Distributed electric vehicle energy management and heat management cooperative control method and system

Abstract

The invention relates to the technical field of electric vehicle energy management, and discloses a distributed electric vehicle energy management and heat management cooperative control method and system, and the method comprises the steps: constructing a vehicle longitudinal dynamics model and a heat management system model; a hierarchical control framework is designed based on a CEM-SAC algorithm, and the hierarchical control framework is applied to a vehicle longitudinal dynamics model and a thermal management system model; wherein the hierarchical control framework comprises an upper-layer TDS and a lower-layer TMM; the upper-layer TDS is used for calculating an optimal torque distribution coefficient based on a vehicle longitudinal dynamics model; and the lower-layer TMM is used for controlling the battery temperature and the motor temperature based on the thermal management system model and according to the real-time thermal state and the optimal torque distribution coefficient output by the upper-layer TDS. The temperature control precision of the battery and the motor can be improved, the driving energy consumption and the energy consumption of a thermal management system are reduced, and the health attenuation of the battery can be effectively relieved.

Description

Technical Field

[0001] This invention belongs to the field of electric vehicle energy management technology, and relates to a distributed electric vehicle energy management and thermal management collaborative control method and system. Background Technology

[0002] Electric vehicles (EVs), as a core vehicle for the carbon neutrality strategy, are rapidly replacing traditional gasoline vehicles due to their advantages such as zero emissions and high energy efficiency. Currently, most EVs on the market use single-motor drive, but this has inherent drawbacks such as low energy efficiency and poor fault tolerance. Four-wheel in-wheel motor independently driven electric vehicles (4WID-EVs), with their simplified chassis structure, rapid torque response, and precise control execution, have become one of the industry's recognized future directions for electric vehicle development. Thanks to an advanced vehicle control architecture, the free distribution of torque among the four wheels provides more flexible space for optimizing vehicle performance, allowing torque to be arbitrarily distributed to the four drive motors according to actual road conditions and vehicle driving status.

[0003] However, the application of 4WID-EVs still faces many challenges, such as short driving range, high battery cost, and thermal safety of batteries and motors: (1) The short driving range of 4WID-EVs not only limits its market promotion, but also has an indirect impact on the environment. Therefore, it is urgent to optimize the torque distribution strategy (TDS), improve motor efficiency, reduce energy waste, thereby extending the vehicle's driving range, reducing the indirect impact on the environment, and promoting the sustainable development of 4WID-EVs. (2) Batteries are not only expensive, but their performance and lifespan directly affect the performance and cost of 4WID-EVs. Therefore, how to reduce battery cost is a problem that must be solved for the promotion and application of 4WID-EVs. (3) The batteries and motors of 4WID-EVs generate a lot of heat when running under high load. If the heat cannot be dissipated in time, it will lead to excessively high battery and motor temperatures. High temperatures will accelerate the performance degradation of batteries and motors, reduce their lifespan, and may even cause thermal runaway, bringing serious safety hazards. The current thermal management system is insufficient in dealing with the complex thermal interaction of multiple motors running independently, and cannot effectively ensure the thermal safety of batteries and motors. Therefore, developing an efficient thermal management system (TMS) to achieve precise control of battery and motor temperatures is key to ensuring the thermal safety of 4WID-EVs. Summary of the Invention

[0004] The purpose of this invention is to provide a distributed electric vehicle energy management and thermal management collaborative control method and system, which can improve the accuracy of battery and motor temperature control, reduce drive energy consumption and thermal management system energy consumption, and effectively alleviate battery health degradation.

[0005] To solve the above-mentioned technical problems, the present invention is implemented using the following technical solution.

[0006] In a first aspect, the present invention proposes a distributed electric vehicle energy management and thermal management coordinated control method, comprising:

[0007] Construct a longitudinal dynamics model and a thermal management system model for the vehicle;

[0008] A hierarchical control framework is designed based on the CEM-SAC algorithm. The hierarchical control framework is applied to the vehicle longitudinal dynamics model and thermal management system model to reduce drive energy consumption and thermal management system energy consumption, maintain battery temperature, maintain motor temperature, and slow down the rate of battery health degradation.

[0009] The hierarchical control framework includes an upper-level TDS and a lower-level TMM. The upper-level TDS is used to calculate the optimal torque distribution coefficient based on the vehicle's longitudinal dynamics model. The lower-level TMM is used to control the battery temperature and motor temperature based on the thermal management system model and the optimal torque distribution coefficient output by the upper-level TDS according to the real-time thermal state.

[0010] In conjunction with the first aspect, the collaborative control method of the present invention further includes conducting simulation experiments based on the hierarchical control framework to verify the performance of the design strategy of the hierarchical control framework.

[0011] In conjunction with the first aspect, the method for constructing the vehicle longitudinal dynamics model is as follows:

[0012] The torque requirement of an electric vehicle is calculated as follows:

[0013] ;

[0014] The above equation is the longitudinal dynamics model of the vehicle, where, Indicates demand-driven torque. Indicates the rotating mass conversion factor. For vehicle quality, To accelerate the vehicle, The acceleration due to gravity is taken as 9.8 m / s². 2 , The road slope angle, The rolling resistance coefficient, The air drag coefficient, The frontal area of ​​the vehicle. For vehicle speed, The radius of the wheel;

[0015] The torques of the four hub motors are as follows:

[0016] ;

[0017] ;

[0018] ;

[0019] ;

[0020] In the formula, , , , These represent the torques of the left front, right front, left rear, and right rear hub motors, respectively. Wheel speed; This refers to the wheelbase; This is the direct yaw moment; This is the torque distribution coefficient for the left front and rear wheels. The torque distribution coefficient for the right front and rear wheels is given by, where .

