New energy automobile thermal management intelligent control method and system based on AI technology

Abstract

The invention relates to the technical field of new energy vehicles, and particularly discloses a new energy vehicle thermal management intelligent control method and system based on an AI technology. The method comprises the following steps: S1, building a thermal management system; S2, collecting and preprocessing data; S3, building an AI algorithm module; S4, carrying out intelligent temperature control analysis, S5, carrying out energy optimization control analysis, and S6, carrying out fault diagnosis control analysis. According to the invention, the thermal management system is established, data acquisition and preprocessing are carried out on the thermal management system, and a temperature prediction model, an energy optimization model and a fault diagnosis model are respectively constructed through an AI algorithm, so that intelligent control of the thermal management system of the new energy vehicle is realized, temperature change can be accurately predicted, energy distribution can be optimized, and the state of the system can be monitored in real time. And the system has fault early warning and fault-tolerant capabilities. According to the system, the heat management efficiency and the energy utilization rate are remarkably improved, the endurance mileage of the vehicle is prolonged, and meanwhile the reliability and safety of the system are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of new energy vehicles, and particularly to an intelligent control method and system for the thermal management of new energy vehicles based on AI technology. Background Art

[0002] The thermal management system of new energy vehicles is crucial for the temperature control of batteries, motors, electronic control systems, and cockpits. Traditional thermal management systems usually adopt fixed threshold control or simple feedback control, which are difficult to cope with complex and changeable working conditions and environmental conditions, resulting in energy waste, low system efficiency, and even affecting vehicle safety and endurance. With the development of AI technology, it becomes possible to use algorithms such as machine learning and deep learning to intelligently control the thermal management system, enabling more accurate temperature prediction and dynamic adjustment. Summary of the Invention

[0003] In order to overcome the above-mentioned defects of the prior art, embodiments of the present invention provide an intelligent control method and system for the thermal management of new energy vehicles based on AI technology to solve the problems raised in the above background art.

[0004] To achieve the above object, the present invention provides the following technical solution: An intelligent control method for the thermal management of new energy vehicles based on AI technology, comprising the following steps:

[0005] S1: Build a thermal management system: Build a test bench and sensor network for the thermal management system of the target new energy vehicle, and deploy an AI control unit in the system;

[0006] S2: Data acquisition and preprocessing: Real-time collect comprehensive data of the thermal management system of the target new energy vehicle through the sensor network and preprocess it;

[0007] S3: Build an AI algorithm module: Based on AI technology, build an AI algorithm module, and the AI algorithm module includes building a temperature prediction model, building an energy optimization model, and building a fault diagnosis model;

[0008] S4: Intelligent temperature control analysis: Based on the temperature prediction output result of the temperature prediction model, perform intelligent temperature control on the thermal management system of the target new energy vehicle, and dynamically adjust the working states of the heat pump system and the heat exchange network;

[0009] S5: Energy optimization control analysis: Based on the energy optimization output result of the energy optimization model, perform energy optimization control on the thermal management system of the target new energy vehicle, and optimize the energy distribution strategy;

[0010] S6: Fault diagnosis control analysis: Based on the fault diagnosis output result of the fault diagnosis model, perform fault diagnosis control on the thermal management system of the target new energy vehicle, and real-time monitor the state of the thermal management system and start the fault tolerance control strategy.

[0011] Preferably, the implementation method of building the thermal management system is as follows:

[0012] Build a thermal management system test bench and sensor network for target new energy vehicles, and deploy an AI control unit in the system;

[0013] The thermal management system test bench includes a thermal management system, a heat exchange coupling system and a multi-source heat pump system; the sensor network includes but is not limited to air temperature sensors, water temperature sensors, humidity sensors, and current sensors; the AI ​​control unit integrates a temperature prediction model, an energy optimization model, and a fault diagnosis model, and performs intelligent control of the thermal management system based on an AI algorithm, including temperature prediction, energy distribution, and dynamic adjustment.

