-time magnetic induction intensity; then, uses the empirical mode algorithm to decompose the source-end traction current data and station-end induced current data; then, trains an induced current prediction model based on the decomposed source-end traction current data, station-end induced current data, and theoretical induced current data, wherein the theoretical induced current data is calculated according to the real-time magnetic induction intensity, and the induced current prediction model is constructed based on the BP neural network model; finally, the real-time traction current data of the source end is measured based on the real-time position of the train, the real-time traction current data of the source end is input into the trained induced current prediction model, and the real-time induced current of the station end is output. That is to say, when predicting the induced current of urban rail transit, the traction current at the source end of the train and the induced current at the station end are monitored, and the EMD algorithm is used to analyze the source and station end data to determine the characteristic quantities of the source end traction current and the station end induced current. The fitting relationship of the induced current is trained based on multiple sets of data on the distance between the train and the equivalent loop and the size of the train traction current. The BP neural network model can predict the size of the induced current for the possible equivalent loop based on the distance and the size of the traction current, providing an important reference for the design of the equivalent loop of the transmission line. BRIEF DESCRIPTION OF THE DRAWINGS
[0029] Figure 1 A diagram showing the mechanism of the induction current intrusion into the transformer caused by a subway line in one embodiment;
[0030] Figure 2 It is a flow chart of a method for predicting induced current in urban rail transit based on source-end and station-end synchronous monitoring in one embodiment;
[0031] Figure 3 A schematic diagram of a digital twin for optimizing the layout of a power transmission channel considering subway induced current in an embodiment;
[0032] Figure 4 It is a flowchart of specific steps of a method for predicting induced current of urban rail transit based on source-end-station-end synchronous monitoring in one embodiment;
[0033] Figure 5 A set of measured data decomposition diagrams in one embodiment;
[0034] Figure 6 A schematic diagram of the structure of a vehicle-mounted source terminal current and positioning monitoring system in one embodiment;
[0035] Figure 7 Schematic diagram of the structure of a transformer station-end current monitoring and control system in one embodiment.
[0036] Explanation of the reference numerals: 1. traction power supply station; 2. rectifier circuit; 3. rail; 4. pantograph; 5. contact network; 6. train; 7. on-board signal transmission; 8. station-end three-phase transformer; 9. transformer grounding drainage network; 10. grounding equivalent branch; 11. soil layer; 12. overhead tower; 13. overhead transmission line; 14. equivalent loop; 15. induced magnetic field; 16. magnetic induction intensity sensor; 17. multi-channel magnetometer; 18. signal transmission instrument; 19. traction power 1. Current collection sensor; 20. On-board inverter module; 21. Traction motor; 22. On-board source-end current and positioning monitoring system; 23. Transformer high-voltage incoming line terminal; 24. Transformer low-voltage and neutral point outgoing line terminal; 25. Hall sensor; 26. AC mutual inductor; 27. Transformer winding; 28. Winding equivalent circuit; 29. Zero-sequence current sensor; 30. Multi-channel current data acquisition instrument; 31. Transformer grounding system; 32. Transformer station-end current monitoring and control system. DETAILED DESCRIPTION
[0037] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.
[0038] In one embodiment, Figure 2 As shown, a method for predicting induced current in urban rail transit based on source-end station-end synchronous monitoring is provided, comprising the following steps:
[0039] S201: Acquire source-end traction current data, station-end induced current data, and real-time magnetic induction intensity, wherein the source-end traction current data includes the real-time position of the train and its corresponding traction current.
[0040] In the embodiment of the present application, first, the source-end traction current data, the station-end induced current data, and the real-time magnetic induction intensity are obtained. Among them, the source-end traction current data includes the real-time position of the train and its corresponding traction current, that is, based on the current position information of the train, the corresponding traction current size in the closed traction current loop formed by the "transmission line-transformer-ground loop" is collected. The station-end induced current data refers to the induced current that invades the transformer station end during the train's travel. The real-time magnetic induction intensity refers to the magnetic induction intensity in the closed traction current loop formed during the train's travel.
[0041] S203: Decomposing the source-end traction current data and the station-end induced current data using an empirical mode algorithm.