[0021] In conjunction with the first aspect, the method for constructing the thermal management system model is as follows:

[0022] The thermal management system model includes a motor model, a battery model, a pump model, a compressor model, and a fan model;

[0023] The method for constructing the motor model is as follows:

[0024] The output power of the four hub motors is calculated using the following formula:

[0025] ;

[0026] ;

[0027] In the above formula, For the first Motor power; For the first Individual motor efficiency; This represents the total power of the four motors. For the first The speed of each motor; , respectively corresponding ;

[0028] The heat generated by the motor can be calculated using the following formula based on the motor power and motor efficiency:

[0029] ;

[0030] In the formula, For the first Heat generated by each motor;

[0031] No. Temperature changes of individual motors Calculated using the following formula:

[0032] ;

[0033] In the formula, Indicates the specific heat capacity of the motor; Indicates the mass of the motor; Indicates the first Heat dissipation of each motor; and These represent the heat transfer coefficient and heat exchange area between the motor and the motor coolant, respectively. Indicates the temperature of the motor coolant. Indicates the first individual motor temperatures;

[0034] The battery model is constructed as follows:

[0035] Battery power and vehicle traction power are calculated using the following formula:

[0036] ;

[0037] ;

[0038] In the formula, For vehicle traction power, This represents the total energy consumption of the TMS, which is the sum of the energy consumption of the battery pump, motor pump, fan, and compressor. Indicates battery power;

[0039] Battery voltage is calculated using the following formula:

[0040] ;

[0041] In the formula, Open circuit voltage, Polarization voltage, Battery current, This refers to the battery's internal resistance. Battery voltage, Indicates polarization capacitor, Indicates polarization resistance;

[0042] Battery current is calculated using the following formula:

[0043] ;

[0044] The total heat generation rate of the battery is calculated by the following formula:

[0045] ;

[0046] In the formula, This represents the total heat generation rate of the battery. Battery temperature; It is the entropy heat coefficient;

[0047] Battery heat dissipation via convection is calculated using the following formula:

[0048] ;

[0049] In the formula, For battery heat convection heat dissipation, The equivalent heat transfer coefficient, This represents the battery surface area. This refers to the battery coolant temperature.

[0050] Battery thermal changes It can be expressed by the following formula:

[0051] ;

[0052] In the formula, For battery quality, This refers to the specific heat capacity of the battery.

[0053] Battery The descent is represented by the following formula:

[0054] ;

[0055] In the formula, For working hours; for Within a time Decrease value; This represents the percentage of battery capacity loss. For time; express Battery current at all times; The equivalent number of cycles before the battery system reaches the end of its lifespan;

[0056] The percentage of battery capacity loss is calculated using the following formula:

[0057] ;

[0058] This represents the percentage of battery capacity loss. Pre-exponential factors; It is the ideal gas constant; Power-law factor; Ampere-hour throughput;

[0059] activation energy Calculated by the following formula:

[0060] ;

[0061] In the formula, The charge / discharge rate of the battery;

[0062] The method for constructing the pump model is as follows:

[0063] Pump power It can be calculated using the following formula:

[0064] ;

[0065] In the formula, The flow rate of the coolant. For the mechanical efficiency of the pump, The density of the coolant;

[0066] The method for constructing the fan model is as follows:

[0067] Fan power Calculated by the following formula:

[0068] ;

[0069] In the formula, This refers to the fan speed. The coefficients are those of the fit.

[0070] The method for constructing the compressor model is as follows:

[0071] Compressor power Calculated by the following formula:

[0072] ;

[0073] ;

[0074] ;

[0075] In the formula, The specific enthalpy of discharge for the compressor, It is the specific enthalpy of suction. It is the specific enthalpy of isentropic discharge. For isentropic efficiency, For compressor coolant flow rate, suction density, This refers to the compressor displacement. This refers to the compressor speed. For compressor efficiency.

[0076] In conjunction with the first aspect, the hierarchical control framework designed based on the CEM-SAC algorithm further includes:

[0077] CEM is introduced based on the SAC algorithm;

[0078] Design a hierarchical control framework;

[0079] The introduction of CEM based on the SAC algorithm includes the following steps:

[0080] Initialize the population actor network, critic network, and policy covariance matrix;

[0081] By sampling the population with Gaussian noise around the current mean strategy, half of the individuals in the population are evaluated and their fitness is calculated.

[0082] The critic network is updated by the other half of the individuals, and then the policy itself updates the population according to the gradient direction of the critic, resulting in the updated population.

[0083] The updated population is sorted by fitness, and the best-performing population is selected. Each individual is considered an elite, and the selected elite individuals are used to update the distribution parameters.

[0084] In conjunction with the first aspect, the design strategy of the hierarchical control framework is further as follows:

[0085] The hierarchical control framework includes an upper-layer TDS and a lower-layer TMM;

[0086] Define the state space, action space, and reward function of the upper-level TDS;

[0087] Define the state space, action space, and reward function of the lower-level TMM;

[0088] The state space of the upper-level TDS is defined as follows:

[0089] ;

[0090] In the formula, The state of the upper-level TDS; Indicates the vehicle's traction power; These represent the motor power of the left front motor, the left rear motor, the right front motor, and the right rear motor, respectively. Indicates wheel speed;

[0091] The action space of the upper-level TDS is defined as follows:

[0092] ;

[0093] In the formula, For the actions of the upper-level TDS; , These represent the torque distribution coefficients for the left and right front and rear wheels, respectively.