[0014] Preferably, the data collection and preprocessing are performed as follows:

[0015] The comprehensive data of the target new energy vehicle thermal management system is collected in real time through the sensor network, and the comprehensive data includes environmental data, vehicle operation data, component status data, system status data, vehicle demand data, sensor data and system operation parameters. The collected comprehensive data is preprocessed, including data cleaning and removal of outliers and noise data.

[0016] Preferably, the contents of constructing the AI ​​algorithm module are as follows:

[0017] Based on historical data and real-time comprehensive data, a temperature prediction model is established using machine learning algorithms;

[0018] The formula of the temperature prediction model is as follows:

[0019] T t =Activation(W in ×I t +W pr ×H t-1 +Bh)×W ou +B ou , where T t surface

[0020] represents the temperature prediction value at time t, that is, the final output result of the temperature prediction model, Activation represents the activation function, I t represents the input vector at time t, H t-1 represents the state vector of the hidden layer at time t-1, W in Represents the weight matrix from the input layer to the hidden layer, W pr represents the weight matrix from hidden layer to hidden layer, Bh represents the bias vector of hidden layer, W ouDenote the weight matrix from the hidden layer to the output layer, B ou Denote the bias vector of the output layer, with a dimension of 1. Let t represent the serial number of each moment, used to mark different moments, and t takes positive integer values.

[0021] Preferably, the formula of the fault diagnosis model is specifically as follows:

[0022] Among them, P(F k |X) represents the probability that the system has the k-th fault category under the condition of the given input feature vector X, that is, the result finally output by the fault diagnosis model. X represents the input feature vector of the fault diagnosis model, and F k represents the k-th fault category, k represents the serial number of each fault category, k = 1, 2, 3,..., M, and M represents the total number of fault categories. W k represents the weight vector corresponding to the k-th fault category from the hidden layer to the output layer, represents the transpose of the weight vector W k T represents the transpose, and W j represents the weight vector corresponding to the j-th fault category from the hidden layer to the output layer, represents the transpose of the weight vector W j Among them, j represents the index in the summation process of the activation function, and j traverses from 1 to the total number of fault categories M. ReLU represents the rectified linear unit activation function. W1 represents the weight matrix from the input layer to the hidden layer, with the number of rows equal to the number of neurons in the hidden layer and the number of columns equal to the dimension of the input feature vector X. b1 represents the bias vector of the hidden layer, with a length consistent with the number of neurons in the hidden layer, and b k2 represents the bias term corresponding to the k-th fault category from the hidden layer to the output layer, and b j2 represents the bias term corresponding to the j-th fault category from the hidden layer to the output layer, and exp represents the exponential function.

[0023] Preferably, the execution method of the intelligent temperature control analysis is specifically as follows:

[0024] Receive the temperature prediction model for constructing the AI algorithm module in S3, and perform intelligent temperature control on the target new energy vehicle thermal management system based on the temperature prediction output result of the temperature prediction model, and dynamically adjust the working states of the heat pump system and the heat exchange network;

[0025] In a low-temperature environment, preferentially utilize the waste heat of the motor and the electric control system to heat the battery and the passenger compartment;

[0026] In a high-temperature environment, start the cooling function of the heat pump system and quickly dissipate heat through the heat exchange network.

[0027] Preferably, the execution method of the energy optimization control analysis is specifically as follows:

[0028] In step S3, an energy optimization model of the AI algorithm module is constructed. Based on the energy optimization output result of the energy optimization model, energy optimization control is performed on the target new energy vehicle thermal management system, the energy distribution strategy is optimized, the energy usage of the heat pump system and the heat exchange network is dynamically allocated, and when the vehicle brakes or decelerates, the heat generated by the motor is used to heat the battery, reducing external energy consumption.

[0029] Preferably, the execution method of the fault diagnosis control analysis is specifically as follows:

[0030] In step S3, a fault diagnosis model of the AI algorithm module is constructed. Based on the fault diagnosis output result of the fault diagnosis model, fault diagnosis control is performed on the target new energy vehicle thermal management system, the status of the thermal management system is monitored in real time and the fault tolerance control strategy is started, the abnormal status of the thermal management system is detected in time, and when a fault occurs, the fault tolerance control strategy is started.