[0042] In the embodiment of the present application, an empirical mode algorithm (EMD) is used to decompose the source-end traction current data and the station-end induced current data to determine the characteristic quantities of the source-end traction current and the station-end induced current, wherein the specific steps of the empirical mode algorithm are: all the maximum points and minimum points in the original data X(t) are counted respectively, and then they are fitted by spline interpolation functions to form two envelopes, and the mean of the two envelopes is recorded as Y1(t), and the original data is subtracted from Y1(t) to obtain a new data waveform:
[0043] X1(t)=X(t)-Y1(t)
[0044] Among them, X1(t) represents the new data, X(t) represents the initial data, and Y1(t) represents the mean of the envelope of the two extreme value bars.
[0045] If the new data X1(t) still has negative maxima and positive minima, repeat the above process until the decomposition satisfies the eigenfunction conditions: (1) the number of zero crossing points and local extreme points of the function is equal or differs by at most one; (2) at any time point, the average value of the two envelopes is zero.
[0046] S205: Training an induced current prediction model based on the decomposed source-end traction current data, station-end induced current data, and theoretical induced current data, wherein the theoretical induced current data is calculated according to the real-time magnetic induction intensity, and the induced current prediction model is constructed based on a BP neural network model.
[0047] In the embodiment of the present application, the decomposed source-end traction current data, station-end induced current data and theoretical induced current data are input into the induced current prediction model to train the model, wherein the theoretical induced current data is calculated based on the real-time magnetic induction intensity. Specifically, the training refers to fitting the induced current data, the traction current magnitude and the train running position through data normalization to obtain the error back propagation (BP) neural network model of the induced current about the traction current value and the distance between the train and the equivalent loop of the transmission line. The relationship between the source-end traction current magnitude, the station-end induced current signal and the position fitting is as follows:
[0048]
[0049] Among them, λ is the starting coefficient, is the traction current and position signal at the source end, k and α are the fitting coefficients.
[0050] The BP neural network model includes: input layer, CNN-XGBoost training layer, prediction model, and output layer. The input layer receives the source-end traction current data, station-end induced current data, and theoretical induced current data. The CNN-XGBoost training layer trains the received historical data through convolution operations and pooling processing. Specifically, the waveform characteristic values of the intrusion induced current measured at the station end are compared in real time with the traction current size of different lines and the characteristic values of the theoretical calculation results after the measured magnetic induction intensity is compared. The source of the induced current is traced back in an integrated manner, and the obtained data is repeatedly trained to generate the CNN-XGBoost training layer to determine the expected value of the induced current near the intersection of subway lines. Finally, the prediction model uses the trained XGBoost regression model to perform regression prediction on the new data and output it through the output layer.
[0051] S207: Based on the real-time position of the train, measure the real-time traction current data of the source end, input the real-time position of the train and its corresponding real-time traction current data of the source end into the trained induced current prediction model, and output the real-time induced current of the station end.
[0052] In the embodiment of the present application, the real-time traction current data of the source end is measured based on the real-time position of the train, and the measured train traction current and position relationship are input into the trained induced current prediction model, i.e., the BP neural network model, to predict the magnitude of the induced current at the transformer station end within the possible equivalent area in real time. Figure 3 As shown in the figure, it is a digital twin schematic diagram of optimizing the layout of the transmission channel under the consideration of subway induced current. The digital twin model is imported based on the historical data of the source end to preliminarily design the overhead line corridor. According to the measured train traction current and position relationship, the model predicts the magnitude of the induced current at the transformer station end within the possible equivalent area. After multiple iterations and repeated training, the transmission line corridor structure with the smallest proportion of induced current is obtained, which also provides an important reference for the conversion of overhead lines to buried lines.
[0053] In the above-mentioned urban rail transit induced current prediction method based on source-end and station-end synchronous monitoring, first, the source-end traction current data, the station-end induced current data, and the real-time magnetic induction intensity are obtained; then, the source-end traction current data and the station-end induced current data are decomposed by using an empirical mode algorithm; then, an induced current prediction model is trained based on the decomposed source-end traction current data, the station-end induced current data, and the theoretical induced current data, wherein the theoretical induced current data is calculated according to the real-time magnetic induction intensity, and the induced current prediction model is constructed based on the BP neural network model; finally, the real-time traction current data of the source end is measured based on the real-time position of the train, the real-time traction current data of the source end is input into the trained induced current prediction model, and the real-time induced current of the station end is output. That is to say, when predicting the induced current of urban rail transit, the traction current at the source end of the train and the induced current at the station end are monitored, and the EMD algorithm is used to analyze the source and station end data to determine the characteristic quantities of the source end traction current and the station end induced current. The fitting relationship of the induced current is trained based on multiple sets of data on the distance between the train and the equivalent loop and the size of the train traction current. The BP neural network model can predict the size of the induced current for the possible equivalent loop based on the distance and the size of the traction current, providing an important reference for the design of the equivalent loop of the transmission line.
