New energy automobile global intelligent thermal management system based on multi-parameter fusion

Abstract

The invention provides a new energy automobile global intelligent thermal management system based on multi-parameter fusion, and relates to the technical field of new energy automobile thermal management, and the system comprises a data processing module which carries out the data fusion processing of a multi-source data set to form a system state vector. And the execution module is used for performing signal conversion and power driving processing on the global cooperative control instruction and outputting a driving signal, and the execution module outputs the driving signal to a corresponding thermal management actuator. According to the new energy vehicle global intelligent thermal management system based on multi-parameter fusion, multi-source data from a plurality of vehicle systems such as a battery, a motor and an environment are integrated, data standardization and fusion are carried out by adopting a Kalman filtering algorithm, and forward prediction calculation is carried out by combining a thermal load prediction model; and accurate prediction of the multi-scale thermal load of the electric vehicle is realized. And real-time thermal management of the thermal management system is ensured, and the thermal stability and energy efficiency of the system are improved.

Description

Technical Field

[0001] This invention relates to the field of thermal management technology for new energy vehicles, specifically to a new energy vehicle full-domain intelligent thermal management system based on multi-parameter fusion. Background Technology

[0002] Current thermal management systems in new energy vehicles generally adopt a distributed architecture, with independent thermal management loops and dedicated controllers for the battery system, motor cooling system, and passenger compartment air conditioning system. The battery thermal management system is responsible for maintaining the operating temperature range of the lithium-ion battery, focusing on preventing performance degradation at low temperatures and avoiding the risk of thermal runaway at high temperatures. For battery systems using lithium hexafluorophosphate electrolyte, thermal safety management requirements are particularly stringent. The motor cooling system mainly undertakes the task of heat dissipation for the drive assembly, while the passenger compartment air conditioning system focuses on environmental comfort adjustment. Each subsystem independently controls itself based on local parameters collected by its own sensors; typically, the battery management system makes decisions solely based on battery temperature and current data.

[0003] Existing technologies suffer from the core problem of low energy efficiency due to system isolation. Each thermal management subsystem forms an independent information silo, lacking data interaction and collaborative control mechanisms. The overall vehicle thermal management process cannot achieve optimal energy allocation and comprehensive utilization. In low-temperature scenarios where the battery needs heating, the waste heat generated by the motor circulation cannot be effectively utilized by the battery system, and the air conditioning system simultaneously consumes battery power for heating. This inefficient energy utilization directly leads to increased vehicle energy consumption, severely restricting the driving range of new energy vehicles, especially under extreme environmental conditions. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a new energy vehicle full-domain intelligent thermal management system based on multi-parameter fusion. The technical problem this invention aims to solve is: how to address the issues of information silos and lack of collaborative control among various thermal management subsystems through a multi-parameter fusion-based heat load prediction model and a full-domain collaborative control mechanism.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a new energy vehicle full-domain intelligent thermal management system based on multi-parameter fusion, comprising: Parameter acquisition module: The parameter acquisition module collects comprehensive thermal parameters of the whole vehicle in real time to form a multi-source dataset; Data processing module: The data processing module performs data fusion processing on the multi-source dataset to form a system state vector; Heat load prediction module: The heat load prediction module receives future driving condition prediction data from the vehicle navigation system. The heat load prediction module performs forward prediction calculations on the system state vector and the future driving condition prediction data to form a future heat load prediction value. The forward prediction calculation adopts a heat load prediction model. The collaborative control module performs multi-objective optimization processing on the predicted future heat load based on the vehicle operation mode to generate a global collaborative control command. The global collaborative control command includes coordinating motor waste heat recovery, battery thermal management, and passenger cabin air conditioning. Execution module: performs signal conversion and power drive processing on the global coordinated control command and outputs a drive signal. The execution module outputs the drive signal to the corresponding thermal management actuator. Status feedback module: The status feedback module collects the actual operating status of the thermal management actuator in real time, and integrates the actual operating status with the measured values ​​of the vehicle's comprehensive thermal condition parameters to form a system performance feedback vector; Model calibration module: The model calibration module calculates the error between the system performance feedback vector and the future heat load prediction value to generate a deviation value. The model calibration module performs online self-calibration of the parameters of the heat load prediction model based on the deviation value to form a closed-loop control.

