Method, apparatus, device and computer readable storage medium for vehicle thermal management

By acquiring the status data of the thermal management system and inputting it into the trained strategy function, and adjusting the parameters of the temperature control components in combination with mapping rules, the safety and accuracy issues of the thermal management system for new energy vehicles are solved, and effective management and safety control of the thermal management object are achieved.

CN117021895BActive Publication Date: 2026-07-14CHONGQING CHANGAN TECH CO LTD

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
CHONGQING CHANGAN TECH CO LTD
Filing Date
2023-09-14
Publication Date
2026-07-14

AI Technical Summary

Technical Problem

In existing technologies, rule-based thermal management control methods are difficult to adapt to the complex operating conditions of new energy vehicles and have poor accuracy, while methods based on machine learning algorithms cannot judge the rationality of control logic and pose safety risks.

Method used

By acquiring the status data of the thermal management system, inputting the trained thermal management strategy function, and combining the mapping rules between the status data and the temperature control components, the operating parameters of the temperature control components are adjusted to achieve effective management of the thermal management object.

Benefits of technology

It improves the operational safety of temperature control components and the management efficiency of thermal management objects, reduces safety risks under unsuitable operating conditions, and enhances the training accuracy and generalization of thermal management strategy functions.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117021895B_ABST
    Figure CN117021895B_ABST
Patent Text Reader

Abstract

The application provides a vehicle thermal management method, device, equipment and computer readable storage medium, the method comprises the following steps: obtaining the first state data of the thermal management system; inputting the first state data into the trained thermal management strategy function to obtain the first control parameter value; determining the second control parameter value corresponding to the first state data according to the mapping rule of the state data of the thermal management system and the control parameter value of the temperature control component on the temperature regulation loop; adjusting the working parameter of the temperature control component of the corresponding thermal management object according to the relationship between the first control parameter value and the second control parameter value, so as to adjust the temperature of the corresponding thermal management object. In this way, by determining the relationship between the first control parameter value and the second control parameter value, the working parameter of the temperature control component of the corresponding thermal management object can be determined within a reasonable range, the safety of the operation of the temperature control component is improved, and effective management of the thermal management object is realized.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This application relates to the field of vehicle thermal management technology, specifically to a method, apparatus, device, and computer-readable storage medium for vehicle thermal management. Background Technology

[0002] The thermal management system is one of the most important systems in new energy vehicles besides the battery, electric drive, and electronic control systems. It typically includes five main parts: cabin thermal management, electric drive thermal management, battery thermal management, controller thermal management, body thermal management, and engine thermal management. Controller thermal management includes the thermal management of the autonomous driving controller, intelligent cockpit controller, and other domain controllers. Body thermal management includes the control of vehicle cooling airflow and the thermal insulation or thermal protection control of vehicle windows (including sunroofs or intelligent glass canopies). Electric drive thermal management includes heat-generating components such as drive motors, inverters, chargers, DC-DC converters, and DC-AC converters. Cabin thermal management mainly involves the heating and cooling of the passenger compartment, temperature zone control, air quality, and vibration and noise (NVH). Battery thermal management mainly involves heating, cooling, and temperature equalization. Engine thermal management includes engine cooling and turbocharger intercooling. By implementing thermal management for the components in the vehicle, user needs can be met, the energy efficiency of the thermally managed components can be improved, and the management costs of the thermally managed components can be reduced.

[0003] In related technologies, rule-based thermal management control methods and machine learning algorithm-based thermal management control methods are commonly used to control vehicle thermal management. However, rule-based thermal management control methods are difficult to adapt to the complex thermal management systems of new energy vehicles and have poor accuracy. Although machine learning algorithm-based thermal management control methods improve the accuracy of the control parameters of the thermal management system, they cannot determine the rationality of their control logic, which may lead to safety risks under unsuitable operating conditions. Summary of the Invention

[0004] This application provides a method, apparatus, device, and computer-readable storage medium for vehicle thermal management. The method can improve the operational safety of the temperature control components of the thermally managed object and achieve effective management of the thermally managed object.

[0005] The technical solution of this application is implemented as follows:

[0006] This application provides a method for vehicle thermal management, including:

[0007] Acquire the first state data of the thermal management system, which includes the temperature of the thermally managed object and the state parameter values ​​of the temperature regulation loop.

[0008] The first state data is input into the trained thermal management strategy function to obtain the first control parameter value; wherein, the trained thermal management strategy function is pre-trained based on the state data sample set of the thermal management system;

[0009] Based on the pre-constructed mapping rule between the status data of the thermal management system and the control parameter values ​​of the temperature control components on the temperature regulation loop, the second control parameter value corresponding to the first status data is determined.

[0010] Based on the relationship between the first control parameter value and the second control parameter value, the operating parameters of the temperature control component of the corresponding thermal management object are adjusted to regulate the temperature of the corresponding thermal management object.

[0011] It is understood that in the vehicle thermal management method provided in this application embodiment, firstly, first state data of the thermal management system is acquired; then, the first state data is input into a trained thermal management strategy function to obtain a first control parameter value; according to a pre-constructed mapping rule between the state data of the thermal management system and the control parameter values ​​of the temperature control components on the temperature regulation loop, a second control parameter value corresponding to the first state data is determined; finally, according to the relationship between the first control parameter value and the second control parameter value, the operating parameters of the temperature control components of the corresponding thermal management object are adjusted to regulate the temperature of the corresponding thermal management object. Thus, by using the relationship between the first control parameter value obtained based on the trained thermal management strategy function and the second control parameter value obtained based on the mapping rule between the state data of the thermal management system and the control parameter values ​​of the temperature control components on the temperature regulation loop, the operating parameters of the temperature control components of the determined corresponding thermal management object can be kept within a reasonable range, improving the safety of the temperature control component operation and achieving effective management of the thermal management object.

[0012] In some embodiments of this application, adjusting the operating parameters of the temperature control component of the corresponding thermal management object according to the relationship between the first control parameter value and the second control parameter value includes: obtaining the safety factor of the temperature control component; determining a control parameter reference value according to the safety factor and the second control parameter; and adjusting the operating parameters of the temperature control component according to the control parameter reference value and a preset control parameter value range.

[0013] It is understandable that the first control parameter value can be constrained by the safety factor of the temperature control component and the reference value of the second control parameter, as well as the preset control parameter value range, so that the operating parameters of the temperature control component are kept within a reasonable range, thus ensuring the safety of the temperature control component's operation.

[0014] In some embodiments, adjusting the operating parameters of the temperature control component according to the control parameter reference value and the preset control parameter value range includes: if the first control parameter value is within the preset control parameter value range and the first control parameter value is less than or equal to the control parameter reference value, adjusting the operating parameters of the temperature control component based on the first control parameter value.

[0015] It is understandable that if the value of the first control parameter is within the preset range of control parameter values, and the value of the first control parameter is less than or equal to the reference value of the control parameter, it means that the value of the first control parameter determined based on the trained thermal management strategy function has no obvious deviation. Therefore, after adjusting the working parameters of the temperature control component based on the value of the first control parameter, it can be ensured that the temperature control component can operate safely.

[0016] In some embodiments, adjusting the operating parameters of the temperature control component according to the control parameter reference value and the preset control parameter value range includes: if the first control parameter value is outside the preset control parameter value range, and / or the first control parameter value is greater than the control parameter reference value, adjusting the operating parameters of the temperature control component based on the second control parameter value.

[0017] It is understandable that if the value of the first control parameter is outside the preset range of control parameter values, and / or the value of the first control parameter is greater than the reference value of the control parameter, it indicates that there is a deviation in the value of the first control parameter determined based on the trained thermal management strategy function, which deviates from the normal operating parameter range of the temperature control component. In this case, adjusting the operating parameters of the temperature control component based on the value of the second control parameter can enable the temperature control component to work normally.

[0018] In some embodiments, adjusting the operating parameters of the temperature control component of the corresponding thermal management object according to the relationship between the first control parameter value and the second control parameter value includes: determining the absolute value of the difference between the first control parameter value and the second control parameter value; and adjusting the operating parameters of the temperature control component of the thermal management object based on the absolute value of the difference, the mapping rule, and the preset temperature range of the thermal management object.

[0019] It is understandable that by determining the absolute value of the difference between the first control parameter value and the second control parameter value, and based on this absolute value of the difference, the mapping rule between the status data of the thermal management system and the control parameter values ​​of the temperature control components in the temperature regulation loop, and the preset temperature range of the thermal management object, the control parameter values ​​of the temperature control components can be constrained, thereby ensuring that the obtained control parameter values ​​of the temperature control components are within the normal range and guaranteeing the safe operation of the temperature control components.

[0020] In some embodiments, adjusting the operating parameters of the temperature control component of the thermal management object based on the absolute value of the difference, the mapping rule, and the preset temperature range of the thermal management object includes: determining a first reference temperature of the thermal management object managed by the temperature control component corresponding to the first control parameter value based on the mapping rule; if the first reference temperature is within the preset temperature range and the absolute value of the difference is less than or equal to the difference threshold, adjusting the operating parameters of the corresponding temperature control component based on the first control parameter value.

[0021] It is understandable that by determining the first reference temperature of the thermal management object managed by the temperature control component corresponding to the first control parameter value based on the mapping rule, and when the first reference temperature is determined to be within the preset temperature range, and the absolute value of the difference between the first control parameter value and the second control parameter value is less than or equal to the difference threshold, the operating parameters of the corresponding temperature control component are adjusted based on the first control parameter value, so that the temperature control component can operate safely, and after adjusting the operating parameters of the temperature control component based on the first control parameter value, the temperature of the managed thermal management object can be maintained within the normal temperature range.

[0022] In some embodiments, adjusting the operating parameters of the temperature control component of the thermal management object based on the absolute value of the difference, the mapping rule, and the preset temperature range of the thermal management object includes: determining a first reference temperature of the thermal management object managed by the temperature control component corresponding to the first control parameter value based on the mapping rule; if the first reference temperature is outside the preset temperature range, and / or the absolute value of the difference is greater than a difference threshold, determining a second reference temperature of the thermal management object managed by the temperature control component corresponding to the second control parameter value based on the mapping rule; if the second reference temperature is within the preset temperature range, adjusting the operating parameters of the corresponding temperature control component based on the second control parameter value.

[0023] Understandably, if the first reference temperature of the thermal management object managed by the temperature control component corresponding to the first control parameter value is outside the preset range, and / or the absolute value of the difference between the first control parameter value and the second control parameter value is greater than the difference threshold, it indicates that the first control parameter value cannot guarantee the safe operation of the temperature control component, and the temperature control component may fail to manage the corresponding thermal management object based on the first control parameter value. In this case, if the second reference temperature of the thermal management object managed by the temperature control component corresponding to the second control parameter value determined based on the mapping rule is within the preset temperature range, the operating parameters of the corresponding temperature control component can be adjusted based on the second control parameter value, so that after adjusting the operating parameters of the temperature control component based on the second control parameter value, the temperature of the managed thermal management object can be maintained within the normal temperature range.

[0024] In some embodiments, the vehicle thermal management method further includes: if the first control parameter value meets a first condition, outputting a warning message to prompt the trained thermal management strategy function to be trained again to obtain a new thermal management strategy function, and determining the first control parameter value based on the new thermal management strategy function; wherein, the first condition includes at least one of the following: the first control parameter value is outside the preset control parameter value range; the first control parameter value is greater than the control parameter reference value; the first reference temperature corresponding to the first control parameter value is outside the preset temperature range; the absolute value of the difference between the first control parameter value and the second control parameter value is greater than the difference threshold.

[0025] It is understandable that if the value of the first control parameter does not meet the first condition, it means that the value of the first control parameter is unreasonable and may not be able to guarantee the safe operation of the temperature control component. Furthermore, the temperature control component may fail to manage the corresponding thermal management object based on the value of the first control parameter. In this case, by outputting a warning message, the trained thermal management strategy function can be subjected to regression testing and iterative training, so that the thermal management control strategy can undergo intelligent self-evolution.

[0026] In some embodiments, the vehicle thermal management method further includes: determining a first target operating mode based on the first state data; and determining the temperature control component corresponding to the first target operating mode and the trained thermal management strategy function.

[0027] It is understandable that by determining the first target working mode based on the first state parameter, and based on the first target working mode, the temperature control component corresponding to the thermal management object and the trained thermal management strategy function are determined. This allows the temperature corresponding to the current thermal management object and the state parameter value corresponding to the temperature control component corresponding to the current thermal management object to be input into the trained thermal management strategy function without needing to input the relevant parameters of other thermal management objects. This enables the rapid acquisition of the first control parameter value corresponding to the temperature control component.

