Temperature monitoring system of high-speed rail traction water cooling system based on RBF neural network

Abstract

The invention relates to a temperature monitoring system of a high-speed rail traction water cooling system based on an RBF neural network, belongs to the field of IoT and intelligent sensors, and solves the problems of low sensor precision and low inspection efficiency. Comprising a plurality of water-cooling system temperature sensors which are installed at water outlets of traction converter water-cooling systems of a high-speed rail traction system and used for monitoring the temperatures of the water-cooling systems in real time and obtaining the current temperatures of the water-cooling systems; the processor is used for performing nonlinear error temperature compensation on the current temperature of each water cooling system monitored by the temperature sensor by using an RBF neural network based on an improved particle swarm to obtain compensated temperature values, judging whether all the compensated temperature values are within a preset temperature range or not, and if not, sending an early warning instruction to the early warning module; and the early warning module is used for performing corresponding early warning based on the early warning instruction. The temperature of the water cooling system is accurately and efficiently monitored.

Description

Technical Field

[0001] The present invention belongs to the technical field of IoT (Internet of Things) and intelligent sensors, and in particular to a temperature monitoring system for a high-speed rail traction water cooling system based on an RBF (Radial Basis Function) network. Background Art

[0002] With the development of intelligent sensing systems and intelligent sensor technology, intelligent sensing units have been widely used, including but not limited to high-speed rail operation, industrial production, logistics and transportation, environmental monitoring, medical diagnosis and other business fields. With the continuous development of high-speed rail technology, the safe operation and performance stability of high-speed rail traction systems have received more and more attention. As one of the key components of the traction system, the stable operation of the water cooling system is crucial to ensure the normal operation of the entire traction system. The temperature of the water cooling system is one of the important parameters that directly affects its performance and reliability.

[0003] At present, the temperature monitoring of the water cooling system in the high-speed rail traction system mainly relies on traditional temperature sensors, but this monitoring method has some defects. Traditional temperature monitoring systems can only provide local temperature information and cannot fully and accurately reflect the working status of the water cooling system; and the sensor's measurement data is often affected by environmental conditions, equipment aging and other factors, resulting in instability and inaccuracy in the monitoring results. Therefore, there is an urgent need for an intelligent sensor that can improve the temperature measurement accuracy.

[0004] In summary, the present application provides a temperature monitoring system for a high-speed rail traction water cooling system based on an RBF neural network, which is used to solve the technical problems of low accuracy, low warning and high false alarm rate of existing water cooling system temperature sensors. Summary of the invention

[0005] In view of the above analysis, an embodiment of the present invention aims to provide a temperature monitoring system for a high-speed rail traction water cooling system based on an RBF neural network, so as to solve the technical problems of low accuracy, low warning and high false alarm rate of the temperature sensor of the water cooling system.

[0006] The embodiment of the present invention provides a temperature monitoring system for a high-speed railway traction water cooling system based on an RBF neural network, comprising:

[0007] Multiple water cooling system temperature sensors are installed at the water outlet of the water cooling system of each traction converter of the high-speed railway traction system to monitor the temperature of the water cooling system in real time and obtain the current temperature of the water cooling system;

[0008] The processor uses an improved particle swarm-based RBF neural network to perform nonlinear error temperature compensation on the current temperature of each water cooling system monitored by the temperature sensor to obtain a compensated temperature value, and determines whether all the compensated temperature values ​​are within a preset temperature range. If not, a warning instruction is sent to the warning module;

[0009] The early warning module is used to issue a corresponding early warning based on the early warning instruction.

[0010] Furthermore, the system further comprises:

[0011] A data storage module, used to store the temperature data and system operation log of the water cooling system, and the indexing table data of the pre-stored temperature sensor;

[0012] User visualization module, used to display the real-time temperature data of the water cooling system and provide an editing interface for preset temperature ranges.

[0013] Furthermore, the water cooling system temperature sensor transmits the current temperature of the water cooling system collected in real time to the processor through a built-in Wi-Fi module.

[0014] Furthermore, the nonlinear error temperature compensation of the current temperature of the water cooling system using the improved particle swarm-based RBF neural network to obtain the compensated temperature value includes:

[0015] 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;

[0016] 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 water cooling system temperature sensor compensation network;

[0017] Based on the water cooling system temperature sensor compensation network, the temperature signals measured by each water cooling system temperature sensor are compensated to obtain a compensated temperature value.

