Control method and apparatus for performance optimization of levitation guidance system, and electronic device

Abstract

The present application discloses a control method and apparatus for performance optimization of a levitation guidance system, and an electronic device. The method comprises: obtaining a plurality of performance indicators and a plurality of system variables of a levitation guidance system; on the basis of the system variables, constructing a multi-objective optimization model regarding each of the performance indicators; and using the multi-objective optimization model to process a sample population, so as to obtain at least one optimal solution, the optimal solution comprising variable values respectively corresponding to each of the system variables, and the variable values in the optimal solution enabling indicator values corresponding to the performance indicators to meet an optimization condition. The sample population comprises a plurality of sample groups, and each of the sample groups comprises sample values respectively corresponding to each of the system variables.

Description

Control method, device and electronic equipment for optimizing performance of suspension guidance system

[0001] This application claims priority to the Chinese patent application filed with the China Patent Office on December 11, 2023, with application number 202311696659.0 and invention name “Control method, device and electronic equipment for performance optimization of suspension guidance system”, the entire contents of which are incorporated by reference into this application. Technical Field

[0002] The present application relates to the field of intelligent customer service technology, and in particular to a control method, device, and electronic equipment for optimizing the performance of a suspension guidance system. Background Art

[0003] The suspension and guidance system of the superconducting electric levitation train includes multiple components such as the suspension frame, superconducting magnets, and ground suspension and guidance coils.

[0004] The performance of a suspension guidance system is dependent on multiple performance parameters of each component. Traditional design methods rely on the experience of each component's designers, requiring multiple rounds of iterative optimization by multiple designers across multiple components to achieve a high-performance suspension guidance system. This results in excessive human resources being consumed in optimizing the performance of the suspension guidance system, resulting in excessively time-consuming optimization and low performance optimization efficiency.

[0005] Therefore, there is an urgent need for a technical solution that can efficiently optimize the performance of the suspension guidance system.

[0006] Summary of the Invention

[0007] In view of this, the present application provides a control method, device, and electronic device for optimizing the performance of a suspension guidance system, which are used to solve the technical problem of low efficiency in optimizing the performance of a suspension guidance system in the prior art, as follows:

[0008] A control method for optimizing the performance of a suspension guidance system, the method comprising:

[0009] Obtaining multiple performance indicators and multiple system variables of the suspension guidance system;

[0010] Constructing a multi-objective optimization model for each of the performance indicators based on the system variables;

[0011] Processing the sample population using the multi-objective optimization model to obtain at least one optimal solution, wherein the optimal solution includes variable values ​​corresponding to each of the system variables, and the variable values ​​in the optimal solution make the indicator value corresponding to the performance indicator meet the optimization condition;

[0012] The sample population includes a plurality of sample groups, and each of the sample groups includes a sample value corresponding to each of the system variables.

[0013] The above method preferably constructs a multi-objective optimization model for each of the performance indicators based on the system variables, including:

[0014] Among the plurality of system variables, at least one target variable is determined for each performance indicator, and an association condition is satisfied between the target variable and its corresponding performance indicator;

[0015] A multi-objective optimization model is constructed based on the target variables corresponding to each of the performance indicators. The multi-objective optimization model takes the indicator values ​​corresponding to the performance indicators satisfying the optimization conditions as the optimization objectives.

[0016] In the above method, preferably, the target variable and its corresponding performance indicator satisfy the association conditions, including:

[0017] The sensitivity index value between the target variable and the performance index is ranked in the top N among all the system variables, where N is a positive integer greater than or equal to 1;

[0018] The value of the sensitivity index between the system variable and the performance index represents the degree of influence of the change of the system variable on the performance index.

[0019] In the above method, preferably, the sensitivity index value between the target variable and the performance index is determined based on at least a change state of the performance index caused by a change state of the system variable.

[0020] In the above method, preferably, the indicator value corresponding to the performance indicator satisfies the optimization condition and at least includes:

[0021] The index value corresponding to the performance index makes the target value output by the multi-objective optimization model the minimum among all candidate values;

[0022] The candidate value is: after the sample value in the sample group is substituted into the corresponding system variable of the suspension guidance system, the index value corresponding to each performance index of the suspension guidance system makes the data value output by the multi-objective optimization model.

[0023] The above method preferably uses the multi-objective optimization model to process the sample population to obtain at least one optimal solution, including:

[0024] Using the multi-objective optimization model and based on a multi-objective optimization algorithm, processing the sample groups in the sample population to obtain an optimal solution that makes the indicator value corresponding to the performance indicator meet the optimization condition;

[0025] The optimal solution comes from the sample population or the optimal solution comes from a sample group expanded from the sample population, and the multi-objective optimization algorithm is a multi-objective evolutionary algorithm with adaptive constraints based on collaborative evolution.

[0026] Preferably, the above method further comprises, before processing the sample population using the multi-objective optimization model to obtain at least one optimal solution:

[0027] Normalization processing is performed on the sample values ​​contained in each of the sample groups in the sample population.