[0006] Preferably, the comprehensive thermal operating parameters of the vehicle include the internal temperature and current of the battery pack, the temperature and power of the motor, the set temperature of the passenger compartment, the ambient temperature, and the future driving condition prediction data from the vehicle navigation system. The future driving condition prediction data includes path elevation data, expected average vehicle speed, and ambient temperature forecast data.

[0007] Preferably, the data fusion processing includes standardizing the multi-source dataset to form a standard dataset, and fusing the standard dataset to form the system state vector. The fusion processing uses a Kalman filter algorithm, and the system state vector includes the battery average temperature, battery heat generation rate, motor temperature, and motor heat generation rate.

[0008] Preferably, the future driving condition prediction data includes path elevation, expected average vehicle speed, and ambient temperature forecast; the heat load prediction model includes a thermodynamic heat generation model and a fluid dynamic heat dissipation model; and the forward prediction calculation steps are as follows: S1. The future driving condition prediction data and the system state vector are fused to form an enhanced prediction input vector; S2. Input the enhanced prediction input vector into the thermodynamic heat generation model and the fluid dynamic heat dissipation model. The thermodynamic heat generation model outputs the total heat generation power, and the fluid dynamic heat dissipation model outputs the maximum heat dissipation power. S3. Calculate the difference between the total heat generation power and the maximum heat dissipation power to form a net heat load demand sequence. The difference calculation is based on the component heat capacity corresponding to the temperature data in the system state vector. The net heat load demand sequence includes multiple future time points. S4. The uncertainty of the net heat load demand sequence is quantified to form the future heat load forecast value. The forward forecast calculation adopts a heat load forecasting model, and the calculation formula of the heat load forecasting model is:

[0009] .

[0010] in, The predicted future heat load is dimensionless. The mass of the working fluid is expressed in kg. Standard mass, unit is kg. This refers to the specific heat capacity of the working fluid, expressed in J / (kg·K). This refers to the standard specific heat capacity, expressed in J / (kg·K). The rate of change of temperature is dimensionless. The heat transfer coefficient is expressed in W / (m²·K). This is the standard heat transfer coefficient, expressed in W / (m²·K). This is the temperature difference, dimensionless. The first predictor coefficient is dimensionless. The second prediction coefficient is dimensionless and satisfies... .

[0011] Preferably, the forward prediction calculation is a multi-scale prediction, which includes macroscopic heat load trend prediction and microscopic heat load fluctuation prediction. The time range of the macroscopic heat load trend prediction is the next 30 minutes, and the time range of the microscopic heat load fluctuation prediction is the next 5 minutes. The future heat load prediction value is time-stamped.

[0012] Preferably, the multi-objective optimization process includes constructing an optimization function, the constraints of which include battery operating temperature, motor operating temperature, and passenger compartment temperature, and the optimization objectives of which include minimizing total system energy consumption, minimizing battery life degradation rate, and maximizing passenger compartment comfort. The optimization function is then solved to obtain a Pareto optimal solution set. The solution process employs a non-dominated sorting genetic algorithm to select the optimal solution from the Pareto optimal solution set according to the vehicle operating mode, thereby generating the global cooperative control command.

[0013] Preferably, the non-dominated sorting genetic algorithm has a population size of 100, an iteration count of 200, a crossover probability of 0.8, and a mutation probability of 0.1. The vehicle operation modes include range-priority mode, performance-priority mode, and comfort-priority mode.

[0014] Preferably, the execution module includes a signal conversion unit and a power drive unit. The signal conversion unit converts the digital control quantity in the global collaborative control command into a corresponding analog voltage signal. The power drive unit amplifies the analog voltage signal to generate the drive signal. The thermal management actuator includes a motor waste heat recovery valve, a battery cooling pump, and a PTC heater.

[0015] Preferably, the signal conversion unit uses a digital-to-analog converter with a resolution ≥12 bits and a conversion rate ≥100kSPS. The power drive unit uses different drive circuits for different types of thermal management actuators. The drive circuits include a pulse width modulation drive circuit for the motor waste heat recovery valve, a three-phase inverter drive circuit for the battery cooling pump, and a solid-state relay drive circuit for the PTC heater.