[0018] Furthermore, the optimization of the expansion speed parameters of the RBF neural network model based on the improved particle swarm algorithm training includes:

[0019] Initialize the parameters in the improved particle swarm algorithm;

[0020] Initialize the initial position and initial velocity of individuals in the population;

[0021] 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;

[0022] updating the individual velocity based on the individual inertia weight, and then updating the individual position;

[0023] 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.

[0024] Further, the calculating of individual fitness values ​​based on individual positions by using a 10-fold cross validation method includes:

[0025] 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;

[0026] 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;

[0027] 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.

[0028] Furthermore, the optimal position of the group is obtained in the following way:

[0029] 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;

[0030] 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;

[0031] 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.

[0032] Furthermore, the inertia weight of the individual is dynamically adjusted as follows:

[0033]

[0034] 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 minare the individual's maximum and minimum inertia weights, σ ​​is the inertia adjustment factor, and p and q are random numbers.

[0035] Further, the individual speed is updated based on the inertia weight of the individual, as follows:

[0036] v t+1 =wv t +c1r1(p best -x t )+c2r2(g best -x t )

[0037] 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;

[0038] Further, the particle position is updated as follows:

[0039] x t+1 =x t +v t+1 *1=x t +v t+1

[0040] 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. Compared with the prior art, the present invention can achieve at least one of the following beneficial effects:

[0041] 1. Improve the accuracy of water cooling system temperature monitoring: Through the temperature compensation of RBF neural network, the temperature of the water cooling system can be predicted more accurately; as a powerful nonlinear model, RBF neural network has strong approximation and generalization capabilities, and can better handle complex temperature monitoring problems. By using the improved particle swarm algorithm to train and optimize the RBF neural network, the system's perception of temperature changes and prediction accuracy can be improved, thereby achieving more accurate and reliable monitoring of the water cooling system temperature. The temperature of the water cooling system can be maintained within the preset temperature range to ensure the safe operation of high-speed trains;

[0042] 2. Improve the temperature monitoring efficiency of the water cooling system: The application of the improved particle swarm algorithm can find the optimal solution of the temperature compensation parameters more quickly, thereby improving the efficiency and response speed of the temperature monitoring system;

[0043] 3. Low false alarm rate: The RBF neural network based on the improved particle swarm is used to perform nonlinear error temperature compensation on the current temperature of the water cooling system monitored by each temperature sensor to obtain the compensated temperature value, and to determine whether all compensated temperature values ​​are within the preset temperature range for early warning, which greatly improves the accuracy and reliability of the early warning, effectively reduces the false alarm rate, and ensures the system's timely discovery and accurate judgment of temperature anomalies, thereby more effectively ensuring the safe and stable operation of the high-speed rail system.

[0044] 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

[0045] 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.

[0046] Figure 1 A schematic diagram of a temperature monitoring system for a high-speed rail traction water cooling system based on an RBF neural network;

[0047] Figure 2 Schematic diagram of RBF temperature compensation method for improving particle swarm;

[0048] Figure 3 It is the structural diagram of RBF neural network model;

[0049] Figure 4 This is the principle diagram of nonlinear temperature error compensation. DETAILED DESCRIPTION

[0050] 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.

[0051] like Figure 1 As shown, a specific embodiment of the present invention discloses a temperature monitoring system for a high-speed railway traction water cooling system based on an RBF neural network, the system comprising:

[0052] Multiple water cooling system temperature sensors are installed at the water outlet of the water cooling system of each traction converter of the high-speed railway traction system to monitor the temperature of the water cooling system in real time and obtain the current temperature of the water cooling system;

[0053] The processor uses an improved particle swarm-based RBF neural network to perform nonlinear error temperature compensation on the current temperature of each water cooling system monitored by the temperature sensor to obtain a compensated temperature value, and determines whether all the compensated temperature values ​​are within a preset temperature range. If not, a warning instruction is sent to the warning module;

[0054] The early warning module is used to issue a corresponding early warning based on the early warning instruction.

[0055] During implementation, the water cooling system temperature sensor is attached to the water outlet of the water cooling system of the traction inverter of the high-speed railway traction system using thermal conductive silicone grease.

[0056] 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 temperature value result of the temperature compensation of the RBF neural network.

[0057] For the preset temperature range, illustratively, it is preset to [5°C, 40°C], which can be set immediately according to actual business needs.

