Vehicle thermal management control method and device, vehicle, electronic equipment and storage medium

By using prediction models based on genetic algorithms and long and short-term memory networks in the thermal management system of new energy vehicles, the problems of reaction lag, temperature overshoot and high energy consumption are solved, and more refined thermal management control and more efficient energy are achieved. use.

CN118927926APending Publication Date: 2024-11-12DEEPAL AUTOMOBILE TECH CO LTD
View PDF 0 Cites 2 Cited by

Patent Information

Application Number
CN202411134239.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-19
Publication Date
2024-11-12

AI Technical Summary

Technical Problem

The range problem of the thermal management system of new energy vehicles at low temperatures and the problems of lag in response, overshooting of temperature and excessive energy consumption in traditional thermal management systems.

Method used

Using a long and short-term memory network prediction model trained based on historical monitoring data and genetic algorithms, the operation of the future vehicle thermal management system is predicted by obtaining current monitoring data and ideal control data, and real-time adjustments are made based on the prediction results and current control data, so as to Optimize thermal management strategies.

Benefits of technology

It improves the control precision of the vehicle thermal management system, reduces temperature overshoot and energy consumption, and improves the system's response speed and efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118927926A_ABST
    Figure CN118927926A_ABST
Patent Text Reader

Abstract

The invention provides a vehicle thermal management control method and device, a vehicle, electronic equipment and a storage medium, and the method comprises the steps: obtaining current monitoring data and ideal control data of a vehicle thermal management system, the current monitoring data comprising current control data and current state data; inputting the current monitoring data into the heat management prediction model for prediction to obtain prediction control data; determining actual control data according to the prediction control data, the ideal control data and the current control data so as to control the vehicle thermal management system; wherein the thermal management prediction model is obtained by training a long short-term memory network based on historical monitoring data and a genetic algorithm, and the vehicle thermal management system comprises at least one of a passenger compartment thermal management system, an electric drive thermal management system and a battery thermal management system; according to the invention, the operation condition of the vehicle thermal management system can be accurately predicted through the thermal management prediction model, so that the control precision of the vehicle thermal management system is improved, and temperature overshoot and energy consumption are reduced.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of vehicle thermal management technology, and in particular to a vehicle thermal management control method, device, vehicle, electronic equipment and storage medium. Background Art

[0002] In recent years, the new energy vehicle industry, led by pure electric vehicles, has been developing rapidly, and it will be an inevitable trend for new energy vehicles to replace fuel vehicles. However, the thermal management system of traditional vehicles is not suitable for new energy vehicles. The thermal management system of new energy vehicles includes multiple systems such as passenger compartment, battery, motor, and electronic control, and the multiple systems affect each other, which puts higher requirements on the control strategy. In addition, the range of electric vehicles at low temperatures also poses a more severe challenge to the thermal management system of the whole vehicle.

[0003] The thermal management system mainly relies on the coordinated action of actuators such as water pumps, fans, compressors, and valve bodies to control parameters such as the temperature of the three-electric system, the temperature of the cockpit, and the energy consumption of the system. This is a complex multi-input and multi-output problem. At present, the commonly used control methods include switch control and proportional-integral-derivative control (PID). The principle of switch control is simple. By comparing the input value and the set value at the current moment, the start and stop of the actuator are controlled. It is mainly used for components such as fans and valve bodies, but frequent start and stop will affect the life of components and driving safety. In contrast, PID control can provide more precise control, automatically adjusting the output speed of the cooling component according to the difference between the actual output value and the set target value, effectively reducing the energy consumption of the system, but there is a problem of untimely response to changes in working conditions. The comprehensive PID feedback control and feedforward control method can improve the control of the cooling system, but the system responsiveness speed still has defects, and more intelligent predictive control needs to be considered.

[0004] The operation of the vehicle thermal management system is affected by many factors such as the external environment, vehicle performance, and driving status, and has the characteristics of instantaneous and dynamic changes. Especially when the temperature and working conditions change greatly, the current control method has some problems, such as delayed response, temperature overshoot, and excessive energy consumption.

[0005] For example, CN117673580B discloses a temperature control device for new energy batteries, which relates to the field of battery pack management technology, wherein the device includes a data collector and a controller; the data collector is used to collect the operating condition data of each battery cell in the new energy battery pack, and combine multiple historical operating condition data to form battery condition time series data, and construct the input data of the corresponding battery cell according to the battery condition time series data and the spatial position data of the battery cell; based on the input data corresponding to each battery cell, an input data sequence is constructed; the controller is used to use the spatiotemporal recurrent neural network to predict the thermal management risk probability of each battery cell corresponding to the future time period, and determine the corresponding temperature control strategy according to the thermal management risk probability. This scheme determines the temperature adjustment direction based on the prediction results of the spatiotemporal recurrent neural network, and then determines the parameter adjustment amplitude according to the genetic algorithm, but there is still a problem of delayed response, which leads to temperature overshoot and excessive energy consumption. Summary of the invention

[0006] The present invention provides a vehicle thermal management control method, device, vehicle, electronic equipment and storage medium to solve the technical problems of delayed response, temperature overshoot and excessive energy consumption in the control of the above thermal management system.