[0016] Preferably, the actual operating state includes the actual opening degree of the motor waste heat recovery valve, the actual speed of the battery cooling pump, and the actual heating power of the PTC heater. The system performance feedback vector includes the battery temperature prediction error, the motor temperature prediction error, and the crew cabin temperature prediction error. The online self-correction adopts the recursive least squares method, and the forgetting factor of the recursive least squares method ranges from 0.95 to 0.99.

[0017] This invention provides a new energy vehicle full-domain intelligent thermal management system based on multi-parameter fusion. It has the following beneficial effects:

[0018] This new energy vehicle full-domain intelligent thermal management system, based on multi-parameter fusion, integrates multi-source data from multiple vehicle systems such as the battery, motor, and environment. It employs a Kalman filter algorithm for data standardization and fusion, and combines this with a heat load prediction model for forward prediction calculations, achieving accurate multi-scale heat load prediction for electric vehicles. This ensures real-time thermal management, improving the system's thermal stability and energy efficiency.

[0019] A non-dominated sorting genetic algorithm was employed for multi-objective optimization. The system coordinated and controlled thermal management aspects such as motor waste heat recovery, battery thermal management, and passenger compartment air conditioning, achieving the lowest system energy consumption, the lowest battery life degradation rate, and the best passenger compartment comfort. This multi-objective optimization strategy improved the overall thermal management efficiency and user experience of electric vehicles. Attached Figure Description

[0020] Figure 1This is a schematic diagram of a system structure for implementing an invention; Figure 2 This is a flowchart illustrating the data processing and heat load prediction process for implementing the invention. Figure 3 It is a multi-objective optimization control flowchart for realizing an invention; Figure 4 This is a schematic diagram of the execution module drive circuit structure for implementing the invention; Figure 5 This is a flowchart of a closed-loop correction control loop for implementing the invention. Detailed Implementation

[0021] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] Example 1 like Figure 1-5 As shown, this embodiment of the invention provides a new energy vehicle full-domain intelligent thermal management system based on multi-parameter fusion, including a parameter acquisition module: the parameter acquisition module collects comprehensive thermal condition parameters of the whole vehicle in real time to form a multi-source dataset. The comprehensive thermal condition parameters of the whole vehicle include the internal temperature and current of the battery pack, the temperature and power of the motor, the set temperature of the passenger compartment, the ambient temperature, and the future driving condition prediction data from the vehicle navigation system. The future driving condition prediction data includes path elevation data, expected average vehicle speed, and ambient temperature forecast data.

[0023] Data Processing Module: The data processing module performs data fusion processing on multi-source datasets to form a system state vector. Data fusion processing includes standardizing the multi-source datasets to form a standard dataset, and then fusing the standard dataset to form the system state vector. The fusion processing uses a Kalman filter algorithm. The system state vector includes the battery average temperature, battery heat generation rate, motor temperature, and motor heat generation rate.

[0024] Heat Load Prediction Module: This module receives future driving condition prediction data from the vehicle navigation system. It performs forward prediction calculations on the system state vector and the future driving condition prediction data to generate the predicted future heat load value. The forward prediction calculation uses a heat load prediction model. The future driving condition prediction data includes path elevation, expected average vehicle speed, and ambient temperature forecast. The heat load prediction model includes a thermodynamic heat generation model and a fluid dynamics heat dissipation model. The forward prediction calculation steps are as follows:

[0025] S1. The future driving condition prediction data and the system state vector are fused to form an enhanced prediction input vector.

[0026] S2. Input the enhanced prediction input vector into the thermodynamic heat generation model and the fluid dynamics heat dissipation model. The thermodynamic heat generation model outputs the total heat generation power, and the fluid dynamics heat dissipation model outputs the maximum heat dissipation power.

[0027] S3. Calculate the difference between the total heat generation power and the maximum heat dissipation power to form a net heat load demand sequence. The difference calculation is based on the component heat capacity corresponding to the temperature data in the system state vector. The net heat load demand sequence includes multiple future time points.