[0094] The reward function of the upper-level TDS is shown in the following formula:

[0095] ;

[0096] In the formula, The stability coefficient; This is a reward for the upper-level TDS; This represents the total power of the four motors.

[0097] The state space of the lower-level TMM is defined as follows:

[0098] ;

[0099] In the formula, This indicates the status of the three-way valve. This refers to the state of the lower-level TMM; These represent the motor temperatures of the left front motor, left rear motor, right front motor, and right rear motor, respectively. Total energy consumption of the thermal management system (TMS);

[0100] The action space of the lower-level TMM is defined as follows:

[0101] ;

[0102] In the formula, For the actions of the lower-level TMM, The power of the motor pump, For the power of the battery pump, For the fan power, This refers to the compressor's power.

[0103] The reward function for the lower-level TMM is shown in the following formula:

[0104] ;

[0105] In the formula, This is a reward for the lower-level TMM; The target temperature set for the battery; The target temperature set for the motor; Here are the weighting coefficients, where .

[0106] Secondly, this invention proposes a distributed electric vehicle energy management and thermal management collaborative control system, comprising:

[0107] The model building module is configured to build vehicle longitudinal dynamics models and thermal management system models.

[0108] The hierarchical control framework design module is configured to design a hierarchical control framework based on the CEM-SAC algorithm, and apply the hierarchical control framework to the vehicle longitudinal dynamics model and thermal management system model to reduce driving energy consumption, maintain battery temperature, maintain motor temperature, and slow down the rate of battery health degradation.

[0109] The hierarchical control framework includes an upper-level TDS and a lower-level TMM. The upper-level TDS is used to calculate the optimal torque distribution coefficient based on the vehicle's longitudinal dynamics model. The lower-level TMM is used to control the battery temperature and motor temperature based on the thermal management system model and the optimal torque distribution coefficient output by the upper-level TDS according to the real-time thermal state.

[0110] Thirdly, the present invention proposes a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described distributed electric vehicle energy management and thermal management coordinated control method.

[0111] Fourthly, the present invention provides a computer device comprising:

[0112] Memory, used to store computer programs;

[0113] A processor is used to execute the computer program to implement the steps of the above-described distributed electric vehicle energy management and thermal management coordinated control method.

[0114] Fifthly, the present invention proposes a computer program product, including a computer program that, when executed by a processor, implements the steps of the above-described distributed electric vehicle energy management and thermal management coordinated control method.

[0115] Compared with the prior art, the beneficial effects achieved by the present invention are as follows:

[0116] (1) This invention proposes a TDS under a hybrid DRL framework for fusion rules for 4WID-EVs. By combining rule-based prior knowledge with DRL through state constraints, it effectively alleviates the problem of low exploration efficiency faced by DRL in global optimization, realizes the optimization of torque distribution of 4WID-EVs, can improve the accuracy of battery and motor temperature control, reduce drive energy consumption and thermal management system energy consumption, and can effectively alleviate battery health degradation.

[0117] (2) The present invention combines the thermal models of the battery and the motor, integrates the motor temperature, battery temperature and battery health status, and enhances the temperature control and energy saving capabilities of the system by dynamically coordinating the key components in the system through DRL, while reducing the battery degradation rate and the system energy consumption.

[0118] (3) This invention designs a hybrid algorithm based on CEM and SAC. By integrating the global exploration capability of CEM, it effectively solves the exploration efficiency problem of SAC in sparse reward environment. CEM maximizes long-term reward by directly optimizing the distribution of policy parameters, overcomes the limitation of traditional SAC relying on dense reward signals, significantly improves the stability and sample efficiency of policy search, and reduces the sensitivity to hyperparameters. Attached Figure Description

[0119] Figure 1 This is a schematic diagram of the framework of the distributed electric vehicle energy management and thermal management collaborative control method in Embodiment 1 of the present invention;

[0120] Figure 2 This is a schematic diagram of the TMS architecture in Embodiment 1 of the present invention;

[0121] Figure 3 This is a schematic diagram of a distributed electric vehicle in Embodiment 1 of the present invention;

[0122] Figure 4 This is a schematic diagram of the high and low torque partitioning of the motor in Embodiment 1 of the present invention. Detailed Implementation

[0123] The technical solution of the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present invention and the specific features in the embodiments are detailed descriptions of the technical solution of the present invention, rather than limitations thereof. In the absence of conflict, the embodiments of the present invention and the technical features in the embodiments can be combined with each other.

[0124] The term "and / or" simply describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone. Additionally, the character " / " generally indicates that the preceding and following related objects have an "or" relationship.

[0125] Example 1

[0126] like Figure 1 , Figure 2 and Figure 3 As shown in this embodiment, the distributed electric vehicle energy management and thermal management coordinated control method includes the following steps:

[0127] Step 1: Construct the vehicle longitudinal dynamics model and thermal management system model (TMS model). Specifically, this involves establishing the vehicle longitudinal dynamics model and the TMS model. For example... Figure 2As shown, the thermal management system (TMS) includes a battery circuit, a motor circuit, an air conditioning circuit, a four-way valve, and a three-way valve. The battery circuit includes a battery, a battery pump, and a cooler connected in series. The motor circuit includes a motor controller, a motor pump, and a radiator, with the motor controller and motor pump connected in series. The air conditioning circuit includes a condenser, an expansion valve, a cooler, and a compressor connected in series. The four ports of the four-way valve are connected to the battery, cooler, motor pump, and three-way valve, respectively. Two ports of the three-way valve are connected to the four-way valve and the radiator, respectively, and the third port is connected between the motor controller and the radiator. The thermal management system in this embodiment also includes a fan. The condenser is located in front of the radiator, and the fan is also located in front of the condenser, accelerating heat dissipation by blowing air. This invention only studies the cooling performance in high-temperature weather. The four-way valve is connected in parallel to the battery circuit and the motor circuit. The three-way valve is connected to the four-way valve, the motor controller, and the radiator, respectively, determining whether the radiator participates in the motor cooling process. The battery circuit, motor circuit, and air conditioning circuit are controlled by the vehicle control unit (VCU). The four hub motors are connected to their respective motor controllers via high-voltage wiring harnesses, and the components of the four motor controllers are integrated into a single high-voltage controller.