[0007] In one embodiment of the present application, the present application provides a vehicle thermal management control method, including: obtaining current monitoring data and ideal control data of a vehicle thermal management system, the current monitoring data including current control data and current state data; inputting the current monitoring data into a thermal management prediction model for prediction to obtain predicted control data; determining actual control data based on the predicted control data, the ideal control data and the current control data to control the vehicle thermal management system; wherein the thermal management prediction model is obtained by training a long short-term memory network based on historical monitoring data and a genetic algorithm, and the vehicle thermal management system includes at least one of a passenger compartment thermal management system, an electric drive thermal management system and a battery thermal management system.

[0008] In one embodiment of the present application, before inputting the current monitoring data into the thermal management prediction model for prediction, the method also includes: obtaining historical monitoring data of the vehicle thermal management system; performing cluster analysis on the historical monitoring data to obtain cluster monitoring data; generating an initial population based on the cluster monitoring data and preset thermal management system constraint relationships, wherein the initial population is used to characterize different network configurations in the long short-term memory network; and iteratively training the long short-term memory network based on the initial population and the historical monitoring data to obtain a thermal management prediction model.

[0009] In one embodiment of the present application, an initial population is generated based on the cluster monitoring data and the preset thermal management system constraint relationship, including: randomly generating initial monitoring data based on the cluster monitoring data and the preset thermal management system constraint relationship; solving the relationship of the initial monitoring data according to the ant colony algorithm to obtain a set of random solutions; and converting the set of random solutions into an initial population corresponding to the genetic algorithm.

[0010] In one embodiment of the present application, the long short-term memory network is iteratively trained according to the initial population and the historical monitoring data to obtain a thermal management prediction model, including: dividing the historical monitoring data into training set data and test set data according to a preset ratio; inputting the training set data into the long short-term memory network corresponding to each initial individual in the initial population for network training to obtain the training loss function value and network parameters of each initial individual; determining the long short-term memory network corresponding to the optimal individual and the optimal parameters as the initial prediction model, and verifying the initial prediction model according to the test set data to obtain the verification loss function value, the optimal individual is obtained based on the initial individual corresponding to the minimum training loss function value, and the optimal parameters are used to characterize the network parameters corresponding to the optimal individual; if the genetic iteration meets the preset iteration condition, the initial prediction model corresponding to the minimum verification loss function value in all genetic iterations is used as the thermal management prediction model; if the genetic iteration does not meet the preset iteration condition, the initial prediction model corresponding to the minimum verification loss function value in all genetic iterations is used as the thermal management prediction model. The initial individuals are selected, crossed and mutated according to the sorting to obtain the next generation population, the next generation population is used as the initial population, and the initial prediction model is used as the long short-term memory network, and the training set data is repeatedly input into the long short-term memory network corresponding to each initial individual in the initial population for network training to obtain the training loss function value and network parameters of each initial individual, the long short-term memory network corresponding to the optimal individual and the optimal parameter is determined as the initial prediction model, and the initial prediction model is verified according to the test set data to obtain the verification loss function value, if the genetic iteration does not meet the preset iteration condition, then the initial individuals are selected, crossed and mutated according to the sorting of all training loss function values ​​to obtain the next generation population, the next generation population is used as the initial population, and the initial prediction model is used as the long short-term memory network step, until the genetic iteration meets the preset iteration condition, and the initial prediction model corresponding to the minimum verification loss function value in all genetic iterations is used as the thermal management prediction model.

[0011] In one embodiment of the present application, the training set data is input into a long short-term memory network corresponding to an initial individual for network training to obtain a training loss function value and network parameters of an initial individual, including: inputting the training set data in batches into a long short-term memory network corresponding to an initial individual for network training to obtain network parameters and prediction set data that meet preset training conditions; and using the mean square error between the training set data and the prediction set data as the training loss function value of an initial individual.

[0012] In one embodiment of the present application, actual control data is determined based on the predicted control data, the ideal control data and the current control data, including: correcting the predicted control data based on a first difference between the predicted control data and the current control data to obtain corrected control data; and determining the actual control data based on a second difference between the corrected control data and the ideal control data.

[0013] In one embodiment of the present application, the predictive control data includes at least one of a predicted motor temperature set corresponding to the electric drive thermal management, a predicted battery temperature corresponding to the battery thermal management, a predicted passenger compartment temperature corresponding to the passenger compartment thermal management, and a predicted air volume.

[0014] In one embodiment of the present application, the present application provides a vehicle thermal management control device, including: a data acquisition module, used to acquire current monitoring data and ideal control data of a vehicle thermal management system, wherein the current monitoring data includes current control data and current status data; a control prediction module, used to input the current monitoring data into a thermal management prediction model for prediction to obtain predicted control data; a control adjustment module, used to determine actual control data based on the predicted control data, the ideal control data and the current control data to control the vehicle thermal management system; wherein the thermal management prediction model is obtained by training a long short-term memory network based on historical monitoring data and a genetic algorithm, and the vehicle thermal management system includes at least one of a passenger compartment thermal management system, an electric drive thermal management system and a battery thermal management system.

[0015] In one embodiment of the present application, the present application provides a vehicle including: a vehicle thermal management control device as described in the above embodiment.

[0016] In one embodiment of the present application, the present application provides an electronic device, which includes: one or more processors; a storage device for storing one or more programs, when the one or more programs are executed by the one or more processors, the electronic device implements the vehicle thermal management control method as described in any of the above embodiments.