[0028] S4. Uncertainty quantification of the net heat load demand sequence is performed to generate future heat load forecasts. The forward forecast calculation uses a heat load forecasting model, and the calculation formula for the heat load forecasting model is as follows:

[0029] .

[0030] in, This is a dimensionless forecast of future heat load. The mass of the working fluid is expressed in kg. Standard mass, unit is kg. This refers to the specific heat capacity of the working fluid, expressed in J / (kg·K). This refers to the standard specific heat capacity, expressed in J / (kg·K). The rate of change of temperature is dimensionless. The heat transfer coefficient is expressed in W / (m²·K). This is the standard heat transfer coefficient, expressed in W / (m²·K). This is the temperature difference, dimensionless. The first predictor coefficient is dimensionless. The second prediction coefficient is dimensionless and satisfies... .

[0031] Forward forecasting is a multi-scale forecast, which includes macro heat load trend forecasting and micro heat load fluctuation forecasting. The time range for macro heat load trend forecasting is the next 30 minutes, and the time range for micro heat load fluctuation forecasting is the next 5 minutes. The future heat load forecast values ​​are time-stamped.

[0032] The collaborative control module performs multi-objective optimization processing on the predicted future heat load based on the vehicle's operating mode to generate global collaborative control commands. These commands coordinate motor waste heat recovery, battery thermal management, and passenger compartment air conditioning. The multi-objective optimization process involves constructing an optimization function with constraints including battery operating temperature, motor operating temperature, and passenger compartment temperature. The optimization objectives are to minimize total system energy consumption, minimize battery life degradation rate, and maximize passenger compartment comfort. Solving the optimization function yields a Pareto optimal solution set using a non-dominated sorting genetic algorithm. The optimal solution is selected from the Pareto optimal solution set based on the vehicle's operating mode to generate the global collaborative control commands. The non-dominated sorting genetic algorithm has a population size of 100, 200 iterations, a crossover probability of 0.8, and a mutation probability of 0.1. The vehicle operating modes include range-priority mode, performance-priority mode, and comfort-priority mode.

[0033] The execution module performs signal conversion and power drive processing on the global coordinated control commands, outputting drive signals. The execution module then outputs these drive signals to the corresponding thermal management actuators. The execution module includes a signal conversion unit and a power drive unit. The signal conversion unit converts the digital control quantities in the global coordinated control commands into corresponding analog voltage signals. The power drive unit amplifies the analog voltage signals to generate drive signals. The thermal management actuators include a motor waste heat recovery valve, a battery cooling pump, and a PTC heater. The signal conversion unit uses a digital-to-analog converter (DAC) with a resolution ≥12 bits and a conversion rate ≥100 kSPS. The power drive unit employs different drive circuits for different types of thermal management actuators. These drive circuits include a pulse width modulation drive circuit for the motor waste heat recovery valve, a three-phase inverter drive circuit for the battery cooling pump, and a solid-state relay drive circuit for the PTC heater.

[0034] Status Feedback Module: The status feedback module collects the actual operating status of the thermal management actuator in real time. It integrates the actual operating status with the measured values ​​of the vehicle's comprehensive thermal parameters to form a system performance feedback vector. The actual operating status includes the actual opening degree of the motor waste heat recovery valve, the actual speed of the battery cooling pump, and the actual heating power of the PTC heater. The system performance feedback vector includes battery temperature prediction error, motor temperature prediction error, and passenger compartment temperature prediction error. Online self-calibration uses recursive least squares method, with the forgetting factor ranging from 0.95 to 0.99.

[0035] Model calibration module: The model calibration module calculates the error between the system performance feedback vector and the future heat load prediction value to generate a deviation value. Based on the deviation value, the model calibration module performs online self-calibration of the parameters of the heat load prediction model to form a closed-loop control.

[0036] The intelligent thermal management system based on multi-parameter fusion optimizes thermal load management, reduces energy waste, improves the working efficiency of batteries and motors, extends battery life, and maintains the comfort of the passenger cabin.

[0037] By combining real-time feedback and model correction modules, the system can dynamically adjust its thermal management strategy according to actual operating conditions, thereby improving the system's reliability and stability and ensuring optimal thermal management performance under different driving conditions.