[0054] In one embodiment of the present application, obtaining source end traction current data includes:
[0055] The real-time position of the train is obtained, the traction current of the train is collected based on the real-time position of the train, and the traction current is reduced in voltage and filtered.
[0056] In one embodiment of the present application, the current position information of the train is obtained. When the train enters a preset transmission line equivalent loop monitoring range, the current traction current of the train is collected, and the traction current is subjected to voltage reduction and filtering. The voltage reduction refers to the use of a voltage reduction circuit, i.e., a BUCK circuit, for voltage reduction, and the filtering refers to the FTT-DFT filtering of the measured waveform glitches.
[0057] In one embodiment of the present application, obtaining the station-side induced current data includes:
[0058] S301: Collect the induced current and neutral point current at the input end of the transformer at the station end.
[0059] S303: Determine the station-end induced current based on the neutral point current and the induced current.
[0060] In one embodiment of the present application, due to the unbalanced intrusion of the three-phase induced current, a current value will be generated at the zero-sequence position. The zero-sequence current contains interference errors such as DC bias magnetism. After processing the input EMD, a characteristic quantity is obtained, and the zero-sequence interference is eliminated to determine whether an intrusion transformer induced current is generated. The intrusion transformer induced current is further determined by combining the error of the transformer input induced current and the zero-sequence current at the neutral point to obtain the measured induced current waveform.
[0061] In one embodiment of the present application, the step of obtaining the source-end traction current data, the station-end induced current data, and the real-time magnetic induction intensity includes:
[0062] S401: Correcting the real-time magnetic induction intensity based on the traction current signal.
[0063] S403: Determine a theoretical induced current based on the equivalent circuit, the equivalent power transmission circuit area and the real-time magnetic induction intensity.
[0064] In one embodiment of the present application, when trains are running on the subway line, they frequently repeat starting, acceleration, inertia, braking and other working conditions, and there may be multiple trains running at the same time in the same power supply section. In addition, the traction control strategies of subway trains are diverse. These factors will cause the traction current of the train to change randomly, and the magnetic field will change with the change of the traction current measured at the source end. The real-time measurement and monitoring of the real-time magnetic induction intensity signal from the subway is compared and analyzed with the traction current signal, that is, the interference magnetic field is eliminated by normalizing and calculating the energy error of the two signals to obtain the corrected real-time magnetic induction intensity. After that, the preset equivalent transmission loop area is imported to calculate the theoretical induced current. The specific calculation formula is as follows.
[0065]
[0066] Among them, N represents the number of equivalent circuits (generally N is 1 for a power transmission circuit), B represents the measured magnetic induction intensity, and S represents the equivalent power transmission circuit area.
[0067] In one embodiment of the present application, the method further includes:
[0068] Based on the source-end traction current data and the station-end induced current data, an empirical mode algorithm is used to determine the source and proportion of the induced current generated on the grid side.
[0069] In one embodiment of the present application, the station-side induced current data is decomposed using the empirical mode algorithm (EMD) to obtain X0,n, and the source-side traction current signal is decomposed using the empirical mode algorithm (EMD) to obtain X m , n.
[0070] X0,n=[IMF 0,1 IMF 0,2...IMF0,n]
[0071]
[0072] By solving the matrix inverse, the proportion coefficients k1 to kn can be obtained, and the source and proportion of the induced current generated on the grid side can be obtained.
[0073] In one embodiment of the present application, the method further includes:
[0074] An error signal is constructed based on the station-end induced current and the theoretical induced current, and interference at the station end is eliminated based on the error signal.
[0075] In one embodiment of the present application, since there is an amplitude difference between the actual measurement and the theoretical calculation, but there is a similarity between the two signal waveforms, the station-end induced current and the theoretical induced current calculated by the magnetic field are analyzed and compared to eliminate the interference of the station end, i.e., the input end of the transformer, such as the influence of geomagnetic explosions (GIC) caused by the action of solar wind and the geomagnetic field. Specifically, the station-end induced current i1(t) and the theoretical induced current i2(t) calculated by the magnetic field are signal processed. The error signal is constructed, and the formula is as follows.
