Thermal management control quantity optimization method and device, electronic equipment, storage medium and vehicle

Through iterative solution and genetic algorithm optimization of thermal management control quantity parameters, the problem of inefficiency when multiple control quantity in the prior art is solved, and the accuracy and speed of thermal management control are improved.

CN119002250APending Publication Date: 2024-11-22BEIJING CO WHEELS TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202310552047.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2023-05-16
Publication Date
2024-11-22

AI Technical Summary

Technical Problem

The existing thermal management control method takes a long time to find the optimal solution when multiple control quantities, resulting in poor control accuracy.

Method used

By obtaining the current working condition status of the vehicle and the control target, initializing multiple sets of control quantity parameters, iteratively solving the optimal set of control quantity parameters, and using genetic algorithms to cross and mutate the remaining sets of control quantity parameters to improve search efficiency.

Benefits of technology

Improve the efficiency of the solver, ensure the accuracy and speed of thermal management control, and avoid overshoot or undershoot problems.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119002250A_ABST
    Figure CN119002250A_ABST
Patent Text Reader

Abstract

The invention discloses a thermal management control quantity optimization method and device, electronic equipment, a storage medium and a vehicle. The method comprises the steps of obtaining state data of a current working condition state of a vehicle and a control target; initializing multiple groups of control quantity parameters; iteratively solving the optimal group of control quantity parameters, and in each iteration process, respectively inputting the state data and the plurality of groups of control quantity parameters into a physical prediction model, and predicting to obtain a plurality of prediction targets; calculating a fitness value based on the prediction target and the control target, and selecting a group of control quantity parameters with the minimum fitness value as an optimal group of control quantity parameters of the iteration process; and if an iteration ending condition is met, outputting an optimal group control quantity parameter, otherwise, performing genetic operation on the rest groups of control quantity parameters by using the optimal group control quantity parameter, and executing a next iteration process. According to the method, when a plurality of control quantity parameters exist in one group of control quantity parameters, the optimal solution search is performed by using the genetic algorithm, so that the search efficiency is improved, and the efficiency of a solver is further improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle-related technologies, and in particular to a method, device, electronic equipment, storage medium and vehicle for optimizing thermal management control quantities. Background Art

[0002] In the existing thermal management control strategy, the thermal management of the vehicle adopts a control method based on rules and table (MAP) calibration. Through a large number of experiments, the specific values ​​of the control quantities of various control targets are calibrated, and then in the actual vehicle, according to the control target, based on the rules, the specific control quantity parameters are obtained by looking up the table of control quantities to control the vehicle. For example, for hybrid vehicles, the integrated control of the engine, water heater (Water Positive Temperature Coefficient, WPTC), and power battery thermal process, the current strategy is a control method based on rules and table calibration.

[0003] However, the control method based on rules and table calibration has poor control accuracy, consumes too much energy on thermal management components, has a slow cooling rate, and may even cause overshoot problems. At the same time, this control method requires a lot of calibration work, which adds a lot of workload to automobile development.

[0004] To this end, the prior art proposes the use of a thermal management model predictive control (MPC) architecture, which uses a physical prediction model to predict the state, uses a solver to perform iterative solutions, and inputs the state data of the vehicle's current operating state and multiple groups of control quantity parameters into the physical prediction model respectively. Based on the comparison between the prediction target and the control target output by the physical prediction model, the output control quantity sequence is determined, and then the output control quantity sequence is used to perform thermal management control on the vehicle.

[0005] The iterative solution is as follows Figure 1 As shown, including:

[0006] Step S101, in each iteration process, points are scattered to obtain control quantity parameters;

[0007] Step S102, inputting the control quantity parameters obtained by scattering points into the physical prediction model in sequence for prediction calculation to obtain the prediction target corresponding to the control quantity parameters;

[0008] Step S103: compare the predicted targets corresponding to all control quantity parameters in this iteration with the control targets to be achieved by the controlled thermal management parameters;

[0009] Step S104, recording the control quantity parameter that is closest to the predicted target and the control target;

[0010] After the iteration, the optimal control quantity parameters of the controlled thermal management parameters are obtained, the control quantities of all the controlled thermal management parameters are calculated cyclically, and finally the optimal control quantity parameters of all the controlled thermal management parameters are combined into an output control quantity sequence.

[0011] However, when there are multiple control quantities for adjusting the same controlled management parameter, for example, for temperature adjustment, it is necessary to simultaneously control the three-way valve opening, four-way valve opening, water heater (Water Positive Temperature Coefficient, WPTC) power, battery water pump speed, fan speed, etc. If an exhaustive permutation and combination method is used to traverse all possible values ​​of all control quantities, it will take a lot of time and computing power, resulting in a long time to find the optimal solution, which in turn leads to poor control accuracy. Summary of the invention

[0012] Based on this, it is necessary to provide a thermal management control quantity optimization method, device, electronic device, storage medium and vehicle to address the technical problem that the thermal management control method of the prior art takes a long time to find the optimal solution when there are multiple control quantities, which in turn leads to poor control accuracy.

[0013] The present invention provides a method for optimizing a thermal management control quantity, comprising:

[0014] Acquire status data and control targets of the current working state of the vehicle, wherein the control targets are targets that the controlled thermal management parameters are expected to achieve under the current working conditions;

[0015] Initializing multiple groups of control quantity parameters, each group of control quantity parameters includes control quantity parameters of multiple control quantities;

[0016] Iteratively solve the optimal group control quantity parameters. In each iteration process:

[0017] Input the state data and multiple groups of control quantity parameters of this iteration process into the physical prediction model respectively, and predict multiple prediction targets, each of which is predicted by a group of control quantity parameters and the state data based on the physical prediction model;

[0018] Calculating the fitness value of each group of control quantity parameters based on the prediction target and the control target, and selecting a group of control quantity parameters with the smallest fitness value as the optimal group of control quantity parameters for this iteration process;

[0019] If the iteration end condition is met, the optimal group control amount parameters are output; otherwise, the optimal group control amount parameters are used to perform genetic operations on the remaining group control amount parameters, and then the next iteration process is executed.

[0020] Furthermore, after the genetic operation is performed on the remaining control amount parameters by using the optimal group control amount parameters, the next iteration process is performed, which specifically includes:

[0021] Replacing some control amount parameters in the remaining control amount parameters except the optimal group control amount parameters with corresponding control amount parameters in the optimal group control amount parameters to obtain the remaining control amount parameters after the crossover operation;

[0022] Multiplying the control amount parameters in the remaining groups of control amount parameters after the crossover operation by a preset variation coefficient to obtain the remaining groups of control amount parameters after the variation operation;

[0023] The next iteration process is performed with the remaining groups of control quantity parameters and the optimal group of control quantity parameters after the mutation operation.

[0024] Furthermore, multiplying the control amount parameters in the remaining groups of control amount parameters after the crossover operation by a preset coefficient of variation specifically includes:

[0025] The control quantity parameters in the remaining groups of control quantity parameters after the crossover operation are multiplied by the preset coefficient of variation of the control quantity corresponding to the control quantity parameters.