[0038] This invention can optimize thermal management according to the vehicle's operating mode, enabling the system to be flexibly adjusted to achieve multiple optimization goals such as minimum energy consumption and maximum comfort.

[0039] By dynamically optimizing the thermal management of the battery, passenger compartment, and motors, safety hazards caused by high or low temperatures are avoided, ensuring the comfort and safety of passengers.

[0040] By combining the Kalman filter algorithm and real-time status feedback, the system can accurately predict and self-adjust its heat load, reduce failures caused by overheating or overcooling, and improve the system's fault tolerance.

[0041] By optimizing the control strategy through a non-dominated sorting genetic algorithm, the system can effectively control according to preset objectives and learn and adjust itself based on environmental changes and real-time data to adapt to different thermal management needs.

[0042] Example 2 This embodiment is a new energy vehicle full-domain intelligent thermal management system based on multi-parameter fusion. Through a heat load prediction module, combined with multi-source data fusion and a heat load prediction model, it calculates the heat load demand of the new energy vehicle in the next 30 minutes to optimize the thermal management system. The specific implementation method is as follows:

[0043] 1. Data Fusion and Enhancement Future driving condition forecast data from vehicle navigation systems: Driving condition forecast data provided by vehicle navigation systems and weather services provides essential input for heat load forecasting. Specific data is as follows:

[0044] Table 1: Forecast data of future driving conditions.

[0045] Vehicle sensor data: Thermal condition data collected in real time by vehicle sensors. Specific data is as follows: Table 2: Onboard Sensor Data Sheet.

[0046] The temperature in the crew cabin is set at 22°C.

[0047] After standardization, the above data is fused using a Kalman filter algorithm to generate the system's state vector, which includes information such as average battery temperature, battery heat generation rate, motor temperature, and motor heat generation rate, providing input for subsequent heat load prediction.

[0048] 2. Predicted Input and Heat Load Calculation Based on the fused enhanced prediction input vector, the data is input into the thermodynamic heat generation model and the fluid dynamics heat dissipation model for heat load calculation: Thermodynamic heat production model: The calculated total heat generation power of the battery is 200W, 210W, 205W, 220W, and 215W, and the total heat generation power of the motor is 300W, 310W, 305W, 320W, and 315W.

[0049] Fluid dynamics heat dissipation model: The calculated maximum heat dissipation power is 150W, 160W, 155W, 150W, 145W, and the maximum heat dissipation capacity of the radiator is 180W, 190W, 185W, 180W, 175W.

[0050] 3. Calculation of net heat load demand The net heat load demand sequence is obtained by calculating the difference between total heat production power and maximum heat dissipation power: The calculation result is: .

[0051] The net heat load demand sequence represents the net heat load required by vehicles at different points in time within the next 30 minutes.

[0052] 4. Quantification of Uncertainty and Future Prediction Uncertainty quantification of the net heat load demand sequence yields the heat load forecast for the next 30 minutes: The predicted macro heat load trend for the next 30 minutes is 120W, 130W, 125W, 130W, and 135W.

[0053] The predicted micro heat load fluctuations for the next 5 minutes are 115W, 125W, 120W, 135W, and 130W.

[0054] The above predictions are time-stamped, representing possible changes in heat load within the next 30 minutes and 5 minutes, respectively.

[0055] 5. Heat load prediction calculation formula The forward prediction calculation uses a heat load prediction model, and the calculation formula for the heat load prediction model is as follows: .

[0056] in, This is a dimensionless forecast of future heat load. The mass of the working fluid is expressed in kg. Standard mass, unit is kg. This refers to the specific heat capacity of the working fluid, expressed in J / (kg·K). This refers to the standard specific heat capacity, expressed in J / (kg·K). The rate of change of temperature is dimensionless. The heat transfer coefficient is expressed in W / (m²·K). This is the standard heat transfer coefficient, expressed in W / (m²·K). This is the temperature difference, dimensionless. The first predictor coefficient is dimensionless. The second prediction coefficient is dimensionless and satisfies... .

