CN-122008798-A - Intelligent control method for automobile thermal management and domain controller

CN122008798ACN 122008798 ACN122008798 ACN 122008798ACN-122008798-A

Abstract

The application provides an intelligent control method for automobile heat management and a domain controller, belongs to the field of automobile heat management, and is used for solving the problems of high energy consumption and insufficient cooperative control capability of a heat management system in the related art. The method realizes layered intelligent coordination through three control stages, wherein the first control stage solves the whole vehicle power distribution strategy based on a multi-objective optimization algorithm, the second control stage performs cooperative adjustment on a thermal management loop based on model predictive control and neural network prediction, and the third control stage performs self-adaptive optimization on an execution component based on particle swarm optimization and reinforcement learning. And the information interaction and the overall optimization are realized through key parameter transmission in each stage. The corresponding domain controller comprises a processing unit for executing an algorithm, a signal acquisition module, a driving output module and a vehicle network communication module. The application realizes the global optimization of the energy consumption, safety and comfort of the thermal management system, and improves the endurance mileage and environmental adaptability of the whole vehicle.

Inventors

LING JIAN
WANG BIAO
ZHAO CHUNMING
ZHOU NENGHUI
LI LEI

Assignees

天津易鼎丰动力科技有限公司

Dates

Publication Date: 20260512
Application Date: 20260403

Claims (10)

1. A method for intelligent control of automotive thermal management, the method comprising: The first control stage is based on a multi-objective optimization algorithm, and the power distribution strategy of the range extender and the power battery is obtained by solving with the whole vehicle energy consumption, the battery charge state and the battery temperature as optimization targets, and the working mode of a whole vehicle thermal management system is determined; The second control stage is used for carrying out coupling dynamic adjustment on the battery thermal management loop and the air conditioner thermal management loop based on a model prediction control algorithm according to the working mode and the power distribution strategy, wherein constraint conditions of the model prediction control algorithm are fused with passenger cabin heat demand levels obtained based on neural network prediction; A third control stage, responding to at least one environmental disturbance parameter of the environmental temperature, the solar radiation intensity and the altitude, solving a control parameter combination of each energy consumption component in the heat management system based on a particle swarm optimization algorithm, and dynamically adjusting the operation parameters of the particle swarm optimization algorithm based on a reinforcement learning algorithm; The power distribution strategy output by the first control stage provides input constraint or objective function parameters for the model predictive control algorithm of the second control stage, and the cabin temperature control target output by the second control stage provides a formation basis for the fitness function of the particle swarm optimization algorithm of the third control stage.
2. The intelligent control method for heat management of an automobile according to claim 1, wherein the first control phase comprises: Establishing a multi-objective optimization problem for reducing fuel consumption of the range extender, maintaining the state of charge of the battery in a preset high-efficiency interval and restricting the temperature of the battery in a safe working window as an optimization objective; And adopting a multi-objective optimization algorithm to carry out iterative solution on the multi-objective optimization problem, and outputting a power distribution sequence of the range extender and the power battery.
3. The method of intelligent control of automotive thermal management according to claim 2, wherein the objective functions of the multi-objective optimization problem include a first sub-objective function for evaluating fuel consumption, a second sub-objective function for evaluating battery state of charge deviating from a desired interval, and a third sub-objective function for evaluating battery temperature exceeding a safe operating range.
4. The intelligent control method for heat management of an automobile according to claim 1, wherein the second control stage comprises: establishing a thermodynamic coupling state space model for representing the dynamic change of the battery temperature and the passenger cabin temperature; Solving a rolling optimization problem with constraint conditions based on the thermodynamic coupling state space model, the current system state and the passenger cabin heat demand level in a prediction time domain in each control period, wherein the rolling optimization problem aims at reducing battery temperature tracking errors and actuator control cost; and outputting a control instruction for adjusting at least one actuator in the battery thermal management loop and the air conditioner thermal management loop according to the solving result.
5. The intelligent control method for heat management of an automobile according to claim 4, wherein the constraint condition fusion is based on a passenger cabin heat demand level predicted by a neural network, and comprises: Inputting the environmental temperature, the current passenger cabin temperature, the vehicle speed and the historical heat demand data into a trained neural network model, and predicting to obtain passenger cabin heat demand level in the future time domain; and when the predicted passenger cabin heat requirement level is higher than a preset threshold, activating a constraint condition on the passenger cabin temperature rising rate in the rolling optimization problem.
6. The intelligent control method for heat management of an automobile according to claim 1, wherein the third control stage is based on a particle swarm optimization algorithm to solve a control parameter combination of each energy consumption component in the heat management system, and comprises: Taking control parameters of at least one thermal management system execution component affecting passenger cabin temperature and system energy consumption as particle position vectors; taking the weighted sum of the time period of reaching the target temperature of the passenger cabin and the total energy consumption of the system as a fitness function; and obtaining a group of control parameter combinations of the energy consumption components through iterative search of a particle swarm optimization algorithm.
7. The method of claim 6, wherein dynamically adjusting the operating parameters of the particle swarm optimization algorithm based on a reinforcement learning algorithm comprises: Taking the environmental disturbance parameter as a state of the reinforcement learning intelligent agent; taking the action of adjusting the inertia weight or the learning factor of the particle swarm optimization algorithm as the action of the intelligent agent; constructing a reward function according to the degree of reduction of the total energy consumption of the system relative to the reference value and the degree of deviation of the battery temperature from the target value; and outputting the adjustment quantity of the running parameters of the particle swarm optimization algorithm through the interactive learning of the intelligent agent and the environment.
8. A domain controller for performing the intelligent control method for thermal management of an automobile according to any one of claims 1 to 7, characterized by comprising a hardware platform comprising: the processing unit is configured to execute algorithms of the first control stage, the second control stage and the third control stage; the signal acquisition module is configured to acquire analog signals from the temperature sensor, the pressure sensor and the humidity sensor; the driving output module is configured to output digital control signals for driving the water pump, the valve, the heater and the compressor; the vehicle network communication module is configured to perform data interaction with the whole vehicle controller, the battery management system and the air conditioner controller.
9. The domain controller of claim 8, wherein the vehicle network communication module comprises at least two controller area network bus interfaces, wherein at least one controller area network bus interface supports a specified frame wakeup function at low power consumption.
10. The domain controller of claim 8, wherein the signal acquisition module comprises a plurality of analog signal acquisition channels, wherein at least one analog signal acquisition channel is configured to connect to a high-precision battery temperature sensor.