[0076] Ev=∑[i1-A·i2] 2 =Σi1 2 -2AΣi1·i2+A 2 i2 2
[0077] Ev=E1-2AΣi1·i2+A 2 E2
[0078] When A = Σi1·i2 / E2, the error energy is minimum.
[0079] Define the correlation coefficient C = Σi1·i2, then the error energy is Ev = E1-C 2 / E2. The smaller the correlation coefficient, the greater the error energy, indicating that the magnetic field change at this time does not come from the main contribution of the traction current, but from geomagnetic storms or other factors. The larger the correlation coefficient, the smaller the error energy, that is, the smaller the station interference, and the more similar the two signals. When the error energy is 0, it means that the two signals are completely consistent and not subject to station interference.
[0080] The following is a specific example to illustrate the specific implementation steps of the urban rail transit induced current prediction method based on source-end station-end synchronous monitoring of the present application. Figure 4As shown, first, S501, obtain the source-end traction current data, the station-end induced current data, and the real-time magnetic induction intensity, wherein the source-end traction current data includes the real-time position of the train and its corresponding traction current data. Specifically, S503, obtain the real-time position of the train, collect the train traction current based on the real-time position of the train, and perform voltage reduction and filtering on the traction current. S505-S507, collect the induced current and neutral point current at the input end of the station-end transformer; determine the station-end induced current based on the neutral point current and the induced current.
[0081] Then, in S509-S511, the real-time magnetic induction intensity is corrected based on the traction current signal; the theoretical induced current is determined based on the equivalent loop, the equivalent transmission loop area and the real-time magnetic induction intensity. In S513, an error signal is constructed based on the station-end induced current and the theoretical induced current, and interference at the station-end is eliminated based on the error signal.
[0082] Afterwards, S515, the source-end traction current data and the station-end induced current data are decomposed using an empirical mode algorithm. Afterwards, S517, an induced current prediction model is trained based on the decomposed source-end traction current data, the station-end induced current data and the theoretical induced current data, wherein the theoretical induced current data is calculated based on the real-time magnetic induction intensity, and the induced current prediction model is constructed based on a BP neural network model. Finally, S519, the source-end real-time traction current data is measured based on the real-time position of the train, the real-time position of the train and its corresponding source-end real-time traction current data are input into the trained induced current prediction model, and the station-end real-time induced current is output. S521, based on the source-end traction current data and the station-end induced current data, an empirical mode algorithm is used to determine the source and proportion of the induced current generated on the power grid side.
[0083] In one embodiment of the present application, Figure 5 As shown, given a collected induced current signal OriginalSignal, the above method is used to decompose it into five intrinsic mode functions IMFs and a monotonic residual Res, namely:
[0084]
[0085] Where X(t) represents the original signal waveform, IMFn(t) represents the intrinsic mode function, and Res(t) represents the residual.
[0086] For other actual measured current waveforms, the number of repetitions of the intrinsic function IMF is determined using their respective standard deviations SD:
[0087]
[0088] Among them, SD krepresents the standard deviation of each repetition, X k is the waveform data, ε is the threshold, which is generally 0.2 to 0.3 of the modulus of the eigenfunction of the initial data. k When it is less than the threshold, the group of current signals stops screening.
[0089] IMF5 uses the terminal divergence effect of EMD as the threshold for ending the number of repetitions of the intrinsic function. The waveform characteristics of IMF1-4 can be used for feature recognition, and the “station end-source end” data feature recognition can be achieved using techniques such as pattern recognition.
[0090] Based on the same inventive concept, the embodiment of the present application also provides an urban rail transit induced current prediction system based on source end-station synchronous monitoring based on the above-mentioned urban rail transit induced current prediction method based on source end-station synchronous monitoring. The implementation scheme for solving the problem provided by the system is similar to the implementation scheme recorded in the above-mentioned method, so the specific limitations in one or more embodiments of the urban rail transit induced current prediction system based on source end-station synchronous monitoring provided below can be referred to the above limitations on the urban rail transit induced current prediction method based on source end-station synchronous monitoring, and will not be repeated here.