[0128] Step 2: Design a hierarchical control framework based on the CEM-SAC algorithm, and apply it to the vehicle longitudinal dynamics model and TMS model constructed in Step 1. This aims to reduce drive energy consumption and thermal management system energy consumption, maintain battery and motor temperatures, and slow down battery health degradation. Specifically:

[0129] Based on the SAC (Soft Actor-Critic) algorithm, a CEM (Cross-Entropy Method) is introduced to design a hierarchical control framework. This framework consists of an upper-level TDS (Torque Distribution Strategies) and a lower-level TMM (Thermal Management Method). Its design includes defining the state space and action space of the upper-level TDS and the lower-level TMM, as well as designing the reward function. The aim is to reduce drive energy consumption (the upper-level TDS achieves torque distribution through...). , To achieve Optimal operation), maintaining stable motor and battery temperatures (the lower-level TMM achieves this by controlling the three-way valve status, fan power, compressor power, and pump power), and slowing down the rate of battery health degradation (the lower-level TMM optimizes this process). To achieve the goals of slowing down the rate of battery health degradation and reducing the energy consumption of key components in the thermal management system (including compressors and fans) (the lower-level TMM achieves this by optimizing the power of the compressor and fan to reduce the energy consumption of key components in the thermal management system).

[0130] Step 3: Conduct simulation experiments based on the hierarchical control framework design strategy to verify the strategy performance. Simulation experiments are performed based on the hierarchical control framework design strategy to verify the strategy performance and examine the optimality and adaptability of the design strategy. Simultaneously, hardware-in-the-loop testing further verifies the real-time performance of the energy management and thermal management coordinated control system of this invention in a real vehicle control unit.

[0131] In a preferred embodiment of the present invention, step one specifically includes the following steps:

[0132] Step S11, the vehicle longitudinal dynamics model is established, and the required driving torque of 4WID-EVs (four-wheel hub motor independently driven electric vehicles) is calculated as follows:

[0133] ;

[0134] In the formula, Indicates demand-driven torque. Indicates the rotating mass conversion factor. For vehicle quality, To accelerate the vehicle, The acceleration due to gravity is taken as 9.8 m / s². 2 , The road slope angle, The rolling resistance coefficient, The air drag coefficient, The frontal area of ​​the vehicle. For vehicle speed, The radius is the wheel radius.

[0135] Assume the torque distribution coefficient of the left front and rear wheels of the vehicle is The torque distribution coefficient of the right front and rear wheels is ,in The torques of the four hub motors are as follows:

[0136] ;

[0137] ;

[0138] ;

[0139] ;

[0140] In the formula, , , , These represent the torques of the left front, right front, left rear, and right rear hub motors, respectively. For wheel speed, Wheelbase This is the direct yaw moment.

[0141] Step S12, establishing the TMS model. The TMS model includes a motor model, a battery model, a pump model (pumps refer to motor pumps and battery pumps), a compressor model, and a fan model:

[0142] Step S121, establishing the motor model. In this embodiment, four identical permanent magnet synchronous motors are selected. The efficiency of the motors is calculated using a quasi-static model. When the vehicle traction force is positive, the output power of the four hub motors is calculated by the following formula:

[0143] ;

[0144] ;

[0145] In the above formula, For the first Motor power, For the first individual motor efficiency. The total power of the four motors, For the first Motor speed, , respectively corresponding ,Right now represent , represent , represent , represent .

[0146] The motor model also includes a motor thermal model, which is used to calculate the heat generation and dissipation of the motor. This invention simplifies the motor thermal model, directly calculating the motor's heat generation based on the motor power and efficiency using the following formula:

[0147] ;

[0148] In the formula, For the first Each motor generates heat.

[0149] According to the first law of thermodynamics, the... Temperature changes of individual motors It can be calculated using the following formula:

[0150] ;

[0151] In the formula, Indicates the specific heat capacity of the motor; Indicates the mass of the motor; Indicates the first Heat dissipation of each motor; and These represent the heat transfer coefficient and heat exchange area between the motor and the motor coolant, respectively. Indicates the temperature of the motor coolant. Indicates the first Motor temperature.

[0152] Step S122, establishing the battery model. The power requirements of the power battery and the vehicle's traction power are calculated using the following formula:

[0153] ;

[0154] ;

[0155] In the formula, For vehicle traction power, This represents the total energy consumption of the TMS, which is the sum of the energy consumption of the battery pump, motor pump, fan, and compressor. This indicates the battery power.

[0156] Battery voltage can be calculated using the following formula:

[0157] ;

[0158] In the formula, Open circuit voltage, Polarization voltage, Battery current, This refers to the battery's internal resistance. Battery voltage, Indicates polarization capacitor, This represents the polarization resistance. The battery current can be expressed as:

[0159] .

