High-speed rail traction system motor temperature monitoring device based on RBF neural network

Abstract

The invention relates to a high-speed rail traction system motor temperature monitoring device based on an RBF neural network, belongs to the field of intelligent sensing, and solves the problems of low sensor precision and high early warning and false alarm rate. Comprising a plurality of traction motor temperature sensors which are distributed in each motor and used for measuring the temperature of the motor in real time, obtaining the current motor temperature and transmitting the current motor temperature to a microprocessor; the microprocessor performs nonlinear error temperature compensation on the current motor temperature monitored by the sensor by using an RBF neural network based on an improved particle swarm to obtain compensated motor temperature values, judges whether all the compensated temperature values are smaller than a first temperature threshold value or not, and sends a first early warning instruction to the early warning module if not; a speed reduction instruction is sent to a high-speed rail centralized control center, and a signal is sent to the air feeder corresponding to the motor with the compensated temperature value exceeding the preset temperature threshold value for cooling; if any one of all the compensated temperature values is larger than or equal to a second temperature threshold value, a stopping instruction is sent; and the early warning module performs early warning based on the early warning instruction. And accurate temperature monitoring is realized.

Description

Technical Field

[0001] The present invention belongs to the field of intelligent sensing systems, and in particular to a high-speed rail traction system motor temperature monitoring device based on a RBF (Radial Basis Function) neural network. Background Art

[0002] With the rapid development of smart sensors, smart sensor components have been widely used in many industries. Temperature sensors are a common sensor technology application on high-speed trains. In order to ensure the normal operation of the main components of the train and timely and effective protection when abnormalities occur, temperature sensors are installed in the main components of the train traction system, such as traction transformers, traction converters, and traction motors. Usually, the traction motor is forced air-cooled, using independent traction fans and ventilation ducts for cooling.

[0003] The traction motor is not only the power source of the high-speed train, but also a vulnerable part of the train. The normal operation of the traction motor is an important guarantee for the safe and punctual operation of the train. If the traction motor overheats, the efficiency will decrease, affecting the power output capacity of the vehicle, which will lead to a decrease in the acceleration of the high-speed rail. Once discovered, the EMU must stop for inspection during operation, which seriously affects the order of train operation. This fault has attracted enough attention from relevant departments.

[0004] At present, the methods for monitoring the temperature of traction motors mainly rely on traditional temperature sensors and monitoring systems. Traditional temperature monitoring systems usually use linear models for temperature estimation and prediction. However, due to the complex working conditions and large environmental changes of traction motors, linear models have limitations in accuracy and stability and cannot fully meet actual needs. On the other hand, traditional temperature monitoring systems often lack the ability to process and dynamically adjust temperature data in real time. Once a temperature abnormality occurs, manual intervention or alarm processing based on static thresholds is often required, and timely response and early warning to temperature changes cannot be achieved. Therefore, there is an urgent need for an intelligent sensing system to achieve accurate detection of motor temperature.

[0005] In summary, the present application improves a high-speed rail traction system motor temperature monitoring device based on RBF neural network to solve the technical problems of low accuracy of existing high-speed rail traction motor temperature sensors and high false alarm rate due to failure to consider the motor temperature change trend warning. Summary of the invention

[0006] In view of the above analysis, an embodiment of the present invention aims to provide a high-speed railway traction system motor temperature monitoring device based on RBF neural network, so as to solve the technical problems that the temperature sensor of the traction motor of the high-speed railway vehicle has low accuracy and high false alarm rate due to failure to consider the temperature change trend of the motor.

[0007] The present invention discloses a high-speed railway traction system motor temperature monitoring device based on RBF neural network, comprising:

[0008] Multiple traction motor temperature sensors are distributed in each traction motor of the high-speed train, and are used to measure the motor temperature of each traction motor in real time, obtain the current motor temperature of the traction motor, and transmit it to the microprocessor;

[0009] The microprocessor uses an improved particle swarm-based RBF neural network to perform nonlinear error temperature compensation on the current motor temperature of the traction motor monitored by the traction motor temperature sensor to obtain a compensated motor temperature value, and determines whether all compensated motor temperature values ​​are less than a first temperature threshold. If not, a first warning instruction is sent to the warning module, and a speed reduction instruction is sent to the high-speed rail control center at the same time, and a control signal is sent to the traction motor blower corresponding to the traction motor whose compensated motor temperature value exceeds a preset temperature threshold, so as to start the blower to cool the traction motor; if any one of all compensated motor temperature values ​​is greater than or equal to a second temperature threshold, a stop instruction is sent; and the second temperature threshold is greater than the first temperature threshold;

[0010] The early warning module is used to issue a corresponding early warning based on the early warning instruction.

[0011] Furthermore, the system further comprises:

[0012] The intelligent center module is used to analyze the stored real-time motor temperature data in real time. If it is found that the motor temperature continues to rise within a preset time period and the current motor temperature approaches the first temperature threshold, a second warning instruction is sent to the warning module; if a traction motor temperature sensor exceeds the normal operating temperature range, a third warning instruction is sent to the warning module;

[0013] A visualization module, used to display the real-time motor data of each traction motor, and provide an editing interface for presetting the first and second temperature thresholds, the preset time period, the normal operating temperature range, and the first, second, and third warning instruction templates;

[0014] A voice prompt module is used for the warning voice prompt of the warning module.