[0017] In one embodiment of the present application, the present application provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor of a computer, the computer executes the vehicle thermal management control method described in any of the above embodiments.

[0018] Beneficial effects of the present invention: The present invention provides a vehicle thermal management control method, device, vehicle, electronic device and storage medium. The present invention obtains predictive control data by inputting the current monitoring data into a thermal management prediction model for prediction. The thermal management prediction model is obtained based on the global optimization capability of the genetic algorithm and the powerful modeling capability of LSTM for time series data. It can accurately predict the operation of the vehicle thermal management system, thereby improving the control precision of the vehicle thermal management system, reducing temperature overshoot and reducing energy consumption.

[0019] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The drawings herein are incorporated into the specification and constitute a part of the specification, showing embodiments consistent with the present application, and together with the specification, are used to explain the principles of the present application. Obviously, the drawings described below are only some embodiments of the present application, and for those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0021] Figure 1 A schematic diagram showing an exemplary system architecture to which the technical solution of the embodiments of the present application can be applied;

[0022] Figure 2 A schematic flow chart of a vehicle thermal management control method according to an embodiment of the present application is shown;

[0023] Figure 3 A schematic diagram of the internal structure of an LSTM according to an embodiment of the present application is shown;

[0024] Figure 4 A schematic diagram of a GA-LSTM model training according to an embodiment of the present application is shown;

[0025] Figure 5 A schematic diagram showing the principle of model predictive control according to an embodiment of the present application is shown;

[0026] Figure 6 A block diagram of a vehicle thermal management control device according to an embodiment of the present application is shown;

[0027] Figure 7 A schematic diagram of the structure of a computer system suitable for implementing an electronic device of an embodiment of the present application is shown. DETAILED DESCRIPTION

[0028] The following describes the embodiments of the present application through specific examples, and those skilled in the art can easily understand other advantages and effects of the present application from the contents disclosed in this specification. The present application can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present application. It should be noted that the following embodiments and features in the embodiments can be combined with each other without conflict.

[0029] It should be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present application, and thus the drawings only show components related to the present application rather than being drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component may be changed at will, and the component layout may also be more complicated.

[0030] In the following description, a large number of details are discussed to provide a more thorough explanation of the embodiments of the present application. However, it is obvious to those skilled in the art that the embodiments of the present application can be implemented without these specific details. In other embodiments, well-known structures and devices are shown in the form of block diagrams rather than in detail to avoid making the embodiments of the present application difficult to understand.

[0031] First of all, it should be noted that electric vehicle thermal management is crucial to the performance and safety of the vehicle. Traditional thermal management methods often rely on empirical rules or simple control strategies, which are difficult to effectively adapt to the changing environment and vehicle operating conditions. In recent years, with the rapid development of artificial intelligence and machine learning technologies, the Long Short-Term Memory (LSTM) network in neural networks has been widely used in time series prediction and control tasks. However, traditional LSTM models may have challenges such as slow convergence and local optimal solution problems.

[0032] See also Figure 1 , Figure 1 A schematic diagram of an exemplary system architecture to which the technical solution of the embodiment of the present application can be applied is shown. Figure 1 As shown, the system architecture may include a vehicle thermal management system 101 and a computer device 102. The computer device 102 may be at least one of a general-purpose computer, a microcontroller, a neural network computer, etc. The vehicle thermal management system 101 is used to transmit current monitoring data and ideal control data to the computer device 102, and the computer device 102 is used to perform control prediction in the future time domain based on the current monitoring data and the ideal control data, thereby obtaining actual control data.

[0033] Exemplarily, after the computer device 102 obtains the current monitoring data and ideal control data of the vehicle thermal management system 101, the current monitoring data is input into the thermal management prediction model for prediction to obtain predicted control data; the actual control data is determined based on the predicted control data, the ideal control data and the current control data to control the vehicle thermal management system; wherein the thermal management prediction model is obtained based on the training of the long short-term memory network based on historical monitoring data and genetic algorithms, the vehicle thermal management system includes at least one of the passenger compartment thermal management system, the electric drive thermal management system and the battery thermal management system, and the current monitoring data includes current control data and current status data.

[0034] In the related art, especially when the temperature and working conditions vary greatly, the current control method has problems such as delayed response, temperature overshoot and excessive energy consumption.

[0035] In order to solve the above technical problems, the present application provides a vehicle thermal management control method, device, vehicle, electronic device and storage medium. The implementation details of the technical solution of the embodiment of the present application are elaborated in detail below.

[0036] See also Figure 2 , Figure 2 FIG. 1 is a flow chart of a vehicle thermal management control method according to an embodiment of the present application. Figure 2 As shown, in an exemplary embodiment, the vehicle thermal management control method includes at least steps S210 to S230, which are described in detail as follows:

[0037] Step S210, obtaining current monitoring data and ideal control data of the vehicle thermal management system.

[0038] Among them, the current monitoring data includes current control data and current status data, and the vehicle thermal management system includes at least one of a passenger compartment thermal management system, an electric drive thermal management system and a battery thermal management system.

[0039] In one embodiment of the present application, the vehicle thermal management system of a pure electric vehicle is divided into three parts: a passenger compartment thermal management system, an electric drive thermal management system, and a battery thermal management system, and the current monitoring data required by each part is sorted out.