[0160] In practical applications, to accurately calculate the heat generation rate of a single battery cell, the battery heat generation rate is simplified. A simplified battery heat generation model is established for theoretical value calculation. The calculation formula for the simplified battery heat generation model is as follows:

[0161] ;

[0162] In the formula, This represents the total heat generation rate of the battery. Battery temperature; Let be the entropy heat coefficient. Battery heat dissipation can be expressed by the following formula:

[0163] ;

[0164] In the formula, For battery heat convection heat dissipation, The equivalent heat transfer coefficient, This represents the battery surface area. This refers to the battery coolant temperature. Therefore, the battery's thermal changes... It can be expressed by the following formula:

[0165] ;

[0166] In the formula, For battery quality, This refers to the specific heat capacity of the battery.

[0167] The decrease in SOH of the battery can be expressed by the following formula:

[0168] ;

[0169] In the formula, For working hours, for The decrease in SOH over time The percentage of battery capacity loss. For time, express Battery current at all times This refers to the equivalent number of cycles before the battery system reaches the end of its lifespan. It also represents the percentage of battery capacity loss. and The charge / discharge rate (i.e., the battery's charge / discharge rate) is related to the battery's internal temperature, and is calculated using the following formula:

[0170] ;

[0171] In the formula, The percentage of battery capacity loss. The pre-exponential factor is obtained by fitting experimental data. Let be the ideal gas constant, and its value is . , The power-law factor is equal to 0.55. This refers to the ampere-hour throughput. The activation energy is calculated using the following formula:

[0172] ;

[0173] In the formula, This refers to the charge / discharge rate of the battery.

[0174] when A 20% decrease in battery power indicates that the battery has reached the end of its lifespan. According to this definition, and It can be represented as follows:

[0175] ;

[0176] ;

[0177] In the formula, This represents the equivalent number of cycles in the total lifetime.

[0178] Step S123: Establishing the pump (motor pump or battery pump) model, compressor model, and fan model. The pump (motor pump or battery pump) power... It can be calculated using the following formula:

[0179] ;

[0180] In the formula, This refers to the flow rate of the coolant (when the pump is an electric motor pump, the coolant refers to the motor coolant; when the pump is a battery pump, the coolant refers to the battery coolant). For the mechanical efficiency of the pump, The density of the coolant; when the pump is an electric motor pump, the pump power. use This indicates the pump's power when the pump is a battery pump. use express.

[0181] Fan power It can be calculated using the following formula:

[0182] ;

[0183] In the formula, This refers to the fan speed. The coefficients are the fitting coefficients, in this embodiment .

[0184] Compressor power It can be calculated using the following formula:

[0185] ;

[0186] ;

[0187] ;

[0188] In the formula, The specific enthalpy of discharge for the compressor, It is the specific enthalpy of suction. It is the specific enthalpy of isentropic discharge. For isentropic efficiency, For compressor coolant flow rate, suction density, This refers to the compressor displacement. This refers to the compressor speed. For compressor efficiency.

[0189] In a preferred embodiment of the present invention, step two specifically includes the following steps:

[0190] Step S21, introducing CEM based on the SAC algorithm, includes the following steps:

[0191] Step S211: Initialize the population actor network. The critic network and the strategy covariance matrix .

[0192] Step S212 involves sampling the population with Gaussian noise around the current mean strategy, as shown in the following equation, and directly evaluating and calculating the fitness of half of the individuals in the population.

[0193]

[0194] In the formula, For individuals within a population, For the individual's serial number, This is a mean-based strategy for the population. Additionally, Figure 2 middle Indicates the population size.

[0195] Step S213: Update the critic network using the other half of the individuals. Then, the strategy itself updates the population based on the gradient direction of the critic network, resulting in the updated population:

[0196] ;

[0197] In the formula, For the parameters of the policy network, For learning rate, In the state At that time, take action Action value function, For the parameters of the value function, In the state Take action below The probability is given by the policy network. For temperature factor, To represent the state From the experience replay pool After sampling, noise The expectation.

[0198] Step S214: Sort the updated population by fitness and select the best performing population. Each individual is considered an elite, and selected elite individuals are used. To update the parameters of the distribution:

[0199] ;

[0200] ;

[0201] ;

[0202] In the above formula, This is the updated population mean. These are the weighting coefficients. For the selected elite individuals, The population mean before the update. The updated covariance matrix, This is a regularization term.

[0203] Step S215: The steps performed by all individuals in the environment are evaluated and stored in the experience replay pool for training the critic network.

[0204] Step S22: A hierarchical control framework was designed, including an upper-level TDS and a lower-level TMM. The design process of the upper-level TDS control method is as follows:

[0205] This approach combines rule-based prior knowledge with Deep Reinforcement Learning (DRL) using state constraints to address the challenge that DRL-based Time-Based Strategy Detection (TDS) struggles to obtain sufficient positive reward signals to guide the discovery of effective optimization strategies during exploration. In the straight-line driving of 4WID-EVs, when torque is low, TDS cannot effectively improve motor efficiency to achieve energy savings. Therefore, the motor efficiency graph is divided into low-torque and high-torque regions. Figure 4 Considering the increased dynamic load on the rear axle during acceleration in practice, the rear wheels can transmit greater driving force with a smaller slip ratio, thereby reducing rolling and slip losses. Therefore, by imposing state constraints in the DRL, rear-wheel drive is implemented in the low-torque region, while the load is distributed through the DRL in the high-torque region, thus solving the problem of sparse positive reward signals during exploration.

[0206] The state space selected by the upper-level TDS aims to accurately describe the operating state of 4WID-EVs during driving, and is defined as follows:

[0207] ;

[0208] In the formula, The state of the upper-level TDS; Indicates wheel speed.

[0209] Action is defined as:

[0210] ;

[0211] In the formula, For the actions of the upper-level TDS; , These represent the torque distribution coefficients for the front and rear wheels on the left and right sides, respectively.