[0091] In one embodiment, an urban rail transit induced current prediction system based on source-end station-end synchronous monitoring is provided, including a vehicle-mounted source-end current and positioning monitoring system, a transformer station-end current monitoring and control system, a transmission line equivalent circuit magnetic induction intensity monitoring system, and a data processing and fitting system, wherein:
[0092] The vehicle-mounted source-end current and positioning monitoring system includes a vehicle-mounted GPS positioning module, a Hall sensor module, a buck circuit module, a signal filtering module, and a communication module, which are used to transmit subway location signals and measure traction current data in real time, and send working signals to the transformer station-end current monitoring and control system, the transmission line equivalent circuit magnetic induction intensity monitoring system, and the data processing and fitting system;
[0093] A transformer station-end current monitoring and control system, including a signal receiving module, a transformer input-end current transformer and a Hall sensor, a multi-channel data acquisition instrument, a transformer input-end current detection device, and a transformer neutral point zero-sequence current detection device, for measuring AC and DC components and neutral point current in real time after receiving a working signal, and determining the measured induced current based on the AC and DC components and the neutral point current;
[0094] The transmission line equivalent circuit magnetic induction intensity monitoring system is installed at the transmission circuit and is used to measure the real-time magnetic induction intensity from the subway in real time after receiving the working signal;
[0095] The data processing and fitting system includes a digital control terminal, a signal receiving module, a signal processing module, a characteristic waveform extraction module and a waveform comparison module. After receiving the working signal, the measured induced current data, traction current data and train running position are processed and fitted to obtain a digital twin model of the induced current about the traction current value and the equivalent loop distance between the train and the transmission line.
[0096] In one embodiment of the present application, an urban rail transit induced current prediction system based on source-end station-end synchronous monitoring includes a vehicle-mounted source-end current and positioning monitoring system, a transformer station-end current monitoring and control system, a transmission line equivalent circuit magnetic induction intensity monitoring system, and a data processing and fitting system, wherein:
[0097] like Figure 6 As shown in the figure, it is a structural schematic diagram of the vehicle-mounted source-end current and positioning monitoring system, which includes a vehicle-mounted GPS positioning module, a Hall sensor module, a buck circuit module, a signal filtering module and a communication module, which are used to transmit subway position signals and measure traction current data in real time, and send working signals to the transformer station-end current monitoring and control system, the transmission line equivalent circuit magnetic induction intensity monitoring system and the data processing and fitting system. Among them, the vehicle-mounted GPS positioning module is used to obtain the current position information of the train. When the train enters the preset transmission line equivalent circuit monitoring range, it sends a collection instruction to the Hall sensor module and to the transformer station-end monitoring system corresponding to the transmission line equivalent circuit. In this way, the vacuum period of equipment operation can be effectively reduced, and the economic benefits and service life of the equipment can be improved. After receiving the collection instruction, the Hall sensor module collects the current traction current of the train and sends it to the buck circuit module. The buck circuit module is used to perform buck processing on the traction current and send the processed traction current to the signal filtering module. The signal filtering module is used to filter the received traction current and send the obtained traction current to the communication module. The communication module is used to receive monitoring range parameters from the digital processing system and forward them to the vehicle-mounted GPS positioning module, and send the received traction current to the digital processing system.
[0098] like Figure 7As shown, it is a schematic diagram of the structure of the transformer station-end current monitoring and control system, including a signal receiving module, a current transformer and a Hall sensor at the transformer input end, a multi-channel data acquisition instrument, a current detection device at the transformer input end, and a transformer neutral point zero-sequence current detection device, which is used to measure the AC and DC components and the neutral point current in real time after receiving the working signal, and determine the measured induced current based on the AC and DC components and the neutral point current. Among them, the signal receiving module starts working after receiving the source end signal system. The current transformer and the Hall sensor measure the AC and DC components at the transformer input end respectively, through the multi-channel data acquisition instrument. After the multi-channel data acquisition instrument collects the induced current data from the transformer input end, it sends the data to the current detection device at the transformer input end. The current at the neutral point is measured by the zero-sequence current detection device.
[0099] The transmission line equivalent circuit magnetic induction intensity monitoring system is installed at the transmission circuit and is used to measure the real-time magnetic induction intensity from the subway after receiving the working signal.
[0100] The data processing and fitting system includes a digital control terminal, a signal receiving module, a signal processing module, a characteristic waveform extraction module and a waveform comparison module. After receiving the working signal, the measured induced current data, traction current data and train running position are processed and fitted to obtain a digital twin model of the induced current about the traction current value and the equivalent loop distance between the train and the transmission line.
[0101] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.
[0102] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.
[0103] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0104] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.