[0028] In some embodiments, the vehicle thermal management method further includes: performing fault diagnosis and enabling processing on the temperature control component to determine that the temperature control component can start normally and work normally; and adjusting the operating parameters of the temperature control component corresponding to the thermal management object according to the relationship between the first control parameter value and the second control parameter value.

[0029] It is understandable that by performing fault diagnosis and enabling processing on the temperature control component, it can be ensured that the temperature control component can start up and work normally before the operating parameters of the temperature control component are adjusted. This ensures the accuracy of adjusting the operating parameters of the temperature control component of the corresponding thermal management object based on the relationship between the first control parameter value and the second control parameter value.

[0030] In some embodiments, the training process of the trained thermal management strategy function includes: determining at least one second target operating mode based on the state data sample set; determining a reference temperature control component corresponding to the i-th second target operating mode, and a state data sample set corresponding to the reference temperature control component; wherein i is greater than 0 and less than or equal to the total number of second target operating modes; and training a preset thermal management strategy function based on the state data sample set corresponding to the reference temperature control component to obtain the trained thermal management strategy function corresponding to the i-th second target operating mode.

[0031] Understandably, by determining the second target operating mode of the thermal management system based on the state data sample set, as well as the corresponding reference temperature control component and its corresponding state data sample set, the complexity of the thermal management system can be reduced. This allows for easier training of the preset thermal management strategy function based on the state data sample set of the reference temperature control component, reducing the dimensionality of the input and output data for training the preset thermal management strategy function, shortening the training time, and improving the control accuracy of vehicle thermal management. Furthermore, dividing the operating modes to train the thermal management strategy function for each mode simplifies the complex problem by operating mode, improving the training accuracy, efficiency, and generalization of the thermal management strategy function.

[0032] In some embodiments, training a preset thermal management strategy function based on a state data sample set corresponding to the reference temperature control component to obtain a trained thermal management strategy function corresponding to the i-th second target operating mode includes: obtaining a first reward value determined by the state data sample set corresponding to the reference temperature control component and a preset reward function; inputting the state data sample set corresponding to the reference temperature control component into the preset thermal management strategy function to obtain candidate control parameter values; determining a second reward value based on the candidate control parameter values ​​and the preset reward function; and performing backpropagation training on the preset thermal management strategy function based on the first reward value and the second reward value until a convergence condition is met to obtain the trained thermal management strategy function corresponding to the i-th second target operating mode.

[0033] It is understandable that by inputting the state data sample set corresponding to the reference temperature control component into the preset thermal management strategy function to obtain candidate control parameter values, and determining the second reward value and the first reward value determined by the state data sample set corresponding to the reference temperature control component based on the candidate control parameter values, the preset thermal management strategy function is trained by backpropagation. When the convergence condition is met, the trained thermal management strategy function corresponding to the i-th second target working mode can be obtained, so that the first control parameter value of the temperature control component can be obtained subsequently based on the first state data of the thermal management system and the trained thermal management strategy function.

[0034] This application provides a vehicle thermal management device, comprising:

[0035] The first acquisition module is used to acquire the first state data of the thermal management system, the first state data including the temperature of the thermal management object and the state parameter values ​​of the temperature regulation loop.

[0036] The second acquisition module inputs the first state data into the trained thermal management strategy function to obtain the first control parameter value; wherein, the trained thermal management strategy function is pre-trained based on the state data sample set of the thermal management system;

[0037] The first determining module is used to determine the second control parameter value corresponding to the first state data according to the mapping rule between the state data of the thermal management system and the control parameter values ​​of the temperature control component on the temperature regulation loop.

[0038] The adjustment module is used to adjust the operating parameters of the temperature control component of the corresponding thermal management object according to the relationship between the first control parameter value and the second control parameter value, so as to adjust the temperature of the corresponding thermal management object.

[0039] This application provides a vehicle thermal management device, including:

[0040] Memory used to store instructions that can execute vehicle thermal management;

[0041] The processor, when executing executable vehicle thermal management instructions stored in the memory, implements the method provided in the embodiments of this application.

[0042] This application provides a computer-readable storage medium storing computer-executable instructions configured to perform the above-described vehicle thermal management method.

[0043] It is understood that the vehicle thermal management method provided in this application, on the one hand, achieves a rationality verification of the first control parameter value by using the relationship between the first control parameter value obtained based on the trained thermal management strategy function and the second control parameter value obtained based on the mapping rule between the state data of the thermal management system and the control parameter values ​​of the temperature control components on the temperature regulation loop. When the rationality verification fails, the operating parameters of the temperature control components of the corresponding thermal management object are corrected based on the second control parameter value, so that the operating parameters of the temperature control components of the corresponding thermal management object are within a reasonable range, thereby improving the safety of the operation of the temperature control components and enabling effective management of the thermal management object. On the other hand, during the training of the thermal management strategy function, each second target working mode of the thermal management system is determined based on the state data sample set, that is, the working modes of the thermal management system are divided, and the thermal management strategy functions corresponding to different working modes are trained separately. This simplifies the complex problem according to the working mode, improving the training accuracy, efficiency and generalization of the thermal management strategy function. Attached Figure Description

[0044] Figure 1 A schematic flowchart illustrating a vehicle thermal management method provided in an embodiment of this application;

[0045] Figure 2 A schematic diagram illustrating a reasonable range of compressor speed values ​​provided in an embodiment of this application;

[0046] Figure 3 A flowchart illustrating a method for intelligent thermal management of new energy vehicles provided in this application embodiment;

[0047] Figure 4 A flowchart illustrating a method for collaborative processing between an intelligent thermal management algorithm training platform and an automotive thermal management processing unit, provided in an embodiment of this application;

[0048] Figure 5 A logical diagram illustrating the evaluation of the rationality of the output value of a reinforcement learning training control algorithm for a thermal management system, provided in an embodiment of this application;

[0049] Figure 6 A schematic diagram of a safety control architecture for an intelligent control strategy of a thermal management system provided in this application embodiment;

[0050] Figure 7 A flowchart illustrating another method for intelligent thermal management of new energy vehicles provided in this application embodiment;

[0051] Figure 8 A schematic diagram of the structural composition of a vehicle thermal management device provided in an embodiment of this application;

[0052] Figure 9This is a schematic diagram of the composition structure of a vehicle thermal management device provided in an embodiment of this application. Detailed Implementation

[0053] The technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings.

[0054] To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limitations on this application. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0055] In the following description, references to “some embodiments” or “other embodiments” describe a subset of all possible embodiments. However, it is understood that “some embodiments” or “other embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict.

[0056] In the following description, the terms "first" and "second" are used merely to distinguish similar objects and do not represent a specific ordering of objects. It is understood that "first" and "second" may be interchanged in a specific order or sequence where permitted, so that the embodiments of this application described herein can be implemented in an order other than that illustrated or described herein.

[0057] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0058] The thermal management system of new energy vehicles is significantly more complex than that of traditional fuel vehicles. In order to improve energy efficiency, new integrated thermal management systems usually involve the recovery and utilization of waste heat from batteries, electric drives and engines, as well as technologies such as heat pumps, battery pulse heating, and motor stall heating, in order to improve the energy efficiency of the thermal management system and reduce system costs.

[0059] In the field of intelligent thermal management of new energy vehicles, Chinese patent CN114704370A discloses an energy efficiency optimization algorithm for a thermal management system based on online parameter learning and MPC. The purpose is to address the lack of rolling optimization based on future prediction data and the poor control effect when disturbances occur in the energy efficiency optimization of the cooling system. Chinese patent CN115840987A discloses a method for generating thermal management strategies for hybrid vehicles based on deep reinforcement learning; Chinese patent CN115130266A discloses a reinforcement learning model for generating thermal management control models based on vehicle thermal state data and heat generation power data, and uses the reinforcement learning model to determine thermal management operations corresponding to planned speed data; Chinese patent CN111597723A discloses an intelligent control method for electric vehicle air conditioning systems based on improved intelligent model predictive control. This method utilizes a coupled thermal model of the vehicle air conditioning system and passenger compartment, along with a matching model predictive controller. Based on the theory of human thermal comfort parameters—Predicted Mean Vote (PMV)—and adaptive algorithms, it establishes adaptors for different individual thermal habits. By combining vehicle speed prediction and thermal comfort adaptation, a complete intelligent controller for the vehicle air conditioning system is established, suitable for multi-input multi-output control systems and air conditioning control systems suitable for personalized driving.

[0060] The intelligent control strategies for vehicle thermal management systems disclosed in current patents, which are similar to online, purely data-driven intelligent algorithms, have several limitations, primarily including: For artificial intelligence algorithms, the attributes of AI models are somewhat fuzzy, parameters are constantly changing, and there is no fixed standard to evaluate the correctness of the next calculation result, thus lacking a specific basis for functional safety. The uncertainty and error rate of artificial intelligence are unavoidable. Training data cannot guarantee its completeness and comprehensiveness, and there will be deviations between actual runtime data and original training data. Differences between the training environment and the actual operating environment may also lead to deviations in output results, potentially causing unexpected failures of the thermal management system. Highly integrated electric vehicle thermal management systems have complex and numerous operating modes. In other non-defined modes, the model is not accurate enough, or if all operating modes are integrated into one model, the model complexity is high, and the optimization algorithm converges slowly. The use of machine learning methods such as reinforcement learning is end-to-end control, not rule-based control, making it impossible to judge the rationality of its control logic. This poses safety risks under unsuitable operating conditions, such as battery thermal management failure and vehicle defrosting / defogging failure. Furthermore, the calculation speed is slow, and the consumption of controller hardware resources is high.

[0061] Based on the problems existing in related technologies, this application provides a method for vehicle thermal management. This method, applied to an in-vehicle terminal, can improve the safety of vehicle thermal management. Figure 1The diagram shown is a flowchart illustrating a vehicle thermal management method according to an embodiment of this application. The method includes the following steps:

[0062] S101. Obtain the first status data of the thermal management system.

[0063] It should be noted that the first state data includes the temperature of the thermally managed object and the state parameter values ​​of the temperature regulation loop. The thermally managed object can be a passenger compartment, electric drive system, battery, controller, vehicle body, and engine, etc.; the temperature regulation loop can be a refrigerant loop, coolant loop, etc., used to regulate the temperature of the thermally managed object; the state parameter values ​​of the temperature regulation loop can be the operating parameter values ​​of the temperature control components within the loop, such as compressors, evaporators, and expansion valves. For example, if the temperature regulation loop is a refrigerant loop, the corresponding state parameter values ​​include compressor speed, condenser temperature, evaporator temperature and pressure, and expansion valve opening; if the temperature regulation loop is a coolant loop, the corresponding state parameter values ​​include expansion valve opening, water pump speed and duty cycle, and heat exchanger coolant inlet and outlet temperatures, etc.

[0064] In some embodiments, the first state data may be acquired in real time. The first state data may include signals from thermal management system related items and signals from the thermal management system components themselves. Thermal management system related items may include, for example, the passenger compartment, battery, electric drive system, and domain controller. The signals corresponding to these related items include passenger compartment temperature-related signals, battery temperature-related signals, electric drive system temperature-related signals, and domain controller temperature-related signals. Thermal management system components may include compressors, condensers, evaporators, expansion valves, etc. The corresponding system component signals include compressor speed, pressure, condenser temperature, evaporator temperature, pressure, and expansion valve opening. In some embodiments, the first state data may also include operating parameter values ​​of the refrigerant circuit, coolant circuit, and various temperature components in the passenger compartment, as well as the target temperature of the thermal management object that is manually input or set.

[0065] S102. Input the first state data into the trained thermal management strategy function to obtain the first control parameter value.

[0066] In some embodiments, the trained thermal management strategy function is pre-trained based on a state data sample set of the thermal management system. The thermal management strategy function is trained by continuously interacting with a training environment (system simulation model) using an agent model (neural network) constructed using mainstream deep reinforcement learning algorithms such as Deep Q Network (DQN) and Proximal Policy Optimization (PPOA). The state data sample set of the thermal management system can be a pre-obtained dataset consisting of state parameter values ​​of the thermal management object and its temperature regulation loop. This state data sample set can be pre-collected and stored on a workstation or cloud platform server. The state data sample set corresponds to a one-dimensional model of the thermal management system, which may include one-dimensional dynamic models of the air conditioning system, passenger compartment system, and corresponding temperature control components of the battery, motor, and engine systems. This one-dimensional dynamic model is modeled using software such as Simulink, AMESim, or KuLI and calibrated using real vehicle or bench data.