[0212] The goal of upper-level torque distribution is optimal energy efficiency. For 4WID-EVs, changes in battery SOC directly reflect energy consumption during driving. Therefore, setting a reward function based on SOC consumption is a common approach. However, the SOC change at any given moment during driving is relatively small, requiring a scaling factor to balance the instantaneous reward at different times. This adds hyperparameters to the DRL. Therefore, to reduce the difficulty of parameter adjustment, this invention designs the following reward function based on the overall energy utilization efficiency of the vehicle:

[0213] ;

[0214] In the formula, To ensure stability and avoid a denominator of 0, its value is 0.1; This indicates the reward from the upper-level TDS.

[0215] Step S23: A hierarchical control framework was designed, and the design process of the lower-level TMM control method is as follows:

[0216] The lower-level TMM aims to control the battery and motor temperatures to maintain them at set target values, while reducing battery degradation rate and TMM energy consumption. The state space chosen in this invention is designed to accurately describe the operating states within the TMM:

[0217] ;

[0218] In the formula, This indicates the status of the three-way valve. This refers to the state of the lower-level TMM.

[0219] Action is defined as:

[0220] ;

[0221] In the formula, For the actions of the lower-level TMM, The power of the motor pump, For the power of the battery pump, For the fan power, This refers to the compressor's power.

[0222] ;

[0223] In the formula, This is a reward for the lower-level TMM; The target temperature set for the battery is 30°C; The target temperature set for the motor is 60℃; Here are the weighting coefficients, where Their values ​​are respectively .

[0224] In a preferred embodiment of the present invention, step three specifically includes the following steps:

[0225] Step S31: The training cycle is a mixed cycle including WLTC (Globally Harmonized Light Vehicle Test Cycle) and HWFET (Highway Fuel Economy Test), and the test cycle is a mixed cycle including UDDS (Urban Dynamics Drive Cycle) and NEDC (New European Driving Cycle).

[0226] Step S32, the hardware-in-the-loop test system includes a host, USB-CAN, VCU and power supply.

[0227] Step S321: Construct the vehicle longitudinal dynamics model and TMS model through the simulation platform.

[0228] Step S322: The trained hierarchical control framework is brushed into the VCU.

[0229] In step S323, the VCU communicates with the simulation platform via USB-CAN to achieve torque distribution and the expected action control of the components.

[0230] The principle of the hierarchical control framework of this invention is as follows:

[0231] Due to the significant nonlinear coupling between vehicle longitudinal dynamics and the thermal management system, and their fundamentally different dynamic response characteristics, the torque distribution process requires rapid dynamic response to meet real-time driving demands. However, the thermal management process, affected by thermal inertia, exhibits significant hysteresis, with a dynamic response speed much slower than that of the torque distribution process, belonging to different time scales. Furthermore, torque commands directly affect the transient heat generation of the motor and battery, making existing integrated control strategies a complex optimization problem with high dimensions and strong nonlinearity, posing severe challenges in controller design and real-time implementation.

[0232] Therefore, this invention employs a hierarchical control framework to achieve effective system decoupling. The upper-level controller (i.e., the upper-level TDS) focuses on vehicle longitudinal motion control, calculating the optimal torque distribution strategy at a high frequency to improve drive efficiency. The lower-level controller (i.e., the lower-level TMM) focuses on thermal management objectives, adjusting the torque command output by the upper-level TDS based on the real-time thermal state (i.e.,... The system uses seven states (as defined in the diagram) to control the temperature of the battery and motor, ensuring safe system operation and optimizing overall energy efficiency. This functional decomposition significantly reduces the design complexity of the controller and enhances the modularity and reliability of the system. The hierarchical strategy of this invention achieves synergistic optimization of power performance and thermal safety while ensuring real-time dynamic response of the vehicle.

[0233] Example 2

[0234] This embodiment further illustrates the distributed electric vehicle energy management and thermal management coordinated control method of the present invention using more specific data.

[0235] Under training conditions, the hierarchical control framework proposed in this invention reduces drive energy consumption by 4.79% compared with the rule-based control strategy, achieving 93.5% of the theoretically optimal DP algorithm optimization effect. The average efficiency of the four motors is increased by 4.99% compared with the rule-based control strategy, TMS energy consumption is reduced by 59.02%, motor temperature RMSE decreases from 0.67 to 0.16, battery temperature decreases from 0.77 to 0.16, and battery SOH decreases from 6.9e-5 to 6.7e-5.

[0236] In hardware-in-the-loop testing, the hierarchical control framework proposed in this invention reduced drive energy consumption by 7.1% compared with rule-based control strategies, achieving 97.3% of the theoretically optimal DP algorithm optimization effect. TMS energy consumption was reduced by 24.7%, motor temperature RMSE decreased from 0.71 to 0.23, battery temperature decreased from 0.82 to 0.21, and battery SOH decreased from 5.28e-5 to 5.16e-5.

[0237] Example 3

[0238] Based on the same inventive concept as Embodiment 1, this embodiment introduces a distributed electric vehicle energy management and thermal management coordinated control system, including:

[0239] The model building module is configured to build vehicle longitudinal dynamics models and thermal management system models.

[0240] The hierarchical control framework design module is configured to design a hierarchical control framework based on the CEM-SAC algorithm, and apply the hierarchical control framework to the vehicle longitudinal dynamics model and thermal management system model to reduce drive energy consumption and thermal management system energy consumption, maintain battery temperature, maintain motor temperature, and slow down the rate of battery health degradation.