[0040] In one embodiment of the present application, the current monitoring data corresponding to the passenger compartment thermal management system includes the current passenger compartment temperature, the current ambient temperature, the current compressor speed, the current fan speed, the current vehicle speed, the current electronic expansion valve (EXV) opening, the current positive temperature coefficient (PTC) heater power, and the current blower speed, so as to output the predicted passenger compartment temperature and predicted air volume in the future time domain.

[0041] In one embodiment of the present application, the current monitoring data corresponding to the battery thermal management system includes the current battery temperature, the current battery SOC, the current vehicle speed, the current compressor speed, the current PTC heater power, and the battery cooling water pump speed, so as to output the predicted battery temperature in the future time domain.

[0042] In one embodiment of the present application, the current monitoring data corresponding to the electric drive thermal management system includes the current DC-DC converter (DCDC) temperature, the current charger temperature, the current DC-AC converter (DCAC) temperature, the current motor stator temperature, the current motor IGBT temperature, the current electric drive cooling water pump speed, and the current fan speed, so as to output the predicted motor temperature set in the future time domain. Among them, the predicted motor temperature set includes the predicted charger temperature, the predicted motor stator temperature, and the predicted motor IGBT temperature.

[0043] In one embodiment of the present application, the control data is used to represent data that needs to be predicted in the future time domain, such as the current control data includes the current passenger compartment temperature, and the current state data includes the current ambient temperature.

[0044] Step S220, inputting the current monitoring data into the thermal management prediction model for prediction to obtain prediction control data.

[0045] Among them, the thermal management prediction model is obtained based on historical monitoring data and genetic algorithm training of long short-term memory network.

[0046] In one embodiment of the present application, before inputting the current monitoring data into the thermal management prediction model for prediction, the method also includes: obtaining historical monitoring data of the vehicle thermal management system; performing cluster analysis on the historical monitoring data to obtain cluster monitoring data; generating an initial population based on the cluster monitoring data and preset thermal management system constraint relationships, the initial population being used to characterize different network configurations in the long short-term memory network; iteratively training the long short-term memory network based on the initial population and the historical monitoring data to obtain a thermal management prediction model.

[0047] In one embodiment of the present application, the type of historical monitoring data is the same as the type of current monitoring data shown in the above embodiments.

[0048] In one embodiment of the present application, the data sources of historical monitoring data are controller area network (CAN) data, vehicle terminal equipment (Telematics Box, TBOX) data, sensor data, and controller data during vehicle testing, experimentation and driving.

[0049] In one embodiment of the present application, all historical monitoring parameters in the historical monitoring data are normalized to achieve data enhancement and prevent model overfitting. The normalization is performed by the Min-Max method. For any historical monitoring parameter x before normalization, the normalized historical monitoring parameter x1 is shown in formula (1):

[0050]

[0051] Among them, x1 is the historical monitoring parameter after normalization, x is the historical monitoring parameter before normalization, and x min Indicates the minimum value of all historical monitoring parameters, x max Indicates the maximum value of all historical monitoring parameters.

[0052] In one embodiment of the present application, an initial population is generated based on clustered monitoring data and preset thermal management system constraint relationships, including: randomly generating initial monitoring data based on clustered monitoring data and preset thermal management system constraint relationships; solving the relationship of the initial monitoring data based on an ant colony algorithm to obtain a set of random solutions; and converting a set of random solutions into an initial population corresponding to a genetic algorithm.

[0053] In one embodiment of the present application, an initial population is randomly generated by cluster analysis of historical monitoring data, wherein the initial population includes 200 initial individuals.

[0054] In one embodiment of the present application, the network configuration includes the number of LSTM layers, the dropout ratio, the hyperparameters of the Adam optimizer, the learning rate, and the number of units of the LSTM layer.

[0055] Iteratively train the long short-term memory network according to the initial population and historical monitoring data to obtain a thermal management prediction model, including: dividing the historical monitoring data into training set data and test set data according to a preset ratio; inputting the training set data into the long short-term memory network corresponding to each initial individual in the initial population for network training to obtain the training loss function value and network parameters of each initial individual; determining the long short-term memory network corresponding to the optimal individual and the optimal parameters as the initial prediction model, and verifying the initial prediction model according to the test set data to obtain the verification loss function value, the optimal individual is obtained based on the initial individual corresponding to the minimum training loss function value, and the optimal parameters are used to characterize the network parameters corresponding to the optimal individual; if the genetic iteration meets the preset iteration conditions, the initial prediction model corresponding to the minimum verification loss function value in all genetic iterations is used as the thermal management prediction model; if the genetic iteration does not meet the preset iteration conditions, the initial prediction model corresponding to the minimum verification loss function value in all genetic iterations is sorted according to the order of all training loss function values. The initial individuals are selected, crossed and mutated to obtain the next generation population, the next generation population is used as the initial population, and the initial prediction model is used as the long short-term memory network. The training set data is repeatedly input into the long short-term memory network corresponding to each initial individual in the initial population for network training to obtain the training loss function value and network parameters of each initial individual, the long short-term memory network corresponding to the optimal individual and the optimal parameters is determined as the initial prediction model, and the initial prediction model is verified according to the test set data to obtain the verification loss function value. If the genetic iteration does not meet the preset iteration condition, the initial individuals are selected, crossed and mutated according to the ranking of all training loss function values ​​to obtain the next generation population, the next generation population is used as the initial population, and the initial prediction model is used as the long short-term memory network step until the genetic iteration meets the preset iteration condition, and the initial prediction model corresponding to the minimum verification loss function value in all genetic iterations is used as the thermal management prediction model.