[0067] In some embodiments, different thermal management strategy functions correspond to different operating modes. In practice, the first state data can be analyzed to determine the first target operating mode corresponding to the first state data, and the first state data can be input into the trained thermal management strategy function corresponding to the first target operating mode. The model predicts the first state data to obtain the first control parameter value. The first control parameter value corresponds to the temperature control component of the thermally managed object. If the temperature control component of the thermally managed object includes multiple components, the corresponding first control parameter values ​​also include multiple values. One temperature control component corresponds to one or more first control parameter values. Therefore, the first control parameter value in the embodiments of the application represents a set of multiple first control parameter values. For example, if the temperature control component corresponding to the thermally managed object includes a compressor, expansion valve, blower, etc., then the corresponding first control parameter value may include compressor speed, expansion valve opening, blower speed, etc.

[0068] S103. Based on the mapping rule between the state data of the pre-built thermal management system and the control parameter values ​​of the temperature control components on the temperature regulation loop, determine the second control parameter value corresponding to the first state data.

[0069] In some embodiments, the mapping rule between the state data of the pre-built thermal management system and the control parameter values ​​of the temperature control components in the temperature regulation loop can be the correspondence between the state data of the thermal management system and the control parameter values ​​of the temperature control components in the temperature regulation loop, which are obtained in advance through experiments or bench calibration. For example, the correspondence between the target temperature of the battery and the water pump duty cycle and the opening degree of the expansion valve can be determined through this correspondence. When the battery reaches the target temperature, the corresponding water pump duty cycle, or coolant temperature, flow rate and expansion valve opening degree can be determined.

[0070] In some embodiments, the mapping rules between the state data of the pre-built thermal management system and the control parameter values ​​of the temperature control components in the temperature regulation loop may include the temperatures of multiple thermal management objects and the control parameter values ​​of the corresponding temperature control components in the temperature regulation loop, such as the mapping relationship between battery temperature and the control parameter values ​​of the temperature control components managing the battery, the mapping relationship between passenger compartment temperature and the control parameter values ​​of the temperature control components managing the passenger compartment, the mapping relationship between electric drive temperature and the control parameter values ​​of the temperature control components managing the electric drive, etc.

[0071] In some embodiments, after pre-establishing mapping rules between the temperatures of different thermal management objects and the control parameter values ​​of the temperature control components on the corresponding temperature regulation loops, the correspondence between the temperatures of the thermal management objects and the control parameter values ​​of the temperature control components on the corresponding temperature regulation loops can be stored in the form of graphs, tables, etc. After obtaining the first state data of the thermal management system, the temperatures of the thermal management objects included in the first state data can be matched with the temperatures of the preset thermal management objects in the mapping rules. If it is determined that there is a reference thermal management object in the mapping rules with the same temperature as the thermal management object, the control parameter value of the temperature control component corresponding to the reference thermal management object can be determined as the second control parameter value. Similar to the first control parameter value, if the temperature control component corresponding to the reference thermal management object includes multiple components, and each temperature control component corresponds to one or more second control parameter values, then the second control parameter value in this embodiment represents a set of second control parameter values.

[0072] S104. Based on the relationship between the first control parameter value and the second control parameter value, adjust the operating parameters of the temperature control component of the corresponding thermal management object to adjust the temperature of the corresponding thermal management object.

[0073] In some embodiments, after obtaining the first control parameter value based on the first state data and the trained thermal management strategy function, and obtaining the second control parameter value based on the first state data and the mapping rule, the target control parameter value of the temperature control component can be further determined according to the relationship between the first control parameter value and the second control parameter value, and the operating parameters of the temperature control component of the thermal management object can be adjusted based on the target control parameter value, thereby realizing the adjustment of the temperature of the thermal management object.

[0074] In some embodiments, the operating parameters of the temperature control components corresponding to the same temperature control component can be adjusted based on the relationship between the first control parameter value and the second control parameter value. That is, the first control parameter value and the second control parameter value correspond to the same type of temperature control component and belong to the same type of control parameter. For example, if the temperature control component is a compressor, both the first control parameter value and the second control parameter value can be the compressor speed; if the temperature control component is a water pump, both the first control parameter value and the second control parameter value can be the water pump flow rate; if the temperature control component is an expansion valve, both the first control parameter value and the second control parameter value can be the opening degree of the expansion valve. In practice, if the temperature control component of the thermal management object includes multiple components, the operating parameters of the corresponding temperature control components can be adjusted based on the relationship between the first control parameter value and the second control parameter value (which belong to the same parameter type) for each temperature control component. After adjusting the operating parameters of each temperature control component, the temperature of the thermal management object can be regulated. The first control parameter value and the second control parameter value described elsewhere in the embodiments of this application all refer to the same type of operating parameter corresponding to the same temperature control component, and will not be elaborated further in this application.

[0075] In some embodiments, the target control parameter value may be a first control parameter value or a second control parameter value. In practice, since the first control parameter value obtained based on the trained thermal management strategy function may deviate from the range of operating parameters corresponding to the normal operation of the temperature control component, thus leading to safety risks, it is also necessary to combine it with the second control parameter value determined based on the mapping rule to obtain a target control parameter value that conforms to the actual operating conditions of the temperature control component, so as to achieve effective regulation of the temperature of the thermally managed object.

[0076] In some embodiments, the relationship between the first control parameter value and the second control parameter value may include the magnitude relationship between the first control parameter value and the second control parameter value. This magnitude relationship constrains the first control parameter value. Here, the first control parameter value may also be compared with a corresponding preset control parameter range, and the temperature value of the thermally managed object managed by the temperature control component corresponding to the first control parameter may be compared with a preset temperature threshold of the thermally managed object to determine the rationality of the first control parameter value. In practice, if the first control parameter is determined to be rational, the operating parameters of the temperature control component of the corresponding thermally managed object can be adjusted based on the first control parameter value. If the first control parameter is determined to be unreasonable, but the second control parameter value is reasonable, the operating parameters of the temperature control component of the corresponding thermally managed object can be adjusted based on the second control parameter value, thereby enabling the corresponding temperature control component to operate normally. By adjusting the operating parameter value of the temperature control component to the target control parameter value, correct temperature control of the thermally managed object can be achieved.

[0077] In this embodiment, firstly, first state data of the thermal management system is acquired; then, the first state data is input into a trained thermal management strategy function to obtain a first control parameter value; based on a pre-built mapping rule between the state data of the thermal management system and the control parameter values ​​of the temperature control components in the temperature regulation loop, a second control parameter value corresponding to the first state data is determined; finally, based on the relationship between the first and second control parameter values, the operating parameters of the temperature control components of the corresponding thermal management object are adjusted to regulate the temperature of the corresponding thermal management object. Thus, by using the relationship between the first control parameter value obtained based on the trained thermal management strategy function and the second control parameter value obtained based on the mapping rule between the state data of the thermal management system and the control parameter values ​​of the temperature control components in the temperature regulation loop, the operating parameters of the temperature control components of the determined thermal management object can be kept within a reasonable range, improving the safety of the temperature control component operation and achieving effective management of the thermal management object.

[0078] In some embodiments of this application, the step S104 of "adjusting the working parameters of the temperature control component of the corresponding thermal management object according to the relationship between the first control parameter value and the second control parameter value" can be achieved by the following steps S201 to S203, and each step will be described below.

[0079] S201. Obtain the safety factor of the temperature control component.

[0080] It should be noted that the safety factor of the temperature control component can be preset, for example, it can be 0.7, 0.8, 1.2, 1.5, etc. The safety factor corresponding to different temperature control components can be the same or different.

[0081] S202. Determine the reference value of the control parameter based on the safety factor and the value of the second control parameter.

[0082] In some embodiments, the safety factor can be a rational number greater than 0 and less than 1, such as 0.6, 0.8, 0.9, etc. In practice, the product of the safety factor and the second control parameter value can be used as a control parameter reference value. The control parameter reference value can be used to constrain the first control parameter value to avoid safety hazards caused by excessively large target control parameter values ​​of the temperature control component.

[0083] In some embodiments, if the value of the first control parameter is less than the value of the second control parameter, then the value of the first control parameter will be less than the reference value of the control parameter, indicating that the value of the first control parameter is within a reasonable range; if the value of the first control parameter is greater than the value of the second control parameter, then the value of the first control parameter may be greater than, less than or equal to the reference value of the control parameter. In this case, if it is determined that the value of the first control parameter is less than or equal to the reference value of the control parameter, then it indicates that the value of the first control parameter is within a reasonable range.

[0084] S203. Adjust the operating parameters of the temperature control component according to the reference value of the control parameter and the preset range of control parameter values.

[0085] In some embodiments, the control parameter reference value and the preset control parameter value range can be used simultaneously to determine the rationality of the control parameter value used to adjust the temperature control component. Based on the relationship between the determined first control parameter value and the control parameter reference value, and whether the upper limit of the control parameter is within the preset control parameter value range, it can be determined whether the first control parameter value is reasonable. If the first control parameter value is determined to be reasonable, the operating parameters of the temperature control component can be adjusted based on the first control parameter value. If the first control parameter value is determined to be unreasonable, the operating parameters of the temperature control component can be adjusted based on the second control parameter value.

[0086] In some embodiments, the preset control parameter value range can be a range determined by an upper limit and a lower limit of the control parameter. The upper limit of the control parameter can be the maximum operating parameter value corresponding to the normal operation of the temperature control component, and the lower limit of the control parameter can be the minimum operating parameter value corresponding to the normal operation of the temperature control component. For example, the upper limit of the control parameter for a compressor might be 8000 rpm, the lower limit might be 800 rpm, the maximum pressure might be 2.7 MPa, and the minimum pressure might be 0.15 MPa. That is, in practice, due to the different characteristics and operating parameter types of different temperature control components, the preset control parameter value ranges corresponding to different temperature control components are different, and the preset control parameter value ranges for different temperature control components are determined at the factory. For example, as shown... Figure 2 As shown, Figure 2 The solid black line in the figure represents the output speed of the compressor corresponding to the single-passenger compartment refrigeration system. The reasonable range for compressor speed includes not only the upper limit of 8000 r / min, but also the lower limit of 800 r / min. Figure 2 The shaded area within the upper and lower limits is within a reasonable range.

[0087] It is understandable that the first control parameter value can be constrained by the safety factor of the temperature control component and the reference value of the second control parameter, as well as the preset control parameter value range, so that the operating parameters of the temperature control component are kept within a reasonable range, thus ensuring the safety of the temperature control component's operation.

[0088] In some embodiments of this application, the operating parameters of the temperature control component are adjusted according to the control parameter reference value and the preset control parameter value range. That is, step S203 can be implemented by step S2031A or step S2031B as described below. Step S2031A and step S2031B as described below will be explained respectively.

[0089] S2031A. If the value of the first control parameter is less than the preset range of control parameter values, and the value of the first control parameter is less than or equal to the reference value of the control parameter, the operating parameters of the temperature control component are adjusted based on the value of the first control parameter.

[0090] In some embodiments, a first control parameter value within a preset control parameter value range indicates that the first control parameter value is greater than the lower limit of the control parameter and less than the upper limit of the control parameter, or the first control parameter value is equal to the lower limit of the control parameter, or the first control parameter value is equal to the upper limit of the control parameter. If it is determined that the first control parameter value is within the preset control parameter value range, it means that there will be no safety risk in the corresponding temperature control component operating with the first control parameter value. Furthermore, if it is determined that the first control parameter value is less than or equal to the control parameter reference value, it means that the first control parameter value is within a reasonable range. Therefore, the operating parameters of the temperature control component can be adjusted based on the first control parameter value, that is, the current operating parameter value of the temperature control component can be adjusted to the first control parameter value. For example, if the first control parameter value is a compressor speed of 3000 r / min and the current compressor speed is 2000 r / min, then the compressor speed per minute will be increased by 1000 r / min.

[0091] It is understandable that if the value of the first control parameter is within the preset range of control parameter values, and the value of the first control parameter is less than or equal to the reference value of the control parameter, it means that the value of the first control parameter determined based on the trained thermal management strategy function has no obvious deviation. Therefore, after adjusting the working parameters of the temperature control component based on the value of the first control parameter, it can be ensured that the temperature control component can operate safely.

[0092] S2031B If the value of the first control parameter is outside the preset range of control parameter values, and / or the value of the first control parameter is greater than the reference value of the control parameter, the operating parameters of the temperature control component are adjusted based on the value of the second control parameter.

