Thermal management method and system of vehicle, vehicle and storage medium

Abstract

The invention discloses a thermal management method and system of a vehicle, the vehicle and a storage medium. The method comprises the steps that multi-source data collected by a vehicle are obtained, and the multi-source data at least comprise vehicle state data, environment data and driving behavior data; performing clustering analysis on the multi-source data, and determining a user driving behavior portrait; based on the user driving behavior portrait and a multi-objective optimization decision algorithm, generating a thermal management control parameter of the vehicle; and performing thermal management on the vehicle based on the thermal management control parameters. According to the invention, the technical problem of low reliability of the thermal management method of the vehicle in the related art is solved.

Description

Technical Field

[0001] This application relates to the field of vehicles, and more specifically, to a thermal management method, system, vehicle, and storage medium for vehicles. Background Technology

[0002] In the era of rapid development of new energy vehicles, improving energy management systems has become crucial for enhancing vehicle performance. Traditional energy management systems rely heavily on preset rules and lack the ability to dynamically perceive the vehicle's environment. Especially under high loads or extreme conditions, this static management system often struggles to achieve optimal energy flow distribution, leading to lower reliability of related vehicle thermal management methods.

[0003] There is currently no effective solution to the above problems. Summary of the Invention

[0004] This application provides a thermal management method, system, vehicle, and storage medium for vehicles, to at least address the technical problem of low reliability in related technologies for vehicle thermal management.

[0005] According to one aspect of the embodiments of this application, a vehicle thermal management method is provided, comprising: acquiring multi-source data collected from the vehicle, wherein the multi-source data includes at least: vehicle status data, environmental data, and driving behavior data; performing cluster analysis on the multi-source data to determine a user driving behavior profile; generating vehicle thermal management control parameters based on the user driving behavior profile and a multi-objective optimization decision algorithm; and performing thermal management on the vehicle based on the thermal management control parameters.

[0006] Furthermore, cluster analysis is performed on multi-source data to determine user driving behavior profiles, including: preprocessing multi-source data to obtain preprocessed data; extracting features from the preprocessed data to obtain driving behavior indicator features; and performing cluster analysis on the driving behavior indicator features to obtain user driving behavior profiles.

[0007] Furthermore, the multi-source data is preprocessed to obtain preprocessed data, including: filtering the acceleration data in the multi-source data to obtain filtered data; synchronizing the timestamps of the filtered data to obtain synchronized data; and calibrating the spatial coordinates of the synchronized data to obtain preprocessed data.

[0008] Furthermore, feature extraction is performed on the preprocessed data to obtain driving behavior index features, including: a rapid acceleration frequency feature based on the acceleration change rate in the preprocessed data, wherein the rapid acceleration frequency feature characterizes the number of times the acceleration change rate exceeds a threshold within a preset time period; a braking attack intensity feature based on deceleration and braking frequency in the preprocessed data, wherein the braking attack intensity feature characterizes the braking force or braking depth of the brake pedal; a speed fluctuation index feature based on vehicle speed in the preprocessed data, wherein the speed fluctuation index characterizes the frequency and amplitude of vehicle speed changes; and a steering variability feature based on steering wheel angle in the preprocessed data, wherein the steering variability feature characterizes the rate of change of steering wheel angle within a preset time period. Based on the rapid acceleration frequency feature, braking attack intensity feature, speed fluctuation index feature, and steering variability feature, the driving behavior index features are obtained.

[0009] Furthermore, based on user driving behavior profiles and multi-objective optimization decision-making algorithms, thermal management control parameters for the vehicle are generated, including: generating initial control parameters for the vehicle based on user driving behavior profiles and multi-objective optimization decision-making algorithms; constructing a performance degradation model for the vehicle based on user driving behavior profiles, wherein the performance degradation model is used to characterize the correlation between user driving behavior profiles and vehicle performance degradation; and adjusting the initial control parameters based on historical vehicle data and the performance degradation model to obtain thermal management control parameters.

[0010] Furthermore, user driving behavior profiles are used to characterize driving behavior patterns. Based on user driving behavior profiles and multi-objective optimization decision-making algorithms, initial control parameters for the vehicle are generated, including: using a pre-defined correlation model to predict the heat load demand of the user driving behavior profile to obtain the target heat load demand, wherein the pre-defined correlation model is used to characterize the quantitative relationship between driving behavior and thermal management demand; constructing the objective function of the multi-objective optimization decision-making algorithm based on the classification type of driving behavior patterns and the target heat load demand; constructing constraints based on battery temperature, pump control signal, and total power; and solving the objective function using a parameter mapping algorithm and constraints to obtain the initial control parameters.