[0026] Furthermore, the state data and multiple groups of control quantity parameters of this iteration process are respectively input into the physical prediction model to predict multiple prediction targets, specifically including:

[0027] In the first iteration process, the state data and all group control quantity parameters of this iteration process are respectively input into the physical prediction model to predict multiple prediction targets; or

[0028] In the iteration process other than the first iteration process, all group control quantity parameters of this iteration process are divided into excellent solution domain and inferior solution domain;

[0029] Inputting the state data and one or more groups of control quantity parameters included in the excellent solution domain into a physical prediction model respectively, and if the physical prediction model can predict a prediction target based on at least one group of control quantity parameters in the excellent solution domain, predicting one or more prediction targets;

[0030] Otherwise, the state data and one or more groups of control quantity parameters included in the inferior solution domain are respectively input into the physical prediction model. If the physical prediction model can predict the prediction target based on at least one group of control quantity parameters in the inferior solution domain, one or more prediction targets are predicted. Otherwise, the iteration is terminated and the optimal group of control quantity parameters obtained in the last iteration process is output.

[0031] Furthermore, if the iteration end condition is met, the optimal group control amount parameter is output, which specifically includes:

[0032] If the number of iterations is greater than or equal to a preset number threshold or the optimal group control quantity parameters obtained in this iteration process are consistent with the optimal group control quantity parameters obtained in the previous iteration process, it is determined that the iteration end condition is met and the optimal group control quantity parameters are output.

[0033] Furthermore, the initializing of multiple groups of control quantity parameters specifically includes:

[0034] A plurality of groups of control quantity parameters are randomly initialized, and each control quantity parameter of each group of control quantity parameters is a random value.

[0035] Further, the calculating of the fitness value of each group of control amount parameters based on the prediction target and the control target specifically includes:

[0036] The fitness value of each group of control quantity parameters is calculated as the absolute value of the difference between the corresponding prediction target and the control target.

[0037] The present invention provides a thermal management control quantity optimization device, comprising:

[0038] A target acquisition module is used to acquire the status data of the current working state of the vehicle and the control target, wherein the control target is the target that the controlled thermal management parameter is expected to achieve under the current working condition;

[0039] An initialization module, used for initializing multiple groups of control quantity parameters, each group of control quantity parameters including control quantity parameters of multiple control quantities;

[0040] Iteration module is used to iteratively solve the optimal group control quantity parameters. In each iteration process:

[0041] Input the state data and multiple groups of control quantity parameters of this iteration process into the physical prediction model respectively, and predict multiple prediction targets, each of which is predicted by a group of control quantity parameters and the state data based on the physical prediction model;

[0042] Calculating the fitness value of each group of control quantity parameters based on the prediction target and the control target, and selecting a group of control quantity parameters with the smallest fitness value as the optimal group of control quantity parameters for this iteration process;

[0043] If the iteration end condition is met, the optimal group control amount parameters are output; otherwise, the optimal group control amount parameters are used to perform genetic operations on the remaining group control amount parameters, and then the next iteration process is executed.

[0044] The present invention provides an electronic device, comprising:

[0045] at least one processor; and,

[0046] a memory communicatively connected to at least one of the processors; wherein,

[0047] The memory stores instructions that can be executed by at least one of the processors. The instructions are executed by at least one of the processors to enable the at least one of the processors to perform the thermal management control amount optimization method as described above.

[0048] The present invention provides a storage medium, wherein the storage medium stores computer instructions. When a computer executes the computer instructions, the storage medium is used to execute all steps of the thermal management control quantity optimization method as described above.

[0049] The present invention provides a vehicle, comprising the vehicle thermal management control device as described above, or the electronic device as described above.

[0050] The present invention solves the optimal group of control quantity parameters by iteratively solving multiple groups of control quantity parameters. In each iterative process, the state data and multiple groups of control quantity parameters of this iterative process are respectively input into the physical prediction model to predict multiple prediction targets. Then, the fitness value of each group of control quantity parameters is calculated based on the prediction target and the control target, and the group of control quantity parameters with the smallest fitness value is selected as the optimal group of control quantity parameters of this iterative process; if the iteration end condition is met, the optimal group of control quantity parameters is output, otherwise, the optimal group of control quantity parameters are genetically operated on the remaining groups of control quantity parameters, and the next iterative process is executed. When a group of control quantity parameters has multiple control quantity parameters, the present invention uses a genetic algorithm to search for the optimal solution, thereby improving the search efficiency and thus improving the efficiency of the solver. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 A flowchart of a control quantity parameter determination method of the prior art;

[0052] Figure 2 A flowchart of a method for optimizing a thermal management control quantity according to an embodiment of the present invention;

[0053] Figure 3 This is a working flow chart of a method for optimizing a thermal management control quantity according to another embodiment of the present invention;

[0054] Figure 4 A working flow chart of vehicle thermal management control according to the best embodiment of the present invention;

[0055] Figure 5 It is a schematic diagram of iterative calculation of the optimal group control amount parameters of the best embodiment of the present invention;

[0056] Figure 6 A schematic diagram of a thermal management control quantity optimization device according to an embodiment of the present invention;

[0057] Figure 7 The figure is a schematic diagram of the hardware structure of an electronic device of the present invention. DETAILED DESCRIPTION

[0058] The specific embodiments of the present invention are further described below in conjunction with the accompanying drawings. The same components are represented by the same reference numerals. It should be noted that the words "front", "rear", "left", "right", "upper" and "lower" used in the following description refer to directions in the accompanying drawings, and the words "inner" and "outer" refer to directions toward or away from the geometric center of a specific component, respectively.

[0059] In order to solve the technical problems existing in the prior art, the present invention provides a thermal management control quantity optimization method, device, electronic equipment, storage medium and vehicle.

[0060] like Figure 2 The figure shows a working flow chart of a method for optimizing a thermal management control quantity according to an embodiment of the present invention, comprising:

[0061] Step S210, obtaining status data of the current working state of the vehicle and a control target, wherein the control target is a target that the controlled thermal management parameter is expected to achieve under the current working condition;

[0062] Step S220, initializing multiple groups of control quantity parameters, each group of control quantity parameters includes control quantity parameters of multiple control quantities;

[0063] Step S230, iteratively solve the optimal group control quantity parameters. In each iteration process:

[0064] Step S231, inputting the state data and multiple groups of control quantity parameters of this iteration process into a physical prediction model respectively, predicting multiple prediction targets, each of which is predicted by a group of control quantity parameters and the state data based on the physical prediction model;

[0065] Step S232, calculating the fitness value of each group of control quantity parameters based on the prediction target and the control target, and selecting a group of control quantity parameters with the smallest fitness value as the optimal group of control quantity parameters for this iteration process;

[0066] Step S233: if the iteration end condition is met, the optimal group control amount parameters are output; otherwise, the optimal group control amount parameters are used to perform genetic operations on the remaining group control amount parameters, and then the next iteration process is executed.

[0067] Specifically, the present invention can be applied to electronic devices in a vehicle that have processing capabilities, such as an electronic control unit (ECU) or an extended domain control unit (XCU).

[0068] First, the electronic device executes step S210 to obtain the data and parameters required by the physical prediction model under the current working state of the vehicle as state data. At the same time, one or more control targets are determined according to the actual vehicle conditions. The control target is the target that the controlled thermal management parameter is expected to achieve under the current working condition. The current working state is determined according to the actual vehicle conditions.

[0069] For example, for the vehicle parameter of the battery target temperature, the battery target temperature is set to 26 degrees, and 26 degrees is the control target of the controlled thermal management parameter of the battery target temperature.