[0028] In the above method, preferably, the performance indicators include any number of the following:

[0029] Vehicle lateral fluctuation, vehicle vertical fluctuation, vehicle stability index, vehicle vertical vibration acceleration, vehicle lateral vibration acceleration;

[0030] The system variables include any number of the following: electromagnetic parameters of the vehicle-mounted magnet, shape parameters of the vehicle-mounted magnet, electromagnetic parameters of the ground suspension guide coil, shape parameters of the ground suspension guide coil, vehicle primary suspension parameters, and vehicle secondary suspension parameters;

[0031] Furthermore, the system variables correspond to variable constraints.

[0032] A control device for optimizing the performance of a suspension guidance system, the device comprising:

[0033] a data acquisition unit, configured to obtain a plurality of performance indicators and a plurality of system variables of the suspension guidance system;

[0034] A model building unit, configured to build a multi-objective optimization model for each of the performance indicators according to the system variables;

[0035] an optimal obtaining unit, configured to process the sample population using the multi-objective optimization model to obtain at least one optimal solution, wherein the optimal solution includes variable values ​​corresponding to each of the system variables, and the variable values ​​in the optimal solution make the indicator value corresponding to each of the performance indicators meet the optimization condition;

[0036] The sample population includes a plurality of sample groups, and each of the sample groups includes a sample value corresponding to each of the system variables.

[0037] An electronic device, comprising:

[0038] A memory for storing computer programs and data generated by the execution of the computer programs;

[0039] A processor is configured to execute the computer program to achieve the following: obtaining a plurality of performance indicators and a plurality of system variables of the suspension guidance system; constructing a multi-objective optimization model for each of the performance indicators based on the system variables; and processing a sample population using the multi-objective optimization model to obtain at least one optimal solution, wherein the optimal solution includes variable values ​​corresponding to each of the system variables, the variable values ​​in the optimal solution making the indicator value corresponding to each of the performance indicators satisfy an optimization condition; wherein the sample population includes a plurality of sample groups, each of the sample groups includes sample values ​​corresponding to each of the system variables.

[0040] It can be seen from the above technical solutions that in the control method, device and electronic equipment for optimizing the performance of a suspension guidance system disclosed in this application, after obtaining multiple performance indicators and multiple system variables of the suspension guidance system, a multi-objective optimization model for each performance indicator is constructed based on these system variables, and then the constructed multi-objective optimization model is used to process a sample population containing multiple sample groups, so that an optimal solution can be obtained. The variable value corresponding to each system variable in the optimal solution can make the indicator value corresponding to each performance indicator meet the optimization conditions, thereby achieving performance optimization of the suspension guidance system. It can be seen that in this application, there is no need for excessive designer participation. The optimal solution can be obtained by constructing a multi-objective optimization model, thereby saving human resources and not causing the optimization performance to take too long, thereby achieving efficient performance optimization control. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0042] FIG1 is a flow chart of a control method for optimizing the performance of a suspension guidance system provided in Example 1 of the present application;

[0043] FIG2 is a partial flow chart of a control method for optimizing the performance of a suspension guidance system provided in Example 1 of the present application;

[0044] FIG3 is a schematic structural diagram of a control device for optimizing the performance of a suspension guidance system provided in Example 2 of the present application;

[0045] FIG4 is a schematic structural diagram of an electronic device provided in Example 3 of the present application;

[0046] FIG5 is a flowchart of the performance optimization of the suspension guidance system in a scenario where the present application is applied to the performance optimization of a superconducting electric suspension train;

[0047] FIG6 is a schematic diagram of the reproduction process of the sample population by ACMCA in an embodiment of the present application. DETAILED DESCRIPTION

[0048] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0049] Referring to Figure 1, there is shown a flowchart illustrating a control method for optimizing the performance of a suspension guidance system, as provided in Example 1 of this application. This method can be applied to electronic devices capable of data processing, such as computers or servers. The technical solution in this embodiment is primarily used to improve the efficiency of optimizing the performance of a suspension guidance system.

[0050] Specifically, the method in this embodiment may include the following steps:

[0051] Step 101: Obtain multiple performance indicators and multiple system variables of the suspension guidance system.

[0052] The levitation and guidance system is the levitation and guidance system of the superconducting electric levitation train, and includes multiple components such as the levitation frame, superconducting magnets, and ground levitation and guidance coils. The performance indicators and system variables of the levitation and guidance system are related to the components in the levitation and guidance system.

[0053] Specifically, the performance indicators of the suspension guidance system include any number of the following:

[0054] Vehicle lateral fluctuation, vehicle vertical fluctuation, vehicle stability index, vehicle vertical vibration acceleration, vehicle lateral vibration acceleration.

[0055] Among them, the vehicle's lateral fluctuation is represented by X1, the vehicle's vertical fluctuation is represented by X2, the vehicle's stability index is represented by X3, the vehicle's vertical vibration acceleration is represented by X4, and the vehicle's lateral vibration acceleration is represented by X5.