[0031] To achieve the above object, the present invention provides the following technical solution: A new energy vehicle thermal management intelligent control system based on AI technology, implementing the above-mentioned new energy vehicle thermal management intelligent control method based on AI technology, includes:

[0032] Build a thermal management system module: used to build the thermal management system bench and sensor network of the target new energy vehicle, and deploy an AI control unit in the system;

[0033] Data acquisition and preprocessing module: used to collect the comprehensive data of the target new energy vehicle thermal management system in real time through the sensor network and preprocess it;

[0034] Build an AI algorithm module: establish an AI algorithm module based on AI technology, and the AI algorithm module includes establishing a temperature prediction model, establishing an energy optimization model, and establishing a fault diagnosis model;

[0035] Intelligent temperature control analysis module: based on the temperature prediction output result of the temperature prediction model, perform intelligent temperature control on the target new energy vehicle thermal management system, and dynamically adjust the working status of the heat pump system and the heat exchange network;

[0036] Energy optimization control analysis module: based on the energy optimization output result of the energy optimization model, perform energy optimization control on the target new energy vehicle thermal management system, and optimize the energy distribution strategy;

[0037] Fault diagnosis control analysis module: based on the fault diagnosis output result of the fault diagnosis model, perform fault diagnosis control on the target new energy vehicle thermal management system, monitor the status of the thermal management system in real time and start the fault tolerance control strategy.

[0038] As described above, a new energy vehicle thermal management intelligent control method and system based on AI technology provided by the present invention has at least the following beneficial effects:

[0039] A new energy vehicle thermal management intelligent control method and system based on AI technology provided by the present invention, by building a thermal management system, collecting and preprocessing data of the new energy vehicle thermal management system, respectively constructing a temperature prediction model, an energy optimization model and a fault diagnosis model through AI algorithms, and being able to realize intelligent control of the new energy vehicle thermal management system based on the models. The present invention realizes intelligent control of the thermal management system with the help of AI algorithms, greatly improving the accuracy and response speed of temperature control; relying on the energy optimization model, it can dynamically allocate energy, effectively reduce energy waste, and thus extend the vehicle's cruising range; using the fault diagnosis model to monitor the system status in real time, it can detect and handle potential faults in time, enhancing the system reliability; at the same time, it has strong environmental adaptability, can calmly handle complex and changeable working conditions and environmental conditions, and comprehensively improve the overall performance of the vehicle. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] The present invention is further described with reference to the accompanying drawings, but the embodiments in the drawings do not constitute any limitation to the present invention. For those of ordinary skill in the art, other drawings can also be obtained according to the following drawings without creative efforts.

[0041] Figure 1 It is a schematic flowchart of a new energy vehicle thermal management intelligent control method based on AI technology of the present invention.

[0042] Figure 2 It is a schematic structural diagram of a new energy vehicle thermal management intelligent control system based on AI technology of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0043] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art without creative efforts based on the embodiments of the present invention belong to the protection scope of the present invention.

[0044] Embodiment 1

[0045] Please refer to Figure 1 As shown, the present invention provides a new energy vehicle thermal management intelligent control method based on AI technology, including the following steps:

[0046] S1: Build a thermal management system: Build a thermal management system bench and a sensor network for the target new energy vehicle, and deploy an AI control unit in the system;

[0047] In this embodiment, it should be specifically noted that the implementation method of building the thermal management system is as follows:

[0048] Build the test bench and sensor network of the thermal management system for the target new energy vehicle, and deploy the AI control unit in the system;

[0049] The test bench of the thermal management system includes a thermal management system, a heat exchange coupling system, and a multi-source heat pump system; the sensor network includes, but is not limited to, an air temperature sensor, a water temperature sensor, a humidity sensor, and a current sensor; the AI control unit integrates a temperature prediction model, an energy optimization model, and a fault diagnosis model, and performs intelligent control on the thermal management system based on the AI algorithm, including temperature prediction, energy distribution, and dynamic adjustment.