[0070] Then, step S220 is executed to initialize multiple groups of control quantity parameters, each group of control quantity parameters includes control quantity parameters of multiple control quantities.

[0071] Each group of control quantity parameters includes a plurality of control quantity parameters, and the control quantity parameters are values ​​of control quantities used to adjust the controlled thermal management parameters.

[0072] For example, with the battery target temperature as the controlled thermal management parameter, the three-way valve opening, four-way valve opening, wptc power, battery water pump speed, fan speed, etc. can be selected as the control quantity. The specific values ​​of the three-way valve opening, four-way valve opening, wptc power, battery water pump speed, and fan speed are the control quantity parameters.

[0073] Then, step S230 is executed to iteratively solve the optimal group control quantity parameters.

[0074] In each iteration process, step S231 to step S233 are executed in sequence.

[0075] First, execute step S231, input the state data and multiple groups of control quantity parameters of this iteration process into the physical prediction model respectively, and predict multiple prediction targets. Each prediction target is predicted by a group of control quantity parameters and the state data based on the physical prediction model.

[0076] Specifically, the physical prediction model is consistent with the actual vehicle structure. The prediction of the physical prediction model can be implemented through a solver, and a subthread can be created, which calls the solver in a loop. The physical prediction model is called in the solver, and the state data and multiple sets of control quantity parameters are input into the physical prediction model to obtain multiple prediction targets.

[0077] Among them, each prediction target is predicted by the state data and a set of control quantity parameters through the physical prediction model.

[0078] The physical prediction model outputs a prediction target of the controlled thermal management parameter according to the input state data and one or more control quantity parameters. By inputting multiple groups of control quantity parameters into the physical prediction model, multiple prediction targets of the controlled thermal management parameter are obtained.

[0079] Then, step S232 is executed to calculate the fitness value of each group of control quantity parameters based on the prediction target and the control target, and select a group of control quantity parameters with the smallest fitness value as the optimal group of control quantity parameters for this iterative process.

[0080] In one embodiment, the calculating the fitness value of each group of control amount parameters based on the prediction target and the control target specifically includes:

[0081] The fitness value of each group of control quantity parameters is calculated as the absolute value of the difference between the corresponding prediction target and the control target.

[0082] Specifically, the fitness value can be calculated as: fitness = abs(T_dest-T_pre), where abs is the absolute value function, T_pre is the predicted target obtained by the physical prediction model under the control quantity, T_dest is the control target, and fitness is the fitness value. By selecting a group of control quantity parameters with the smallest fitness value as the optimal group of control quantity parameters for this iteration process, a group of control quantity parameters with the closest prediction target to the control target is selected as the optimal group of control quantity parameters.

[0083] Then, execute step S233 to determine whether the iteration end condition is met. If the iteration end condition is met, output the optimal group control quantity parameters. Otherwise, after performing genetic operations on the remaining group control quantity parameters with the optimal group control quantity parameters, execute steps S231 to S233 to execute the next iteration process.

[0084] For a control objective, the optimal group of control quantity parameters of the control objective is output through iterative solution, and then the optimal group of control quantity parameters of other control objectives are calculated using the same algorithm.

[0085] The optimal group of control quantity parameters corresponding to each control target are combined into an output control quantity sequence. Then, the output control quantity sequence is used to perform thermal management control on the vehicle.

[0086] In some embodiments, within one cycle, the sub-thread outputs multiple intermediate control quantity sequences within a preset time and an output control quantity sequence that reaches the preset time. Each time the main thread obtains the intermediate control quantity sequence or the output control quantity sequence, it uses the intermediate control quantity sequence or the output control quantity sequence to control the vehicle.

[0087] The intermediate control quantity sequence includes one or more groups of intermediate control quantity parameters. When there is only one control target, the intermediate control quantity sequence includes one group of intermediate control quantity parameters, and when there are multiple control targets, the intermediate control quantity sequence includes multiple groups of the intermediate control quantity parameters, that is, the output control quantity sequence includes the intermediate control quantity parameters of the controlled thermal management parameters corresponding to each control target.

[0088] In the case where the controlled thermal management parameter corresponds to one control quantity, each group of intermediate control quantity parameters includes one intermediate control quantity parameter.

[0089] When there are multiple control quantities corresponding to the controlled thermal management parameter, each group of intermediate control quantity parameters includes multiple intermediate control quantity parameters of different control quantities.

[0090] The control quantity parameter is the value of the corresponding control quantity, and the corresponding control quantity parameter is achieved by changing the control quantity, thereby controlling the controlled thermal management parameter to achieve the control target.

[0091] Specifically, the physical prediction model outputs the predicted target of the controlled thermal management parameter at a preset time based on the input state data and one or more control quantity parameters, and the physical prediction model also outputs multiple intermediate control quantity parameters within the preset time. The control quantity parameter input to the physical prediction model is the value of the control quantity when the physical prediction model reaches the predicted target after the preset time, and the intermediate control quantity parameter is a plurality of values ​​of the control quantity before reaching the control quantity parameter. After the optimal group of control quantity parameters is selected, the physical prediction model will output a plurality of intermediate control quantity parameters when the control quantity reaches the optimal group of control quantity parameters. Among them, the control quantity parameters of the optimal group of control quantity parameters are the optimal control quantity parameters of the corresponding control quantity.

[0092] For example, the controlled thermal management parameter is the battery temperature, and the target temperature, i.e. the control target, is 26 degrees. The control quantities are the three-way valve opening, the four-way valve opening, the wptc power, the battery water pump speed, and the fan speed. The physical prediction model predicts that when the battery temperature is at the optimal control quantity parameters of the three-way valve opening, the four-way valve opening, the wptc power, the battery water pump speed, and the fan speed, the temperature will be 34 degrees after 30 seconds. The prediction target of the physical prediction model is 34 degrees. At the same time, the physical prediction model will also output the control quantity parameters corresponding to each control quantity every second within 30 seconds as the intermediate control quantity parameters of each control quantity.

[0093] Output the output control quantity parameters and the intermediate control quantity parameters to the main thread. As in the above example, the main thread will obtain multiple control quantity parameters for each second within 30 seconds, and the multiple control quantity parameters at 30 seconds are the optimal control quantity parameters. The main thread controls the controlled thermal management parameters adjusted by the multiple control quantities in the vehicle according to the multiple control quantity parameters for each second, that is, controls the components corresponding to the controlled thermal management parameters. In this example, the main thread controls the three-way valve opening, four-way valve opening, wptc power, battery water pump speed, and fan speed as the multiple control quantity parameters obtained every second, and controls the three-way valve opening, four-way valve opening, wptc power, battery water pump speed, and fan speed as the optimal control quantity parameters at 30 seconds, so that the battery temperature reaches 34 degrees after 30 seconds.

[0094] The present invention solves the optimal group of control quantity parameters by iteratively solving multiple groups of control quantity parameters. In each iterative process, the state data and multiple groups of control quantity parameters of this iterative process are respectively input into the physical prediction model to predict multiple prediction targets. Then, the fitness value of each group of control quantity parameters is calculated based on the prediction target and the control target, and the group of control quantity parameters with the smallest fitness value is selected as the optimal group of control quantity parameters of this iterative process; if the iteration end condition is met, the optimal group of control quantity parameters is output, otherwise, the optimal group of control quantity parameters are genetically operated on the remaining groups of control quantity parameters, and the next iterative process is executed. When a group of control quantity parameters has multiple control quantity parameters, the present invention uses a genetic algorithm to search for the optimal solution, thereby improving the search efficiency and thus improving the efficiency of the solver.