[0057] Predictive coefficient: Predictive coefficient and These are set according to the design requirements of the heat load prediction model. The prediction coefficients reflect the relative importance of the two main factors considered by the heat load prediction model when calculating the heat load.

[0058] The thermodynamic heat generation model has a greater impact, while the fluid dynamics heat dissipation model has a relatively smaller impact, but it still has some influence on the predicted heat load. Therefore, the following settings are made: , .

[0059] The mass of the working fluid is determined based on the actual working fluid used in the thermal management system. Based on the actual coolant flow rate and system capacity, the working fluid mass is determined to be 1.2 kg. A standard mass is typically a reference value used for comparison with the working fluid mass. A standard reference mass of 1.0 kg is provided by laboratory testing or during the engineering design phase.

[0060] Specific heat capacity: Based on the specific composition of the coolant and experimental data, the specific heat capacity is confirmed to be 1000 J / (kg·K). The standard specific heat capacity is 950 J / (kg·K) based on the laboratory standard for the selected coolant.

[0061] Temperature change rate: Based on the response time and heat change rate of the thermal management system, the temperature change rate is determined to be 0.02, dimensionless.

[0062] The heat transfer coefficient was determined to be 200 W / (m²·K) based on the actual cooling system design. The standard heat transfer coefficient was obtained as 180 W / (m²·K) under idealized heat exchange conditions.

[0063] Temperature difference: Based on the temperature difference between the radiator and the environment, the temperature difference is set to 5.

[0064] Substitute the above data into the heat load prediction formula: Therefore, the predicted future heat load is 2.24W.

[0065] The calculation results show that the predicted heat load demand is 2.24W, which provides a scientific basis for the optimization of the thermal management system, supports new energy vehicles to achieve efficient thermal management under different operating conditions, and improves system stability and energy utilization efficiency.

[0066] Example 3 This embodiment is based on a multi-parameter fusion-based intelligent thermal management system for new energy vehicles. It optimizes the thermal management system through a collaborative control module, improving system energy efficiency, extending battery life, and enhancing passenger comfort. The specific implementation method is as follows:

[0067] 1. Data Input and Fusion In this embodiment, the collaborative control module receives data from different sensors and systems to achieve comprehensive optimization of the vehicle thermal management system.

[0068] Future heat load forecast data comes from the vehicle navigation system and heat load forecasting module: Future driving condition prediction data: The route elevation is 300m, the expected average vehicle speed is 60km / h, and the ambient temperature is 20℃, all based on real-time traffic and meteorological data predictions.

[0069] Vehicle thermal management status is obtained from the vehicle comprehensive thermal condition parameter acquisition module: The battery temperature is 28℃, the battery current is 50A, the motor temperature is 75℃, the motor power is 100kW, the passenger compartment temperature is set at 22℃, and the ambient temperature is 19℃. All of these data are collected in real time by on-board sensors.

[0070] The above data is fused using the Kalman filter algorithm to generate a system state vector: System state vector: The average battery temperature is 28℃, and the battery heat generation rate is 150W. The motor temperature is 75℃, and the motor heat generation rate is 200W.

[0071] 2. Multi-objective optimization calculation The collaborative control module constructs an optimization function based on the fused system state vector, performs multi-objective optimization, and generates corresponding control commands.

[0072] Optimize constraints: The battery operating temperature range is 20-30℃, the motor operating temperature range is 70-80℃, and the passenger cabin temperature range is 20-24℃.

[0073] Optimization goal: The system minimizes total energy consumption to 50Wh, minimizes battery life degradation rate to 2% per hour, and maximizes passenger cabin comfort to 8.5 / 10.

[0074] The optimized control instructions are obtained by solving the problem using a non-dominated sorting genetic algorithm: Optimized control commands: The opening of the motor waste heat recovery valve is adjusted to 60%, the speed of the battery cooling pump is set to 1500 rpm, and the temperature of the passenger cabin air conditioning is set to 21°C.

[0075] 3. Execution and Driving The execution module performs signal conversion and power drive based on the optimized control commands, and controls each thermal management actuator.

[0076] Signal conversion and power drive: The digital control signal is converted into an analog voltage signal by a digital-to-analog converter. The signal range is 0-5V, the resolution is ≥12 bits, and the conversion rate is ≥100kSPS.