[0093] In some embodiments, if it is determined that the value of the first control parameter is outside the preset range of control parameter values, it indicates that the value of the first control parameter exceeds the maximum operating parameter value (upper limit of control parameter) of the corresponding temperature control component, or the value of the first control parameter is less than the minimum operating parameter value (lower limit of control parameter) of the corresponding temperature control component. If the operation of the corresponding temperature control component based on the first control parameter value may cause damage or failure of the temperature control component, in this case, the operating parameters of the corresponding temperature control component can be adjusted based on the second control parameter value.

[0094] In some embodiments, if the value of the first control parameter is outside the preset range of control parameter values, it indicates that the value of the first control parameter may be too large or too small. If the corresponding temperature control component operates based on the value of the first control parameter, it may not be able to effectively control the temperature of the thermal management object managed by the temperature control component. For example, if the value of the first control parameter is the compressor speed, the corresponding thermal management object is a battery, the target compressor speed is 2000 r / min, and the value of the first control parameter is 2500 r / min, then when operating based on the value of the first control parameter, more cooling capacity may be generated, thereby causing the actual temperature of the battery to be lower than the target temperature, and effective management of the battery temperature cannot be achieved.

[0095] In other embodiments, if the first control parameter value is outside the preset control parameter value range and the first control parameter value is greater than the control parameter reference value, it indicates that the first control parameter value exceeds the actual allowable range. If the corresponding temperature control component operates based on the first control parameter value, there is a high probability of safety risks. Therefore, in this case, the operating parameters of the corresponding temperature control component can be adjusted based on the second control parameter value, that is, the current operating parameter value of the corresponding temperature control component is adjusted to the second control parameter value.

[0096] In other embodiments, the correspondence between the actual physical meaning and numerical value of the control parameter value of the temperature control component can be varied. For example, when the temperature control component is an electronic expansion valve, the control parameter value corresponding to the electronic expansion valve is the valve opening degree. If the preset control parameter value range corresponding to the valve opening degree is [0, 100%], one correspondence between the valve opening degree and the value is as follows: a valve opening degree of 0 indicates the minimum opening degree at which the electronic expansion valve can work normally, and a valve opening degree of 100% indicates the maximum opening degree at which the electronic expansion valve can work normally. In this case, if the first control parameter value corresponding to the electronic expansion valve is within the preset control parameter value range of [0, 100%], and the first control parameter value is less than or equal to the control parameter reference value (the second control parameter value and safety factor corresponding to the electronic expansion valve are determined), then the opening degree of the electronic expansion valve is adjusted based on the first control parameter value; if the first control parameter value is greater than the control parameter reference value, then the opening degree of the electronic expansion valve is adjusted based on the second control parameter value. Another possible correspondence between the expansion valve opening degree and its numerical value is as follows: an opening degree of 0 represents the maximum opening degree at which the electronic expansion valve can operate normally, and an opening degree of 100% represents the minimum opening degree at which the electronic expansion valve can operate normally. In this case, if the first control parameter value corresponding to the electronic expansion valve is within the preset control parameter value range of [0, 100%], and the first control parameter value is greater than the control parameter reference value (the second control parameter value and safety factor corresponding to the electronic expansion valve are determined), then the opening degree of the electronic expansion valve is adjusted based on the first control parameter value; if the first control parameter value is less than or equal to the control parameter reference value, then the opening degree of the electronic expansion valve is adjusted based on the second control parameter value. That is, in practice, because the numerical order corresponding to the control parameter range is different, the preconditions for adjusting the working parameters of the temperature control component based on the first control parameter value are also different.

[0097] It is understandable that if the first control parameter value is outside the preset control parameter value range, and / or the first control parameter value is greater than the control parameter reference value, it indicates that the first control parameter value determined based on the trained thermal management strategy function has a deviation, deviating from the normal operating parameter range of the temperature control component. In this case, adjusting the operating parameters of the temperature control component based on the second control parameter value can enable the temperature control component to work normally.

[0098] In some embodiments of this application, the step S104 of "adjusting the working parameters of the temperature control component of the corresponding thermal management object according to the relationship between the first control parameter value and the second control parameter value" can also be achieved by the following steps S301 to S302, and each step is described below.

[0099] S301. Determine the absolute value of the difference between the first control parameter value and the second control parameter value.

[0100] In some embodiments, the first control parameter value and the second control parameter value are operating parameter values ​​of the same temperature control component, such as the compressor speed, expansion valve opening, evaporator temperature, water pump duty cycle, etc. The absolute value of the difference between the first control parameter value and the second control parameter value can be the absolute value of the difference between the first control parameter value and the second control parameter value.

[0101] For example, if the first control parameter is a compressor speed of 5000 r / min and the second control parameter is 4500 r / min, then the absolute value of the difference between the first and second control parameter values ​​is 500 r / min; if the first control parameter is an expansion valve opening of 150 steps and the second control parameter is an expansion valve opening of 200 steps, then the absolute value of the difference between the first and second control parameter values ​​is 50 steps; if the first control parameter is an evaporator temperature of 18°C ​​and the second control parameter is an evaporator temperature of 12°C, then the absolute value of the difference between the first and second control parameter values ​​is 6°C; if the first control parameter is a water pump duty cycle of 1 m / s and the second control parameter is 0.7 m / s, then the absolute value of the difference between the first and second control parameter values ​​is 0.3 m / s. The first and second control parameter values, and the absolute values ​​of the differences between them, are merely illustrative examples and are not intended to limit the scope of this application.

[0102] S302. Adjust the operating parameters of the temperature control component of the thermal management object based on the absolute value of the difference, mapping rules, and the preset temperature range of the thermal management object.

[0103] In some embodiments, the preset temperature range of the thermal management object can be the temperature range corresponding to when the thermal management object can operate normally. In practice, the preset temperature ranges of different thermal management objects may be different or the same. For example, the preset temperature range of a vehicle battery may be [0°C, 40°C], and the preset temperature range of a vehicle motor may be [0°C, 65°C]. The thermal management object and its preset temperature range are merely illustrative examples and are not intended to limit the scope of this application.

[0104] In some embodiments, based on mapping rules, the temperature that the temperature control component can reach when operating based on a first control parameter value is determined for the corresponding managed thermal object. The absolute value of the difference between the first and second control parameter values ​​can be used to determine the degree of deviation of the first control parameter value. The preset temperature range can determine whether the temperature of the corresponding thermal object is within a reasonable range under the control of the first control parameter value. By comparing the absolute value of the difference with a difference threshold, and comparing the temperature of the thermal object determined based on the mapping rules with the preset temperature range, the target control parameter value of the temperature control component used to manage the thermal object can be determined, and the operating parameters of the temperature control component of the thermal object can be adjusted based on the target control parameter value.

[0105] It is understandable that by determining the absolute value of the difference between the first control parameter value and the second control parameter value, and based on this absolute value of the difference, the mapping rule between the status data of the thermal management system and the control parameter values ​​of the temperature control components in the temperature regulation loop, and the preset temperature range of the thermal management object, the control parameter values ​​of the temperature control components can be constrained, thereby ensuring that the obtained control parameter values ​​of the temperature control components are within the normal range and guaranteeing the safe operation of the temperature control components.

[0106] In some embodiments of this application, the operating parameters of the temperature control component of the thermal management object are adjusted based on the absolute value of the difference, the mapping rule, and the preset temperature range of the thermal management object. That is, the above step S302 can also be implemented by the following steps S3021A to S3022A. The following describes steps S3021A to S3022A.

[0107] S3021A. Based on the mapping rules, determine the first reference temperature of the thermal management object managed by the temperature control component corresponding to the first control parameter value.

[0108] In some embodiments, the mapping rule between the status data of the thermal management system and the control parameter values ​​of the temperature control component in the temperature regulation loop includes the correspondence between the temperature of the thermally managed object and the control parameter values ​​of the temperature control component in the temperature regulation loop. Therefore, when the first control parameter value is known, the temperature of the thermally managed object corresponding to the first control parameter value, i.e., the first reference temperature, can be determined based on this mapping rule. It should be noted that the first reference temperature represents the temperature that the thermally managed object managed by the temperature control component can reach when the temperature control component operates at the first control parameter value.

[0109] For example, if the first control parameter value is an expansion valve opening of 300 steps and a water pump duty cycle of 50%, then based on the mapping rule, it can be determined that when the expansion valve opening is 200 steps and the water pump duty cycle is 40%, the temperature of the corresponding thermal management object, the passenger compartment, can reach 24°C. That is, the temperature control components corresponding to the first control parameter value are the expansion valve and the water pump, the thermal management object managed by the temperature control components is the passenger compartment, and the first reference temperature corresponding to the thermal management object is 24°C. If the first control parameter value is a compressor speed of 4000 r / min and an expansion valve opening of 200 steps, then based on the mapping rule, it can be determined that when the compressor speed is 4000 r / min and the expansion valve opening is 200 steps, the temperature of the corresponding thermal management object, the battery, can reach 20°C. That is, the temperature control components corresponding to the first control parameter value are the compressor and the expansion valve, the thermal management object managed by the temperature control components is the battery, and the first reference temperature corresponding to the thermal management object is 20°C. The mapping relationship between the first control parameter value, the temperature control component, the thermal management object, and the temperature of the thermal management object and the corresponding control parameter value of the temperature control component is merely an illustrative example and is not limited in this application.

[0110] S3022A: If the first reference temperature is within the preset temperature range and the absolute value of the difference is less than or equal to the difference threshold, adjust the working parameters of the corresponding temperature control component based on the first control parameter value.

[0111] In some embodiments, the difference threshold can be the maximum difference between the first control parameter value and the second control parameter value. The difference thresholds for different types of control parameter values ​​may be the same or different. If the first reference temperature range is determined to be within the preset temperature range, it means that the thermally managed object can remain within the normal temperature range under the management of the corresponding temperature control component based on the first control parameter value, and there will be no safety risks or poor comfort caused by excessively low or high temperatures. Furthermore, if the absolute value of the difference between the first control parameter value and the second control parameter value is determined to be less than or equal to the difference threshold, it means that the first control parameter value will not be too large or too small. Therefore, the operating parameters of the corresponding temperature control component can be adjusted based on the first control parameter value, that is, the current operating parameters of the corresponding temperature control component can be adjusted to be the same as the first control parameter value.

[0112] It is understandable that by determining the first reference temperature of the thermal management object managed by the temperature control component corresponding to the first control parameter value based on the mapping rule, and when the first reference temperature is determined to be within the preset temperature range, and the absolute value of the difference between the first control parameter value and the second control parameter value is less than or equal to the difference threshold, the operating parameters of the corresponding temperature control component are adjusted based on the first control parameter value, so that the temperature control component can operate safely, and after adjusting the operating parameters of the temperature control component based on the first control parameter value, the temperature of the managed thermal management object can be maintained within the normal temperature range.

[0113] In some embodiments of this application, the operating parameters of the temperature control component of the thermal management object are adjusted based on the absolute value of the difference, the mapping rule, and the preset temperature range of the thermal management object. That is, the above step S302 can also be implemented by the following steps S3021B to S3023B. The following describes steps S3021B to S3023B.

[0114] S3021B. Based on the mapping rules, determine the first reference temperature of the thermal management object managed by the temperature control component corresponding to the first control parameter value.

[0115] In some embodiments, when the value of the first control parameter is known, the temperature of the thermally managed object corresponding to the first control parameter value, i.e., the first reference temperature, can be determined based on mapping rules. It should be noted that the first reference temperature represents the temperature that the thermally managed object managed by the temperature control component can reach when the temperature control component operates at the first control parameter value. For example, if the first control parameter value is an expansion valve opening of 300 steps and a water pump duty cycle of 50%, then based on the mapping rule, it can be determined that when the expansion valve opening is 300 steps and the water pump duty cycle is 50%, the temperature of the corresponding thermal management object, the passenger compartment, can reach 24°C. That is, the temperature control components corresponding to the first control parameter value are the expansion valve and the water pump, the thermal management object managed by the temperature control components is the passenger compartment, and the first reference temperature corresponding to the thermal management object is 24°C. If the first control parameter value is a compressor speed of 4000 r / min and an expansion valve opening of 200 steps, then based on the mapping rule, it can be determined that when the compressor speed is 4000 r / min and the expansion valve opening is 200 steps, the temperature of the corresponding thermal management object, the battery, can reach 20°C. That is, the temperature control components corresponding to the first control parameter value are the compressor and the expansion valve, the thermal management object managed by the temperature control components is the battery, and the first reference temperature corresponding to the thermal management object is 20°C.