[0241] The hierarchical control framework includes an upper-level TDS and a lower-level TMM. The upper-level TDS is used to calculate the optimal torque distribution coefficient based on the vehicle's longitudinal dynamics model. The lower-level TMM is used to control the battery temperature and motor temperature based on the thermal management system model and the optimal torque distribution coefficient output by the upper-level TDS according to the real-time thermal state.

[0242] Example 4

[0243] Based on the same inventive concept as other embodiments, this embodiment introduces a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described distributed electric vehicle energy management and thermal management coordinated control method.

[0244] Example 5

[0245] Based on the same inventive concept as other embodiments, this embodiment introduces a computer device, including: a memory for storing a computer program; and a processor for executing the computer program to implement the steps of the above-described distributed electric vehicle energy management and thermal management coordinated control method.

[0246] Example 6

[0247] Based on the same inventive concept as other embodiments, this embodiment introduces a computer program product, including a computer program that, when executed by a processor, implements the steps of the above-described distributed electric vehicle energy management and thermal management coordinated control method.

[0248] This invention focuses on distributed four-wheel independent drive electric vehicles, constructing a vehicle longitudinal dynamics model and a thermal management system model. Addressing the complex nonlinear coupling between these two systems, a hierarchical control framework integrating CEM and SAC is designed. The upper layer optimizes drive efficiency through torque distribution, while the lower layer focuses on the thermal management of the vehicle's battery and motor. By controlling key components, it actively manages battery and motor temperature changes, reducing battery health degradation and energy consumption. The hierarchical control framework based on the CEM-SAC algorithm demonstrates superior performance in optimization comparison experiments with different algorithms, showing stronger overall performance compared to offline algorithms DDPG and TD3. Simulation and HIL test results show that the proposed hierarchical control framework achieves approximately 93.5% DP (Power Demand) energy saving, reduces TMS (Total Power Surge) energy consumption by approximately 59.02%, improves the accuracy of battery and motor temperature control, and effectively mitigates battery health degradation. Furthermore, HIL test results further verify the hierarchical control framework's excellent real-time performance and generalization ability, highlighting its strong potential in engineering applications.

[0249] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0250] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0251] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0252] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0253] The embodiments of the present invention have been described above with reference to the accompanying drawings. However, the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other modifications under the guidance of the present invention, and these modifications are all within the protection scope of the present invention.

Claims

1. A method for coordinated control of energy management and thermal management in distributed electric vehicles, characterized in that, include: Construct a longitudinal dynamics model and a thermal management system model for the vehicle; A hierarchical control framework is designed based on the CEM-SAC algorithm. The hierarchical control framework is applied to the vehicle longitudinal dynamics model and thermal management system model to reduce drive energy consumption and thermal management system energy consumption, maintain battery temperature, maintain motor temperature, and slow down the rate of battery health degradation. The hierarchical control framework includes an upper-level TDS and a lower-level TMM. The upper-level TDS is used to calculate the optimal torque distribution coefficient based on the vehicle's longitudinal dynamics model. The lower-level TMM is used to control the battery temperature and motor temperature based on the thermal management system model and the optimal torque distribution coefficient output by the upper-level TDS according to the real-time thermal state.

2. The distributed electric vehicle energy management and thermal management coordinated control method according to claim 1, characterized in that: It also includes conducting simulation experiments based on the hierarchical control framework to verify the performance of the design strategy of the hierarchical control framework.

3. The distributed electric vehicle energy management and thermal management coordinated control method according to claim 1, characterized in that: The method for constructing the vehicle longitudinal dynamics model is as follows: The torque requirement of an electric vehicle is calculated as follows: ； The above equation is the longitudinal dynamics model of the vehicle, where, Indicates demand-driven torque. Indicates the rotating mass conversion factor. For vehicle quality, To accelerate the vehicle, The acceleration due to gravity is taken as 9.8 m / s². 2 , The road slope angle, The rolling resistance coefficient, The air drag coefficient, The frontal area of ​​the vehicle. For vehicle speed, The radius of the wheel; The torques of the four hub motors are as follows: ； ； ； ； In the formula, , , , These represent the torques of the left front, right front, left rear, and right rear hub motors, respectively. Wheel speed; Wheelbase; This is the direct yaw moment; This is the torque distribution coefficient for the left front and rear wheels. The torque distribution coefficient for the right front and rear wheels is given by, where .