[0077] The battery cooling pump is driven by a three-phase inverter and is set to a speed of 1500 rpm.

[0078] The PTC heater is driven by a solid-state relay and is set to 950W to maintain a comfortable temperature in the cabin.

[0079] 4. Status Feedback and Correction The status feedback module collects the actual operating status of the actuator in real time and compares it with the vehicle's thermal parameters to form a system performance feedback vector. Feedback data: The actual opening degree of the motor waste heat recovery valve is 55%, the actual speed of the battery cooling pump is 1450 rpm, and the actual heating power of the PTC heater is 950W.

[0080] Prediction error: The actual measured battery temperature was 30℃, and the predicted temperature was 28℃, with a prediction error of 2℃. The actual measured motor temperature was 78℃, and the predicted temperature was 75℃, with a prediction error of 3℃. The actual measured passenger compartment temperature was 23.5℃, and the predicted temperature was 22℃, with a prediction error of 1.5℃.

[0081] Online calibration using the recursive least squares method is employed to adjust the parameters of the heat load prediction model, thereby improving prediction accuracy. The adjusted prediction parameters will continue to be used in the heat load prediction for the next cycle.

[0082] 5. Closed-loop control Following status feedback and correction, the collaborative control module performs the next round of optimization based on the updated heat load prediction data, continuously refining the control strategy to ensure the system remains in optimal operating condition under various driving conditions. This optimization process forms a closed-loop control system, monitoring, providing feedback, and correcting the system status in real time to ensure that each component is in its optimal state.

[0083] In summary, the collaborative control module dynamically adjusts the thermal management strategy based on real-time thermal data and driving condition predictions, ensuring that the system achieves a balance between optimal energy efficiency, temperature, and comfort. Through the application of Kalman filtering, non-dominated sorting genetic algorithms, and recursive least squares methods, the system continuously optimizes and responds to different driving conditions, achieving adaptive adjustment of the thermal management system and improving the overall performance and user experience of new energy vehicles.

[0084] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A new energy vehicle full-domain intelligent thermal management system based on multi-parameter fusion, characterized in that, include: Parameter acquisition module: The parameter acquisition module collects comprehensive thermal parameters of the whole vehicle in real time to form a multi-source dataset; Data processing module: The data processing module performs data fusion processing on the multi-source dataset to form a system state vector; Heat load prediction module: The heat load prediction module receives future driving condition prediction data from the vehicle navigation system. The heat load prediction module performs forward prediction calculations on the system state vector and the future driving condition prediction data to form a future heat load prediction value. The forward prediction calculation adopts a heat load prediction model. The collaborative control module performs multi-objective optimization processing on the predicted future heat load based on the vehicle operation mode to generate a global collaborative control command. The global collaborative control command includes coordinating motor waste heat recovery, battery thermal management, and passenger cabin air conditioning. Execution module: performs signal conversion and power drive processing on the global coordinated control command and outputs a drive signal. The execution module outputs the drive signal to the corresponding thermal management actuator. Status feedback module: The status feedback module collects the actual operating status of the thermal management actuator in real time, and integrates the actual operating status with the measured values ​​of the vehicle's comprehensive thermal condition parameters to form a system performance feedback vector; Model calibration module: The model calibration module calculates the error between the system performance feedback vector and the future heat load prediction value to generate a deviation value. The model calibration module performs online self-calibration of the parameters of the heat load prediction model based on the deviation value to form a closed-loop control.

2. The new energy vehicle full-domain intelligent thermal management system based on multi-parameter fusion as described in claim 1, characterized in that: The comprehensive thermal operating parameters of the vehicle include the internal temperature and current of the battery pack, the temperature and power of the motor, the set temperature of the passenger compartment, the ambient temperature, and the future driving condition prediction data from the vehicle navigation system. The future driving condition prediction data includes path elevation data, expected average vehicle speed, and ambient temperature forecast data.