[0056] The system variables of the suspension guidance system can include any of the following:

[0057] Electromagnetic parameters of vehicle-mounted magnets, external parameters of vehicle-mounted magnets, electromagnetic parameters of ground suspension guide coils, external parameters of ground suspension guide coils, primary suspension parameters of vehicles, and secondary suspension parameters of vehicles.

[0058] Moreover, the system variables of the suspension guidance system correspond to variable constraints.

[0059] For example, the electromagnetic parameters of the on-board magnet include at least the current and the number of turns, represented by Y1 and Y2, respectively. The external parameters of the on-board magnet include at least the thickness and length, represented by Y3 and Y4, respectively. The electromagnetic parameters of the ground suspension guide coil include at least the number of turns, represented by Y5. The external parameters of the ground suspension guide coil include at least the distance between the upper and lower coils and the height, represented by Y6 and Y7, respectively. The primary suspension parameters of the vehicle include at least the stiffness and damping, represented by Y8 and Y9, respectively. The secondary suspension parameters of the vehicle include at least the stiffness and damping, represented by Y10 and Y11, respectively.

[0060] Among them, the variable constraints corresponding to Y1 to Y11 are shown in Table 1:

[0061] Table 1 Variable constraints

[0062] It should be noted that when the value of any system variable changes, the value of at least one performance indicator changes. In other words, the value of the system variable affects the value of the performance indicator.

[0063] Step 102: Construct a multi-objective optimization model for each performance indicator based on the system variables.

[0064] Taking the performance indicators X1 to X5 as an example, the constructed multi-objective optimization model can be expressed by the following formula (1): min f(x) = (f1(x), f2(x), f3(x), f4(x), f5(x)) (1)

[0065] Where 0≤x j ≤1, j=1,2,3,…,11, refers to the corresponding system variables. Based on this, f1(x) is the optimization submodule for X1, f2(x) is the optimization submodule for X2, f3(x) is the optimization submodule for X3, f4(x) is the optimization submodule for X4, and f5(x) is the optimization submodule for X5. f(x) is a multi-objective optimization model for X1, X2, X3, X4, and X5. The optimization goal of f(x) is to minimize the output value so that the index value corresponding to each performance index in X1, X2, X3, X4, and X5 meets the optimization conditions.

[0066] Step 103: Process the sample population using a multi-objective optimization model to obtain at least one optimal solution.

[0067] The sample population includes multiple sample groups, each of which includes sample values ​​corresponding to each of the system variables. Based on this, the optimal solution obtained in this embodiment includes variable values ​​corresponding to each of the system variables, and the variable values ​​in the optimal solution ensure that the indicator values ​​corresponding to the performance indicators meet the optimization conditions.

[0068] In one implementation, in this embodiment, the sample values ​​in each sample group in the sample population can be substituted into the corresponding system variables in the suspension guidance system in turn, thereby obtaining the index value corresponding to each performance index. In this way, the candidate value of its output can be obtained through the multi-objective optimization model, and the target value that meets the optimization conditions can be screened out from these candidate values, and the sample group corresponding to the target value is determined as the optimal solution.

[0069] In another implementation, in this embodiment, the sample population is multiplied, such as taking the index value corresponding to the performance index to meet the optimization condition as the reproduction goal to obtain an expanded sample group, and the expanded sample group is added to the sample population. In this way, according to the previous method, the sample values ​​in each sample group in the expanded sample population are substituted into the corresponding system variables in the suspension guidance system in turn, thereby obtaining the index value corresponding to each performance index. In this way, the candidate value of its output can be obtained through the multi-objective optimization model, and the target value that meets the optimization condition is screened out from these candidate values, and the sample group corresponding to the target value is determined as the optimal solution.

[0070] As can be seen from the above technical solutions, in a control method for optimizing the performance of a suspension guidance system provided in Example 1 of the present application, after obtaining multiple performance indicators and multiple system variables of the suspension guidance system, a multi-objective optimization model for each performance indicator is constructed based on these system variables, and then the constructed multi-objective optimization model is used to process a sample population containing multiple sample groups, so that an optimal solution can be obtained. In the optimal solution, the variable value corresponding to each system variable can make the indicator value corresponding to each performance indicator meet the optimization conditions, thereby achieving performance optimization of the suspension guidance system. It can be seen that in this embodiment, there is no need for excessive designer participation. The optimal solution can be obtained by constructing a multi-objective optimization model, thereby saving human resources and not causing the optimization performance to take too long, thereby achieving efficient performance optimization control.

[0071] In one implementation, in step 103, when the multi-objective optimization model is used to process the sample population to obtain at least one optimal solution, it can be implemented in the following manner:

[0072] Using a multi-objective optimization model and based on a multi-objective optimization algorithm, the sample groups in the sample population are processed to obtain the optimal solution that makes the index values ​​corresponding to the performance index meet the optimization conditions;

[0073] In which, the optimal solution comes from the sample population or the optimal solution comes from a sample group extended from the sample population, and the multi-objective optimization algorithm is a multi-objective evolutionary algorithm with adaptive constraints based on co-evolution, and the multi-objective evolutionary algorithm with adaptive constraints based on co-evolution is represented by ACMCA (An Adaptive Constrained Multi-Objective Co-Evolutionary Algorithm).