[0015] Furthermore, the traction motor temperature sensor transmits the current motor temperature of the traction motor collected in real time to the microprocessor through a built-in Wi-Fi module.

[0016] Furthermore, the intelligent center module also pre-stores the indexing table data of the traction motor temperature sensor, and the indexing table data serves as a sample data set.

[0017] Furthermore, the nonlinear error temperature compensation is performed on the current motor temperature of the high-speed train monitored by each traction motor temperature sensor using the improved particle swarm-based RBF neural network to obtain the compensated motor temperature value, which includes:

[0018] Using an improved particle swarm algorithm to optimize the expansion speed parameters of the RBF neural network model, obtaining the optimal position of the group and passing it to the RBF neural network model to update the expansion speed parameters, and using the training set in the sample data set to train the RBF neural network model after the updated parameters;

[0019] Using the test set in the sample data set to verify, if the verification fails, returning to the step of optimizing the expansion speed parameters of the RBF neural network model using the improved particle swarm algorithm until the verification passes, and obtaining the trained RBF neural network model as the traction motor temperature sensor compensation network;

[0020] Based on the traction motor temperature sensor compensation network, the temperature signal measured by each traction motor temperature sensor is compensated to obtain a compensated motor temperature value.

[0021] Furthermore, the optimization of the expansion speed parameters of the RBF neural network model based on the improved particle swarm algorithm training includes:

[0022] Initialize the parameters in the improved particle swarm algorithm;

[0023] Initialize the initial position and initial velocity of individuals in the population;

[0024] In each iteration, the individual fitness value is calculated based on the individual position and the sample data set using a 10-fold cross validation method; the inertia weight of the individual is dynamically adjusted;

[0025] updating the individual velocity based on the individual inertia weight, and then updating the individual position;

[0026] When the maximum number of iterations is reached, the optimal position of the group is returned, and the expansion speed parameters of the RBF neural network model are optimized and updated based on the optimal position of the group.

[0027] Further, the calculating of individual fitness values ​​based on individual positions by using a 10-fold cross validation method includes:

[0028] The individual position is input into the model as the expansion speed parameter of the RBF neural network model; the sample data set is randomly divided into 10 sample subsets;

[0029] In each cross-validation fitting, 9 sample subsets are used as training subsets and the remaining 1 sample subset is used as a validation subset. The RBF neural network is trained using the 9 training subsets, and then the sum of the absolute errors between the temperature fitting value obtained after the validation subset is input into the RBF neural network and the corresponding standard temperature value is calculated;

[0030] Perform 10 cross validations, and add up the sums of the 10 absolute errors obtained to obtain the average value as the individual fitness value.

[0031] Furthermore, the optimal position of the group is obtained in the following way:

[0032] At each iteration, the individual fitness value of each individual is compared with its historical individual optimal position, and if it is better than the historical individual optimal position, the individual optimal position is updated;

[0033] Determine whether the current individual optimal position of each individual is better than the group optimal position, and if so, update the group optimal position to the individual optimal position;

[0034] During the particle swarm iteration, it is determined whether the individual position exceeds a preset value range, and if so, the individual position is randomized.

[0035] Furthermore, the inertia weight of the individual is dynamically adjusted as follows:

[0036]

[0037] Among them, w is the inertia weight of the individual, m is the current number of iterations, M is the maximum number of iterations, and w max 、w min are the individual's maximum and minimum inertia weights, σ ​​is the inertia adjustment factor, and p and q are random numbers.

[0038] Further, the individual speed is updated based on the inertia weight of the individual, as follows:

[0039] v t+1 =wv t +c1r1(p best -x t )+c2r2(g best -x t )

[0040] Among them, v t+1 is the speed of the individual at time t+1, v t is the speed of the individual at time t, r1 and r2 are random numbers in the range of 0-1, c1 is the individual learning factor, c2 is the global learning factor, and x t is the individual’s current position, p best is the optimal position of the individual, gbest is the optimal position for the group;

[0041] Update the particle position as follows:

[0042] x t+1 =x t +v t+1 *1=x t +v t+1

[0043] Among them, x t+1 is the position of the individual at time t+1, x t is the position of the individual at time t, v t+1 is the speed of the individual at time t+1.

[0044] Compared with the prior art, the present invention can achieve at least one of the following beneficial effects:

[0045] 1. Improve the accuracy of motor temperature control: Through the temperature compensation of RBF neural network, the temperature can be predicted and controlled more accurately, so that the temperature of the motor of the high-speed rail traction system can be kept within the set first temperature threshold range;

[0046] 2. Realize remote monitoring and adjustment: Using the visualization module, train operation and maintenance personnel or train staff can monitor the temperature of the motor of the high-speed train traction system anytime and anywhere, and perform remote duct cooling, thereby realizing remote intelligent management, avoiding manual inspections, and saving human resource costs;

[0047] 3. Improve the efficiency of motor temperature control: The application of improved particle swarm algorithm can find the optimal solution of temperature compensation parameters more quickly, thereby improving the efficiency and response speed of the motor temperature monitoring system of the high-speed rail traction system;

[0048] 4. Low false alarm rate: The nonlinear error temperature compensation of the current motor temperature of the motor monitored by the temperature sensor of the traction system motor is performed using the RBF neural network based on the improved particle swarm to obtain the compensated motor temperature value, and determine whether all compensated motor temperature values ​​are less than the first temperature threshold. If not, a temperature drop warning is issued and a duct cooling command is sent. It is determined whether all compensated motor temperature values ​​are less than the second temperature threshold. If not, a stop command is sent for parking and maintenance, which greatly reduces the problem of high false alarm rate.