3. The new energy vehicle full-domain intelligent thermal management system based on multi-parameter fusion according to claim 1, characterized in that: The data fusion processing includes standardizing the multi-source dataset to form a standard dataset, and fusing the standard dataset to form the system state vector. The fusion processing uses the Kalman filter algorithm. The system state vector includes the battery average temperature, battery heat generation rate, motor temperature, and motor heat generation rate.

4. The new energy vehicle full-domain intelligent thermal management system based on multi-parameter fusion according to claim 1, characterized in that: The future driving condition prediction data includes path elevation, expected average vehicle speed, and ambient temperature forecast. The heat load prediction model includes a thermodynamic heat generation model and a fluid dynamics heat dissipation model. The forward prediction calculation steps are as follows: S1. The future driving condition prediction data and the system state vector are fused to form an enhanced prediction input vector; S2. Input the enhanced prediction input vector into the thermodynamic heat generation model and the fluid dynamic heat dissipation model. The thermodynamic heat generation model outputs the total heat generation power, and the fluid dynamic heat dissipation model outputs the maximum heat dissipation power. S3. Calculate the difference between the total heat generation power and the maximum heat dissipation power to form a net heat load demand sequence. The difference calculation is based on the component heat capacity corresponding to the temperature data in the system state vector. The net heat load demand sequence includes multiple future time points. S4. The uncertainty of the net heat load demand sequence is quantified to form the predicted value of the future heat load.

5. The new energy vehicle full-domain intelligent thermal management system based on multi-parameter fusion according to claim 1, characterized in that: The forward prediction calculation is a multi-scale prediction, which includes macro-scale heat load trend prediction and micro-scale heat load fluctuation prediction. The time range of the macro-scale heat load trend prediction is the next 30 minutes, and the time range of the micro-scale heat load fluctuation prediction is the next 5 minutes. The future heat load prediction value is time-stamped.

6. The new energy vehicle full-domain intelligent thermal management system based on multi-parameter fusion according to claim 1, characterized in that: The multi-objective optimization process includes constructing an optimization function, the constraints of which include battery operating temperature, motor operating temperature, and passenger compartment temperature. The optimization objectives of the optimization function include minimizing total system energy consumption, minimizing battery life degradation rate, and maximizing passenger compartment comfort. The optimization function is solved to obtain a Pareto optimal solution set. The solution process uses a non-dominated sorting genetic algorithm to select the optimal solution from the Pareto optimal solution set according to the vehicle operating mode to generate the global cooperative control command.

7. The new energy vehicle full-domain intelligent thermal management system based on multi-parameter fusion according to claim 6, characterized in that: The non-dominated sorting genetic algorithm has a population size of 100, an iteration count of 200, a crossover probability of 0.8, and a mutation probability of 0.

1. The vehicle operation modes include range-priority mode, performance-priority mode, and comfort-priority mode.

8. The new energy vehicle full-domain intelligent thermal management system based on multi-parameter fusion according to claim 1, characterized in that: The execution module includes a signal conversion unit and a power drive unit. The signal conversion unit converts the digital control quantity in the global collaborative control command into a corresponding analog voltage signal. The power drive unit amplifies the analog voltage signal to generate the drive signal. The thermal management actuator includes a motor waste heat recovery valve, a battery cooling pump, and a PTC heater.

9. The new energy vehicle full-domain intelligent thermal management system based on multi-parameter fusion according to claim 8, characterized in that: The signal conversion unit uses a digital-to-analog converter with a resolution ≥12 bits and a conversion rate ≥100kSPS. The power drive unit uses different drive circuits for different types of thermal management actuators. The drive circuits include a pulse width modulation drive circuit for the motor waste heat recovery valve, a three-phase inverter drive circuit for the battery cooling pump, and a solid-state relay drive circuit for the PTC heater.

10. The new energy vehicle full-domain intelligent thermal management system based on multi-parameter fusion according to claim 1, characterized in that: The actual operating status includes the actual opening degree of the motor waste heat recovery valve, the actual speed of the battery cooling pump, and the actual heating power of the PTC heater. The system performance feedback vector includes the battery temperature prediction error, the motor temperature prediction error, and the crew cabin temperature prediction error. The online self-correction adopts the recursive least squares method, and the forgetting factor of the recursive least squares method ranges from 0.95 to 0.99.

Images