[0116] S3022B If the first reference temperature is outside the preset temperature range, and / or the absolute value of the difference is greater than the difference threshold, the second reference temperature of the thermal management object managed by the temperature control component corresponding to the second control parameter value is determined based on the mapping rule.

[0117] In some embodiments, a first reference temperature outside a preset temperature range means that the first reference temperature is less than the minimum value corresponding to the preset temperature range, or the first reference temperature is greater than the maximum value corresponding to the preset temperature range. For example, if the preset temperature range is [0℃, 50℃], then the first reference temperature is less than 0℃ or greater than 50℃. If the first reference temperature is outside the preset temperature range, it indicates that if the temperature control component manages the corresponding thermal management object based on the first control parameter value, the thermal management object will deviate from its normal state. For example, it may cause the battery to explode due to excessive temperature, or cause the battery to be damaged due to excessively low temperature, preventing it from supplying power to other components, or cause the temperature of the passenger compartment to be much higher or lower than the target temperature expected by the user, thereby reducing the user's comfort experience. In this case, it indicates that the first control parameter value is unreasonable, and the second reference temperature of the thermal management object managed by the temperature control component corresponding to the second control parameter value can be determined based on mapping rules.

[0118] In some embodiments, if the absolute value of the difference between the first control parameter value and the second control parameter value is greater than the difference threshold, it indicates that the first control parameter value and the second control parameter value differ significantly, and the first control parameter value is too large or too small. In this case, the second reference temperature of the thermal management object managed by the temperature control component corresponding to the second control parameter value can be further determined based on the mapping rule.

[0119] In other embodiments, if the first reference temperature is outside the preset temperature range, and the absolute value of the difference between the first control parameter value and the second control parameter value is greater than the difference threshold, it indicates that there will be a safety risk if the temperature control component manages the thermal management object based on the first control parameter value. Therefore, the rationality of the second control parameter value can be further judged. Based on the mapping rule, the second reference temperature of the thermal management object managed by the temperature control component corresponding to the second control parameter value is determined, and the rationality of the second control parameter value is determined based on the relationship between the second reference temperature and the preset temperature range.

[0120] S3023B: If the second reference temperature is within the preset temperature range, adjust the operating parameters of the corresponding temperature control component based on the second control parameter value.

[0121] In some embodiments, a second reference temperature within a preset temperature range means that the second reference temperature is greater than or equal to the minimum value corresponding to the preset temperature range, and less than or equal to the maximum value corresponding to the preset temperature range. For example, if the preset temperature range is [10℃, 60℃], then the second reference temperature is greater than or equal to 10℃ and less than or equal to 60℃. If it is determined that the second reference temperature is within the preset temperature range, it means that when the temperature control component operates based on the second control parameter value, the temperature of the thermally managed object managed by the temperature control component can be kept within the normal range. At this time, the operating parameters of the corresponding temperature control component can be adjusted based on the second control parameter value, that is, the current operating parameters of the corresponding temperature control component can be adjusted to be the same as the second control parameter value.

[0122] Understandably, if the first reference temperature of the thermally managed object corresponding to the first control parameter value is outside the preset temperature range, and / or the absolute value of the difference between the first control parameter value and the second control parameter value is greater than the difference threshold, it indicates that the first control parameter value cannot guarantee the safe operation of the corresponding temperature control component, and the temperature control component may fail to manage the corresponding thermally managed object based on the first control parameter value. In this case, if the second reference temperature of the thermally managed object corresponding to the second control parameter value determined based on the mapping rule is within the preset temperature range, the operating parameters of the corresponding temperature control component can be adjusted based on the second control parameter value, so that after adjusting the operating parameters of the temperature control component based on the second control parameter value, the temperature of the managed thermally managed object can be maintained within the normal temperature range.

[0123] In some embodiments of this application, after determining the first reference temperature of the thermal management object managed by the temperature control component corresponding to the first control parameter value based on the mapping rule, i.e., step S3021A, the following step S401 can also be executed, which will be described below.

[0124] S401. If the value of the first control parameter meets the first condition, output a warning message.

[0125] In some implementations, warning messages are used to indicate the need to retrain the trained thermal management strategy function to obtain a new thermal management strategy function, based on which the first control parameter value is determined. The first condition can be at least one of the following: the first control parameter value is outside a preset control parameter value range; the first control parameter value is greater than a control parameter reference value; the first reference temperature corresponding to the first control parameter value is outside a preset temperature range; or the absolute value of the difference between the first control parameter value and the second control parameter value is greater than a difference threshold.

[0126] In some embodiments, the first condition indicates that the value of the first control parameter is unreasonable or fails the reasonableness check, indicating that there is a problem with the previously trained thermal management strategy, and it is necessary to perform regression testing and iterative training to enable the thermal management control strategy to intelligently self-evolve. The strategy is then retrained until a new thermal management strategy function is obtained, and the value of the first control parameter is redetermined based on the new thermal management strategy function.

[0127] It is understandable that if the value of the first control parameter does not meet the first condition, it means that the value of the first control parameter is unreasonable and may not be able to guarantee the safe operation of the temperature control component. Furthermore, the temperature control component may fail to manage the corresponding thermal management object based on the value of the first control parameter. In this case, by outputting a warning message, the trained thermal management strategy function can be subjected to regression testing and iterative training, so that the thermal management control strategy can undergo intelligent self-evolution.

[0128] In some embodiments of this application, after obtaining the first state data of the thermal management system, i.e., step S101, the following steps S501 to S502 can also be performed. Steps S501 to S502 will be described below.

[0129] S501. Determine the first target working mode based on the first state data.

[0130] In some embodiments, the first target operating mode can be the required operating mode of the thermal management object, including passenger compartment cooling mode, passenger compartment heating mode, battery cooling, battery heating, electric drive cooling, passenger compartment dehumidification, defogging mode, maintenance mode, or multiple combinations of passenger compartment and battery cooling or heating modes. The first state data may include the current temperature and target temperature of the thermal management object, the current operating parameters of the temperature control component, etc. By analyzing the current temperature, target temperature, and current operating parameters of the temperature control component in the first state data, the required operating mode of the thermal management object, i.e., the first target operating mode, can be determined. For example, if the current temperature of the passenger compartment is 27°C and the driver's set temperature is 23°C, then the required mode of the passenger compartment is passenger compartment cooling. In other embodiments, it can also be determined whether the thermal management object can enter the corresponding required operating mode based on the vehicle's status, ambient temperature, etc. For example, if the pressure of the refrigeration circuit system is very low, the compressor cannot work, and cooling will not be possible.

[0131] S502. Determine the temperature control component and the trained thermal management strategy function corresponding to the first target working mode.

[0132] In some embodiments, controlling the corresponding thermal management object through temperature control components can enable the thermal management object to achieve the purpose of a first target operating mode. After determining the first target operating mode, it can be determined which temperature control components are needed to manage the thermal management object under this first target operating mode. For example, if the determined first target operating mode is passenger compartment cooling, the corresponding temperature control components may include compressors, expansion valves, etc.; if the determined first target operating mode is battery temperature equalization, the corresponding temperature control components may include expansion valves, water pumps, etc. The first target operating mode and the temperature control components of the thermal management object determined based on the first target operating mode are merely illustrative examples, and this application does not limit them.

[0133] It is understandable that by determining the first target working mode based on the first state parameter, and based on the first target working mode, the temperature control component corresponding to the thermal management object and the trained thermal management strategy function are determined. This allows the temperature corresponding to the current thermal management object and the state parameter value corresponding to the temperature control component corresponding to the current thermal management object to be input into the trained thermal management strategy function without needing to input the relevant parameters of other thermal management objects. This enables the rapid acquisition of the first control parameter value corresponding to the temperature control component.

[0134] In some embodiments of this application, after determining the second control parameter value corresponding to the first state data according to the mapping rule between the state data of the pre-built thermal management system and the control parameter value of the temperature control component on the temperature regulation loop, i.e., after step S103, the following steps S601 to S602 can also be executed. Steps S601 to S602 will be described below.

[0135] S601. Perform fault diagnosis and enable processing on the temperature control component to ensure that the temperature control component can start normally and work normally.

[0136] In some embodiments, fault diagnosis of the temperature control component can involve detecting whether a fault exists in the temperature control component, and enabling the temperature control component can involve determining whether the temperature control component can start or stop normally. In practice, fault diagnosis of the temperature control component includes system rationality diagnosis, component rationality diagnosis, sensor circuit-level diagnosis, and motor drive diagnosis; enabling the temperature control component includes protection and start / stop control of the compressor, expansion valve, etc.

[0137] In some embodiments, if the temperature control component is found to be faulty after fault diagnosis and enable processing, such as being unable to turn on or off normally, or the operating parameters of the temperature control component being outside the normal operating parameter range, then the temperature control component needs to be adjusted or repaired to eliminate the fault and ensure that the temperature control component can start and work normally.

[0138] S602. Adjust the operating parameters of the temperature control component of the corresponding thermal management object according to the relationship between the first control parameter value and the second control parameter value.

[0139] In some embodiments, after determining that the temperature control component is fault-free, can start normally, and can work normally, the target operating parameter value of the temperature control component of the corresponding thermal management object can be determined based on the relationship between the second control parameter value and the second control parameter value. The operating parameter of the temperature control component of the corresponding thermal management object can be adjusted based on the target operating parameter value, for example, adjusting the operating parameter of the temperature control component of the corresponding thermal management object to be the same as the first control parameter value, or adjusting the operating parameter of the temperature control component of the corresponding thermal management object to be the same as the second control parameter value.

[0140] It is understandable that by performing fault diagnosis and enabling processing on the temperature control component, it can be ensured that the temperature control component can start up and work normally before the operating parameters of the temperature control component are adjusted. This ensures the accuracy of adjusting the operating parameters of the temperature control component of the corresponding thermal management object based on the relationship between the first control parameter value and the second control parameter value.

[0141] In some embodiments of this application, the training process of the trained thermal management strategy function can be implemented through the following steps S701 to S703, and each step is described below.

[0142] S701. Based on the state data sample set, determine at least one second target working mode.

[0143] In some embodiments, the state data sample set may be pre-obtained and includes one or more thermal management object samples. Each thermal management object sample includes the temperature of the thermal management object and the state parameter values ​​of the corresponding temperature regulation loop. The second target operating mode may be the required operating mode of the thermal management object in the state data sample set. Different thermal management objects may have different required operating modes. For example, the occupant compartment may need to be heated, while the battery and motor may need to be cooled.

[0144] In some embodiments, by analyzing the state data sample set, one or more thermal management objects can be identified. Further, based on the current temperature or target temperature of the thermal management object, or the state parameter values ​​in the temperature regulation loop corresponding to the thermal management object, multiple required operating modes of the thermal management system, i.e., the second target operating mode, can be further determined.

[0145] S702. Determine the reference temperature control component corresponding to the i-th second target working mode, and the state data sample set corresponding to the reference temperature control component.

[0146] It should be noted that i is a positive integer, greater than 0 and less than or equal to the total number of second target operating modes. In some embodiments, the temperature control loops corresponding to different second target operating modes are different (the reference temperature control components in the temperature control loops are different), and the different temperature control loops can operate in series or in parallel. The reference temperature control component is the temperature control component in the temperature control loop corresponding to the i-th second target operating mode. The reference temperature control component can be an actuator such as a compressor, expansion valve, evaporator, condenser, or water pump. In practice, once the second target operating mode is determined, the corresponding reference temperature control component can be determined accordingly. For example, if the determined i-th second target operating mode is battery cooling, the corresponding reference temperature control component can be an expansion valve, water pump, etc.

[0147] In some embodiments, after determining the reference temperature control component corresponding to the i-th second target operating mode, a state data sample set corresponding to the reference temperature control component can be determined. This state data sample set includes the operating parameters of the reference temperature control component. In practice, a thermal management object may have one or more reference temperature control components. Therefore, the state data sample set may include the operating parameters of one or more reference temperature control components. For example, when the reference temperature control component includes a compressor, an expansion valve, and a water pump, the corresponding state data sample set of the reference temperature control component includes the compressor speed, the opening degree of the expansion valve, the duty cycle of the water pump, etc.

[0148] S703. Based on the state data sample set corresponding to the reference temperature control component, train the preset thermal management strategy function to obtain the trained thermal management strategy function corresponding to the i-th second target working mode.