[0050] In this embodiment, it should be specifically noted that the thermal management system, the heat exchange coupling system, and the multi-source heat pump system are important components of the thermal management of new energy vehicles, and play a key role in ensuring vehicle performance, improving energy utilization efficiency, and optimizing the driving experience;

[0051] The thermal management system is one of the core systems of new energy vehicles and is responsible for precisely regulating the temperatures of the battery, motor, electric control system, and passenger compartment. It consists of a battery thermal management circuit, a motor and electric control thermal management circuit, and a passenger compartment thermal management circuit;

[0052] The core function of the heat exchange coupling system is to achieve heat transfer between different thermal management circuits and promote the efficient use of energy. For example, it can recover the waste heat generated by the motor and electric control circuits and apply it to battery heating or heating the passenger compartment, thereby reducing additional energy consumption;

[0053] The multi-source heat pump system is mainly composed of an air source and a water source heat pump system and can work normally within a relatively wide temperature range (10°C to -20°C).

[0054] S2: Data collection and preprocessing: Real-time collect the comprehensive data of the thermal management system of the target new energy vehicle through the sensor network and preprocess it;

[0055] In this embodiment, it should be specifically noted that the implementation method of the data collection and preprocessing is as follows:

[0056] Real-time collect the comprehensive data of the thermal management system of the target new energy vehicle through the sensor network. The comprehensive data includes environmental data, vehicle operation data, component status data, system status data, vehicle demand data, sensor data, and system operation parameters, and preprocess the collected comprehensive data, including data cleaning, removing outliers and noise data;

[0057] The environmental data includes, but is not limited to, environmental temperature, humidity, and light intensity. Environmental factors will directly or indirectly affect the heat dissipation and temperature rise of vehicle components;

[0058] The vehicle operation data includes, but is not limited to, vehicle driving speed, acceleration, motor power, battery charge and discharge current. The operation state of the vehicle determines the heat generation of each component;

[0059] The component state data includes, but is not limited to, battery temperature, motor temperature, coolant temperature, and coolant flow rate. These data reflect the current thermal state of the components;

[0060] The system state data includes, but is not limited to, the working state (on / off, power) of the heat pump, the flow rate of the heat exchange network, and the temperature and pressure of the coolant;

[0061] The vehicle demand data includes, but is not limited to, the temperature setting of the cockpit and the working temperature requirements of the battery and motor;

[0062] The sensor data includes, but is not limited to, the data collected by temperature sensors, pressure sensors, and flow sensors;

[0063] The system operation parameters include, but are not limited to, the current, voltage, and speed of the heat pump, and the speed of the water pump.

[0064] S3: Construct an AI algorithm module: Based on AI technology, establish an AI algorithm module. The AI algorithm module includes establishing a temperature prediction model, an energy optimization model, and a fault diagnosis model;

[0065] In this embodiment, it should be specifically noted that the content of constructing the AI algorithm module is as follows:

[0066] Based on historical data and real-time comprehensive data, use machine learning algorithms (such as LSTM neural network) to establish a temperature prediction model to predict the future temperature changes of the battery, motor, electric control system, and cockpit;

[0067] The formula of the temperature prediction model is specifically as follows:

[0068] T t =Activation(W in ×I t +W pr ×H t-1 +Bh)×W ou +B ou , where T t represents the temperature prediction value at time t, that is, the final output result of the temperature prediction model, Activation represents the activation function, I

[0069] t ​Denote the input vector at time t, H t-1 Denote the state vector of the hidden layer at time t-1, W in Denote the weight matrix from the input layer to the hidden layer, W pr Denote the weight matrix from the hidden layer to the hidden layer, used to transfer the information of the hidden layer state H at the previous moment t-1 To the information of the hidden layer state at the current moment, Bh denotes the bias vector of the hidden layer, W ou Denote the weight matrix from the hidden layer to the output layer, B ou Denote the bias vector of the output layer, with a dimension of 1, t represents the number of each moment, used to mark different moments, and t takes positive integers;

[0070] Specifically, it should be noted that, I t Denote the input vector at time t, including the collected real-time data, such as environmental temperature, vehicle driving speed, battery current temperature. Among them, I t Denote an n-dimensional vector, I t =[Te t , Vs t , Tb t ,…], Te t Denote the environmental temperature at time t, Vs t Denote the vehicle driving speed at time t, Tb t Denote the battery current temperature at time t.