[0049] In the present invention, the above-mentioned technical solutions can also be combined with each other to achieve more preferred combination solutions. Other features and advantages of the present invention will be described in the subsequent description, and some advantages can become obvious from the description, or can be understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained through the contents particularly pointed out in the description and the drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] The drawings are only for the purpose of illustrating particular embodiments and are not to be considered limiting of the present invention. Like reference symbols denote like components throughout the drawings.

[0051] Figure 1 A schematic diagram of a high-speed rail traction system motor temperature monitoring device based on RBF neural network;

[0052] Figure 2 Schematic diagram of RBF motor temperature compensation method for improved particle swarm;

[0053] Figure 3 It is the structural diagram of RBF neural network model;

[0054] Figure 4 This is the schematic diagram of the nonlinear motor temperature error compensation for the high-speed train traction motor. DETAILED DESCRIPTION

[0055] The preferred embodiments of the present invention are described in detail below in conjunction with the accompanying drawings, wherein the accompanying drawings constitute a part of this application and are used together with the embodiments of the present invention to illustrate the principles of the present invention, but are not used to limit the scope of the present invention.

[0056] like Figure 1 As shown, a specific embodiment of the present invention discloses a high-speed railway traction system motor temperature monitoring device based on RBF neural network, comprising:

[0057] Multiple traction motor temperature sensors are distributed in each traction motor of the high-speed train, and are used to measure the motor temperature of each traction motor in real time, obtain the current motor temperature of the traction motor, and transmit it to the microprocessor;

[0058] The microprocessor uses an improved particle swarm-based RBF neural network to perform nonlinear error temperature compensation on the current motor temperature of the traction motor monitored by the traction motor temperature sensor to obtain a compensated motor temperature value, and determines whether all compensated motor temperature values ​​are less than a first temperature threshold. If not, a first warning instruction is sent to the warning module, and a speed reduction instruction is sent to the high-speed rail control center at the same time, and a control signal is sent to the traction motor blower corresponding to the traction motor whose compensated motor temperature value exceeds a preset temperature threshold, so as to start the blower to cool the traction motor; if any one of all compensated motor temperature values ​​is greater than or equal to a second temperature threshold, a stop instruction is sent; and the second temperature threshold is greater than the first temperature threshold;

[0059] The early warning module is used to issue a corresponding early warning based on the early warning instruction.

[0060] For example, the temperature sensor in the present invention is a semiconductor temperature sensor, which can be debugged within a working range of -50°C to 150°C, and environmental factors such as humidity will not affect the motor temperature value result of the temperature compensation of the RBF neural network.

[0061] Each traction motor in a high-speed train is equipped with a traction motor temperature sensor, which is attached to a base at the bottom of the traction motor housing using thermal grease.

[0062] The bottom of the traction motor housing is a closed space and will not be dislocated as the high-speed train travels at high speed.

[0063] For the preset first temperature threshold and the second temperature threshold, the second temperature threshold is greater than the first temperature threshold. For example, the preset first temperature threshold is 34° C. and the second temperature threshold is 36° C., which can be set immediately according to actual business needs.

[0064] The first warning instruction and the second warning instruction are both motor temperature warning instructions;

[0065] The second warning instruction is a status warning instruction of the traction motor temperature sensor.

[0066] The warning instructions are set according to actual business needs. For example, the first warning instruction can be set as "traction motor ID+traction motor temperature sensor ID+motor real-time temperature+time stamp".

[0067] The intelligent center module is used to analyze the stored real-time motor temperature data in real time. If it is found that the motor temperature continues to rise within a preset time period and the current motor temperature approaches the first temperature threshold, a second warning instruction is sent to the warning module; if a traction motor temperature sensor exceeds the normal operating temperature range, a third warning instruction is sent to the warning module;

[0068] A visualization module, used to display the real-time motor data of each traction motor, and provide an editing interface for presetting the first and second temperature thresholds, the preset time period, the normal operating temperature range, and the first, second, and third warning instruction templates;

[0069] Voice prompt module, used for the warning voice prompt of the warning module

[0070] The preset time period is, for example, set to 30 seconds;

[0071] The second warning instruction is, for example, set to "traction motor ID+traction motor temperature sensor ID+motor temperature approaches+first temperature threshold within a preset time period".

[0072] The third warning instruction is, for example, set to "traction motor ID+traction motor temperature sensor ID+current motor temperature+abnormality", corresponding to the application scenario where the traction motor temperature sensor may fall off.

[0073] Normal working temperature range, the maximum value of the range is less than the first temperature threshold. For example, 30-33°C

[0074] The voice prompt module converts the text warning information issued by the warning module into voice prompts to help train operation and maintenance personnel or train staff to receive warnings immediately and take corresponding measures.