[0056] In one embodiment of the present application, data is sorted and divided, and historical monitoring data is divided into training set data and test set data according to a preset ratio of 3:1 to perform offline training and verify the predictive performance of the model.

[0057] In one embodiment of the present application, LSTM is the earliest proposed recurrent neural network gating algorithm, which has the ability to store data at long time intervals. Its memory unit contains three gating mechanisms: input gate, forget gate and output gate, any of which can be regarded as a traditional artificial neuron, and the total weight is calculated through the activation function. During the data processing process, the input gate is responsible for the data that needs to be saved to the unit state at the current moment, and the forget gate selects and memorizes the input information and retains the important part. The output gate controls the output content.

[0058] In one embodiment of the present application, see Figure 3 , Figure 3FIG. 4 shows a schematic diagram of the internal structure of an LSTM according to an embodiment of the present application. Figure 3 As shown, according to the structure of LSTM, its internal calculation logic is shown in equations (2) to (7), where f t For the forget gate, i t is the input gate, o t is the output gate, C t-1 is the unit state at the previous moment, is the candidate unit state, C t is the current unit state, h t is the output of the current unit, h t-1 is the output of the unit at the previous moment, x t is the input of the current layer.

[0059] In one embodiment of the present application, the forget gate is as shown in formula (2):

[0060] f t =σ(W f ·[h t-1 ,x t ]+b f ) Formula (2)

[0061] Among them, f t is the forget gate, σ is the Sigmoid activation function, W f is the forget weight, h t-1 is the output of the unit at the previous moment, x t is the input of the current layer, b f is the forget bias term.

[0062] In one embodiment of the present application, the input gate is as shown in formula (3):

[0063] i t =σ(W i ·[h t-1 ,x t ]+b i ) Formula (3)

[0064] Among them, i t is the input gate, σ is the Sigmoid activation function, W i is the input weight, h t-1 is the output of the unit at the previous moment, x t is the input of the current layer, b i is the input bias term.

[0065] In one embodiment of the present application, the output gate is as shown in formula (4):

[0066] o t =σ(W o ·[ht-1 ,x t ]+b o ) Formula (4)

[0067] Among them, t is the output gate, σ is the Sigmoid activation function, W o is the output weight, h t-1 is the output of the unit at the previous moment, x t is the input of the current layer, b o is the output bias term.

[0068] In one embodiment of the present application, the candidate unit state is as shown in formula (5):

[0069]

[0070] in, is the candidate unit state, tanh is the hyperbolic tangent activation function, W c is the current state weight, h t-1 is the output of the unit at the previous moment, x t is the input of the current layer, b c is the current state bias.

[0071] In one embodiment of the present application, C t The current unit state is shown in formula (6):

[0072]

[0073] Among them, C t is the current unit state, f t is the forget gate, C t-1 is the unit state at the previous moment, i t is the input gate, The candidate unit status.

[0074] In one embodiment of the present application, the output of the current unit is as shown in formula (7):

[0075] h t =o t ×tanh(C t ) Formula (7)

[0076] Among them, h t is the output of the current unit, o t is the output gate, tanh is the hyperbolic tangent activation function, C t The current unit status.

[0077] In one embodiment of the present application, the network parameters include a plurality of weight parameters and a plurality of bias item parameters. The weight parameters include a forgetting weight, an input weight, a current state weight, and an output weight; and the bias item parameters include a forgetting bias item, an input bias item, a current state bias item, and an output bias item.

[0078] The training set data is input into a long short-term memory network corresponding to an initial individual for network training to obtain a training loss function value and network parameters of the initial individual, including: inputting the training set data in batches into a long short-term memory network corresponding to an initial individual for network training to obtain network parameters and prediction set data that meet preset training conditions; and using the mean square error between the training set data and the prediction set data as the training loss function value of the initial individual.

[0079] In one embodiment of the present application, the matching degree between multiple training prediction values ​​and corresponding training true values ​​is evaluated by the training loss function value, the training true values ​​are obtained based on the training set data, and the training prediction values ​​are obtained based on the prediction set data.

[0080] In one embodiment of the present application, the training loss function value is obtained based on the mean square error (MSE), as shown in formula (8):

[0081]

[0082] Among them, MSE is the mean square error, n is the number of data in the training set, and Y i is the i-th training true value in the training set data, is the i-th training prediction value in the prediction set data.

[0083] In one embodiment of the present application, the validation loss function value is obtained based on the mean square error between the validation set data and the validation prediction data. The validation prediction data is obtained by validating the initial prediction model based on the test set data.

[0084] In one embodiment of the present application, the LSTM training process sets the number of iterations (epoch) to 200, the initial learning rate lr to 0.001, the Adam optimizer is used to calculate the gradient, and the parameter betas of the Adam optimizer is (0.9, 0.999). The batch size (Batchsize) of the training and testing processes is 4.

[0085] In one embodiment of the present application, the training set data is input into the LSTM in batches, the loss between the training prediction value and the training true value in the LSTM is calculated, and the loss is used as the fitness of the individual in the genetic algorithm, and the network parameters of the LSTM are iteratively updated through the Adam optimization algorithm. Through the three processes of selection, crossover and mutation in the genetic algorithm, the next generation population is generated to obtain a new network configuration.