[0071] The execution method of establishing the energy optimization model is specifically as follows:

[0072] Establish an energy optimization model through a reinforcement learning algorithm (such as Q-learning algorithm), optimize the energy distribution of the heat pump system, heat exchange coupling unit and each heat management loop, and maximize the system efficiency;

[0073] The formula of the energy optimization model is specifically as follows:

[0074] Q t+1 =Q t +α×(r t +γmax a′ Q(S t+1 , a′)-Q(S t , a t ))), where Q t+1 Denote the updated Q value at time t+1, that is, the final output result of the energy optimization model, which is based on the current reward and the estimation of future rewards, and is the result after updating Q t , Q t Denote the Q value at time t, α represents the learning rate, and its value range is between 0 and 1, γ represents the discount factor, S t Denote the state of the heat management system at time t, at denotes the action taken at time t, and Q(S t , a t ) represents the Q-value of taking action a t in state S t , r t denotes the immediate reward obtained by taking action a t at time t, S t+1 represents the state of the thermal management system at time t + 1, that is, the next state entered by the system after taking action a t , a′ represents all possible actions in state S t+1 , and γ max a′ Q(S t+1 , a′) represents the maximum Q-value among all possible actions in state S t+1 ;

[0075] It should be specifically noted that S t represents the state of the thermal management system at time t, which is a vector containing system state data, environmental data, and vehicle demand data; for example: S t represents the state information such as the heat pump operating at a specific power, the flow rate of the heat exchange network being a certain value, the environmental temperature being a certain value, and the cabin temperature being set to a certain temperature at a certain moment;

[0076] a t denotes the action taken at time t, such as adjusting the power of the heat pump or changing the valve opening of the heat exchange network;

[0077] It should be specifically noted that in the Q-learning algorithm of the reinforcement learning algorithm, the Q-value represents the estimated value of the long-term cumulative reward that can be obtained by taking a certain action in a certain state;

[0078] The execution method of establishing the fault diagnosis model is specifically as follows:

[0079] Based on the deep learning algorithm, establish a fault diagnosis model to monitor the operating state of the thermal management system in real time, and timely detect and warn of potential faults;

[0080] The formula of the fault diagnosis model is specifically as follows:

[0081] Among them, P(F k |X) represents the probability that the system appears in the k-th fault category given the input feature vector X, that is, the result finally output by the fault diagnosis model, X represents the input feature vector of the fault diagnosis model, F k represents the k-th fault category, k represents the number of each fault category, k = 1, 2, 3,..., M, M represents the total number of fault categories, and W kIt represents the weight vector corresponding to the k-th fault category from the hidden layer to the output layer. It represents the weight vector W k The transpose of, T represents the transpose, W j It represents the weight vector corresponding to the j-th fault category from the hidden layer to the output layer. It represents the weight vector W j The transpose of, where j represents the index in the summation process of the activation function, j traverses from 1 to the total number of fault categories M, ReLU represents the rectified linear unit activation function, W1 represents the weight matrix from the input layer to the hidden layer, the number of rows of which is the number of neurons in the hidden layer, and the number of columns is equal to the dimension of the input feature vector X, b1 represents the bias vector of the hidden layer, and the length is consistent with the number of neurons in the hidden layer, b k2 It represents the bias term corresponding to the k-th fault category from the hidden layer to the output layer, b j2 It represents the bias term corresponding to the j-th fault category from the hidden layer to the output layer, exp represents the exponential function;

[0082] Specifically, X represents the input vector of the fault diagnosis model, which contains the collected real-time data, such as the data collected by temperature sensors, pressure sensors, and flow sensors, as well as the current, voltage, and rotational speed of the heat pump and the rotational speed of the water pump system operation data; among them, X represents an m-dimensional vector, X = [Ts1, Ts2, Ts3, Ih, Vh,...], Te t It represents the measured value of the temperature sensor, Vs t It represents the measured value of the pressure sensor, Tb t It represents the measured value of the flow sensor, Ih represents the current of the heat pump, and Vh represents the voltage of the heat pump.