Claims
1. A method for predicting induced current in urban rail transit based on source-end and station-end synchronous monitoring, characterized in that: The method comprises: Acquire source-end traction current data, station-end induced current data, and real-time magnetic induction intensity, wherein the source-end traction current data includes the real-time position of the train and its corresponding traction current data; Decomposing the source-end traction current data and the station-end induced current data by using an empirical mode algorithm; Training an induced current prediction model based on the decomposed source-end traction current data, station-end induced current data, and theoretical induced current data, wherein the theoretical induced current data is calculated according to the real-time magnetic induction intensity, and the induced current prediction model is constructed based on a BP neural network model; The real-time traction current data of the source end is measured based on the real-time position of the train, the real-time position of the train and its corresponding real-time traction current data of the source end are input into the trained induced current prediction model, and the real-time induced current of the station end is output.
2. The method for predicting induced current in urban rail transit based on source-end-station synchronous monitoring according to claim 1 is characterized in that: The obtaining of source end traction current data comprises: The real-time position of the train is obtained, the traction current of the train is collected based on the real-time position of the train, and the traction current is reduced in voltage and filtered.
3. The method for predicting induced current in urban rail transit based on source-end-station synchronous monitoring according to claim 1 is characterized in that: The obtaining of station-side induced current data comprises: Collect the induced current and neutral point current at the input end of the transformer at the station end; The station-end induced current is determined based on the neutral point current and the induced current.
4. The method for predicting induced current in urban rail transit based on source-end-station synchronous monitoring according to claim 1 is characterized in that: The step of obtaining the source-end traction current data, the station-end induced current data, and the real-time magnetic induction intensity includes: Correcting the real-time magnetic induction intensity based on the traction current signal; The theoretical induced current is determined based on the equivalent loop, the equivalent power transmission loop area and the real-time magnetic induction intensity.
5. The method for predicting induced current in urban rail transit based on source-end-station synchronous monitoring according to claim 1 is characterized in that: The method further comprises: Based on the source-end traction current data and the station-end induced current data, an empirical mode algorithm is used to determine the source and proportion of the induced current generated on the grid side.
6. The method for predicting induced current in urban rail transit based on source-end-station synchronous monitoring according to claim 1 is characterized in that: The method further comprises: An error signal is constructed based on the station-end induced current and the theoretical induced current, and interference at the station end is eliminated based on the error signal.
7. An urban rail transit induced current prediction system based on source-end and station-end synchronous monitoring, including a vehicle-mounted source-end current and positioning monitoring system, a transformer station-end current monitoring and control system, a transmission line equivalent circuit magnetic induction intensity monitoring system, and a data processing and fitting system, wherein: The vehicle-mounted source-end current and positioning monitoring system includes a vehicle-mounted GPS positioning module, a Hall sensor module, a buck circuit module, a signal filtering module, and a communication module, which are used to transmit subway location signals and measure traction current data in real time, and send working signals to the transformer station-end current monitoring and control system, the transmission line equivalent circuit magnetic induction intensity monitoring system, and the data processing and fitting system; A transformer station-end current monitoring and control system, including a signal receiving module, a transformer input-end current transformer and a Hall sensor, a multi-channel data acquisition instrument, a transformer input-end current detection device, and a transformer neutral point zero-sequence current detection device, for measuring AC and DC components and neutral point current in real time after receiving a working signal, and determining the measured induced current based on the AC and DC components and the neutral point current; The transmission line equivalent circuit magnetic induction intensity monitoring system is installed at the transmission circuit and is used to measure the real-time magnetic induction intensity from the subway in real time after receiving the working signal; The data processing and fitting system includes a digital control terminal, a signal receiving module, a signal processing module, a characteristic waveform extraction module and a waveform comparison module. After receiving the working signal, the measured induced current data, traction current data and train running position are processed and fitted to obtain a digital twin model of the induced current about the traction current value and the equivalent loop distance between the train and the transmission line.
Citation Information
Patent Citations
Combined method wind speed prediction model construction method
CN111353640A
Online prediction method and device for grounding current of transformer iron core, storage medium and fault diagnosis method
CN114925718A
Urban rail transit short-time passenger flow prediction method based on CEEMDAN-LSTM
CN115034457A
Transformer magnetic bias evaluation method for stray current and induced current proportion analysis
CN116400149A
Method and device for detecting running state of current transformer and electronic equipment
CN117290679A