[0095] like Figure 3 The figure shows a working flow chart of a method for optimizing a thermal management control quantity in another embodiment of the present invention, comprising:

[0096] Step S310, obtaining status data and control targets of the current operating state of the vehicle, wherein the control targets are targets that the controlled thermal management parameters are expected to achieve under the current operating conditions.

[0097] Step S320, randomly initializing multiple groups of control quantity parameters, each group of the control quantity parameters includes control quantity parameters of multiple control quantities, and each control quantity parameter of each group of the control quantity parameters is a random value.

[0098] Step S330, iteratively solve the optimal group control quantity parameters. In each iteration process:

[0099] Step S331, input the state data and multiple groups of control quantity parameters of this iteration process into the physical prediction model respectively, and predict multiple prediction targets, each of which is predicted by a group of control quantity parameters and the state data based on the physical prediction model.

[0100] In one embodiment, the state data and multiple groups of control quantity parameters of this iteration process are respectively input into a physical prediction model to predict multiple prediction targets, specifically including:

[0101] In the first iteration process, the state data and all group control quantity parameters of this iteration process are respectively input into the physical prediction model to predict multiple prediction targets; or

[0102] In the iteration process other than the first iteration process, all group control quantity parameters of this iteration process are divided into excellent solution domain and inferior solution domain;

[0103] Inputting the state data and one or more groups of control quantity parameters included in the excellent solution domain into a physical prediction model respectively, and if the physical prediction model can predict a prediction target based on at least one group of control quantity parameters in the excellent solution domain, predicting one or more prediction targets;

[0104] Otherwise, the state data and one or more groups of control quantity parameters included in the inferior solution domain are respectively input into the physical prediction model. If the physical prediction model can predict the prediction target based on at least one group of control quantity parameters in the inferior solution domain, one or more prediction targets are predicted. Otherwise, the iteration is terminated and the optimal group of control quantity parameters obtained in the last iteration process is output.

[0105] Step S332, calculating the fitness value of each group of control quantity parameters as the absolute value of the difference between the corresponding prediction target and the control target, and selecting a group of control quantity parameters with the smallest fitness value as the optimal group of control quantity parameters for this iteration process.

[0106] Step S333: If the number of iterations is greater than or equal to the preset number threshold or the optimal group control parameters obtained in this iteration process are consistent with the optimal group control parameters obtained in the previous iteration process, it is determined that the iteration end condition is met and the optimal group control parameters are output; otherwise, the optimal group control parameters are used to perform genetic operations on the remaining group control parameters, and then the next iteration process is executed.

[0107] In one embodiment, after performing genetic operation on the remaining control amount parameters with the optimal group control amount parameters, executing the next iteration process specifically includes:

[0108] Replacing some control amount parameters in the remaining control amount parameters except the optimal group control amount parameters with corresponding control amount parameters in the optimal group control amount parameters to obtain the remaining control amount parameters after the crossover operation;

[0109] Multiplying the control amount parameters in the remaining groups of control amount parameters after the crossover operation by a preset variation coefficient to obtain the remaining groups of control amount parameters after the variation operation;

[0110] The next iteration process is performed with the remaining groups of control quantity parameters and the optimal group of control quantity parameters after the mutation operation.

[0111] In one embodiment, multiplying the control amount parameters in the remaining groups of control amount parameters after the crossover operation by a preset coefficient of variation specifically includes:

[0112] The control quantity parameters in the remaining groups of control quantity parameters after the crossover operation are multiplied by the preset coefficient of variation of the control quantity corresponding to the control quantity parameters.

[0113] Specifically, first, the electronic device executes step S310 to obtain status data of the current operating state of the vehicle and a control target, where the control target is a target that the controlled thermal management parameter is expected to achieve under the current operating condition.

[0114] Among them, the electronic device can create a main thread, and through the working condition determination module of the main thread, according to the actual vehicle situation, update the current working condition status at a preset interval, and obtain the status data and control target of the current working condition status.

[0115] Then, with the preset time as a cycle, steps S320 to S330 are executed in each cycle to obtain the optimal group control amount parameters.

[0116] In some embodiments, at every preset period, the sub-thread obtains state data and control targets from the main thread. During the period, for each of the control targets, the solver is used to solve the optimal set of control quantity parameters of the output control target.

[0117] Specifically, at the beginning of the cycle, the child thread obtains the state data and the control target from the main thread. Then, in the child thread, firstly, step S320 is executed to randomly initialize multiple groups of control amount parameters, each group of the control amount parameters includes control amount parameters of multiple control amounts, and each control amount parameter of each group of the control amount parameters is a random value.

[0118] Specifically, the initial multiple groups of control quantity parameters CtrlVal_k can be obtained through the strategy of random scattering, and each control quantity parameter in each group of control quantity parameters CtrlVal_k obtains a random value through a random function; (the format of a group of control quantity parameters is [three-way valve opening, four-way valve opening, wptc power, battery water pump speed, fan speed...])

[0119] Then, step S330 is executed to iteratively solve the optimal group control quantity parameters.

[0120] Specifically, the solver calls the physical prediction model, inputs the state data and multiple groups of control quantity parameters into the physical prediction model, obtains multiple prediction targets, and then iteratively solves the optimal group of control quantity parameters. At the same time, the solver obtains the intermediate control quantity parameters corresponding to the optimal control quantity parameters from the physical prediction model. Then, at every preset time interval, such as 1 second, the intermediate control quantity parameters or the optimal group of control quantity parameters are output to the main thread. Then, at the beginning of the next cycle, the child thread obtains the state data and control target from the main thread again. That is, in each cycle, the child thread only obtains the state data and control target from the main thread once.

[0121] The solver iteratively solves the output prediction target of each controlled thermal management parameter. The termination condition of the iterative solution is that the number of iterations is greater than or equal to the preset number threshold or the optimal group control quantity parameters obtained in this iteration process are consistent with the optimal group control quantity parameters obtained in the previous iteration process. When the termination condition of the iterative solution is met, the iteration ends. The output prediction target at the end of the iteration is the output prediction target of the iterative solution, and the optimal group control quantity parameters at the end of the iteration are the optimal group control quantity parameters of the iterative solution.

[0122] In each iteration, steps S331 to S333 are performed:

[0123] Step S331, input the state data and multiple groups of control quantity parameters of this iteration process into the physical prediction model respectively, and predict multiple prediction targets, each of which is predicted by a group of control quantity parameters and the state data based on the physical prediction model.

[0124] In one embodiment, the state data and multiple groups of control quantity parameters of this iteration process are respectively input into a physical prediction model to predict multiple prediction targets, specifically including:

[0125] In the first iteration process, the state data and all group control quantity parameters of this iteration process are respectively input into the physical prediction model to predict multiple prediction targets; or

[0126] In the iteration process other than the first iteration process, all group control quantity parameters of this iteration process are divided into excellent solution domain and inferior solution domain;

[0127] Inputting the state data and one or more groups of control quantity parameters included in the excellent solution domain into a physical prediction model respectively, and if the physical prediction model can predict a prediction target based on at least one group of control quantity parameters in the excellent solution domain, predicting one or more prediction targets;

[0128] Otherwise, the state data and one or more groups of control quantity parameters included in the inferior solution domain are respectively input into the physical prediction model. If the physical prediction model can predict the prediction target based on at least one group of control quantity parameters in the inferior solution domain, one or more prediction targets are predicted. Otherwise, the iteration is terminated and the optimal group of control quantity parameters obtained in the last iteration process is output.