[0086] In one embodiment of the present application, the network configuration includes the number of LSTM layers, the dropout ratio, the hyperparameters of the Adam optimizer, the learning rate, and the number of units of the LSTM layer.

[0087] In one embodiment of the present application, the network parameters include multiple weight parameters and multiple bias item parameters.

[0088] In one embodiment of the present application, selection is used to select excellent individuals from the current population so that they have the opportunity to serve as the parent population for the next generation population update and iteration; crossover is to randomly pair individuals in the group to obtain a new generation of individuals that combine the characteristics of the parents, which can promote population diversity and accelerate convergence; mutation is to change one or more individuals with a lower probability, which can generate new individuals to increase the diversity and search space of the population and prevent falling into a local optimal solution.

[0089] In one embodiment of the present application, based on the training model of the previous LSTM, the LSTM continues to be trained until the number of iterations reaches a preset iteration threshold.

[0090] In one embodiment of the present application, the performance of the initial prediction model obtained in each iteration on the test set data is observed, and the initial prediction model obtained with the optimal number of iterations is selected as the final model, thereby obtaining a thermal management prediction model.

[0091] In one embodiment of the present application, an optimal GA-LSTM model is used to output prediction results, combined with an effective control strategy to achieve accurate predictive control of the vehicle thermal management system.

[0092] In one embodiment of the present application, see Figure 4 , Figure 4 FIG. 4 shows a schematic diagram of a GA-LSTM model training according to an embodiment of the present application. Figure 4As shown in the figure, obtain the data set: obtain the historical monitoring data; generate the initial population: perform cluster analysis on the historical monitoring data to obtain the cluster monitoring data, and generate the initial population according to the cluster monitoring data and the preset thermal management system constraint relationship corresponding to the vehicle thermal management system; LSTM neural network training: input the training set data into the long short-term memory network corresponding to each initial individual in the initial population for network training, obtain the training loss function value and network parameters of each initial individual, and determine the long short-term memory network corresponding to the optimal individual and the optimal parameters as the initial prediction model; calculate the fitness: use the training loss function value of each initial individual as the fitness of each individual; whether the iteration threshold is reached: if the genetic iteration does not meet the preset iteration condition, then enter the selection step; if the genetic iteration meets the preset iteration condition, then enter the step of obtaining the optimal GA-LSTM model; selection: select according to the ranking of all fitness Select a preset first number of initial individuals as multiple direct individuals; crossover: pair a preset second number of initial individuals according to the ranking of all fitnesses to obtain multiple new generation individuals; mutation: mutate a preset third number of initial individuals according to the ranking of all fitnesses to obtain multiple mutant individuals; generate the next generation population: take each direct individual, each new generation individual and each mutant individual as the next generation population, and take the next generation population as the initial population, and re-enter the LSTM neural network training step; obtain the optimal GA-LSTM model: take the initial prediction model corresponding to the minimum verification loss function value in all genetic iterations as the thermal management prediction model, and the verification loss function value is obtained based on the verification of the initial training model corresponding to each generation of genetics; output the prediction result: output the prediction control data to the vehicle thermal management system according to the thermal management prediction model to determine the actual control data.

[0093] In one embodiment of the present application, the prediction accuracy of the thermal management prediction model and the control effect of the vehicle thermal management system are improved by combining the global optimization capability of the genetic algorithm and the sequence modeling capability of the LSTM neural network.

[0094] Step S230, determining actual control data according to the predicted control data, the ideal control data and the current control data to control the vehicle thermal management system.

[0095] In one embodiment of the present application, actual control data is determined based on predicted control data, ideal control data and current control data, including: correcting the predicted control data based on a first difference between the predicted control data and the current control data to obtain corrected control data; and determining the actual control data based on a second difference between the corrected control data and the ideal control data.

[0096] In one embodiment of the present application, see Figure 5 , Figure 5FIG. 2 shows a schematic diagram of the principle of model predictive control according to an embodiment of the present application. Figure 5 As shown, this application is based on the model predictive control principle, including three links: model prediction, rolling optimization and feedback correction. By predicting the behavior of the vehicle thermal management system in the future, the vehicle thermal management system is optimized online. Among them, the prediction model: based on the current monitoring data of the vehicle thermal management system, such as the current control data y(k) at the current time k, the predicted control data y at the future time k+j is predicted by the thermal management prediction model m (k+j); Feedback correction: The optimal control quantity can be found through feedback correction, and the output y m The first difference between (k+j) and y(k) is used to calculate y m (k+j) is corrected to obtain the correction control data y at the future time k+j p (k+j); rolling optimization: output y based on prediction p (k+j) and the reference trajectory y r The actual control variable u(k) is calculated by the second difference of (k+j), and the reference trajectory includes the ideal control data. The present application realizes fine control of the vehicle thermal management system through multi-step prediction.

[0097] See also Figure 6 , Figure 6 A block diagram of a vehicle thermal management control device according to an embodiment of the present application is shown. The device can be applied to Figure 1 The implementation environment shown is specifically configured in the computer device 102. This embodiment does not limit the implementation environment applicable to the device.

[0098] like Figure 6 As shown, a vehicle thermal management control device 600 according to an embodiment of the present application includes: a data acquisition module 601 , a control prediction module 602 and a control adjustment module 603 .