[0058] For early warning, according to actual business needs, for example, multiple levels of early warning can be set.

[0059] The system further comprises:

[0060] A data storage module, used to store the temperature data and system operation log of the water cooling system, and the indexing table data of the pre-stored temperature sensor;

[0061] User visualization module, used to display the real-time temperature data of the water cooling system and provide an editing interface for preset temperature ranges.

[0062] The water cooling system of the traction converter has a key impact on the stable operation and safety of the traction system.

[0063] The water cooling system temperature sensor transmits the current temperature of the water cooling system collected in real time to the processor through a built-in Wi-Fi module.

[0064] The traction motor is a very important component of the high-speed rail traction system. When the train is running, the traction motor converts electrical energy into mechanical energy to drive the train; when the train brakes, the traction motor recovers kinetic energy and regenerates it. Therefore, the safety monitoring of the traction motor is particularly important. When various faults occur in the traction motor, the water temperature and water pressure of the water cooling system of the traction converter in the traction motor will change significantly. The main structure of the traction motor consists of a rotor, bearings, windings and other components. When any component has problems, such as rotor wear, bearing fatigue, insufficient lubrication, etc., it is easy to cause the motor temperature to rise, and then cause the water temperature and water pressure of the water cooling system to change rapidly. These faults can reduce the service life of the traction motor at the least, or cause major safety accidents at the worst. The water cooling system of the traction converter has a key impact on the stable operation and safety of the traction system.

[0065] The water cooling system is a high-speed rail undercarriage equipment, and each carriage is different from the other. For example, the undercarriage equipment of Car No. 1, Car No. 3, Car No. 5, Car No. 7, and Car No. 8 is equipped with a water cooling system water tank.

[0066] The processor obtains the temperature sensor graduation table data from the data storage module as a sample data set;

[0067] The samples in the sample data set include the standard resistance value of the temperature sensor element and the corresponding standard temperature value;

[0068] The standard resistance value is used as an input variable of the RBF neural network model.

[0069] The temperature sensor's graduation table data includes the standard resistance value of the temperature sensor element and the corresponding standard temperature value, and the two are in a one-to-one correspondence;

[0070] Obtain the standard resistance value and the corresponding standard temperature value from the temperature sensor's graduation table as sample data.

[0071] Exemplarily, the sample data set includes more than 10,000 pieces of sample data, and each sample includes a standard resistance value and a corresponding standard temperature value of a temperature sensor element.

[0072] 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.

[0073] Exemplarily, the ratio of the sample size of the training set to the sample size of the test set is 8:2.

[0074] like Figure 2 As shown, the nonlinear error temperature compensation of the current temperature of the water cooling system using the RBF neural network based on the improved particle swarm to obtain the compensated temperature value includes:

[0075] 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;

[0076] 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 water cooling system temperature sensor compensation network;

[0077] Based on the water cooling system temperature sensor compensation network, the temperature signals measured by each water cooling system temperature sensor are compensated to obtain a compensated temperature value.

[0078] The compensated temperature value is obtained as a compensation value of the current temperature of the water cooling system where the temperature sensor is located.

[0079] Initialize the RBF neural network model;

[0080] 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.

[0081] The standard resistance value of the 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 temperature value. The RBF neural network model is built based on this.

[0082] like Figure 3 As shown in , the input layer input data x is 1 input, which is the standard resistance value of the temperature sensor; the output layer output y is the compensated temperature value.

[0083] The nonlinear temperature error of the 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.

[0084] like Figure 4 As shown, the temperature compensation principle is g(R x )=f -1 (t x ).

[0085] f(t x ) is the input-output function of the temperature sensor, i.e., the actual measured output of the temperature sensor, g(R xi ) is the temperature sensor compensation network function, and the output is the compensated temperature value, t xis the actual temperature of the water cooling system being tested, that is, the standard temperature value in the graduation table, t y is the temperature value after compensation.

[0086] The purpose of the temperature sensor compensation network is to ensure that the compensated temperature value is equal to the actual temperature value of the water cooling system being measured, satisfying the requirements of formula (1):

[0087] t y =g(R x )R x =g(R x )f(t x )t x =t x Formula (1)

[0088] Right now

[0089] g(R x )=f -1 (t x ) Formula (2)

[0090] Among them, R x is the resistance value of the temperature sensor of the sample, t x is the actual temperature of the measured environment.