[0129] Specifically, if it is the first iteration process, multiple groups of randomly initialized control quantity parameters are input into the physical prediction model to predict the prediction target of each group of control quantity parameters under the state data.

[0130] If it is an iterative process other than the first iterative process, that is, the second iterative process and the iterative process after the second iterative process, all the groups of control quantity parameters of this iterative process are divided into excellent solution domains and inferior solution domains. The excellent solution domain may not include any group of control quantity parameters, or may include one or more groups of control quantity parameters, and the inferior solution domain may not include any group of control quantity parameters, or may include one or more groups of control quantity parameters.

[0131] In some embodiments, based on preset empirical conditions, all group control quantity parameters in this iteration process are divided into excellent solution domains and poor solution domains.

[0132] Specifically, the empirical condition is a condition pre-set by the developer based on experience according to the current vehicle state. The empirical condition is mainly to exclude obviously inappropriate control parameters.

[0133] For example, in this solution, according to the current vehicle state, the three-way valve needs to be opened, then the empirical condition can be set as: one or more groups of control quantity parameters including the three-way valve opening of 0 are divided into the inferior solution domain, and other groups of control quantity parameters are divided into the excellent solution domain.

[0134] Then, all groups of control quantity parameters contained in the excellent solution domain are preferentially input into the physical prediction model respectively. If the physical prediction model can predict the prediction target based on at least one group of control quantity parameters in the excellent solution domain, the excellent solution domain has a solution, and one or more prediction targets are predicted.

[0135] If the physical prediction model cannot predict the prediction target based on at least one set of control quantity parameters in the excellent solution domain, then the excellent solution domain has no solution, and the state data and one or more sets of control quantity parameters included in the inferior solution domain are respectively input into the physical prediction model. If the physical prediction model can predict the prediction target based on at least one set of control quantity parameters in the inferior solution domain, then the inferior solution domain has a solution, and one or more prediction targets are predicted.

[0136] If the physical prediction model cannot predict the prediction target based on at least one set of control quantity parameters in the inferior solution domain, the inferior solution domain has no solution, the iteration is terminated, and the optimal set of control quantity parameters obtained in the previous iteration process is output.

[0137] In this embodiment, in other iteration processes except the first iteration process, all group control quantity parameters of this iteration process are divided into excellent solution domains and poor solution domains, and the solution is preferentially obtained from the high-quality solution domain, thereby reducing the solution process and improving the solution efficiency.

[0138] After obtaining the predicted target, execute step S332 to calculate the fitness value of each group of control parameters as the absolute value of the difference between the corresponding predicted target and the control target, and select the group of control parameters with the smallest fitness value as the optimal group of control parameters for this iterative process.

[0139] Specifically, the fitness value calculated in this embodiment is: fitness = abs(T_dest-T_pre), where abs is an absolute value function, T_pre is a predicted target obtained by the physical prediction model under the group of control quantities, T_dest is a control target, and fitness is a fitness value. By selecting a group of control quantity parameters with the smallest fitness value as the optimal group of control quantity parameters for this iteration process, a group of control quantity parameters with the closest prediction target to the control target is selected as the optimal group of control quantity parameters.

[0140] Then, step S333 is executed to first determine whether the iteration end condition is met. The iteration end condition is: the number of iterations is greater than or equal to the preset number threshold or the optimal group control amount parameter obtained in this iteration process is consistent with the optimal group control amount parameter obtained in the previous iteration process.

[0141] When the iteration end condition is met, the iteration ends and the optimal group control quantity parameters obtained in this iteration process are output.

[0142] If the iteration end condition is not met, the optimal group control amount parameters are used to perform genetic operations on the remaining group control amount parameters, and then the next iteration process is executed.

[0143] In one embodiment, after performing genetic operation on the remaining control amount parameters with the optimal group control amount parameters, executing the next iteration process specifically includes:

[0144] Replacing some control amount parameters in the remaining control amount parameters except the optimal group control amount parameters with corresponding control amount parameters in the optimal group control amount parameters to obtain the remaining control amount parameters after the crossover operation;

[0145] Multiplying the control amount parameters in the remaining groups of control amount parameters after the crossover operation by a preset variation coefficient to obtain the remaining groups of control amount parameters after the variation operation;

[0146] The next iteration process is performed with the remaining groups of control quantity parameters and the optimal group of control quantity parameters after the mutation operation.

[0147] Specifically, genetic operations include crossover operations and mutation operations.

[0148] Crossover operation: replace some control quantity parameters in the remaining groups of control quantity parameters other than the optimal group of control quantity parameters with corresponding control quantity parameters in the optimal group of control quantity parameters to obtain the remaining groups of control quantity parameters after the crossover operation.

[0149] Specifically, a crossover operation is performed on the remaining groups of control quantity parameters other than the optimal group of control quantity parameters, and a part of the control quantity parameters in the remaining groups of control quantity parameters are replaced with the corresponding control quantity parameters in the optimal group of control quantity parameters, and another part of the control quantity parameters in the remaining groups of control quantity parameters are retained.

[0150] Which part of the control quantity parameters are replaced with the corresponding control quantity parameters in the optimal group of control quantity parameters and which part of the control quantity parameters are retained can be preset or randomly set.

[0151] In some embodiments, the control quantity parameters perform a crossover operation based on the corresponding control quantities, and the crossover operation includes retaining the original control quantity parameters corresponding to some control quantities in the remaining groups of control quantity parameters, and replacing the control quantity parameters corresponding to other control quantities with the control quantity parameters of this type of control quantities in the optimal group of control quantity parameters.

[0152] Which control quantities use the original control quantity parameters and which control quantities use the control quantity parameters of the control quantities in the optimal group of control quantity parameters can be set in advance or randomly.

[0153] Mutation operation: multiply the control quantity parameters in the remaining groups of control quantity parameters by the preset coefficient of variation.

[0154] Specifically, the remaining groups of control quantity parameters other than the optimal group of control quantity parameters perform a variation operation, and the control quantity parameters in the remaining groups of control quantity parameters are multiplied by a preset variation coefficient.

[0155] Among them, the coefficient of variation can be preset.

[0156] This embodiment performs crossover and mutation operations on the control quantity parameters of the remaining groups except the optimal group control quantity parameters. On the one hand, the crossover operation is used to retain the control quantity parameters of some optimal group control quantity parameters in the remaining groups of control quantity parameters. On the other hand, the mutation operation is used to mutate the control quantity parameters of the remaining groups of control quantity parameters. The control is precise and fast. When the control target is not reached, the predicted target can be quickly approached to the control target. When the control target is reached, the genetic algorithm can be used to continuously and accurately approach the optimal solution, thereby maintaining the predicted target without overshoot or undershoot problems.