[0074] Specifically, in this embodiment, according to the multi-objective optimization model's optimization goal of ensuring that the values ​​corresponding to the performance indicators meet the optimization criteria, ACMCA is used to multiply sample groups within the sample population. The multiplication goal is to ensure that, after substituting the sample values ​​in the multiplied sample groups into the corresponding system variables of the suspension guidance system, the values ​​corresponding to each performance indicator of the suspension guidance system meet the optimization criteria. Thus, after ACMCA, the multi-objective optimization model can obtain one or more optimal solutions based on the sample population.

[0075] Furthermore, in order to reduce the impact of different orders of magnitude between various system variables on performance optimization in this embodiment, before using the multi-objective optimization model to process the sample population in step 103 to obtain at least one optimal solution, in this embodiment, the sample values ​​contained in each sample group in the sample population can be normalized first, and then step 103 is executed to use the multi-objective optimization model to process the sample groups in the normalized sample population based on the multi-objective optimization algorithm to obtain the optimal solution in which the index values ​​corresponding to the performance indicators meet the optimization conditions.

[0076] In one implementation, in step 102, when constructing a multi-objective optimization model for each performance indicator based on the system variables, it can be implemented in the following manner, as shown in FIG2 :

[0077] Step 201: Determine at least one target variable for each performance indicator among multiple system variables.

[0078] Among them, the target variable and its corresponding performance indicator meet the association condition.

[0079] For example, each performance indicator may correspond to one or more target variables. The target variables corresponding to different performance indicators may have the same system variables or different system variables.

[0080] Specifically, the target variable and its corresponding performance index satisfy the association condition, which can be: the sensitivity index value between the target variable and the performance index is ranked in the top N among all system variables, where N is a positive integer greater than or equal to 1.

[0081] Among them, the value of the sensitivity index between the system variable and the performance index represents the degree of influence of the change of the system variable on the performance index.

[0082] For example, each system variable and each performance indicator have a corresponding sensitivity index value. The sensitivity index values ​​between different system variables and the same performance indicator can be different or the same, and the sensitivity index values ​​between the same system variable and different performance indicators can be the same or different.

[0083] In one case, a larger sensitivity index value indicates a higher degree of influence of a change in a system variable value on a performance index. A smaller sensitivity index value indicates a lower degree of influence of a change in a system variable value on a performance index. Based on this, the sensitivity index values ​​between the target variable and the performance index are ranked in the top N among all system variables from largest to smallest, where N can be 3. For example, in this embodiment, for each performance index, the sensitivity index value between the performance index and each system variable is obtained, and the sensitivity index values ​​corresponding to all system variables are ranked from largest to smallest. The system variables corresponding to the top three sensitivity index values ​​are determined as the target variables corresponding to the corresponding performance index.

[0084] In another case, a larger sensitivity index value indicates a lower degree of influence of a change in the system variable's value on the performance index, while a smaller sensitivity index value indicates a higher degree of influence of a change in the system variable's value on the performance index. Based on this, the sensitivity index values ​​between the target variable and the performance index are ranked in the top N among all system variables, where N can be 3. For example, in this embodiment, for each performance index, the sensitivity index value between the performance index and each system variable is obtained, and the sensitivity index values ​​corresponding to all system variables are ranked in ascending order. The system variables corresponding to the top three sensitivity index values ​​are then determined as the target variables corresponding to the corresponding performance index.

[0085] In a specific implementation, the sensitivity index value between the system variable and the performance index is determined based on at least a change state of the performance index caused by a change state of the system variable.

[0086] The change state of the system variable includes: the historical variable value before the change and the current variable value after the change, which are represented by P1 and P2 respectively; the change state of the performance index includes the historical index value before the change of the system variable and the current index value after the change of the system variable, which are represented by N1 and N2 respectively. Based on this, the sensitivity index value between the system variable and the performance index is obtained by the following formula (2):

[0087] Step 202: Construct a multi-objective optimization model based on the target variables corresponding to each performance indicator.

[0088] Among them, the multi-objective optimization model takes the index value corresponding to the performance index to meet the optimization conditions as the optimization goal.

[0089] In the specific implementation, the index value corresponding to the performance index satisfies the optimization condition, which can be: the index value corresponding to the performance index makes the target value output by the multi-objective optimization model the minimum among all candidate values,

[0090] The candidate value is: after the sample value in the sample group is substituted into the corresponding system variable of the suspension guidance system, the index value corresponding to each performance index of the suspension guidance system makes the data value output by the multi-objective optimization model.