[0075] In the visualization module, the real-time motor temperature data of each traction motor is displayed:

[0076] (1) Green motor temperature data value: indicates that the motor temperature is within the normal temperature range;

[0077] (2) Yellow motor temperature data value: indicates that the motor temperature is greater than or equal to the first temperature threshold;

[0078] (3) Red motor temperature data value: indicates that the shaft motor temperature is greater than or equal to the second temperature threshold;

[0079] (4) Blue motor temperature data value: indicates that the motor temperature exceeds the normal operating temperature range, the traction motor temperature sensor falls off, and other abnormal faults.

[0080] The intelligent central module also pre-stores the indexing table data of the traction motor temperature sensor, and the indexing table data serves as a sample data set.

[0081] The graduation table stores the standard resistance value of the traction motor temperature sensor element and the corresponding standard temperature value.

[0082] The samples in the sample data set include the standard resistance value of the traction motor temperature sensor element and the corresponding standard temperature value;

[0083] The standard resistance value is used as an input variable of the RBF neural network model.

[0084] The graduation table data of the traction motor temperature sensor includes the standard resistance value of the traction motor temperature sensor element and the corresponding standard temperature value, and the two are in a one-to-one correspondence;

[0085] Obtain the standard resistance value and the corresponding standard temperature value from the graduation table of the traction motor temperature sensor as sample data.

[0086] Exemplarily, the sample data set includes more than 10,000 pieces of sample data, each sample including a standard resistance value and a corresponding standard temperature value of a traction motor temperature sensor element.

[0087] The sample data set is divided into a training set and a test set. The training set is used to train the RBF neural network model, and the test set is used to verify the RBF neural network model.

[0088] Exemplarily, the ratio of the sample size of the training set to the sample size of the test set is 8:2.

[0089] The traction motor temperature sensor transmits the current motor temperature of the traction motor collected in real time to the microprocessor via a built-in Wi-Fi module.

[0090] like Figure 2 As shown, the nonlinear error temperature compensation is performed on the current motor temperature of the high-speed train monitored by each traction motor temperature sensor by using the RBF neural network based on the improved particle swarm to obtain the compensated motor temperature value, which includes:

[0091] Using an improved particle swarm algorithm to optimize the expansion speed parameters of the RBF neural network model, obtaining the optimal position of the group and passing it to the RBF neural network model to update the expansion speed parameters, and using the training set in the sample data set to train the RBF neural network model after the updated parameters;

[0092] Using the test set in the sample data set to verify, if the verification fails, returning to the step of optimizing the expansion speed parameters of the RBF neural network model using the improved particle swarm algorithm until the verification passes, and obtaining the trained RBF neural network model as the traction motor temperature sensor compensation network;

[0093] Based on the traction motor temperature sensor compensation network, the temperature signal measured by each traction motor temperature sensor is compensated to obtain a compensated motor temperature value.

[0094] The compensated motor temperature value is obtained as a compensation value of the current motor temperature of the traction motor where the traction motor temperature sensor is located.

[0095] Initialize the RBF neural network model;

[0096] RBF neural network model, such as Figure 3 As shown in Figure 1, it includes 1 input layer, 1 hidden layer and 1 output layer. The hidden layer is composed of a set of radial basis functions.

[0097] The standard resistance value of the traction motor temperature sensor is used as the input data of the input layer of the RBF neural network model, and the output data of the output layer is the compensated motor temperature value. The RBF neural network model is built based on this.

[0098] like Figure 3As shown in , the input layer input data x is 1 input, which is the standard resistance value of the traction motor temperature sensor; the output layer output y is the compensated motor temperature value.

[0099] The nonlinear temperature error of the traction motor temperature sensor includes the nonlinear temperature error caused by the "temperature drift" of the electronic element and the nonlinear temperature error of the amplifier circuit.

[0100] like Figure 4 As shown in the figure, the motor temperature compensation principle of the traction motor of the high-speed train traction system is g(R x ) = f -1 (t x ).

[0101] f(t x ) is the input-output function of the traction motor temperature sensor, i.e., the actual measured output of the traction motor temperature sensor, g(R xi ) is the compensation network function of the traction motor temperature sensor, and the output is the motor temperature value after compensation, t x is the actual temperature of the traction motor under test, i.e. the standard temperature value in the graduation table, t y is the motor temperature value after compensation.

[0102] The purpose of the traction motor temperature sensor compensation network is to ensure that the compensated motor temperature value is equal to the actual temperature value of the traction motor being measured after compensation, satisfying the requirements of formula (1):

[0103] t y =g(R x )R x =g(R x )f(t x )t x =t x Formula (1)

[0104] Right now

[0105] g(R x ) = f -1 (t x ) Formula (2)

[0106] Among them, R x is the resistance value of the sample traction motor temperature sensor, t x is the actual temperature of the traction motor under test.

[0107] The purpose of the traction motor temperature sensor compensation network is also the predetermined goal of the training of the RBF neural network model.

[0108] What is known is the standard resistance value of the traction motor under the standard temperature value. Now we need to obtain the motor temperature value based on the standard resistance value of the traction motor, which is the relationship between the function and the inverse function.

[0109] Since f(t x ) is a nonlinear function, so g(R x ) is also a nonlinear function. Through the powerful approximation of nonlinear functions by the RBF neural network model, it is possible to approximate g(R) with arbitrary accuracy. x ), thereby completing the nonlinear temperature error compensation for the traction motor temperature sensor.