[0149] In some embodiments, the preset thermal management strategy function can be a deep reinforcement learning network model constructed based on a deep reinforcement learning algorithm to predict the control parameter values ​​corresponding to a one-dimensional dynamic model. The one-dimensional dynamic model is a pre-calibrated simulation model corresponding to a reference temperature control component. The one-dimensional dynamic model can be obtained by calibrating real vehicle data using software such as Simulink, AMESim, or KuLI. For example, if the reference temperature control component is a compressor, then the one-dimensional dynamic model is the simulation model corresponding to the compressor. The thermal management strategy function can be constructed using a deep reinforcement learning algorithm agent combined with the one-dimensional dynamic model. The thermal management strategy function is used to determine the control parameter values ​​corresponding to the one-dimensional dynamic model (or the reference temperature control component corresponding to the one-dimensional dynamic model).

[0150] In some embodiments, a preset thermal management strategy function is trained using a sample set of state data corresponding to a reference temperature control component as training data. The parameters in the preset thermal management strategy function are continuously adjusted until a convergence condition is met, at which point training stops, thus obtaining the trained thermal management strategy function corresponding to the i-th second target operating mode. During the training of the preset thermal management strategy function, the training data includes not only the sample set of state data corresponding to the reference temperature control component, but also comfort parameters, reward values, etc., determined based on the one-dimensional dynamic model corresponding to the reference temperature control component.

[0151] For example, in one possible implementation, a corresponding thermal management strategy function can be trained for each second target operating mode. That is, one second target operating mode corresponds to one trained thermal management strategy function. Therefore, after training the thermal management strategy functions corresponding to each second target operating mode, multiple trained thermal management strategy functions can be obtained, with the number of trained thermal management strategy functions being the same as the number of second target operating modes. The trained thermal management strategy functions are then combined or integrated to obtain a complete trained thermal management strategy function. Before inputting the first state data into the trained thermal management strategy function to determine the first control parameter, the first target operating mode corresponding to the first state data can be determined first. Then, the trained thermal management strategy function corresponding to the first target operating mode can be determined from the complete trained thermal management strategy functions. Based on the trained thermal management strategy function corresponding to the first target operating mode and the first state data, the value of the first control parameter is determined.

[0152] Understandably, by determining the second target operating mode of the thermal management system based on the state data sample set, as well as the corresponding reference temperature control component and its corresponding state data sample set, the complexity of the thermal management system can be reduced. This allows for easier training of the preset thermal management strategy function based on the state data sample set of the reference temperature control component, reducing the dimensionality of the input and output data for training the preset thermal management strategy function, shortening the training time, and improving the control accuracy of vehicle thermal management. Furthermore, dividing the operating modes to train the thermal management strategy function for each mode simplifies the complex problem by operating mode, improving the training accuracy, efficiency, and generalization of the thermal management strategy function.

[0153] In some embodiments of this application, a preset thermal management strategy function is trained based on the state data sample set corresponding to the reference temperature control component to obtain the trained thermal management strategy function corresponding to the i-th second target working mode. That is, step S703 can also be implemented by the following steps S7031 to S7034. Each step is described below.

[0154] S7031. Obtain the first reward value determined by the state data sample set corresponding to the reference temperature control component and the preset reward function.

[0155] In some embodiments, the first reward value may be determined based on a preset reward function and a state data sample set corresponding to the reference temperature control component. The state data sample set corresponding to the reference temperature control component includes, under the second temperature control mode, the energy consumption of the reference temperature control component, the temperature of the thermal management object managed by the reference temperature control component, and the target temperature, etc. The preset reward function may be pre-set and is related to PMV and the energy consumption of the temperature control component. For example, if the reference temperature control component is a compressor, the reward function R can be expressed by the following formula (1):

[0156]

[0157] Among them, E comp For the energy consumption of the compressor, T cab r1 represents the temperature of the crew cabin, where r1 represents the temperature T of the crew cabin. cab and target temperature T target The absolute value of the difference between them, PMV represents the human thermal comfort parameter.

[0158] S7032. Input the state data sample set corresponding to the reference temperature control component into the preset thermal management strategy function to obtain candidate control parameter values.

[0159] In some embodiments, the state dataset corresponding to the reference temperature control component is input into a preset thermal management strategy function. The preset thermal management strategy function can then be used to predict and output candidate control parameter values ​​corresponding to the reference temperature control component. In some embodiments, the parameters input to the preset thermal management strategy function may include, in addition to the state data sample set corresponding to the reference temperature control component, the temperature of the thermal management object corresponding to the i-th second target operating mode, the target temperature, etc.

[0160] S7033. Determine the second reward value based on the candidate control parameter values ​​and the preset reward function.

[0161] In some embodiments, the second reward value can be determined based on a preset reward function and candidate control parameter values. Based on the candidate control parameter values ​​and the one-dimensional dynamic model corresponding to the reference temperature control component, the energy consumption and thermal comfort parameter PMV of the reference temperature control component, as well as the temperature of the thermally managed object managed by the reference temperature control component, can be determined. By substituting the energy consumption of the reference temperature control component, the thermal comfort parameter PMV, the temperature of the thermally managed object, and the target temperature into the preset reward function, the second reward value can be determined.

[0162] S7034. Based on the first reward value and the second reward value, perform backpropagation training on the preset thermal management strategy function until the convergence condition is met, and obtain the trained thermal management strategy function corresponding to the i-th second target working mode.

[0163] In some embodiments, the preset thermal management strategy function can be backpropagated and trained based on the difference between the first reward value and the second reward value. If the output first reward value and the second reward value differ significantly, the difference can be fed back forward to continuously adjust the parameters in the preset network model until the convergence condition is met and the training is completed. Then, the trained thermal management strategy function corresponding to the i-th second target working mode can be obtained.

[0164] In some embodiments, if the preset thermal management strategy function is a deep reinforcement learning model, state data can be constructed based on the state data sample set corresponding to the reference temperature control component. This state data is then input into the deep reinforcement learning model to obtain corresponding actions (candidate control parameter values). The one-dimensional dynamic model corresponding to the reference temperature control component executes the action based on the candidate control parameter values, obtains the corresponding reward, and obtains the next state data after executing the action. The next state data is then input into the preset thermal management strategy function for retraining to obtain new actions and update the reward. A loss function is established based on the reward function. When the loss function is less than a preset threshold, the convergence condition is satisfied, and the trained thermal management strategy function corresponding to the i-th second target working mode is obtained. After obtaining the trained thermal management strategy functions corresponding to each second target working mode, the trained strategy functions can be combined or integrated to obtain the complete trained strategy function of the entire thermal management system.

[0165] It is understandable that by inputting the state data sample set corresponding to the reference temperature control component into the preset thermal management strategy function to obtain candidate control parameter values, and by using the second reward value determined by the candidate control parameter value and the first reward value determined by the state data sample set corresponding to the reference temperature control component to perform backpropagation training on the preset thermal management strategy function, the trained thermal management strategy function corresponding to the i-th second target working mode can be obtained when the convergence condition is met. This allows the first control parameter value of the temperature control component to be obtained subsequently based on the first state data of the thermal management system and the trained thermal management strategy function.

[0166] In the vehicle thermal management method provided in this application embodiment, firstly, first state data of the thermal management system is acquired; then, the first state data is input into a trained thermal management strategy function to obtain a first control parameter value; according to a pre-constructed mapping rule between the state data of the thermal management system and the control parameter values ​​of the temperature control components in the temperature regulation loop, a second control parameter value corresponding to the first state data is determined; finally, according to the relationship between the first control parameter value and the second control parameter value, the operating parameters of the temperature control components of the corresponding thermal management object are adjusted to regulate the temperature of the corresponding thermal management object. Thus, by using the relationship between the first control parameter value obtained based on the trained thermal management strategy function and the second control parameter value obtained based on the mapping rule between the state data of the thermal management system and the control parameter values ​​of the temperature control components in the temperature regulation loop, the operating parameters of the temperature control components of the determined corresponding thermal management object can be kept within a reasonable range, improving the safety of the temperature control component operation and achieving effective management of the thermal management object.

[0167] The implementation process of the application embodiments in practical application scenarios is described below.

[0168] like Figure 3 The diagram shown is a flowchart of a method for intelligent thermal management of new energy vehicles provided in an embodiment of this application. This method can be implemented through the following steps S801 to S805.

[0169] S801. Obtain the input signal of the thermal management system (equivalent to "first state data of the thermal management system" in other embodiments).

[0170] In some embodiments, the thermal management system input signals may include signals related to thermal management system items and signals of the thermal management system components themselves, such as passenger compartment temperature related signals, battery temperature related signals, electric drive system temperature related signals, domain controller temperature related signals, as well as refrigerant circuit (such as compressor speed, evaporator temperature, expansion valve opening, etc.), coolant circuit status at various points, air circuit status at various points, etc.

[0171] S802. Determine the thermal management system demand mode and operating sub-mode (equivalent to the "first target operating mode" in other embodiments) based on the thermal management system input signal.

[0172] In some embodiments, the thermal management system demand mode can be determined based on the system energy state and target energy. The system energy state can be the current state of the vehicle system, such as Auto mode, cooling mode, heating mode, defrosting, dehumidifying, defogging mode, maintenance mode, etc. Further, the thermal management system demand mode is decomposed into cooling modes (including single passenger compartment cooling, single battery cooling, passenger compartment + battery dual cooling, etc.), and similarly also includes cooling dehumidifying mode, heating dehumidifying mode, heating mode (including single passenger compartment heating, single battery heating, passenger compartment + battery dual heating, etc.), defrosting mode, etc.

[0173] Understandably, the purpose of dividing demand patterns is to further clarify the control requirements of the refrigerant circuit, coolant circuit, and air circuit. Most importantly, it can reduce the complexity of the thermal management system, making it easier to use reinforcement learning algorithms to train the vehicle's thermal management system control strategy. This reduces the input and output dimensions of the reinforcement learning control algorithm, shortens training time, and improves control accuracy. For example, when training a thermal management system control strategy for a single passenger compartment cooling mode, the training "actions" only need to include the compressor target speed, the electronic expansion valve target opening, radiator fan control, and passenger compartment damper control, without needing to consider the cooling or heating of the battery, electric drive system, and domain controller.

[0174] For example, in one possible implementation, the complex thermal management system is simplified by dividing its operating modes. The technical solution of this application achieves approximately 700-800 convergence cycles when training the thermal management strategy function, with each cycle taking approximately 5 minutes, for a total time of about 2.6 days. In contrast, related technical solutions achieve approximately 1000-1200 convergence cycles, with each cycle taking about 20 minutes, for a total time of about 20 days. Therefore, the technical solution provided in this application improves the convergence speed of the thermal management strategy function and enhances the efficiency of training it.

[0175] Furthermore, when this application uses the trained thermal management strategy function for real vehicle testing, the stack memory usage is less than 6 bytes (KB), while in related technologies, due to the centralized processing of multiple operating conditions and the large state space, the stack memory usage is greater than 48KB. Therefore, the technical solution of this application has the characteristics of fast calculation speed and low consumption of controller hardware resources.

[0176] S803. Enable and diagnose the thermal management actuator (equivalent to the "temperature control component" in other embodiments).

[0177] In some embodiments, thermal management actuators are used to manage the temperature of thermally managed objects. These actuators include compressors, expansion valves, evaporators, water pumps, etc., and the thermally managed objects include batteries, motors, passenger compartments, etc. Enable and diagnostic processing of the thermal management actuators can include system rationality diagnosis, component rationality diagnosis, sensor circuit-level diagnosis, motor drive diagnosis, and component enable control, such as compressor protection and start-stop control. By performing enable and diagnostic processing on the thermal management actuators, it is possible to determine whether a fault exists in the actuator. Faulty actuators can be repaired to ensure normal operation and achieve effective management of the thermally managed objects.

[0178] S804. The output value of the thermal management system reinforcement learning training control algorithm (equivalent to the "first control parameter value" in other embodiments) is validated for rationality using the output value of the rule-based thermal management system control algorithm (equivalent to the "second control parameter value" in other embodiments).

[0179] In some embodiments, the rule-based thermal management system control algorithm outputs different drive values ​​according to different load states after the system has been matched and calibrated in the early stage, forming a Map to adjust the control. At the same time, in a multi-input multi-output thermal management control system, it is generally necessary to decouple the parameters first and convert them into multiple single-output control loops for regulation, such as logic gate limiting control strategy or PID control strategy.