[0157] In one embodiment, multiplying the control amount parameters in the remaining groups of control amount parameters after the crossover operation by a preset coefficient of variation specifically includes:

[0158] The control quantity parameters in the remaining groups of control quantity parameters after the crossover operation are multiplied by the preset coefficient of variation of the control quantity corresponding to the control quantity parameters.

[0159] Specifically, each control quantity parameter is a value of a control quantity, so a preset coefficient of variation of the control quantity is obtained, and when performing a variation operation, the control quantity parameters in the remaining groups of control quantity parameters are multiplied by the preset coefficient of variation of the control quantity corresponding to the control quantity parameter.

[0160] In this embodiment, a suitable preset coefficient of variation is set according to the control amount corresponding to the control amount parameter, so that the control amount parameters of different control amounts can be multiplied by the preset coefficient of variation of a suitable size to achieve suitable variation.

[0161] After completing the crossover operation and the mutation operation, the optimal group control quantity parameters and the remaining group control quantity parameters are used as the control quantity parameters of each group in the next iteration process, and step S331 is continued until the iteration end condition is met, and the optimal group control quantity parameters are output.

[0162] For a control objective, the optimal group of control quantity parameters of the control objective is output through iterative solution, and then the optimal group of control quantity parameters of other control objectives are calculated using the same algorithm.

[0163] The optimal group of control quantity parameters corresponding to each control target are combined into an output control quantity sequence. Then, the output control quantity sequence is used to perform thermal management control on the vehicle.

[0164] In this embodiment, when a group of control quantity parameters has multiple control quantity parameters, a genetic algorithm is used to search for the optimal solution, and the genetic algorithm is optimized for the solver to further improve the solution efficiency.

[0165] like Figure 4 The flowchart of a vehicle thermal management control process according to the best embodiment of the present invention is shown, which includes: a main thread and a sub-thread for looping and calling a solver, wherein:

[0166] In the main thread:

[0167] Step S401, the working condition determination module updates the current working condition status once per second according to the actual vehicle conditions;

[0168] Step S402, the initialization module updates the data and parameters required by the solver once per second according to the current working state;

[0169] Step S403, the output module sequentially outputs a set of control quantity sequences given by the solver every second;

[0170] Step S404: output the control quantity sequence to control the actual vehicle.

[0171] In the child thread:

[0172] Step S405, each time the solver is called, it obtains the latest data from the initialization module of the main thread, calculates the control quantity sequence every second after 30 seconds, obtains 30 groups of control quantity sequences, and provides the 30 groups of control quantity sequences to the output module of the main thread.

[0173] The above steps are executed repeatedly until the algorithm exits.

[0174] Among them, the function of the solver is to use the physical prediction model of the thermal management model predictive control (MPC) to predict the optimal group of control quantity parameters of each controlled thermal management parameter after a preset time, wherein the optimal group of control quantity parameters are determined by the thermal management control quantity optimization method of the present invention. The optimal group of control quantity parameters are the control quantity parameters after the preset time, and the solver also outputs multiple intermediate control quantity parameters before the optimal control value. Among them, among the 30 groups of control quantity sequences in step S405, the first 29 groups of control quantity sequences are intermediate control quantity sequences containing intermediate control quantity parameters, and the 30th group of control quantity sequences are optimal control quantity sequences containing optimal group of control quantity parameters, that is, output control quantity sequences. When there is only one control target, the control quantity sequence includes a group of optimal group of control quantity parameters, and when there are multiple control targets, the control quantity sequence includes the optimal group of control quantity parameters of the controlled thermal management parameters corresponding to each control target.

[0175] In the main thread, the output module outputs a set of control quantity sequences every second. Since the calculation takes 3 seconds, the output module does not output the control quantity sequence in the first 3 seconds of the entire program. After 3 seconds, the output module of the main thread will continue to output the control quantity sequence.

[0176] The optimal group control quantity parameters are calculated iteratively. The iterative calculation of the optimal group control quantity parameters of each controlled thermal management parameter is as follows: Figure 5 As shown, including:

[0177] Step S501, obtaining the control target T_dest of the controlled thermal management parameter, and then setting the fitness value function: fitness = T_dest - T_pre, where T_pre is the predicted target of the controlled thermal management parameter predicted by the physical prediction model under the group of control quantity parameters.

[0178] As an example, if the controlled thermal management parameter is temperature, the prediction target is the predicted temperature.

[0179] Step S502, by using a random scattering strategy, 50 initial groups of control amount parameters CtrlVal_k are obtained, each control amount parameter in each group of control amount parameters CtrlVal_k is a random value obtained by a random function, and the control amount parameter is the value of the control amount.

[0180] As an example, the format of a set of control quantity parameters is [three-way valve opening, four-way valve opening, wptc power, battery water pump speed, fan speed...], among which the three-way valve opening, four-way valve opening, wptc power, battery water pump speed, and fan speed are the control quantities, and the specific values ​​of the three-way valve opening, four-way valve opening, wptc power, battery water pump speed, and fan speed are the control quantity parameters.

[0181] Step S503, input the 50 groups of control quantity parameters obtained by scattering into the physical prediction model in sequence for prediction calculation, and obtain the prediction target T_pre of the controlled thermal management parameter corresponding to the control quantity.

[0182] Step S504, calculating the fitness values ​​of 50 groups of control parameters, and finding a group of control parameters with the smallest fitness as the optimal group of control parameters CtrlVal_best.

[0183] Step S505, cross-operate the control quantity parameters of the CtrlVal_best group with the remaining 49 groups of control quantity parameters: some of the remaining 49 groups of control quantity parameters use the original control quantity values, and some control quantity parameters use the control quantity values ​​corresponding to the control quantity parameters of the CtrlVal_best group, and after the cross-operation, 49 new control quantity parameters CtrlVal_c are obtained.

[0184] Step S506, mutate the newly generated 49 groups of control quantity parameters CtrlVal_c: multiply each control quantity parameter in each group of control quantity parameters by a coefficient according to the corresponding control quantity, and after the mutation operation, obtain the new 49 groups of control quantity parameters and the optimal group of control quantity parameters as the new 50 groups of control quantity parameters CtrlVal_new.

[0185] Step S507, screening the new 50 groups of control amount parameters CtrlVal_new based on experience, and dividing them into excellent solution domains and inferior solution domains.

[0186] For example, in this solution, according to the current vehicle state, the three-way valve needs to be opened. If any group of control quantity parameters includes a three-way valve opening of 0, it is classified into the inferior solution domain, and other groups of control quantity parameters are classified into the excellent solution domain;

[0187] Step S508: If there is a solution in the excellent solution domain, only the fitness in the excellent solution domain is calculated, otherwise the fitness in the inferior solution domain is calculated.

[0188] There is a solution in the solution domain, and the physical prediction model can predict the prediction target based on at least one set of control quantity parameters in the solution domain.

[0189] Step S509: Find a group of control parameters with the minimum fitness from the excellent solution domain as the optimal group of control parameters CtrlVal_best.

[0190] Step S510, repeat steps S505-S509 until the found CtrlVal_best no longer changes, that is, the optimal solution is found, and the optimal solution is sent to the output module, and the current solver is completed.

[0191] This embodiment has accurate and fast control. When the control target is not reached, the predicted target can be quickly approached to the control target; when the control target is reached, the genetic algorithm can be used to continuously and accurately approach the optimal solution, thereby maintaining the predicted target without overshoot or undershoot. At the same time, by optimizing the genetic algorithm, the search efficiency is improved, thereby improving the efficiency of the solver.