[0110] The spread speed is a key parameter of the RBF neural network model. The value of the spread speed will affect the network fitting effect. Generally speaking, the larger the spread speed, the smoother the final fitted function will be, but the error will be larger, the number of neurons in the intermediate hidden layer will be larger, and the amount of calculation will be larger;

[0111] The smaller the spread value, the better the function approximation effect will be, but the approximation curve will be not smooth. The RBF neural network model needs to obtain a suitable spread value, that is, the input range of the neuron response should cover a large enough range, and each neuron should not overlap.

[0112] For covering a large enough input range: the value of the expansion speed spread should be selected to ensure that the response range of each RBF neuron is large enough to cover the main features in the input space. If the expansion speed spread value is set too small, some input data may not be covered, thus affecting the fitting ability of the RBF neural network model. Therefore, the expansion speed spread value should be large enough to cover the entire distribution range of the input data resistance value;

[0113] To avoid overlap between neurons: At the same time, the value of the spread speed should not be set too large, so as to cause overlap between adjacent neurons. If neurons overlap, the output of the RBF neural network model will be unstable, and it will be difficult to accurately fit the input data. The selection of the spread speed value also needs to ensure that the response ranges of the neurons do not overlap.

[0114] When selecting the expansion speed spread value, it is necessary to comprehensively consider the distribution of input data, neural network fitting, and the mutual influence between neurons, so as to find a suitable expansion speed spread value that can cover the input space and avoid overlapping of the corresponding ranges of neurons, and determine the optimal spread value to obtain the best RBF neural network performance.

[0115] The method of optimizing the expansion speed parameters of the RBF neural network model by using the improved particle swarm algorithm includes:

[0116] Initialize the parameters in the improved particle swarm algorithm;

[0117] Initialize the initial position and initial velocity of individuals in the population;

[0118] In each iteration, the individual fitness value is calculated based on the individual position and the sample data set using a 10-fold cross validation method; the inertia weight of the individual is dynamically adjusted;

[0119] updating the individual velocity based on the individual inertia weight, and then updating the individual position;

[0120] When the maximum number of iterations is reached, the optimal position of the group is returned, and the expansion speed parameters of the RBF neural network model are optimized and updated based on the optimal position of the group.

[0121] Initialize the parameters in the improved particle swarm algorithm; initialize the initial position and initial velocity of individuals in the population;

[0122] Initialization parameters are as follows:

[0123] Particle individual learning factor c1, global learning factor c2;

[0124] The number of individuals in the particle group (total number of particles) N, that is, the group size;

[0125] Particle dimension Dim;

[0126] The number of algorithm iterations M;

[0127] The maximum value of the particle individual position L max , minimum value L min ;

[0128] Particle individual inertia weight w i =(w i1 ,w i2 ,w i3 ,…,w id ), where i represents the i-th particle, d represents the d-th dimension, and each dimension of each particle corresponds to an inertia coefficient;

[0129] The initial inertia weight w max ; Inertia weight w at the maximum number of iterations of the algorithm min ; Inertia adjustment factor σ; Random numbers p, q.

[0130] Among them, the particle dimension parameter of the improved particle swarm algorithm is the number of input variables of the RBF neural network model.

[0131] The input of the RBF neural network model is the standard resistance value of the traction motor temperature sensor element, 1 variable, so the particle dimension Dim = 1;

[0132] The number of algorithm iterations M is the preset maximum number of iterations.

[0133] The input variable of the RBF neural network model is the standard resistance value of the traction motor temperature sensor element, which ensures that the optimal position of the group output by the improved particle swarm algorithm is a value.

[0134] If the individual learning factor c1 = 0, the individual particle only has group experience, and the particle swarm algorithm converges faster, but it is easy to fall into the local optimum;

[0135] If the global learning factor c2 = 0, the particles do not share information as a group. A particle group of size N is equivalent to running N individual particles that act independently. The probability of obtaining a solution is very small. Therefore, c1 = c2 is generally set. In this way, the individual particle experience and the group experience have the same important influence, making the final group optimal position more accurate.

[0136] Exemplarily, c1=c2=2.

[0137] In the initial stage of the algorithm, the individual inertia weight is set to a larger value to promote global search, because a larger inertia weight will make the particle more inclined to move along the original moving direction, thus helping to jump out of the global optimal solution.

[0138] As the number of iterations increases, the inertia weight gradually decreases. At the maximum number of iterations, the inertia weight is set to a smaller value to ensure that the algorithm can adjust the particle movement direction more finely in the later stages of the search and better converge to the global optimal solution.

[0139] For example, w max The default value is 0.9; w min The default value is 0.4.

[0140] Initialize the initial positions and initial velocities of individuals in the population;

[0141] The individual initial position and individual initial velocity are randomly generated to initialize the position and velocity of each individual (ie particle).

[0142] The initial position and initial velocity of the individual are randomly initialized within a certain range to ensure that the initial state of each individual has a certain diversity.