[0091] It should be noted that the sample groups here can be directly derived from the sample population, or from the sample groups expanded from the original sample groups contained in the sample population through ACMCA. Based on this, in step 103, the multi-objective optimization model is used to process the sample groups in the normalized sample population based on ACMCA to obtain the optimal solution in which the indicator values ​​corresponding to the performance indicators meet the optimization conditions.

[0092] Referring to Figure 3, there is a schematic diagram illustrating the structure of a control device for optimizing the performance of a suspension guidance system, according to a second embodiment of the present application. This device can be configured in an electronic device capable of data processing, such as a computer or server. The technical solution in this embodiment is primarily used to improve the efficiency of optimizing the performance of a suspension guidance system.

[0093] Specifically, the device in this embodiment may include the following units:

[0094] A data acquisition unit 301 is used to obtain a plurality of performance indicators and a plurality of system variables of the suspension guidance system;

[0095] A model building unit 302 is configured to build a multi-objective optimization model for each of the performance indicators based on the system variables;

[0096] An optimal obtaining unit 303 is configured to process the sample population using the multi-objective optimization model to obtain at least one optimal solution, wherein the optimal solution includes variable values ​​corresponding to each of the system variables, and the variable values ​​in the optimal solution make the indicator value corresponding to each of the performance indicators meet the optimization condition;

[0097] The sample population includes a plurality of sample groups, and each of the sample groups includes a sample value corresponding to each of the system variables.

[0098] As can be seen from the above technical solution, in a control device for optimizing the performance of a suspension guidance system provided in Example 2 of the present application, after obtaining multiple performance indicators and multiple system variables of the suspension guidance system, a multi-objective optimization model for each performance indicator is constructed based on these system variables. Then, the constructed multi-objective optimization model is used to process a sample population containing multiple sample groups. In this way, an optimal solution can be obtained. The variable value corresponding to each system variable in the optimal solution can make the indicator value corresponding to each performance indicator meet the optimization condition, thereby achieving performance optimization of the suspension guidance system. It can be seen that in this embodiment, there is no need for excessive designer participation. The optimal solution can be obtained by constructing a multi-objective optimization model, thereby saving human resources and not causing the optimization performance to take too long, thereby achieving efficient performance optimization control.

[0099] In one implementation, the model building unit 302 is specifically used to: determine at least one target variable for each of the performance indicators among the multiple system variables, and the target variable and its corresponding performance indicator satisfy an association condition; and construct a multi-objective optimization model based on the target variable corresponding to each of the performance indicators, and the multi-objective optimization model takes the indicator value corresponding to the performance indicator satisfying the optimization condition as the optimization goal.

[0100] Among them, the association conditions between the target variable and its corresponding performance indicator include: the sensitivity index value between the target variable and the performance indicator is ranked in the top N among all the system variables, and N is a positive integer greater than or equal to 1; wherein the size of the sensitivity index value between the system variable and the performance indicator represents the degree of influence of the change of the system variable on the performance indicator.

[0101] Optionally, the sensitivity index value between the target variable and the performance index is determined based at least on a change state of the performance index caused by a change state of the system variable.

[0102] Optionally, the index value corresponding to the performance indicator satisfies the optimization condition, including at least: the index value corresponding to the performance indicator makes the target value output by the multi-objective optimization model the smallest among all candidate values; wherein, the candidate value is: after the sample value in the sample group is substituted into the corresponding system variable of the suspension guidance system, the suspension guidance system makes the data value output by the multi-objective optimization model at each index value corresponding to the performance indicator.

[0103] In one implementation, the optimal acquisition unit 303 is specifically used to: utilize the multi-objective optimization model and, based on the multi-objective optimization algorithm, process the sample group in the sample population to obtain an optimal solution that makes the index value corresponding to the performance index meet the optimization conditions; wherein, the optimal solution comes from the sample population or the optimal solution comes from the sample group extended from the sample population, and the multi-objective optimization algorithm is a multi-objective evolutionary algorithm with adaptive constraints based on collaborative evolution.

[0104] In one implementation, before processing the sample population using the multi-objective optimization model to obtain at least one optimal solution, the optimal obtaining unit 303 is further configured to: perform normalization processing on the sample values ​​included in each sample group in the sample population.

[0105] The performance index includes any number of the following: vehicle lateral fluctuation, vehicle vertical fluctuation, vehicle stability index, vehicle vertical vibration acceleration, and vehicle lateral vibration acceleration;

[0106] The system variables include any number of the following: electromagnetic parameters of the vehicle-mounted magnet, shape parameters of the vehicle-mounted magnet, electromagnetic parameters of the ground suspension guide coil, shape parameters of the ground suspension guide coil, vehicle primary suspension parameters, and vehicle secondary suspension parameters; and the system variables correspond to variable constraints.

[0107] It should be noted that the specific implementation of each unit in this embodiment can refer to the corresponding content in the previous text and will not be described in detail here.