[0099] The data acquisition module 601 is used to acquire the current monitoring data and ideal control data of the vehicle thermal management system, where the current monitoring data includes the current control data and the current state data;

[0100] The control prediction module 602 is used to input the current monitoring data into the thermal management prediction model for prediction to obtain prediction control data;

[0101] a control adjustment module 603, for determining actual control data according to the predicted control data, the ideal control data and the current control data, so as to control the vehicle thermal management system, the vehicle thermal management system including at least one of a passenger compartment thermal management system, an electric drive thermal management system and a battery thermal management system;

[0102] Among them, the thermal management prediction model is obtained based on historical monitoring data and genetic algorithm training of long short-term memory network.

[0103] In one embodiment of the present application, the present application includes a vehicle, and the vehicle includes the vehicle thermal management control device as shown in the above embodiment.

[0104] It should be noted that the vehicle thermal management control device provided in the above embodiment and the vehicle thermal management control method provided in the above embodiment belong to the same concept, wherein the specific manner in which each module and unit performs operations has been described in detail in the method embodiment, and will not be repeated here. In actual applications, the vehicle thermal management control device provided in the above embodiment can allocate the above functions to different functional modules as needed, that is, divide the internal structure of the device into different functional modules to complete all or part of the functions described above, and this is not limited here.

[0105] An embodiment of the present application also provides an electronic device, comprising: one or more processors; a storage device for storing one or more programs, when the one or more programs are executed by one or more processors, the electronic device implements the vehicle thermal management control method provided in the above-mentioned embodiments.

[0106] See also Figure 7 , Figure 7 The structure diagram of the computer system suitable for implementing the electronic device of the embodiment of the present application is shown. It should be noted that: Figure 7 The computer system 700 of the electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0107] like Figure 7 As shown, the computer system 700 includes a central processing unit (CPU) 701, which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) 702 or the program loaded from the storage part 708 to the random access memory (RAM) 703, such as executing the method in the above embodiment. In the RAM 703, various programs and data required for system operation are also stored. The CPU 701, the ROM 702 and the RAM 703 are connected to each other through the bus 704. The input / output (I / O) interface 705 is also connected to the bus 704.

[0108] The following components are connected to the I / O interface 705: an input section 706 including a keyboard, a mouse, etc.; an output section 707 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 708 including a hard disk, etc.; and a communication section 709 including a network interface card such as a LAN (Local Area Network) card, a modem, etc. The communication section 709 performs communication processing via a network such as the Internet. A drive 710 is also connected to the I / O interface 705 as needed. A removable medium 711, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 710 as needed so that a computer program read therefrom is installed into the storage section 708 as needed.

[0109] In particular, according to an embodiment of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a computer program for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication section 709, and / or installed from a removable medium 711. When the computer program is executed by a central processing unit (CPU) 701, various functions defined in the system of the present application are executed.

[0110] It should be noted that the computer-readable medium shown in the embodiment of the present application can be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium can be, for example, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media can include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, wherein a computer-readable computer program is carried. This propagated data signal can take a variety of forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination of the above. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. A computer program contained on a computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wired, etc., or any suitable combination of the above.

[0111] The flowchart and block diagram in the accompanying drawings illustrate the possible architecture, functions and operations of the system, method and computer program product according to various embodiments of the present application. Wherein, each box in the flowchart or block diagram can represent a module, a program segment, or a part of the code, and the above-mentioned module, program segment, or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flowchart, and the combination of boxes in the block diagram or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0112] The units involved in the embodiments described in the present application can be implemented by software or by hardware, and the units described can also be set in a processor. Among them, the names of these units do not constitute a limitation on the units themselves under certain circumstances. Therefore, the technical solution according to the implementation mode of the present application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, including a number of instructions to enable a computing device (which can be a personal computer, a server, a touch terminal, or a network device, etc.) to execute a method according to the implementation mode of the present application.

[0113] Another aspect of the present application further provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor of a computer, the computer executes the vehicle thermal management control method provided in the above embodiments. The computer-readable storage medium may be included in the electronic device described in the above embodiments, or may exist independently without being assembled into the electronic device.

[0114] In the above embodiments, unless otherwise specified, by using serial numbers such as "first" and "second" to describe common objects, it only means that they refer to different instances of the same object, rather than indicating that the objects being described must adopt a given order, whether in time, space, sorting or any other way.

[0115] The above embodiments are merely illustrative of the principles and effects of the present application, and are not intended to limit the present application. Anyone familiar with the technology may modify or change the above embodiments without violating the spirit and scope of the present application. Therefore, all equivalent modifications or changes made by a person of ordinary skill in the art without departing from the spirit and technical ideas disclosed in the present application shall still be covered by the claims of the present application.

Claims

1. A vehicle thermal management control method, characterized in that: The method comprises: Acquire current monitoring data and ideal control data of the vehicle thermal management system, wherein the current monitoring data includes current control data and current state data; Inputting the current monitoring data into a thermal management prediction model for prediction to obtain prediction control data; determining actual control data according to the predicted control data, the ideal control data and the current control data to control the vehicle thermal management system; Among them, the thermal management prediction model is obtained based on historical monitoring data and genetic algorithm training of long short-term memory network, and the vehicle thermal management system includes at least one of a passenger compartment thermal management system, an electric drive thermal management system and a battery thermal management system.