[0143] In each iteration, the individual fitness value is calculated based on the individual position and the sample data set using a 10-fold cross validation method;

[0144] The calculation of individual fitness values ​​based on individual positions using a 10-fold cross validation method includes:

[0145] The individual position is input into the model as the expansion speed parameter of the RBF neural network model; the sample data set is randomly divided into 10 sample subsets;

[0146] In each cross-validation fitting, 9 sample subsets are used as training subsets and the remaining 1 sample subset is used as a validation subset. The RBF neural network is trained using the 9 training subsets, and then the sum of the absolute errors between the temperature fitting value obtained after the validation subset is input into the RBF neural network and the corresponding standard temperature value is calculated;

[0147] Perform 10 cross validations, and add up the sums of the 10 absolute errors obtained to obtain the average value as the individual fitness value.

[0148] Based on the individual fitness value, the individual optimal position p is obtained best and the optimal position of the group g best .

[0149] At the beginning of the iteration, the initial individual fitness value calculated based on the individual initial position is used as the individual optimal position p best , the minimum value of the optimal position of all individuals in the particle swarm, as the initial group optimal position g best .

[0150] In the iterative process, determine the optimal position p of the individual best :Each particle individual records the individual optimal position it has reached in history. If the current individual fitness value is less than the individual optimal position, the individual optimal position is updated to the current position. Individual optimal position p best It is the best position that the individual particle can reach during the search process;

[0151] The optimal position of the group is obtained as follows:

[0152] At each iteration, the individual fitness value of each individual is compared with its historical individual optimal position, and if it is better than the historical individual optimal position, the individual optimal position is updated;

[0153] Determine whether the current individual optimal position of each individual is better than the group optimal position. If so, update the group optimal position to the individual optimal position.

[0154] In short, the individual optimal position is the optimal position of each particle itself, while the group optimal position of the particle swarm is the best position historically achieved by the entire particle swarm.

[0155] These positions are updated continuously during the algorithm iteration process to gradually optimize the search effect of the particle swarm.

[0156] Dynamically adjust the inertia weight of the particle,

[0157] The dynamic adjustment of the individual inertia weight is as shown in formula (3):

[0158]

[0159] Among them, w is the inertia weight of the individual, m is the current number of iterations, M is the maximum number of iterations, and w max 、w min are the individual's maximum and minimum inertia weights, σ ​​is the inertia adjustment factor, and p and q are random numbers.

[0160] The first item w min and the second term (w max -w min ) through the exponential function The inertia weight is large in the early stage, and decreases nonlinearly with the increase of iteration times. The third term σ*betarnd(p,q) uses beta distribution to adjust the overall value distribution of w, and adds the inertia adjustment factor σ to control the deviation of the inertia weight, making the dynamic adjustment of the inertia weight w more reasonable.

[0161] The betarnd(p,q) function generates random numbers that follow a beta distribution with two random parameters p and q.

[0162] The inertia adjustment factor σ is a value between 0 and 1, which is used to control the degree of adjustment of the inertia weight. The specific value of the inertia adjustment factor can be adjusted and optimized according to specific problems and experimental experience.

[0163] For example, when the inertia adjustment factor is set to 0.5, the change of the inertia weight will be relatively balanced and will not tend to decrease too quickly or too slowly.

[0164] Update particle velocity and particle position.

[0165] The individual speed is updated based on the individual inertia weight, as shown in formula (4):

[0166] v t+1 =wv t +c1r1(p best -x t )+c2r2(g best -x t ) Formula (4)

[0167] Among them, v t+1 is the speed of the individual at time t+1, v t is the speed of the individual at time t, r1 and r2 are random numbers in the range of 0-1, c1 is the individual learning factor, c2 is the global learning factor, and x t is the individual’s current position, pbest is the optimal position of the individual, g best The optimal position for the group.

[0168] The velocity of the particle at time t+1 is composed of the velocity inertia of the particle at time t, the self-cognition part velocity and the global cognition part velocity.

[0169] The self-perception part of the speed makes the particle move toward the individual historical optimal position, that is, the best position that the particle believes it has obtained in the past search;

[0170] The global recognition of the partial velocity enables particles to move toward the historical optimal position of the group.

[0171] When updating the speed, particles will take into account their own movement speed inertia, individual historical optimal position and group historical optimal position, so that particles can explore and utilize information more comprehensively in the search space.

[0172] Update the particle position as shown in formula (5):

[0173] x t+1 =x t +v t+1 *1=x t +v t+1 Formula (5)

[0174] Among them, x t+1 is the position of the individual at time t+1, x t is the position of the individual at time t, v t+1 is the velocity of the individual at time t+1.

[0175] The position of the particle at time t+1 is equal to the sum of the position of the particle at time t and the velocity of the particle at time t+1*the movement time, where the movement time is 1s.

[0176] During the particle swarm iteration, it is determined whether the individual position exceeds a preset value range, and if so, the individual position is randomized.

[0177] Determine whether the particle position exceeds the value range [L min , L max ], if exceeded, the individual particle positions will be randomized.

[0178] The updated individual particle position is checked to determine whether it exceeds the preset particle position value range. If it exceeds the range, the particle position is adjusted and the individual particle position is regenerated by randomization. That is, a random position is regenerated for the particle within the specified value range.

[0179] Ensure that the position of the particles is always within a reasonable range, thus ensuring the stability and effectiveness of the algorithm.

[0180] Track and record the best state of particles in the search process, including individual optimal position and group optimal position. By recording the historical optimal position, they are compared in subsequent iterations of the improved particle swarm algorithm to timely update the individual optimal position and group optimal position, thereby guiding the particles to search in a more promising direction.