[0091] The purpose of the temperature sensor compensation network is also the predetermined goal of the training of the RBF neural network model.

[0092] What is known is the standard resistance value under the standard temperature value. Now we need to get the temperature value based on the standard resistance value, which is the relationship between the function and the inverse function.

[0093] 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 temperature sensor.

[0094] 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;

[0095] 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.

[0096] 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;

[0097] 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.

[0098] 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.

[0099] The method of optimizing the expansion speed parameters of the RBF neural network model by using the improved particle swarm algorithm includes:

[0100] Initialize the parameters in the improved particle swarm algorithm;

[0101] Initialize the initial position and initial velocity of individuals in the population;

[0102] 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;

[0103] updating the individual velocity based on the individual inertia weight, and then updating the individual position;

[0104] 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.

[0105] Initialize the parameters in the improved particle swarm algorithm; initialize the initial position and initial velocity of individuals in the population;

[0106] Initialization parameters are as follows:

[0107] Particle individual learning factor c1, global learning factor c2;

[0108] The number of individuals in the particle group (total number of particles) N, that is, the group size;

[0109] Particle dimension Dim;

[0110] The number of algorithm iterations M;

[0111] The maximum value of the particle individual position L max , minimum value L min ;

[0112] 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;

[0113] 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.

[0114] Among them, the particle dimension parameter of the improved particle swarm algorithm is the number of input variables of the RBF neural network model.

[0115] The input of the RBF neural network model is the standard resistance value of the temperature sensor element, 1 variable, so the particle dimension Dim = 1;

[0116] The number of algorithm iterations M is the preset maximum number of iterations.

[0117] The input variable of the RBF neural network model is the standard resistance value of the temperature sensor element, which ensures that the optimal position of the group output by the improved particle swarm algorithm is a value.

[0118] 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;

[0119] 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.

[0120] Exemplarily, c1=c2=2.

[0121] 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.

[0122] 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.

[0123] For example, w max The default value is 0.9; w min The default value is 0.4.

[0124] Initialize the initial positions and initial velocities of individuals in the population;

[0125] The individual initial position and individual initial velocity are randomly generated to initialize the position and velocity of each individual (ie particle).

[0126] 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.

[0127] 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;

[0128] The calculation of individual fitness values ​​based on individual positions using a 10-fold cross validation method includes:

[0129] 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;

[0130] 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;

[0131] 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.

[0132] Based on the individual fitness value, the individual optimal position p is obtained best and the optimal position of the group g best .

[0133] 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 .

[0134] 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;

[0135] The optimal position of the group is obtained as follows:

[0136] 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;

[0137] 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.

[0138] 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.

[0139] These positions are updated continuously during the algorithm iteration process to gradually optimize the search effect of the particle swarm.

[0140] Dynamically adjust the inertia weight of the particle,

[0141] The dynamic adjustment of the individual inertia weight is as shown in formula (3):

[0142]

[0143] 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.

[0144] 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.

[0145] The betarnd(p,q) function generates random numbers that follow a beta distribution with two random parameters p and q.

[0146] 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.

[0147] 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.

[0148] Update particle velocity and particle position.

[0149] The individual speed is updated based on the individual inertia weight, as shown in formula (4):

[0150] v t+1 =wv t +c1r1(p best -x t )+c2r2(g best -x t ) Formula (4)

[0151] 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 The optimal position for the group.

[0152] 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.

[0153] 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;

[0154] The global recognition of the partial velocity enables particles to move toward the historical optimal position of the group.

[0155] 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.

[0156] Update the particle position as shown in formula (5):

[0157] x t+1 =x t +v t+1 *1=x t +v t+1 Formula (5)

[0158] 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.

[0159] 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.

[0160] 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.

[0161] Determine whether the particle position exceeds the value range [L min , L max ], if exceeded, the individual particle positions will be randomized.

[0162] 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.

[0163] Ensure that the position of the particles is always within a reasonable range, thus ensuring the stability and effectiveness of the algorithm.

[0164] 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.

[0165] The improved particle swarm algorithm updates the individual optimal position p of each particle in the iterative processbest and the optimal position of the group g best .

[0166] 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.

[0167] Exemplarily, the maximum number of iterations of the improved particle swarm algorithm is preset to 500 times.

[0168] 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.

[0169] 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;

[0170] 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;

[0171] 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.

[0172] The trained RBF neural network model is used as the compensation network for the temperature sensor of the water cooling system.