4. The distributed electric vehicle energy management and thermal management coordinated control method according to claim 1, characterized in that: The method for constructing the thermal management system model is as follows: The thermal management system model includes a motor model, a battery model, a pump model, a compressor model, and a fan model; The method for constructing the motor model is as follows: The output power of the four hub motors is calculated using the following formula: ； ； In the above formula, For the first Motor power; For the first Individual motor efficiency; This represents the total power of the four motors. For the first The speed of each motor; , respectively corresponding ; The heat generated by the motor can be calculated using the following formula based on the motor power and motor efficiency: ； In the formula, For the first Heat generated by each motor; No. Temperature changes of individual motors Calculated using the following formula: ； In the formula, Indicates the specific heat capacity of the motor; Indicates the mass of the motor; Indicates the first Heat dissipation of each motor; and These represent the heat transfer coefficient and heat exchange area between the motor and the motor coolant, respectively. Indicates the temperature of the motor coolant. Indicates the first individual motor temperatures; The battery model is constructed as follows: Battery power and vehicle traction power are calculated using the following formula: ； ； In the formula, For vehicle traction power, This refers to the total energy consumption of the thermal management system (TMS), which is the sum of the energy consumption of the battery pump, motor pump, fan, and compressor. Indicates battery power; Battery voltage is calculated using the following formula: ； In the formula, Open circuit voltage, Polarization voltage, Battery current, This refers to the battery's internal resistance. Battery voltage, Indicates polarization capacitor, Indicates polarization resistance; The battery current is calculated using the following formula: ； The total heat generation rate of the battery is calculated by the following formula: ； In the formula, This represents the total heat generation rate of the battery. Battery temperature; It is the entropy heat coefficient; Battery heat dissipation via convection is calculated using the following formula: ； In the formula, For battery heat convection heat dissipation, The equivalent heat transfer coefficient, This represents the battery surface area. This refers to the battery coolant temperature. Battery thermal changes It can be expressed by the following formula: ； In the formula, For battery quality, This refers to the specific heat capacity of the battery. battery The descent is represented by the following formula: ； In the formula, For working hours; for Within a time Decrease value; This represents the percentage of battery capacity loss. For time; express Battery current at all times; The equivalent number of cycles before the battery system reaches the end of its lifespan; The percentage of battery capacity loss is calculated using the following formula: ； This represents the percentage of battery capacity loss. Pre-exponential factors; It is the ideal gas constant; Power-law factor; Ampere-hour throughput; activation energy Calculated by the following formula: ； In the formula, The charge / discharge rate of the battery; The method for constructing the pump model is as follows: Pump power It can be calculated using the following formula: ； In the formula, The flow rate of the coolant. For the mechanical efficiency of the pump, The density of the coolant; The method for constructing the fan model is as follows: Fan power Calculated by the following formula: ； In the formula, This refers to the fan speed. The coefficients are those of the fit. The method for constructing the compressor model is as follows: Compressor power Calculated by the following formula: ； ； ； In the formula, The specific enthalpy of discharge for the compressor, It is the specific enthalpy of suction. It is the specific enthalpy of isentropic discharge. For isentropic efficiency, For compressor coolant flow rate, suction density, This refers to the compressor displacement. This refers to the compressor speed. For compressor efficiency.

5. The distributed electric vehicle energy management and thermal management coordinated control method according to claim 1, characterized in that: The hierarchical control framework designed based on the CEM-SAC algorithm includes: CEM is introduced based on the SAC algorithm; Design a hierarchical control framework; The introduction of CEM based on the SAC algorithm includes the following steps: Initialize the population actor network, critic network, and policy covariance matrix; By sampling the population with Gaussian noise around the current mean strategy, half of the individuals in the population are evaluated and their fitness is calculated. The critic network is updated by updating the other half of the individuals, and then the policy itself updates the population according to the gradient direction of the critic, resulting in the updated population. The updated population is sorted by fitness, and the best-performing population is selected. Each individual is considered an elite, and the selected elite individuals are used to update the distribution parameters.

6. The distributed electric vehicle energy management and thermal management coordinated control method according to claim 1, characterized in that: The design strategy of the hierarchical control framework is as follows: The hierarchical control framework includes an upper-layer TDS and a lower-layer TMM; Define the state space, action space, and reward function of the upper-level TDS; Define the state space, action space, and reward function of the lower-level TMM; The state space of the upper-level TDS is defined as follows: ； In the formula, The state of the upper-level TDS; Indicates the vehicle's traction power; These represent the motor power of the left front motor, the left rear motor, the right front motor, and the right rear motor, respectively. Indicates wheel speed; The action space of the upper-level TDS is defined as follows: ； In the formula, For the actions of the upper-level TDS; , These represent the torque distribution coefficients for the left and right front and rear wheels, respectively. The reward function of the upper-level TDS is shown in the following formula: ； In the formula, The stability coefficient; This is a reward for the upper-level TDS; This represents the total power of the four motors. The state space of the lower-level TMM is defined as follows: ； In the formula, This indicates the status of the three-way valve. This refers to the state of the lower-level TMM; These represent the motor temperatures of the left front motor, left rear motor, right front motor, and right rear motor, respectively. Total energy consumption of the thermal management system (TMS); The action space of the lower-level TMM is defined as follows: ； In the formula, For the actions of the lower-level TMM, The power of the motor pump, For the power of the battery pump, For the fan power, This refers to the compressor's power. The reward function for the lower-level TMM is shown in the following formula: ； In the formula, This is a reward for the lower-level TMM; The target temperature set for the battery; The target temperature set for the motor; Here are the weighting coefficients, where .

7. A distributed electric vehicle energy management and thermal management coordinated control system, characterized in that, include: The model building module is configured to build vehicle longitudinal dynamics models and thermal management system models. The hierarchical control framework design module is configured to design a hierarchical control framework based on the CEM-SAC algorithm, and apply the hierarchical control framework to the vehicle longitudinal dynamics model and thermal management system model to reduce drive energy consumption and thermal management system energy consumption, maintain battery temperature, maintain motor temperature, and slow down the rate of battery health degradation. The hierarchical control framework includes an upper-level TDS and a lower-level TMM. The upper-level TDS is used to calculate the optimal torque distribution coefficient based on the vehicle's longitudinal dynamics model. The lower-level TMM is used to control the battery temperature and motor temperature based on the thermal management system model and the optimal torque distribution coefficient output by the upper-level TDS according to the real-time thermal state.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the distributed electric vehicle energy management and thermal management coordinated control method according to any one of claims 1 to 6.

9. A computer device, characterized in that, include: Memory, used to store computer programs; A processor for executing the computer program to implement the steps of the distributed electric vehicle energy management and thermal management coordinated control method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that: When the computer program is executed by the processor, it implements the steps of the distributed electric vehicle energy management and thermal management coordinated control method as described in any one of claims 1 to 6.

Images