[0180] In some embodiments, the output values ​​of the rule-based thermal management system control algorithm and the output values ​​of the thermal management system reinforcement learning training control algorithm can be the working parameter values ​​of the thermal management actuator. The output values ​​of the rule-based thermal management system control algorithm can be determined based on the Map. For example, if the Map records the correspondence between the temperature of the thermal management object and the working parameter values ​​of the corresponding thermal management actuator, the temperature of the thermal management object in the thermal management system can be matched with the Map. After successful matching, the working parameter values ​​of the thermal management actuator corresponding to the thermal management object can be obtained.

[0181] In some embodiments, the output value of the thermal management system reinforcement learning training control algorithm can be determined based on the deep reinforcement learning thermal management policy function. In practice, the deep reinforcement learning thermal management policy function can be trained first, and after training, a trained deep reinforcement learning thermal management policy function (equivalent to the "trained thermal management policy function" in other embodiments) can be obtained. Based on the trained deep reinforcement learning thermal management policy function, the input signal of the thermal management system is predicted to obtain the working parameter value of the thermal management actuator, that is, the output value of the thermal management system reinforcement learning training control algorithm.

[0182] In some embodiments, the derivation of the reinforcement learning training control algorithm for the thermal management system mainly includes the following processes, such as... Figure 4 As shown, firstly, a one-dimensional model of the thermal management system is modeled and calibrated on a simulation platform using test data from a bench or actual vehicle. Simultaneously, reinforcement learning training software is selected, agent hyperparameters are set, and states and rewards are defined. Then, a control strategy training module is obtained by integrating the calibrated one-dimensional model, and a reinforcement learning control algorithm is trained through this module. Finally, the algorithm is integrated with other control modules (such as the thermal management system signal input module, thermal management actuator enable and diagnostic module) to obtain an executable thermal management system control algorithm. The algorithm training is based on different system operating sub-modes, meaning that the deep reinforcement learning thermal management strategy functions corresponding to different system operating sub-modes are different, mainly differing in the network parameters within the model. By dividing the thermal management strategy function training objects into different operating conditions according to the thermal management system's operating modes for deep reinforcement learning training, and then integrating the thermal management strategy functions of each sub-mode into the thermal management processor, multiple trained thermal management strategies are obtained, instead of training all operating modes of all thermal management objects together. This reduces the consumption of controller hardware resources.

[0183] For example, in some embodiments, the deep reinforcement learning algorithm training process includes: constructing a one-dimensional dynamic thermal mathematical model simulating an air conditioning system, a passenger cabin system, and a battery, motor, and engine system; constructing an agent model for predicting control parameters of the one-dimensional dynamic model based on the deep reinforcement learning algorithm; and training the deep reinforcement learning agent using the one-dimensional dynamic model as the training environment to obtain thermal management strategy functions for the air conditioning, passenger cabin, battery, motor, and engine systems.

[0184] In some embodiments, such as Figure 5 As shown, the method for evaluating the rationality of the output value of the reinforcement learning training control algorithm of the thermal management system can be based on the system boundary condition thresholds (battery operating temperature limit, motor temperature rise limit, controller temperature rise limit, compressor protection limit, autopilot controller temperature rise limit, system power limit, etc.), temperature settings or planned values ​​(passenger cabin temperature, battery temperature, controller temperature, system available power and energy, compressor speed, etc.), and comprehensively evaluate it according to four aspects: safety, differentiation, comfort and range. If a weighted coefficient is given and summed, a rationality evaluation value is obtained. When the rationality evaluation value is greater than the set threshold, it means that the rationality evaluation fails; otherwise, it passes.

[0185] For example, regarding compressor speed control, when the thermal management system reinforcement learning training control algorithm outputs the compressor target speed control, several situations may arise, such as: safety: the target speed may cause insufficient cooling capacity of the thermal management system, leading to excessively high inlet temperature of the battery cooling system and potentially causing battery thermal runaway under extreme conditions; secondly, discrepancies: the deviation Δt between the compressor target speed output by the thermal management system reinforcement learning training control algorithm and the speed target calculated by the rule-based thermal management control algorithm is too large, showing significant irrationality compared to normal operating conditions; thirdly, comfort: the current thermal management system reinforcement learning training control algorithm cannot determine issues such as in-vehicle and engine compartment noise caused by compressor speed control, which requires expert experience or real-vehicle calibration to avoid similar NVH problems; fourthly, range: the thermal management system reinforcement learning training control algorithm may lead to excessively high vehicle energy consumption, affecting the vehicle's range. This needs to be evaluated by comparing component control efficiency maps, thermal management system energy efficiency ratio (COP) calculation models, and battery remaining range estimation models to assess the rationality of its control parameters, including the control of the compressor, electric fan, battery, motor, and cooling water pump.

[0186] In some embodiments, if the output value A of the reinforcement learning training control algorithm is less than or equal to the output value B * safety factor f of the rule-based thermal management system control strategy (equivalent to the "control parameter reference value" in other embodiments); and / or, the output value A of the reinforcement learning training control algorithm is less than or equal to the output limit value D of the thermal management system actuator (equivalent to the "control parameter upper limit value" in other embodiments), then the output value of the thermal management system reinforcement learning training control algorithm passes the rationality check (equivalent to "the first control parameter value is within the preset control parameter value range, and the first control parameter value is less than or equal to the control parameter reference value, and the working parameters of the temperature control component are adjusted based on the first control parameter value" in other embodiments).

[0187] S805. If the rationality check passes, the thermal management system actuator will be controlled by the thermal management system reinforcement learning training control algorithm; otherwise, it will be taken over by the rule-based thermal management system control strategy for actuator correction control.

[0188] In some embodiments, if the output value of the thermal management system reinforcement learning training control algorithm passes the rationality check, it indicates that the output value of the thermal management system reinforcement learning training control algorithm is within the normal operating range of the thermal management actuator. The thermal management actuator will not pose a safety risk after operating based on the output value of the thermal management system reinforcement learning training control algorithm. Therefore, the operating parameters of the corresponding thermal management actuator can be adjusted based on this output value to achieve temperature control of the thermally managed object. For example, if the output value of the thermal management system reinforcement learning training control algorithm is the compressor speed, and this compressor speed falls within... Figure 2The shaded area shown indicates that the compressor speed is within a reasonable range, meaning that the output value of the thermal management system reinforcement learning training control algorithm has passed the rationality check.

[0189] In other embodiments, if it is determined that the output value of the thermal management system reinforcement learning training control algorithm fails the rationality check, it indicates that the output value of the thermal management system reinforcement learning training control algorithm is outside the normal working range of the thermal management actuator. If the thermal management actuator runs based on the output value of the thermal management system reinforcement learning training control algorithm, there may be safety risks. In this case, the output value of the rule-based thermal management system control algorithm will be used to adjust the working parameters of the corresponding thermal management actuator to control the temperature of the thermal management object. At the same time, system alarms will be issued, relevant data will be uploaded for regression testing and algorithm training, and the mode strategy function will be updated after completion.

[0190] In some embodiments, such as Figure 3 As shown, after the thermal management system reinforcement learning thermal management strategy function is trained, the corresponding algorithm or software program can be updated to the vehicle thermal management processing unit. The control output of the intelligent thermal management control module and the output of the rule-based soft management control strategy are compared for rationality in the thermal management safety monitoring and comparison module. If the two control outputs are within a reasonable control range, the processing unit executes the control output of the main path; otherwise, it executes the correction output of the redundant path.

[0191] In some embodiments, such as Figure 5 The diagram shown is a safety control architecture diagram of an intelligent control strategy for a thermal management system provided in an embodiment of this application. It mainly includes three parts: a main thermal management processing path 901, a redundant path 902, and a comparator 903. The main path 901 is the automotive thermal management reinforcement learning training control algorithm, the redundant path 902 is the rule-based thermal management system control strategy, and the comparator 903 is the control output of both.

[0192] In some embodiments, such as Figure 7 The diagram shown is a flowchart illustrating another method for intelligent thermal management of new energy vehicles provided in this application embodiment. The following will use... Figure 7 Taking an example, the method for intelligent thermal management of new energy vehicles provided in the embodiments of this application will be described.

[0193] First, step S11 is executed to acquire the input signal of the thermal management system (equivalent to "first state data of the thermal management system" in other embodiments); then, step S12 is executed to determine the thermal management system demand mode based on the thermal management system input signal (equivalent to "second target working mode" in other embodiments); step S13 is executed to further determine the system working sub-mode based on the thermal management system demand mode; step S14 is executed to enable and diagnose the thermal management actuator (equivalent to "fault diagnosis and enabling processing of temperature control components" in other embodiments); step S15 is executed to compare the output value of the rule-based thermal management system control algorithm with the output value of the thermal management system reinforcement learning training control algorithm to determine whether the output value of the thermal management system reinforcement learning training control algorithm is reasonable. If the output value of the thermal management system reinforcement learning training control algorithm is determined to be reasonable, then step S16 is executed to control the thermal management actuator based on the output value of the thermal management system reinforcement learning training control algorithm; otherwise, step S17 is executed to control the thermal management actuator based on the output value of the rule-based thermal management system control algorithm (equivalent to "adjusting the working parameters of the temperature control component of the corresponding thermal management object according to the relationship between the first control parameter value and the second control parameter value" in other embodiments).

[0194] Understandably, by introducing reinforcement learning-based control algorithms into the thermal management system across multiple operating modes, and determining the operating parameters of the thermal management actuators based on the reinforcement learning-based thermal management policy function, the real-time performance of the vehicle's intelligent thermal management system can be improved, and the multi-input, multi-output control requirements of the automotive thermal management system can be met. Furthermore, by using the output values ​​of the rule-based thermal management system control algorithm to verify the rationality of the output values ​​of the reinforcement learning-based control algorithm, rather than directly controlling the thermal management actuators based on the output values ​​of the reinforcement learning-based control algorithm, the safety of the vehicle's intelligent thermal management system can be ensured.

[0195] This application provides a method, apparatus, device, and computer-readable storage medium for vehicle thermal management. Using this technical solution, firstly, first state data of the thermal management system is acquired; then, the first state data is input into a trained thermal management strategy function to obtain a first control parameter value; according to a pre-constructed mapping rule between the state data of the thermal management system and the control parameter values ​​of the temperature control components in the temperature regulation loop, a second control parameter value corresponding to the first state data is determined; finally, based on the relationship between the first and second control parameter values, the operating parameters of the temperature control components of the corresponding thermal management object are adjusted to regulate the temperature of the corresponding thermal management object. Thus, by using the relationship between the first control parameter value obtained based on the trained thermal management strategy function and the second control parameter value obtained based on the mapping rule between the state data of the thermal management system and the control parameter values ​​of the temperature control components in the temperature regulation loop, the operating parameters of the temperature control components of the determined thermal management object can be kept within a reasonable range, improving the safety of the temperature control component operation and achieving effective management of the thermal management object.

[0196] This application provides a vehicle thermal management device. Figure 8 This is a schematic diagram of the structural composition of a vehicle thermal management device provided in an embodiment of this application, as shown below. Figure 8 As shown, the vehicle thermal management device 20 includes:

[0197] The first acquisition module 21 is used to acquire the first state data of the thermal management system, the first state data including the temperature of the thermal management object and the state parameter values ​​of the temperature regulation loop.

[0198] The second acquisition module 22 inputs the first state data into the trained thermal management strategy function to obtain the first control parameter value; wherein, the trained thermal management strategy function is pre-trained based on the state data sample set of the thermal management system;

[0199] The first determining module 23 is used to determine the second control parameter value corresponding to the first state data according to the mapping rule between the state data of the thermal management system and the control parameter values ​​of the temperature control component on the temperature regulation loop.

[0200] The adjustment module 24 is used to adjust the operating parameters of the temperature control component of the corresponding thermal management object according to the relationship between the first control parameter value and the second control parameter value, so as to adjust the temperature of the corresponding thermal management object.

[0201] In some embodiments, the adjustment module 24 includes:

[0202] The first acquisition submodule is used to acquire the safety factor of the temperature control component;

[0203] The first determining submodule is used to determine a reference value for the control parameter based on the safety factor and the second control parameter value;

[0204] The first adjustment submodule is used to adjust the operating parameters of the temperature control component according to the control parameter reference value and the preset control parameter value range.

[0205] In some embodiments, the first adjustment submodule includes: a first adjustment unit, configured to adjust the operating parameters of the temperature control component based on the first control parameter value if the first control parameter value is within the preset control parameter value range and the first control parameter value is less than or equal to the control parameter reference value.

[0206] In some embodiments, the first adjustment submodule further includes: a second adjustment unit, configured to adjust the operating parameters of the temperature control component based on the second control parameter value if the first control parameter value is outside the preset control parameter value range, and / or the first control parameter value is greater than the control parameter reference value.