[0173] like Figure 4 As shown, based on the resistance value of the temperature sensor, the temperature value of the water cooling system temperature sensor after compensation is obtained through the water cooling system temperature sensor compensation network.

[0174] Based on the standard resistance value of the temperature sensor, the temperature sensor compensation network after updating the spread speed parameter spread value predicts or estimates the actual temperature of the water cooling system where the water cooling system temperature sensor is located by learning the relationship between the resistance value and the actual temperature.

[0175] During the training process, the known standard resistance value and its corresponding standard temperature are used as input-output to train the temperature sensor compensation network, so that the temperature sensor compensation network can learn the mapping relationship between the resistance value and the actual temperature.

[0176] Once trained, when the temperature sensor is given a new resistance value, the compensation network can use the learned relationship to predict or estimate the corresponding compensated temperature value.

[0177] In this way, the water cooling system temperature sensor compensation network can be used to obtain the compensated temperature value of the water cooling system temperature sensor, that is, the current temperature of the water cooling system where the temperature sensor is located.

[0178] Compared with the prior art, the present invention can achieve at least one of the following beneficial effects:

[0179] 1. Improve the accuracy of water cooling system temperature monitoring: Through the temperature compensation of RBF neural network, the temperature of the water cooling system can be predicted more accurately; as a powerful nonlinear model, RBF neural network has strong approximation and generalization capabilities, and can better handle complex temperature monitoring problems. By using the improved particle swarm algorithm to train and optimize the RBF neural network, the system's perception of temperature changes and prediction accuracy can be improved, thereby achieving more accurate and reliable monitoring of the water cooling system temperature. The temperature of the water cooling system can be maintained within the preset temperature range to ensure the safe operation of high-speed trains;

[0180] 2. Improve the temperature monitoring efficiency of the water cooling system: The application of the improved particle swarm algorithm can find the optimal solution of the temperature compensation parameters more quickly, thereby improving the efficiency and response speed of the temperature monitoring system;

[0181] 3. Low false alarm rate: The RBF neural network based on the improved particle swarm is used to perform nonlinear error temperature compensation on the current temperature of the water cooling system monitored by each temperature sensor to obtain the compensated temperature value, and to determine whether all compensated temperature values ​​are within the preset temperature range for early warning, which greatly improves the accuracy and reliability of the early warning, effectively reduces the false alarm rate, and ensures the system's timely discovery and accurate judgment of temperature anomalies, thereby more effectively ensuring the safe and stable operation of the high-speed rail system.

[0182] 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.

[0183] 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 temperature monitoring system for high-speed rail traction water cooling system based on RBF neural network, characterized in that: The system comprises: Multiple water cooling system temperature sensors are installed at the water outlet of the water cooling system of each traction converter of the high-speed railway traction system to monitor the temperature of the water cooling system in real time and obtain the current temperature of the water cooling system; The processor uses an improved particle swarm-based RBF neural network to perform nonlinear error temperature compensation on the current temperature of each water cooling system monitored by the temperature sensor to obtain a compensated temperature value, and determines whether all the compensated temperature values ​​are within a preset temperature range. If not, a warning instruction is sent to the warning module; The early warning module is used to issue a corresponding early warning based on the early warning instruction.

2. The system according to claim 1, characterized in that The system further comprises: A data storage module, used to store the temperature data and system operation log of the water cooling system, and the indexing table data of the pre-stored temperature sensor; User visualization module, used to display the real-time temperature data of the water cooling system and provide an editing interface for preset temperature ranges.

3. The system according to claim 2, characterized in that The water cooling system temperature sensor transmits the current temperature of the water cooling system collected in real time to the processor through a built-in Wi-Fi module.

4. The system according to claim 3, characterized in that The method of using the improved particle swarm-based RBF neural network to perform nonlinear error temperature compensation on the current temperature of the water cooling system to obtain a compensated 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 water cooling system temperature sensor compensation network; Based on the water cooling system temperature sensor compensation network, the temperature signals measured by each water cooling system temperature sensor are compensated to obtain a compensated temperature value.

5. The system according to claim 4, 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.

6. The system according to claim 5, 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.

7. The system according to claim 6, 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.

8. The system according to claim 7, 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.

9. The method according to claim 8, 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 The optimal position for the group.

10. The system according to claim 9, characterized in that 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 velocity of the individual at time t+1.

Images