[0083] S4: Intelligent temperature control analysis: Based on the temperature prediction output result of the temperature prediction model, perform intelligent temperature control on the target new energy vehicle thermal management system, and dynamically adjust the working states of the heat pump system and the heat exchange network;

[0084] In this embodiment, specifically, the execution method of the intelligent temperature control analysis is as follows:

[0085] Receive the temperature prediction model of the AI algorithm module in S3, and based on the temperature prediction output result of the temperature prediction model, perform intelligent temperature control on the target new energy vehicle thermal management system, and dynamically adjust the working states of the heat pump system and the heat exchange network;

[0086] In a low-temperature environment, preferentially utilize the waste heat of the motor and the electric control system to heat the battery and the passenger compartment;

[0087] In a high-temperature environment, start the cooling function of the heat pump system and quickly dissipate heat through the heat exchange network.

[0088] S5: Energy Optimization Control Analysis: Based on the energy optimization output results of the energy optimization model, perform energy optimization control on the target new energy vehicle thermal management system to optimize the energy distribution strategy;

[0089] In this embodiment, it should be specifically noted that the execution method of the energy optimization control analysis is as follows:

[0090] Receive the energy optimization model of the AI algorithm module constructed in S3. Based on the energy optimization output results of the energy optimization model, perform energy optimization control on the target new energy vehicle thermal management system to optimize the energy distribution strategy, dynamically allocate the energy usage of the heat pump system and the heat exchange network, and when the vehicle brakes or decelerates, use the heat generated by the motor to heat the battery to reduce external energy consumption.

[0091] S6: Fault Diagnosis Control Analysis: Based on the fault diagnosis output results of the fault diagnosis model, perform fault diagnosis control on the target new energy vehicle thermal management system to monitor the thermal management system status in real time and initiate the fault tolerance control strategy.

[0092] In this embodiment, it should be specifically noted that the execution method of the fault diagnosis control analysis is as follows:

[0093] Receive the fault diagnosis model of the AI algorithm module constructed in S3. Based on the fault diagnosis output results of the fault diagnosis model, perform fault diagnosis control on the target new energy vehicle thermal management system to monitor the thermal management system status in real time and initiate the fault tolerance control strategy, timely detect the abnormal status of the thermal management system, and when a fault occurs, initiate the fault tolerance control strategy to ensure the safe operation of the system.

[0094] Embodiment 2

[0095] Please refer to Figure 2 As shown, the present invention provides a new energy vehicle thermal management intelligent control system based on AI technology, including: a thermal management system module building, a data acquisition and preprocessing module, an AI algorithm module building, an intelligent temperature control analysis module, an energy optimization control analysis module, and a fault diagnosis control analysis module;

[0096] Thermal management system module building: Used to build the thermal management system bench and sensor network of the target new energy vehicle, and deploy the AI control unit in the system;

[0097] Data acquisition and preprocessing module: Used to collect the comprehensive data of the target new energy vehicle thermal management system in real time through the sensor network and preprocess it;

[0098] Building AI algorithm modules: Based on AI technology, build AI algorithm modules, which include building a temperature prediction model, building an energy optimization model, and building a fault diagnosis model;

[0099] Intelligent temperature control analysis module: Based on the temperature prediction output results of the temperature prediction model, perform intelligent temperature control on the target new energy vehicle thermal management system, and dynamically adjust the working states of the heat pump system and the heat exchange network;

[0100] Energy optimization control analysis module: Based on the energy optimization output results of the energy optimization model, perform energy optimization control on the target new energy vehicle thermal management system, and optimize the energy distribution strategy;

[0101] Fault diagnosis control analysis module: Based on the fault diagnosis output results of the fault diagnosis model, perform fault diagnosis control on the target new energy vehicle thermal management system, monitor the state of the thermal management system in real time, and activate the fault tolerance control strategy.