[0207] In some embodiments, the adjustment module 24 further includes:

[0208] The second determining submodule is used to determine the absolute value of the difference between the first control parameter value and the second control parameter value;

[0209] The second adjustment submodule is used to adjust the operating parameters of the temperature control component of the thermal management object based on the absolute value of the difference, the mapping rule, and the preset temperature range of the thermal management object.

[0210] In some embodiments, the second adjustment submodule includes:

[0211] The first determining unit is used to determine, based on the mapping rule, the first reference temperature of the thermal management object managed by the temperature control component corresponding to the first control parameter value;

[0212] The third adjustment unit is used to adjust the operating parameters of the corresponding temperature control component based on the first control parameter value if the first reference temperature is within the preset temperature range and the absolute value of the difference is less than or equal to the difference threshold.

[0213] In some embodiments, the second adjustment submodule further includes:

[0214] The second determining unit is used to determine, based on the mapping rule, the first reference temperature of the thermal management object managed by the temperature control component corresponding to the first control parameter value;

[0215] The third determining unit is used to determine the second reference temperature of the thermal management object managed by the temperature control component corresponding to the second control parameter value based on the mapping rule if the first reference temperature is outside the preset temperature range and / or the absolute value of the difference is greater than the difference threshold.

[0216] The fourth adjustment unit is used to adjust the operating parameters of the corresponding temperature control component based on the second control parameter value if the second reference temperature is within the preset temperature range.

[0217] In some embodiments, the vehicle thermal management device 20 further includes a signal output module, configured to output a warning message if the first control parameter value meets a first condition, prompting the trained thermal management strategy function to be retrained to obtain a new thermal management strategy function, and to determine the first control parameter value based on the new thermal management strategy function; wherein the first condition includes at least one of the following:

[0218] The first control parameter value is outside the preset control parameter value range;

[0219] The first control parameter value is greater than the control parameter reference value;

[0220] The first reference temperature corresponding to the first control parameter value is outside the preset temperature range;

[0221] The absolute value of the difference between the first control parameter value and the second control parameter value is greater than the difference threshold.

[0222] In some embodiments, the vehicle thermal management device 20 further includes:

[0223] The second determining module is used to determine the first target working mode based on the first state data;

[0224] The third determining module is used to determine the temperature control component corresponding to the first target operating mode and the trained thermal management strategy function.

[0225] In some embodiments, the vehicle thermal management device 20 further includes a fourth determining module, used to perform fault diagnosis and enabling processing on the temperature control component, and to determine that the temperature control component can start normally and work normally.

[0226] In some embodiments, the vehicle thermal management device 20 further includes:

[0227] The fifth determining module is used to determine at least one second target working mode based on the state data sample set;

[0228] The sixth determining module is used to determine the reference temperature control component corresponding to the i-th second target working mode, and the state data sample set corresponding to the reference temperature control component; wherein i is greater than 0 and less than or equal to the total number of second target working modes;

[0229] The training module is used to train the preset thermal management strategy function based on the state data sample set corresponding to the reference temperature control component, so as to obtain the trained thermal management strategy function corresponding to the i-th second target working mode.

[0230] In some embodiments, the training module includes:

[0231] The second acquisition submodule is used to acquire a first reward value determined by the state data sample set corresponding to the reference temperature control component and a preset reward function; the third acquisition submodule is used to input the state data sample set corresponding to the reference temperature control component into the preset thermal management strategy function to obtain candidate control parameter values.

[0232] The third determining submodule is used to determine the second reward value based on the candidate control parameter value and the preset reward function;

[0233] The training submodule is used to perform backpropagation training on the preset thermal management strategy function based on the first reward value and the second reward value until the convergence condition is met, so as to obtain the trained thermal management strategy function corresponding to the i-th second target working mode.

[0234] It should be noted that the description of the vehicle thermal management device in this application embodiment is similar to the description of the corresponding method embodiment described above, and has similar beneficial effects as the method embodiment; therefore, it will not be repeated. For technical details not disclosed in this device embodiment, please refer to the description of the method embodiment of this application for understanding.

[0235] It should be noted that, in the embodiments of this application, if the above-described method for determining video route information is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiments of this application, or the part that contributes to related solutions, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), magnetic disks, or optical disks. Thus, the embodiments of this application are not limited to any specific hardware and software combination.

[0236] Accordingly, embodiments of this application provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the vehicle thermal management method provided in the above embodiments.

[0237] This application also provides a device for vehicle thermal management. Figure 9 This is a schematic diagram of the structural composition of a vehicle thermal management device provided in an embodiment of this application, as shown below. Figure 9 As shown, the vehicle thermal management device 30 includes a memory 31 and a processor 32. The memory 31 stores instructions for executing vehicle thermal management; the processor 32 executes the instructions stored in the memory to implement the vehicle thermal management method provided in the above embodiment.

[0238] The descriptions of the vehicle thermal management device and storage medium embodiments above are similar to those of the method embodiments above, and have similar beneficial effects. For technical details not disclosed in the vehicle thermal management device and storage medium embodiments of this application, please refer to the descriptions of the method embodiments of this application for understanding.

[0239] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising at least one…" does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0240] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods, such as: multiple units or components can be combined, or integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the various components shown or discussed can be through some interfaces, and the indirect coupling or communication connection between devices or units can be electrical, mechanical, or other forms.

[0241] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units. They may be located in one place or distributed across multiple network units. Some or all of the units may be selected to achieve the purpose of this embodiment according to actual needs.

[0242] In addition, each functional unit in the various embodiments of this application can be integrated into one processing unit, or each unit can be a separate unit, or two or more units can be integrated into one unit; the integrated unit can be implemented in hardware or in the form of hardware plus software functional units.

[0243] Those skilled in the art will understand that all or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it performs the steps of the above method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, ROMs, magnetic disks, or optical disks.

[0244] Alternatively, if the integrated units described above are implemented as software functional modules and sold or used as independent products, they can also be stored in a computer-readable storage medium. Based on this understanding, the technical solutions of the embodiments of this application, or the parts that contribute to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a product to execute all or part of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, ROMs, magnetic disks, or optical disks.

[0245] The above description is merely an embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for vehicle thermal management, characterized in that, include: Acquire the first state data of the thermal management system, which includes the temperature of the thermally managed object and the state parameter values ​​of the temperature regulation loop. The first state data is input into the trained thermal management strategy function to obtain the first control parameter value; wherein, the trained thermal management strategy function is pre-trained based on the state data sample set of the thermal management system; Based on the pre-constructed mapping rule between the status data of the thermal management system and the control parameter values ​​of the temperature control components on the temperature regulation loop, the second control parameter value corresponding to the first status data is determined. Based on the relationship between the first control parameter value and the second control parameter value, the operating parameters of the temperature control component of the corresponding thermal management object are adjusted to regulate the temperature of the corresponding thermal management object; The step of adjusting the operating parameters of the temperature control component corresponding to the thermal management object based on the relationship between the first control parameter value and the second control parameter value includes: Obtain the safety factor of the temperature control component; determine the reference value of the control parameter based on the safety factor and the second control parameter value; if the first control parameter value is within the preset control parameter value range and the first control parameter value is less than or equal to the reference value of the control parameter, adjust the operating parameters of the temperature control component based on the first control parameter value; or, Determine the absolute value of the difference between the first control parameter value and the second control parameter value; based on the mapping rule, determine the first reference temperature of the thermal management object managed by the temperature control component corresponding to the first control parameter value; if the first reference temperature is within the preset temperature range of the thermal management object, and the absolute value of the difference is less than or equal to the difference threshold, adjust the working parameters of the corresponding temperature control component based on the first control parameter value.

2. The method according to claim 1, characterized in that, The method further includes: If the first control parameter value is outside the preset control parameter value range, and / or the first control parameter value is greater than the control parameter reference value, the operating parameters of the temperature control component are adjusted based on the second control parameter value.

3. The method according to claim 1, characterized in that, The method further includes: If the first reference temperature is outside the preset temperature range, and / or the absolute value of the difference is greater than the difference threshold, the second reference temperature of the thermal management object managed by the temperature control component corresponding to the second control parameter value is determined based on the mapping rule. If the second reference temperature is within the preset temperature range, the operating parameters of the corresponding temperature control component are adjusted based on the second control parameter value.

4. The method according to claim 1, characterized in that, Also includes: If the value of the first control parameter satisfies the first condition, a warning message is output to prompt the trained thermal management strategy function to be retrained to obtain a new thermal management strategy function, and the value of the first control parameter is determined based on the new thermal management strategy function; wherein, the first condition includes at least one of the following: The first control parameter value is outside the preset control parameter value range; The first control parameter value is greater than the control parameter reference value; The first reference temperature corresponding to the first control parameter value is outside the preset temperature range; The absolute value of the difference between the first control parameter value and the second control parameter value is greater than the difference threshold.

5. The method according to any one of claims 1 to 4, characterized in that, Also includes: Based on the first state data, determine the first target working mode; Determine the temperature control component corresponding to the first target operating mode and the trained thermal management strategy function.

6. The method according to any one of claims 1 to 4, characterized in that, Also includes: The temperature control component is subjected to fault diagnosis and enable processing to determine that the temperature control component can start normally and work normally. Based on the relationship between the first control parameter value and the second control parameter value, the operating parameters of the temperature control component of the corresponding thermal management object are adjusted.

7. The method according to any one of claims 1 to 4, characterized in that, The training process of the trained thermal management strategy function includes: Based on the state data sample set, at least one second target working mode is determined; Determine the reference temperature control component corresponding to the i-th second target operating mode, and the state data sample set corresponding to the reference temperature control component; wherein i is greater than 0 and less than or equal to the total number of second target operating modes; Based on the state data sample set corresponding to the reference temperature control component, the preset thermal management strategy function is trained to obtain the trained thermal management strategy function corresponding to the i-th second target working mode.

8. The method according to claim 7, characterized in that, The step of training a preset thermal management strategy function based on the state data sample set corresponding to the reference temperature control component to obtain the trained thermal management strategy function corresponding to the i-th second target working mode includes: Obtain a first reward value determined by the state data sample set corresponding to the reference temperature control component and a preset reward function; Input the state data sample set corresponding to the reference temperature control component into the preset thermal management strategy function to obtain candidate control parameter values; The second reward value is determined based on the candidate control parameter value and the preset reward function; Based on the first reward value and the second reward value, the preset thermal management strategy function is trained by backpropagation until the convergence condition is met, and the trained thermal management strategy function corresponding to the i-th second target working mode is obtained.

9. A vehicle thermal management device, characterized in that, include: The first acquisition module is used to acquire the first state data of the thermal management system, the first state data including the temperature of the thermal management object and the state parameter values ​​of the temperature regulation loop. The second acquisition module inputs the first state data into the trained thermal management strategy function to obtain the first control parameter value; wherein, the trained thermal management strategy function is pre-trained based on the state data sample set of the thermal management system; The first determining module is used to determine the second control parameter value corresponding to the first state data according to the mapping rule between the state data of the thermal management system and the control parameter values ​​of the temperature control component on the temperature regulation loop. An adjustment module is used to adjust the operating parameters of a temperature control component for a corresponding thermal management object based on the relationship between the first control parameter value and the second control parameter value, thereby adjusting the temperature of the corresponding thermal management object. The adjustment of the operating parameters of the temperature control component based on the relationship between the first control parameter value and the second control parameter value includes: obtaining a safety factor for the temperature control component; determining a control parameter reference value based on the safety factor and the second control parameter value; adjusting the operating parameters of the temperature control component based on the first control parameter value if the first control parameter value is within a preset control parameter value range and the first control parameter value is less than or equal to the control parameter reference value; or, determining the absolute value of the difference between the first control parameter value and the second control parameter value; determining a first reference temperature for the thermal management object managed by the temperature control component corresponding to the first control parameter value based on the mapping rule; and adjusting the operating parameters of the corresponding temperature control component based on the first control parameter value if the first reference temperature is within a preset temperature range of the thermal management object and the absolute value of the difference is less than or equal to a difference threshold.

10. A vehicle thermal management device, characterized in that, include: Memory and processor; Memory used to store instructions that can execute vehicle thermal management; A processor, when executing executable vehicle thermal management instructions stored in the memory, implements the method according to any one of claims 1 to 8.

11. A computer-readable storage medium, characterized in that, The device stores instructions for vehicle thermal management, which, when executed by a processor, implement the method as described in any one of claims 1 to 8.