[0192] Based on the same inventive concept, Figure 6 FIG. 1 is a schematic diagram of a thermal management control quantity optimization device according to an embodiment of the present invention, comprising:

[0193] A target acquisition module 601 is used to acquire the state data of the current working state of the vehicle and the control target, wherein the control target is the target that the controlled thermal management parameter is expected to achieve under the current working condition;

[0194] An initialization module 602 is used to initialize multiple groups of control quantity parameters, each group of control quantity parameters includes control quantity parameters of multiple control quantities;

[0195] The iteration module 603 is used to iteratively solve the optimal group control quantity parameters. In each iteration process:

[0196] Input the state data and multiple groups of control quantity parameters of this iteration process into the physical prediction model respectively, and predict multiple prediction targets, each of which is predicted by a group of control quantity parameters and the state data based on the physical prediction model;

[0197] Calculating the fitness value of each group of control quantity parameters based on the prediction target and the control target, and selecting a group of control quantity parameters with the smallest fitness value as the optimal group of control quantity parameters for this iteration process;

[0198] If the iteration end condition is met, the optimal group control amount parameters are output; otherwise, the optimal group control amount parameters are used to perform genetic operations on the remaining group control amount parameters, and then the next iteration process is executed.

[0199] The present invention solves the optimal group of control quantity parameters by iteratively solving multiple groups of control quantity parameters. In each iterative process, the state data and multiple groups of control quantity parameters of this iterative process are respectively input into the physical prediction model to predict multiple prediction targets. Then, the fitness value of each group of control quantity parameters is calculated based on the prediction target and the control target, and the group of control quantity parameters with the smallest fitness value is selected as the optimal group of control quantity parameters of this iterative process; if the iteration end condition is met, the optimal group of control quantity parameters is output, otherwise, the optimal group of control quantity parameters are genetically operated on the remaining groups of control quantity parameters, and the next iterative process is executed. When a group of control quantity parameters has multiple control quantity parameters, the present invention uses a genetic algorithm to search for the optimal solution, thereby improving the search efficiency and thus improving the efficiency of the solver.

[0200] In one embodiment, after performing genetic operation on the remaining control amount parameters with the optimal group control amount parameters, the next iteration process is performed, which specifically includes:

[0201] Replacing some control amount parameters in the remaining control amount parameters except the optimal group control amount parameters with corresponding control amount parameters in the optimal group control amount parameters to obtain the remaining control amount parameters after the crossover operation;

[0202] Multiplying the control amount parameters in the remaining groups of control amount parameters after the crossover operation by a preset variation coefficient to obtain the remaining groups of control amount parameters after the variation operation;

[0203] The next iteration process is performed with the remaining groups of control quantity parameters and the optimal group of control quantity parameters after the mutation operation.

[0204] In one embodiment, multiplying the control amount parameters in the remaining groups of control amount parameters after the crossover operation by a preset coefficient of variation specifically includes:

[0205] The control quantity parameters in the remaining groups of control quantity parameters after the crossover operation are multiplied by the preset coefficient of variation of the control quantity corresponding to the control quantity parameters.

[0206] In one embodiment, the state data and multiple groups of control quantity parameters of this iteration process are respectively input into a physical prediction model to predict multiple prediction targets, specifically including:

[0207] In the first iteration process, the state data and all group control quantity parameters of this iteration process are respectively input into the physical prediction model to predict multiple prediction targets; or

[0208] In the iteration process other than the first iteration process, all group control quantity parameters of this iteration process are divided into excellent solution domain and inferior solution domain;

[0209] Inputting the state data and one or more groups of control quantity parameters included in the excellent solution domain into a physical prediction model respectively, and if the physical prediction model can predict a prediction target based on at least one group of control quantity parameters in the excellent solution domain, predicting one or more prediction targets;

[0210] Otherwise, the state data and one or more groups of control quantity parameters included in the inferior solution domain are respectively input into the physical prediction model. If the physical prediction model can predict the prediction target based on at least one group of control quantity parameters in the inferior solution domain, one or more prediction targets are predicted. Otherwise, the iteration is terminated and the optimal group of control quantity parameters obtained in the last iteration process is output.

[0211] In one embodiment, if the iteration end condition is met, then the optimal group control amount parameter is output, specifically including:

[0212] If the number of iterations is greater than or equal to a preset number threshold or the optimal group control quantity parameters obtained in this iteration process are consistent with the optimal group control quantity parameters obtained in the previous iteration process, it is determined that the iteration end condition is met and the optimal group control quantity parameters are output.

[0213] In one embodiment, the initializing multiple groups of control amount parameters specifically includes:

[0214] A plurality of groups of control quantity parameters are randomly initialized, and each control quantity parameter of each group of control quantity parameters is a random value.

[0215] In one embodiment, the calculating of the fitness value of each group of control amount parameters based on the prediction target and the control target specifically includes:

[0216] The fitness value of each group of control quantity parameters is calculated as the absolute value of the difference between the corresponding prediction target and the control target.

[0217] Regarding the device in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0218] like Figure 7 The figure shows a hardware structure diagram of an electronic device of the present invention, including:

[0219] at least one processor 701; and,

[0220] A memory 702 that is communicatively connected to at least one of the processors 701; wherein,

[0221] The memory 702 stores instructions that can be executed by at least one of the processors. The instructions are executed by at least one of the processors to enable the at least one of the processors to perform the thermal management control quantity optimization method as described above.

[0222] Figure 7 A processor 701 is taken as an example.

[0223] The electronic device may further include: an input device 703 and a display device 704 .

[0224] The processor 701, the memory 702, the input device 703 and the display device 704 may be connected via a bus or other means, and the figure takes the connection via a bus as an example.

[0225] The memory 702 is a non-volatile computer-readable storage medium that can be used to store non-volatile software programs, non-volatile computer executable programs and modules, such as program instructions / modules corresponding to the thermal management control quantity optimization method in the embodiment of the present application, for example, Figure 2 , Figure 3 The processor 701 executes various functional applications and data processing by running the non-volatile software programs, instructions and modules stored in the memory 702, that is, the thermal management control amount optimization method in the above embodiment is implemented.

[0226] The memory 702 may include a program storage area and a data storage area, wherein the program storage area may store an operating system and an application required by at least one function; the data storage area may store data created according to the use of the thermal management control quantity optimization method, etc. In addition, the memory 702 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some embodiments, the memory 702 may optionally include a memory remotely arranged relative to the processor 701, and these remote memories may be connected to the device for executing the thermal management control quantity optimization method via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0227] The input device 703 can receive user clicks and generate signal inputs related to user settings and function control of the thermal management control quantity optimization method. The display device 704 can include display devices such as a display screen.

[0228] The one or more modules are stored in the memory 702 , and when executed by the one or more processors 701 , the thermal management control quantity optimization method in any of the above method embodiments is executed.