[0108] 4 is a schematic diagram of the structure of an electronic device provided in Example 3 of the present application. The electronic device may include the following structures:

[0109] Memory 401, used to store computer programs and data generated by the execution of the computer programs;

[0110] Processor 402 is used to execute the computer program to achieve: obtaining multiple performance indicators and multiple system variables of the suspension guidance system; constructing a multi-objective optimization model for each of the performance indicators based on the system variables; using the multi-objective optimization model, processing the sample population to obtain at least one optimal solution, the optimal solution including variable values ​​corresponding to each of the system variables, the variable values ​​in the optimal solution making the indicator value corresponding to each of the performance indicators meet the optimization conditions; wherein the sample population includes multiple sample groups, each of the sample groups including sample values ​​corresponding to each of the system variables.

[0111] Of course, the electronic device may also include modules such as a communication device and a display device.

[0112] As can be seen from the above technical solution, in an electronic device provided in Example 3 of the present application, after obtaining multiple performance indicators and multiple system variables of the suspension guidance system, a multi-objective optimization model for each performance indicator is constructed based on these system variables. Then, the constructed multi-objective optimization model is used to process a sample population containing multiple sample groups. In this way, an optimal solution can be obtained. The variable value corresponding to each system variable in the optimal solution can make the indicator value corresponding to each performance indicator meet the optimization conditions, thereby achieving performance optimization of the suspension guidance system. It can be seen that in this embodiment, there is no need for excessive designer participation. The optimal solution can be obtained by constructing a multi-objective optimization model, thereby saving human resources and not causing the optimization performance to take too long, thereby achieving efficient performance optimization control.

[0113] Taking the suspension guidance system on a superconducting electric levitation train as an example, this application uses a genetic algorithm to perform multi-objective optimization, which can take into account the performance parameters of each component while taking into account the overall performance of the suspension guidance system. The optimization speed is fast, less human resources are occupied, and the optimal solution is more accurate.

[0114] The basic approach of this application is to first draw upon the experience of each component's designers to obtain initial parameter data for each component, which serve as the design variables to be optimized, referred to as system variables as described above. Next, the desired performance of the suspension and guidance system is determined, defined as the performance indicators described above, such as optimal stability, minimum weight, and lowest price. Then, using a suitable genetic algorithm, such as ACMCA, the performance parameters of each component are optimized, achieving simultaneous optimization of multiple performance parameters of the suspension and guidance system.

[0115] The overall cost of a superconducting maglev system is directly related to tunnel cross-sectional dimensions, track accuracy requirements, and bridge stiffness requirements. To reduce the overall cost of a superconducting maglev system while ensuring safety and comfort during high-speed / ultra-high-speed operation, it is necessary to simultaneously consider the train-track-bridge-tunnel coupling during train design, reduce tunnel cross-sectional dimensions, and lower requirements for track accuracy and bridge stiffness.

[0116] The safety and comfort of a train are directly dependent on the performance of the suspension and guidance system. This includes fluctuations in the levitation and guidance forces caused by the interaction between the onboard magnets and the ground-based suspension and guidance coils during vehicle movement, fluctuations in the clearance between the vehicle, the trackside, and the ground, and vibrations of the suspension frame after the suspension parameters are matched. A maglev train with a high-performance suspension and guidance system is the only way to ensure safety and comfort.

[0117] In order to optimize the performance of the suspension guidance system, multiple system parameters need to be optimized simultaneously. These system parameters cannot be simply evaluated as the larger the better or the smaller the better. Instead, these parameters influence and couple with each other, and need to be synchronized to achieve the best mutual matching parameters.

[0118] Wheel-rail trains do not require aerodynamic coupling optimization due to their relatively low speeds. This application can be applied to high-speed maglev trains exceeding 600 kilometers per hour. Due to these high speeds, local airflow velocities can even exceed the speed of sound in tunnel crossings, strong crosswinds, or tunnel entry and exit conditions, making them transonic or supersonic systems. The interaction between the flow field and vehicle dynamics is more pronounced. Furthermore, wheel-rail trains are free of electromagnetic forces affecting the vehicle, significantly reducing operational complexity.

[0119] The multi-objective optimization control scheme in this application is described in detail below, as shown in FIG5 :

[0120] First, the vehicle magnet designer A, the ground suspension guide coil designer B, and the suspension frame designer C each design components based on their design experience. The output component performance parameters are the system variables mentioned above, such as the vehicle magnet parameters, the ground suspension guide coil parameters, the suspension frame parameters, and other variable parameters.

[0121] Then, the component performance parameters are given to the vehicle-track coupling dynamics calculation engineer to carry out dynamic calculations. By establishing a vehicle-track coupling physical model, vehicle-track coupling dynamics calculations are performed to output the performance parameters of the suspension guidance system, namely the performance indicators mentioned above, such as vehicle lateral fluctuation, vehicle vertical fluctuation, vehicle stability index, vehicle vertical vibration acceleration, vehicle lateral vibration acceleration, etc.