[0181] The improved particle swarm algorithm updates the individual optimal position p of each particle in the iterative process best and the optimal position of the group g best .

[0182] Traverse the individual optimal position of each individual particle, find the optimal fitness value of the group in the particle group, update the optimal position of the group to the position of the particle with the optimal fitness value, and pass the optimal position of the group back to the RBF neural network model as the value of the key parameter spread of the RBF neural network model.

[0183] Exemplarily, the maximum number of iterations of the improved particle swarm algorithm is preset to 800 times.

[0184] The maximum number of iterations may vary in actual applications. The specific value needs to be determined based on the complexity of the problem, the availability of computing resources, and the requirements for the accuracy of the results. The maximum number of iterations can usually be set based on the specific problem and computing resources. The more iterations, the larger the space the algorithm searches, but it also increases the computational cost. The specific value needs to be adjusted based on the complexity of the problem and the limitations of computing resources to minimize the computational time while ensuring the search effect.

[0185] The optimal position of the group is obtained and passed to the RBF neural network model to update the expansion speed parameters, and the RBF neural network model after the updated parameters is trained using the training set in the sample data set; the test set in the sample data set is used for verification, and if the verification fails, the step of optimizing the expansion speed parameters of the RBF neural network model using the improved particle swarm algorithm is returned until the verification passes, and the trained RBF neural network model is obtained as the temperature sensor compensation network;

[0186] Improve the optimal position of the particle swarm algorithm, send it back to the RBF neural network model, and use this value to update the expansion speed parameter spread;

[0187] The training set in the sample data set is used to train the RBF neural network model after updating the parameters, and then the test set is used for verification. If the verification fails, the improved particle swarm algorithm is used to optimize the expansion speed parameters until the verification passes.

[0188] The trained RBF neural network model is used as the temperature sensor compensation network.

[0189] like Figure 4 As shown, based on the resistance value of the traction motor temperature sensor, the motor temperature value after the traction motor temperature sensor is compensated is obtained through the traction motor temperature sensor compensation network.

[0190] Based on the standard resistance value of the traction motor temperature sensor, the traction motor temperature sensor compensation network after updating the spread speed parameter spread value predicts or estimates the actual motor temperature of the traction motor where the high-speed train traction motor temperature sensor is located by learning the relationship between the resistance value and the actual temperature.

[0191] During the training process, the known standard resistance value and its corresponding standard temperature are used as input-output to train the traction motor temperature sensor compensation network, so that the traction motor temperature sensor compensation network can learn the mapping relationship between the resistance value and the actual temperature.

[0192] Once trained, the compensation network can use the learned relationship to predict or estimate the corresponding compensated motor temperature value when the traction motor temperature sensor is given a new resistance value.

[0193] In this way, the traction motor temperature sensor compensation network can be used to obtain the motor temperature value after the traction motor temperature sensor is compensated, that is, the current motor temperature of the high-speed train traction motor where the traction motor temperature sensor is located.

[0194] Compared with the prior art, the present invention can achieve at least one of the following beneficial effects:

[0195] 1. Improve the accuracy of motor temperature control: Through the temperature compensation of RBF neural network, the temperature can be predicted and controlled more accurately, so that the temperature of the motor of the high-speed rail traction system can be kept within the set first temperature threshold range;

[0196] 2. Realize remote monitoring and adjustment: Using the visualization module, train operation and maintenance personnel or train staff can monitor the temperature of the motor of the high-speed train traction system anytime and anywhere, and perform remote duct cooling, thereby realizing remote intelligent management, avoiding manual inspections, and saving human resource costs;

[0197] 3. Improve the efficiency of motor temperature control: The application of improved particle swarm algorithm can find the optimal solution of temperature compensation parameters more quickly, thereby improving the efficiency and response speed of the motor temperature monitoring system of the high-speed rail traction system;

[0198] 4. Low false alarm rate: The nonlinear error temperature compensation of the current motor temperature of the motor monitored by the temperature sensor of the traction system motor is performed using the RBF neural network based on the improved particle swarm to obtain the compensated motor temperature value, and determine whether all compensated motor temperature values ​​are less than the first temperature threshold. If not, a temperature drop warning is issued and a duct cooling command is sent. It is determined whether all compensated motor temperature values ​​are less than the second temperature threshold. If not, a stop command is sent for parking and maintenance, which greatly reduces the problem of high false alarm rate.

[0199] Those skilled in the art will appreciate that all or part of the processes of the above-mentioned embodiments can be implemented by instructing related hardware through a computer program, and the program can be stored in a computer-readable storage medium, wherein the computer-readable storage medium is a disk, an optical disk, a read-only storage memory, or a random access memory, etc.

[0200] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by any technician familiar with the technical field within the technical scope disclosed by the present invention should be covered within the protection scope of the present invention.