[0102] Finally: The above are only the preferred embodiments of the present invention and are not used to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

[0103] The above is only the specific implementation manner of this application, but the protection scope of this application is not limited thereto. Any person skilled in the art can easily think of changes or replacements within the technical scope disclosed in this application, and all of them should be covered by the protection scope of this application. Therefore, the protection scope of this application shall be subject to the protection scope of the claims.

Claims

1. An intelligent control method for the thermal management of new energy vehicles based on AI technology, characterized in that, It includes the following steps: S1: Build a thermal management system: Build a test bench and a sensor network for the thermal management system of the target new energy vehicle, and deploy an AI control unit in the system; S2: Data collection and preprocessing: Collect the comprehensive data of the thermal management system of the target new energy vehicle in real time through the sensor network, and preprocess it; S3: Build an AI algorithm module: Build an AI algorithm module based on AI technology. The AI algorithm module includes building a temperature prediction model, building an energy optimization model, and building a fault diagnosis model; S4: Intelligent temperature control analysis: Based on the temperature prediction output result of the temperature prediction model, perform intelligent temperature control on the thermal management system of the target new energy vehicle, and dynamically adjust the working states of the heat pump system and the heat exchange network; S5: Energy optimization control analysis: Based on the energy optimization output result of the energy optimization model, perform energy optimization control on the thermal management system of the target new energy vehicle, and optimize the energy distribution strategy; S6: Fault diagnosis control analysis: Based on the fault diagnosis output result of the fault diagnosis model, perform fault diagnosis control on the thermal management system of the target new energy vehicle, monitor the state of the thermal management system in real time, and start the fault tolerance control strategy.

2. The intelligent control method for the thermal management of new energy vehicles based on AI technology according to claim 1, wherein: The implementation method of building the thermal management system is specifically as follows: Build a test bench and a sensor network for the thermal management system of the target new energy vehicle, and deploy an AI control unit in the system; The test bench of the thermal management system includes a thermal management system, a heat exchange coupling system, and a multi-source heat pump system; the sensor network includes, but is not limited to, an air temperature sensor, a water temperature sensor, a humidity sensor, and a current sensor; the AI control unit integrates a temperature prediction model, an energy optimization model, and a fault diagnosis model, and performs intelligent control on the thermal management system based on AI algorithms, including temperature prediction, energy distribution, and dynamic adjustment.

3. An intelligent control method for the thermal management of new energy vehicles based on AI technology according to claim 1, characterized in that: The implementation method of the data collection and preprocessing is specifically as follows: Collect the comprehensive data of the thermal management system of the target new energy vehicle in real time through the sensor network. The comprehensive data includes environmental data, vehicle operation data, component status data, system status data, vehicle demand data, sensor data, and system operation parameters, and preprocess the collected comprehensive data, including data cleaning, removing outliers and noise data.

4. A new energy vehicle thermal management intelligent control method based on AI technology according to claim 1, characterized in that: The content of building the AI algorithm module is specifically as follows: Build a temperature prediction model based on historical data and real-time comprehensive data using machine learning algorithms; The formula of the temperature prediction model is specifically as follows: T t = Activation(W in × I t + W pr × H t-1 + Bh) × W ou + B ou , where T t Table The predicted temperature value at time t, which is the final output result of the temperature prediction model. Activation represents the activation function, I t represents the input vector at time t, H t-1 represents the state vector of the hidden layer at time t - 1, W in represents the weight matrix from the input layer to the hidden layer, W pr represents the weight matrix from the hidden layer to the hidden layer. Bh represents the bias vector of the hidden layer, W ou represents the weight matrix from the hidden layer to the output layer, B ou represents the bias vector of the output layer, with a dimension of 1. t represents the number of each time moment, and t takes positive integer values.