[0229] The present invention solves the optimal group of control quantity parameters by iteratively solving multiple groups of control quantity parameters. In each iterative process, the state data and multiple groups of control quantity parameters of this iterative process are respectively input into the physical prediction model to predict multiple prediction targets. Then, the fitness value of each group of control quantity parameters is calculated based on the prediction target and the control target, and the group of control quantity parameters with the smallest fitness value is selected as the optimal group of control quantity parameters of this iterative process; if the iteration end condition is met, the optimal group of control quantity parameters is output, otherwise, the optimal group of control quantity parameters are genetically operated on the remaining groups of control quantity parameters, and the next iterative process is executed. When a group of control quantity parameters has multiple control quantity parameters, the present invention uses a genetic algorithm to search for the optimal solution, thereby improving the search efficiency and thus improving the efficiency of the solver.

[0230] An embodiment of the present invention provides a storage medium, wherein the storage medium stores computer instructions. When a computer executes the computer instructions, the storage medium is used to execute all steps of the thermal management control quantity optimization method as described above.

[0231] In the context of the present disclosure, a storage medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or apparatus. The storage medium may be a machine-readable signal medium or a machine-readable storage medium. Alternatively, the storage medium may be a non-temporary computer-readable storage medium, for example, a non-temporary computer-readable storage medium may be a ROM, a random access memory (Random Access Memory, RAM), a compact disc read-only memory (Compact Disc ROM, CD-ROM), a magnetic tape, a floppy disk, and an optical data storage device, etc.

[0232] An embodiment of the present invention provides a vehicle, including the vehicle thermal management control device as described above, or the electronic device as described above. It is understandable that the vehicle may also include: a processor, a memory, and a computer program. The computer program is stored in the memory and is configured to be executed by the processor to implement the thermal management control quantity optimization method provided in the embodiment of the present disclosure. The processor and the memory are already in Figure 7 The parts described in the illustrated embodiment are not repeated here.

[0233] The above-mentioned embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. A method for optimizing thermal management control quantity, characterized in that: include: Acquire status data of the current working state of the vehicle and a control target, wherein the control target is a target that the controlled thermal management parameter is expected to achieve under the current working condition; Initializing multiple groups of control quantity parameters, each group of control quantity parameters includes control quantity parameters of multiple control quantities; Iteratively solve the optimal group control quantity parameters. In each iteration process: Input the state data and multiple groups of control quantity parameters of this iteration process into the physical prediction model respectively, and predict multiple prediction targets, each of which is predicted by a group of control quantity parameters and the state data based on the physical prediction model; Calculating the fitness value of each group of control quantity parameters based on the prediction target and the control target, and selecting a group of control quantity parameters with the smallest fitness value as the optimal group of control quantity parameters for this iteration process; If the iteration end condition is met, the optimal group control amount parameters are output; otherwise, the optimal group control amount parameters are used to perform genetic operations on the remaining group control amount parameters, and then the next iteration process is executed.

2. The thermal management control quantity optimization method according to claim 1, characterized in that: After the genetic operation is performed on the remaining control parameters by using the optimal control parameter, the next iteration process is performed, which specifically includes: Replacing some control amount parameters in the remaining control amount parameters except the optimal group control amount parameters with corresponding control amount parameters in the optimal group control amount parameters to obtain the remaining control amount parameters after the crossover operation; Multiplying the control amount parameters in the remaining groups of control amount parameters after the crossover operation by a preset variation coefficient to obtain the remaining groups of control amount parameters after the variation operation; The next iteration process is performed with the remaining groups of control quantity parameters and the optimal group of control quantity parameters after the mutation operation.

3. The thermal management control quantity optimization method according to claim 2, characterized in that: The step of multiplying the control amount parameters in the remaining groups of control amount parameters after the crossover operation by a preset coefficient of variation specifically includes: The control quantity parameters in the remaining groups of control quantity parameters after the crossover operation are multiplied by the preset coefficient of variation of the control quantity corresponding to the control quantity parameters.

4. The thermal management control quantity optimization method according to claim 1, characterized in that: The state data and multiple groups of control quantity parameters of this iteration process are respectively input into the physical prediction model to predict multiple prediction targets, specifically including: In the first iteration process, the state data and all group control quantity parameters of this iteration process are respectively input into the physical prediction model to predict multiple prediction targets; or In the iteration process other than the first iteration process, all group control quantity parameters of this iteration process are divided into excellent solution domain and inferior solution domain; Inputting the state data and one or more groups of control quantity parameters included in the excellent solution domain into a physical prediction model respectively, and if the physical prediction model can predict a prediction target based on at least one group of control quantity parameters in the excellent solution domain, predicting one or more prediction targets; Otherwise, the state data and one or more groups of control quantity parameters included in the inferior solution domain are respectively input into the physical prediction model. If the physical prediction model can predict the prediction target based on at least one group of control quantity parameters in the inferior solution domain, one or more prediction targets are predicted. Otherwise, the iteration is terminated and the optimal group of control quantity parameters obtained in the last iteration process is output.

5. The thermal management control quantity optimization method according to claim 1, characterized in that: If the iteration end condition is met, the optimal group control amount parameter is output, which specifically includes: If the number of iterations is greater than or equal to a preset number threshold or the optimal group control quantity parameters obtained in this iteration process are consistent with the optimal group control quantity parameters obtained in the previous iteration process, it is determined that the iteration end condition is met and the optimal group control quantity parameters are output.

6. The method for optimizing the thermal management control amount according to any one of claims 1 to 5, characterized in that: The initialization of multiple groups of control quantity parameters specifically includes: A plurality of groups of control quantity parameters are randomly initialized, and each control quantity parameter of each group of control quantity parameters is a random value.

7. The method for optimizing the thermal management control amount according to any one of claims 1 to 5, characterized in that: The calculating of the fitness value of each group of control amount parameters based on the prediction target and the control target specifically includes: The fitness value of each group of control quantity parameters is calculated as the absolute value of the difference between the corresponding prediction target and the control target.

8. A thermal management control quantity optimization device, characterized in that: include: A target acquisition module is used to acquire the status data of the current working state of the vehicle and the control target, wherein the control target is the target that the controlled thermal management parameter is expected to achieve under the current working condition; An initialization module, used for initializing multiple groups of control quantity parameters, each group of control quantity parameters including control quantity parameters of multiple control quantities; Iteration module is used to iteratively solve the optimal group control quantity parameters. In each iteration process: Input the state data and multiple groups of control quantity parameters of this iteration process into the physical prediction model respectively, and predict multiple prediction targets, each of which is predicted by a group of control quantity parameters and the state data based on the physical prediction model; Calculating the fitness value of each group of control quantity parameters based on the prediction target and the control target, and selecting a group of control quantity parameters with the smallest fitness value as the optimal group of control quantity parameters for this iteration process; If the iteration end condition is met, the optimal group control amount parameters are output; otherwise, the optimal group control amount parameters are used to perform genetic operations on the remaining group control amount parameters, and then the next iteration process is executed.

9. An electronic device, characterized in that: include: at least one processor; as well as, a memory communicatively connected to at least one of the processors; wherein, The memory stores instructions that can be executed by at least one of the processors, and the instructions are executed by at least one of the processors to enable at least one of the processors to perform the thermal management control amount optimization method according to any one of claims 1 to 7.

10. A storage medium, characterized in that: The storage medium stores computer instructions, and when a computer executes the computer instructions, it is used to execute all steps of the thermal management control quantity optimization method according to any one of claims 1 to 7.

11. A vehicle, characterized in that: It includes the thermal management control quantity optimization device as claimed in claim 8, or the electronic device as claimed in claim 9.