[0122] Afterwards, if the designer determines that the performance of the suspension guidance system is not optimal, multi-objective optimization is directly performed through the multi-objective optimization scheme of this application. If it is optimal, the optimal parameters of each component of the suspension guidance system and the optimal performance parameters of the system are output.

[0123] The optimization process of the multi-objective optimization solution of this application is as follows:

[0124] First, refine the suspension and steering performance indicators, that is, the design goals, including vehicle lateral fluctuation (X1), vehicle vertical fluctuation (X2), vehicle stability index (X3), vehicle vertical vibration acceleration (X4), and vehicle lateral vibration acceleration (X5).

[0125] Next, the design variables were refined to determine the design variables to be optimized. These included the electromagnetic parameters of the onboard magnet, the external parameters of the onboard magnet, the electromagnetic parameters of the ground-based suspension guide coil, the external parameters of the ground-based suspension guide coil, the vehicle's primary suspension parameters, and the vehicle's secondary suspension parameters. The constraints for the design variables to be optimized were also determined, as shown in Table 1.

[0126] Based on this, the sensitivity between the five design objectives and the 11 design variables was analyzed, and the obtained sensitivity index values ​​were ranked. The ranking results are shown in Table 2.

[0127] Table 2 Ranking results of sensitivity index values

[0128] In addition, initialize the sample population of the design variables. The process is as follows:

[0129] 1) Set the variable array Div[·] to store feasible samples and determine the maximum number of loops, such as 10 or 20 times;

[0130] 2) Within the value range of the 11 design variables in the second row of Table 1, generate a group of individuals at a fixed interval and store them in Div[·] to avoid excessive concentration of values;

[0131] 3) Traverse the feasible samples stored in Div[·] in sequence and eliminate the same samples;

[0132] 4) Determine whether the maximum number of cycles is met. If not, go to 2), otherwise exit the initialization process.

[0133] At the same time, the optimization algorithm is determined in advance:

[0134] Different genetic algorithms are used to solve different problems. There are many genetic algorithms available in this application, including BP, LSSVM, NCGA, NSGA-II, AMGA, etc. This application preferably applies the ACMCA evolutionary algorithm to the design of the suspension guidance system of the electric suspension train. The algorithm evolves two populations with complementary functions (main population and archive population) and undergoes a double breeding operation. In the first breeding process, the main population realizes the adaptive selection between target optimization and constraint processing through the designed dynamic fitness allocation function. The secondary breeding further enhances the convergence and diversity of the population through cooperation with the archive. The breeding process of the sample population by ACMCA is shown in Figure 6:

[0135] After initializing the sample population, the first reproduction process uses dynamic fitness allocation combined with environmental selection to generate a descendant population from the main population through crossover mutation. The background population is then merged with the main population. A second reproduction process then occurs, using restricted mating selection from the main and archive populations into the mating pool. This process then undergoes crossover mutation to generate a descendant population. The archive population, main population, and background population are then merged. The archive population is then reconstructed using an angle-based selection scheme combined with non-dominated sorting. The main population and background population are then merged, and the main population is reconstructed based on environmental selection for iterative reproduction. This reproduction process continues iteratively until the termination condition is met. If the termination condition is met, the main population is output.

[0136] Based on this, the multi-objective optimization model is determined:

[0137] During the optimization process, to reduce the impact of the magnitude of the design variables, a unified normalization process is performed so that the value range of each design variable in the multi-objective optimization model is [0, 1]. Based on the ACMCA algorithm, the optimization direction is to find the minimum value of each objective function, and a multi-objective optimization model is constructed as shown in Formula (1).

[0138] In formula (1), x is an 11-dimensional variable, corresponding to the 11 design variables in Table 1; f(x) is a 5-dimensional objective function vector, corresponding to the five performance indicators mentioned above, and its mapping relationship with x is determined by the ACMCA algorithm.

[0139] Finally, the optimal solution set is calculated as follows:

[0140] When using ACMCA to optimize the design of the suspension guidance system parameters, the mathematical function in formula (1) is used as the fitness function, the population size is taken as 100, the crossover probability is 0.8, the mutation probability is 0.06, and the maximum evolutionary generation is 200. After the population individuals are quickly non-dominated sorted and the crowding degree is calculated, the crossover mutation is selected for continuous optimization until the predetermined number of iterations is reached and the optimal solution is obtained.

[0141] Furthermore, this application verifies the accuracy of the design target based on the optimized design variables using traditional vehicle-track coupling dynamics calculation methods. If the optimal design target is confirmed, the result is output. If not, all the previous steps are repeated until the optimal design target is output.

[0142] It can be seen that in this application, mathematical models are used to calculate the optimal design variables and the optimal design goals, replacing the traditional method of relying on designer experience to infer. The optimization speed is fast, the number of personnel occupied is small, and the output can be guaranteed to be the optimal result.

[0143] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. Reference can be made to the common and similar parts between the various embodiments. For the devices disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple, and the relevant parts can be referred to the method description.