5. The intelligent control method for the thermal management of new energy vehicles based on AI technology according to claim 1, characterized in that: The formula of the fault diagnosis model is specifically as follows: Among them, P(F k |X) represents the probability that the system exhibits the k-th fault category given the input feature vector X, that is, the result finally output by the fault diagnosis model. X represents the input feature vector of the fault diagnosis model, and F k represents the k-th fault category, k represents the number of each fault category, k = 1, 2, 3,..., M, and M represents the total number of fault categories. W k represents the weight vector corresponding to the k-th fault category from the hidden layer to the output layer. represents the transpose of the weight vector W k , T represents the transpose, and W j represents the weight vector corresponding to the j-th fault category from the hidden layer to the output layer. represents the transpose of the weight vector W j , where j represents the index in the summation process of the activation function, and j traverses from 1 to the total number of fault categories M. ReLU represents the rectified linear unit activation function, W1 represents the weight matrix from the input layer to the hidden layer, the number of its rows is the number of hidden layer neurons, and the number of columns is equal to the dimension of the input feature vector X. b1 represents the bias vector of the hidden layer, and its length is the same as the number of hidden layer neurons. b k2 represents the bias term corresponding to the k-th fault category from the hidden layer to the output layer, and b j2 represents the bias term corresponding to the j-th fault category from the hidden layer to the output layer. exp represents the exponential function.

6. The intelligent control method for the thermal management of new energy vehicles based on AI technology according to claim 1, characterized in that: The implementation method of the intelligent temperature control analysis is specifically as follows: Receive the temperature prediction model of the AI algorithm module built in S3. Based on the temperature prediction output result of the temperature prediction model, perform intelligent temperature control on the thermal management system of the target new energy vehicle, and dynamically adjust the working states of the heat pump system and the heat exchange network; In a low-temperature environment, preferentially use the waste heat of the motor and the electric control system to heat the battery and the passenger compartment; In a high-temperature environment, start the cooling function of the heat pump system, and quickly dissipate heat through the heat exchange network.

7. An intelligent control method for the thermal management of new energy vehicles based on AI technology according to claim 1, characterized in that: The implementation method of the energy optimization control analysis is specifically as follows: In step S3, an energy optimization model of the AI algorithm module is constructed. Based on the energy optimization output result of the energy optimization model, energy optimization control is performed on the target new energy vehicle thermal management system to optimize the energy distribution strategy and dynamically allocate the energy usage of the heat pump system and the heat exchange network.

8. An intelligent control method for the thermal management of new energy vehicles based on AI technology according to claim 1, characterized in that: The execution method of the fault diagnosis control analysis is specifically as follows: In step S3, a fault diagnosis model of the AI algorithm module is received. Based on the fault diagnosis output result of the fault diagnosis model, fault diagnosis control is performed on the target new energy vehicle thermal management system to monitor the thermal management system status in real time and activate the fault tolerance control strategy, timely detect the abnormal status of the thermal management system, and activate the fault tolerance control strategy when a fault occurs.

9. A new energy vehicle thermal management intelligent control system based on AI technology, which is used for implementing the new energy vehicle thermal management intelligent control method according to any one of claims 1-8 above, and is characterized in that, It includes: Construct a thermal management system module: used to build the thermal management system bench and sensor network of the target new energy vehicle, and deploy an AI control unit in the system; Data acquisition and preprocessing module: used to collect the comprehensive data of the target new energy vehicle thermal management system in real time through the sensor network and preprocess it; Construct an AI algorithm module: Based on AI technology, an AI algorithm module is established. The AI algorithm module includes establishing a temperature prediction model, establishing an energy optimization model, and establishing a fault diagnosis model; Intelligent temperature control analysis module: Based on the temperature prediction output result of the temperature prediction model, perform intelligent temperature control on the target new energy vehicle thermal management system to dynamically adjust the working status of the heat pump system and the heat exchange network; Energy optimization control analysis module: Based on the energy optimization output result of the energy optimization model, perform energy optimization control on the target new energy vehicle thermal management system to optimize the energy distribution strategy; Fault diagnosis control analysis module: Based on the fault diagnosis output result of the fault diagnosis model, perform fault diagnosis control on the target new energy vehicle thermal management system to monitor the thermal management system status in real time and activate the fault tolerance control strategy.

Images