Description

Intelligent control method for automobile thermal management and domain controller Technical Field The application relates to the field of automobile heat management, in particular to an intelligent control method for automobile heat management and a domain controller. Background With the rapid development of the automobile industry, the whole car heat management system is used as a key for guaranteeing the safety, cruising and riding comfort of the car, and the energy consumption and the intelligent level of the whole car heat management system are widely focused. Currently, the integrated domain controller replaces the traditional distributed control unit, and an advanced algorithm is introduced to improve energy efficiency, so that the integrated domain controller has become an important development direction of industry. Prior art approaches have attempted to optimize thermal management through hardware integration or multi-algorithm application. For example, the technology realizes multi-mode management through eight-way valve integration, but the core chip and the architecture of the technology depend on foreign technology, and the scheme realizes heat intercommunication of a three-electric system, but the control precision and the intelligent degree are to be improved, and the scheme focuses on an intelligent control algorithm and lacks an independently controllable hardware platform as a support. However, the prior art still has shortcomings. The system depends on non-autonomous hardware or lacks systematic cooperation in an algorithm level, and a complete and self-adaptive closed-loop control system from overall vehicle energy global planning, dynamic coordination of multiple heat source loops and fine tuning of an execution part cannot be formed, so that the overall optimal balance of energy consumption, safety and comfortableness is difficult to realize when facing complex and changeable working conditions. Disclosure of Invention The application provides an intelligent control method for automobile thermal management and a domain controller, which can solve the problems of high energy consumption and insufficient cooperative control capability of the traditional thermal management system by using a layered fusion and cooperative optimization intelligent control system. In a first aspect, the present application provides a method for intelligent control of automotive thermal management. The method comprises a first control stage, a second control stage, a third control stage and a particle swarm optimization algorithm, wherein the first control stage is based on a multi-target optimization algorithm, the power distribution strategy of a range extender and a power battery is obtained by solving the multi-target optimization algorithm by taking the whole vehicle energy consumption, the battery charge state and the battery temperature as optimization targets, the working mode of a whole vehicle thermal management system is determined, the second control stage is based on the model prediction control algorithm to dynamically adjust the coupling of a battery thermal management loop and an air conditioner thermal management loop according to the working mode and the power distribution strategy, the constraint condition of the model prediction control algorithm is fused with the passenger cabin thermal demand level obtained based on neural network prediction, the third control stage is based on a particle swarm optimization algorithm to obtain the control parameter combination of each energy consumption component in the thermal management system, and the operation parameter of the particle swarm optimization algorithm is dynamically adjusted based on a reinforcement learning algorithm, the power distribution strategy output by the first control stage is used for providing input constraint or target function parameters for the model prediction control algorithm of the second control stage, and the cabin control target output by the second control stage is used for providing an adaptive function for the particle swarm optimization algorithm. By adopting the technical scheme, the three-layer algorithm of multi-objective optimization, model prediction control and neural network prediction, particle swarm optimization and reinforcement learning is organically fused and cascaded according to the logic of macroscopic power planning, mesoscopic loop coordination and microscopic component tuning. The output of the upper algorithm is used as the input or constraint of the lower algorithm, so that cooperation is formed on the time scale and the control granularity, global, dynamic and self-adaptive optimal control of the whole vehicle thermal management system is realized, and the comprehensive energy consumption of the system is effectively reduced. Further, the first control stage comprises the steps of establishing a multi-objective optimization problem for reducing fuel consumption of the range extender, enabl