2. The vehicle thermal management control method according to claim 1, characterized in that: Before inputting the current monitoring data into the thermal management prediction model for prediction, the method further includes: Acquiring historical monitoring data of the vehicle thermal management system; Performing cluster analysis on the historical monitoring data to obtain cluster monitoring data; generating an initial population according to the cluster monitoring data and a preset thermal management system constraint relationship, wherein the initial population is used to characterize different network configurations in the long short-term memory network; The long short-term memory network is iteratively trained according to the initial population and the historical monitoring data to obtain a thermal management prediction model.

3. The vehicle thermal management control method according to claim 2, characterized in that: Generating an initial population according to the cluster monitoring data and a preset thermal management system constraint relationship includes: Randomly generating initial monitoring data according to the clustered monitoring data and a preset thermal management system constraint relationship; Solving the relationship of the initial monitoring data according to the ant colony algorithm to obtain a set of random solutions; The set of random solutions is converted into an initial population corresponding to the genetic algorithm.

4. The vehicle thermal management control method according to claim 2, characterized in that: The long short-term memory network is iteratively trained according to the initial population and the historical monitoring data to obtain a thermal management prediction model, including: Dividing the historical monitoring data into training set data and test set data according to a preset ratio; Inputting the training set data into the long short-term memory network corresponding to each initial individual in the initial population for network training to obtain the training loss function value and network parameters of each initial individual; The long short-term memory network corresponding to the optimal individual and the optimal parameters is determined as the initial prediction model, and the initial prediction model is verified according to the test set data to obtain the verification loss function value, the optimal individual is obtained based on the initial individual corresponding to the minimum training loss function value, and the optimal parameters are used to characterize the network parameters corresponding to the optimal individual; If the genetic iteration meets the preset iteration conditions, the initial prediction model corresponding to the minimum verification loss function value in all genetic iterations is used as the thermal management prediction model; If the genetic iteration does not meet the preset iteration conditions, then the initial individuals are selected, crossed and mutated according to the ranking of all training loss function values ​​to obtain the next generation population, the next generation population is used as the initial population, and the initial prediction model is used as the long short-term memory network, and the training set data is repeatedly input into the long short-term memory network corresponding to each initial individual in the initial population for network training to obtain the training loss function value and network parameters of each initial individual, the long short-term memory network corresponding to the optimal individual and the optimal parameters is determined as the initial prediction model, and the initial prediction model is verified according to the test set data to obtain the verification loss function value. If the genetic iteration does not meet the preset iteration conditions, then the initial individuals are selected, crossed and mutated according to the ranking of all training loss function values ​​to obtain the next generation population, the next generation population is used as the initial population, and the initial prediction model is used as the long short-term memory network step, until the genetic iteration meets the preset iteration conditions, and the initial prediction model corresponding to the minimum verification loss function value in all genetic iterations is used as the thermal management prediction model.

5. The vehicle thermal management control method according to claim 4, characterized in that: The training set data is input into a long short-term memory network corresponding to an initial individual for network training to obtain a training loss function value and network parameters of the initial individual, including: Inputting the training set data into a long short-term memory network corresponding to an initial individual in batches for network training to obtain network parameters and prediction set data that meet preset training conditions; The mean square error between the training set data and the prediction set data is used as the training loss function value of an initial individual.

6. The vehicle thermal management control method according to any one of claims 1 to 5, characterized in that: Determining actual control data according to the predicted control data, the ideal control data and the current control data includes: Correcting the predicted control data according to a first difference between the predicted control data and the current control data to obtain corrected control data; The actual control data is determined according to a second difference between the corrected control data and the ideal control data.

7. The vehicle thermal management control method according to any one of claims 1 to 5, characterized in that: The predictive control data includes at least one of a predicted motor temperature set corresponding to the electric drive thermal management, a predicted battery temperature corresponding to the battery thermal management, and a predicted passenger compartment temperature and a predicted air volume corresponding to the passenger compartment thermal management.

8. A vehicle thermal management control device, characterized in that: The device comprises: A data acquisition module, used to acquire current monitoring data and ideal control data of the vehicle thermal management system, wherein the current monitoring data includes current control data and current state data; A control prediction module, used for inputting the current monitoring data into a thermal management prediction model for prediction to obtain prediction control data; a control adjustment module, configured to determine actual control data according to the predicted control data, the ideal control data and the current control data, so as to control the vehicle thermal management system; Among them, the thermal management prediction model is obtained based on historical monitoring data and genetic algorithm training of long short-term memory network, and the vehicle thermal management system includes at least one of a passenger compartment thermal management system, an electric drive thermal management system and a battery thermal management system.

9. A vehicle, characterized in that: The vehicle comprises: the vehicle thermal management control device as claimed in claim 8.

10. An electronic device, characterized in that: The electronic device comprises: one or more processors; A storage device for storing one or more programs, which, when executed by the one or more processors, enables the electronic device to implement the vehicle thermal management control method as described in any one of claims 1 to 7.

11. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and when the computer program is executed by a processor of a computer, the computer is caused to execute the vehicle thermal management control method according to any one of claims 1 to 7.

Citation Information

Cited By

  • Vehicle temperature control method and device, processor and vehicle

    CN119428083A

  • Vehicle thermal management system control method and device and thermal management system

    CN119914513A