Claims

1. A high-speed railway traction system motor temperature monitoring device based on RBF neural network, characterized in that: include: Multiple traction motor temperature sensors are distributed in each traction motor of the high-speed train, and are used to measure the motor temperature of each traction motor in real time, obtain the current motor temperature of the traction motor, and transmit it to the microprocessor; The microprocessor uses an improved particle swarm-based RBF neural network to perform nonlinear error temperature compensation on the current motor temperature of the traction motor monitored by the traction motor temperature sensor to obtain a compensated motor temperature value, and determines whether all compensated motor temperature values ​​are less than a first temperature threshold value. If not, a first warning instruction is sent to the warning module, and a speed reduction instruction is sent to the high-speed rail control center at the same time, and a control signal is sent to the traction motor blower corresponding to the traction motor whose compensated motor temperature value exceeds a preset temperature threshold value, so as to start the blower to cool the traction motor; if any one of all compensated motor temperature values ​​is greater than or equal to the second temperature threshold value, a stop instruction is sent; The second temperature threshold is greater than the first temperature threshold; The early warning module is used to issue a corresponding early warning based on the early warning instruction.

2. The device according to claim 1, characterized in that The system further comprises: The intelligent center module is used to analyze the stored real-time motor temperature data in real time. If it is found that the motor temperature continues to rise within a preset time period and the current motor temperature approaches the first temperature threshold, a second warning instruction is sent to the warning module; if a traction motor temperature sensor exceeds the normal operating temperature range, a third warning instruction is sent to the warning module; A visualization module, used to display the real-time motor data of each traction motor, and provide an editing interface for presetting the first and second temperature thresholds, the preset time period, the normal operating temperature range, and the first, second, and third warning instruction templates; A voice prompt module is used for the warning voice prompt of the warning module.

3. The device according to claim 1, characterized in that The traction motor temperature sensor transmits the current motor temperature of the traction motor collected in real time to the microprocessor via a built-in Wi-Fi module.

4. The device according to claim 2, characterized in that The intelligent central module also pre-stores the indexing table data of the traction motor temperature sensor, and the indexing table data serves as a sample data set.

5. The device according to claim 4, characterized in that The method of using the improved particle swarm-based RBF neural network to perform nonlinear error temperature compensation on the current motor temperature of the high-speed train monitored by each traction motor temperature sensor to obtain the compensated motor temperature value includes: Using an improved particle swarm algorithm to optimize the expansion speed parameters of the RBF neural network model, obtaining the optimal position of the group and passing it to the RBF neural network model to update the expansion speed parameters, and using the training set in the sample data set to train the RBF neural network model after the updated parameters; Using the test set in the sample data set to verify, if the verification fails, returning to the step of optimizing the expansion speed parameters of the RBF neural network model using the improved particle swarm algorithm until the verification passes, and obtaining the trained RBF neural network model as the traction motor temperature sensor compensation network; Based on the traction motor temperature sensor compensation network, the temperature signal measured by each traction motor temperature sensor is compensated to obtain a compensated motor temperature value.

6. The device according to claim 5, characterized in that The optimization of the expansion speed parameters of the RBF neural network model based on the improved particle swarm algorithm training includes: Initialize the parameters in the improved particle swarm algorithm; Initialize the initial position and initial velocity of individuals in the population; In each iteration, the individual fitness value is calculated based on the individual position and the sample data set using a 10-fold cross validation method; the inertia weight of the individual is dynamically adjusted; updating the individual velocity based on the individual inertia weight, and then updating the individual position; When the maximum number of iterations is reached, the optimal position of the group is returned, and the expansion speed parameters of the RBF neural network model are optimized and updated based on the optimal position of the group.

7. The device according to claim 6, characterized in that The calculation of individual fitness values ​​based on individual positions using a 10-fold cross validation method includes: The individual position is input into the model as the expansion speed parameter of the RBF neural network model; the sample data set is randomly divided into 10 sample subsets; In each cross-validation fitting, 9 sample subsets are used as training subsets and the remaining 1 sample subset is used as a validation subset. The RBF neural network is trained using the 9 training subsets, and then the sum of the absolute errors between the temperature fitting value obtained after the validation subset is input into the RBF neural network and the corresponding standard temperature value is calculated; Perform 10 cross validations, and add up the sums of the 10 absolute errors obtained to obtain the average value as the individual fitness value.

8. The device according to claim 7, characterized in that The optimal position of the group is obtained as follows: At each iteration, the individual fitness value of each individual is compared with its historical individual optimal position, and if it is better than the historical individual optimal position, the individual optimal position is updated; Determine whether the current individual optimal position of each individual is better than the group optimal position, and if so, update the group optimal position to the individual optimal position; During the particle swarm iteration, it is determined whether the individual position exceeds a preset value range, and if so, the individual position is randomized.

9. The device according to claim 8, characterized in that The inertia weight of the individual is dynamically adjusted as follows: Among them, w is the inertia weight of the individual, m is the current number of iterations, M is the maximum number of iterations, and w max 、w min are the individual's maximum and minimum inertia weights, σ ​​is the inertia adjustment factor, and p and q are random numbers.

10. The device according to claim 9, characterized in that Update the individual velocity based on the individual inertia weight as follows: v t+1 =wv t +c1r1(p best -x t )+c2r2(g best -x t ) Among them, v t+1 is the speed of the individual at time t+1, v t is the speed of the individual at time t, r1 and r2 are random numbers in the range of 0-1, c1 is the individual learning factor, c2 is the global learning factor, and x t is the individual’s current position, p best is the optimal position of the individual, g best is the optimal position for the group; Update the particle position as follows: x t+1 =x t +v t+1 *1=x t +v t+1 Among them, x t+1 is the position of the individual at time t+1, x t is the position of the individual at time t, v t+1 is the speed of the individual at time t+1.

Images