[0144] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the above description has generally described the components and steps of each example according to their functions. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0145] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein may be implemented directly using hardware, a software module executed by a processor, or a combination of the two. The software module may be placed in a random access memory (RAM), internal memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.

[0146] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present application. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application is not limited to the embodiments shown herein, but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A control method for optimizing the performance of a suspension guidance system, characterized in that: The method comprises: Obtaining a plurality of performance indicators and a plurality of system variables of the suspension guidance system; Constructing a multi-objective optimization model for each of the performance indicators according to the system variables; Using the multi-objective optimization model, processing the sample population to obtain at least one optimal solution, wherein the optimal solution includes variable values ​​corresponding to each of the system variables, and the variable values ​​in the optimal solution make the indicator values ​​corresponding to the performance indicators meet the optimization conditions; The sample population includes a plurality of sample groups, and each of the sample groups includes sample values ​​corresponding to each of the system variables.

2. The method according to claim 1, characterized in that According to the system variables, a multi-objective optimization model for each of the performance indicators is constructed, including: Among the plurality of system variables, at least one target variable is determined for each performance indicator, and a correlation condition is satisfied between the target variable and its corresponding performance indicator; A multi-objective optimization model is constructed according to the target variables corresponding to each of the performance indicators. The multi-objective optimization model takes the indicator values ​​corresponding to the performance indicators satisfying the optimization conditions as the optimization objectives.

3. The method according to claim 2, characterized in that The target variable and its corresponding performance indicator satisfy the association conditions, including: The sensitivity index value between the target variable and the performance index is ranked in the top N among all the system variables, where N is a positive integer greater than or equal to 1; The value of the sensitivity index between the system variable and the performance index represents the degree of influence of the change of the system variable on the performance index.

4. The method according to claim 3, characterized in that The sensitivity index value between the target variable and the performance index is determined based on at least a change state of the performance index caused by a change state of the system variable.

5. The method according to claim 2, characterized in that: The indicator value corresponding to the performance indicator satisfies the optimization condition, which includes at least: The index value corresponding to the performance index makes the target value output by the multi-objective optimization model the smallest among all candidate values; The candidate value is: the sample value in the sample group is substituted into the corresponding value of the suspension guidance system. After responding to the system variables, the index value corresponding to each performance index of the suspension guidance system makes the data value output by the multi-objective optimization model.

6. The method according to claim 1 or 2, characterized in that: Using the multi-objective optimization model, processing the sample population to obtain at least one optimal solution includes: Using the multi-objective optimization model and based on a multi-objective optimization algorithm, the sample group in the sample population is processed to obtain an optimal solution that makes the indicator value corresponding to the performance indicator meet the optimization condition; The optimal solution comes from the sample population or the optimal solution comes from a sample group expanded from the sample population, and the multi-objective optimization algorithm is a multi-objective evolutionary algorithm with adaptive constraints based on collaborative evolution.

7. The method according to claim 6, characterized in that Before using the multi-objective optimization model to process the sample population to obtain at least one optimal solution, the method further includes: The sample values ​​contained in each of the sample groups in the sample population are normalized.

8. The method according to claim 1 or 2, characterized in that: The performance indicators include any of the following: Vehicle lateral fluctuation, vehicle vertical fluctuation, vehicle stability index, vehicle vertical vibration acceleration, vehicle lateral vibration acceleration; The system variables include any number of the following: electromagnetic parameters of vehicle-mounted magnets, shape parameters of vehicle-mounted magnets, electromagnetic parameters of ground suspension guide coils, shape parameters of ground suspension guide coils, primary suspension parameters of vehicles, and secondary suspension parameters of vehicles; Furthermore, the system variables correspond to variable constraints.

9. A control device for optimizing the performance of a suspension guidance system, characterized in that: The device comprises: A data acquisition unit, used to obtain a plurality of performance indicators and a plurality of system variables of the suspension guidance system; A model building unit, used for building a multi-objective optimization model for each of the performance indicators according to the system variables; An optimal obtaining unit, used to process the sample population using the multi-objective optimization model to obtain at least one optimal solution, wherein the optimal solution includes variable values ​​corresponding to each of the system variables, and the variable values ​​in the optimal solution make the indicator value corresponding to each of the performance indicators meet the optimization condition; The sample population includes a plurality of sample groups, each of which includes a sample group corresponding to each Sample values ​​of the system variables.

10. An electronic device, characterized in that: The electronic device comprises: A memory, used to store a computer program and data generated by the execution of the computer program; A processor is used to execute the computer program to achieve: obtaining multiple performance indicators and multiple system variables of the suspension guidance system; constructing a multi-objective optimization model for each of the performance indicators based on the system variables; using the multi-objective optimization model, processing the sample population to obtain at least one optimal solution, the optimal solution includes variable values ​​corresponding to each of the system variables, the variable values ​​in the optimal solution make the indicator value corresponding to each of the performance indicators meet the optimization conditions; wherein the sample population includes multiple sample groups, each of the sample groups includes sample values ​​corresponding to each of the system variables.

Images