[0011] Furthermore, based on the user driving behavior profile, a vehicle performance degradation model is constructed, including: tracking the user driving behavior profile based on a hidden Markov model to determine the changing trend of the user driving behavior profile; constructing a battery capacity degradation model and a cooling system efficiency degradation model based on the changing trend of the user driving behavior profile; and obtaining the performance degradation model based on the battery capacity degradation model and the cooling system efficiency degradation model.

[0012] Furthermore, the initial control parameters are adjusted based on vehicle historical data and performance degradation model to obtain thermal management control parameters, including: compensating the initial control parameters based on performance degradation model to obtain compensated control parameters; and optimizing the compensated control parameters using vehicle historical data to obtain thermal management control parameters.

[0013] According to another aspect of the embodiments of this application, a vehicle thermal management system is also provided, comprising: a data acquisition module for acquiring multi-source data collected from the vehicle, wherein the multi-source data includes at least: vehicle status data, environmental data, and driving behavior data; an analysis module for performing cluster analysis on the multi-source data to determine a user driving behavior profile; a generation module for generating vehicle thermal management control parameters based on the user driving behavior profile and a multi-objective optimization decision algorithm; and a control module for performing thermal management on the vehicle based on the thermal management control parameters.

[0014] Furthermore, the generation module includes: a decision module, used to generate initial control parameters for the vehicle based on the user's driving behavior profile and a multi-objective optimization decision algorithm; and a feedback module, used to construct a performance degradation model for the vehicle based on the user's driving behavior profile, and adjust the initial control parameters based on historical vehicle data and the performance degradation model to obtain thermal management control parameters, wherein the performance degradation model is used to characterize the correlation between the user's driving behavior profile and the vehicle's performance degradation.

[0015] According to another aspect of the embodiments of this application, a vehicle thermal management device is also provided, comprising: a data acquisition module for acquiring multi-source data collected from the vehicle, wherein the multi-source data includes at least: vehicle status data, environmental data, and driving behavior data; an analysis module for performing cluster analysis on the multi-source data to determine a user driving behavior profile; a generation module for generating vehicle thermal management control parameters based on the user driving behavior profile and a multi-objective optimization decision algorithm; and a control module for performing thermal management on the vehicle based on the thermal management control parameters.

[0016] According to another aspect of the embodiments of this application, a vehicle is also provided, including: a memory storing an executable program; and a processor for running the program, wherein the program executes the methods in various embodiments of this application when it runs.

[0017] According to another aspect of the embodiments of this application, a computer-readable storage medium is also provided, the computer-readable storage medium including a stored executable program, wherein, when the executable program is running, it controls the device where the computer-readable storage medium is located to perform the methods of various embodiments of this application.

[0018] According to another aspect of the embodiments of this application, a computer program product is also provided, including a computer program that, when executed by a processor, implements the methods of various embodiments of this application.

[0019] According to another aspect of the embodiments of this application, a computer program product is also provided, including a non-volatile computer-readable storage medium storing a computer program that, when executed by a processor, implements the methods in various embodiments of this application.

[0020] According to another aspect of the embodiments of this application, a computer program is also provided, which, when executed by a processor, implements the methods of the various embodiments of this application.

[0021] In this embodiment, firstly, multi-source data, including vehicle status data, environmental data, and driving behavior data, are acquired. Next, cluster analysis is performed on the multi-source data to determine the user's driving behavior profile. Further, based on the user's driving behavior profile and a multi-objective optimization decision algorithm, thermal management control parameters for the vehicle are generated. Finally, thermal management of the vehicle is implemented based on these parameters. This application first constructs a comprehensive data framework reflecting the actual driving environment and behavioral patterns by collecting multi-source data from all angles. Then, cluster analysis algorithms are used to process the collected multi-source data in depth to determine the user's driving behavior profile. This profile generation transforms abstract driving behavior into concrete and quantifiable features, laying the foundation for personalized thermal management strategies. Based on this, the most suitable thermal management control parameters for the current driving needs are determined using the user's driving behavior profile and a multi-objective optimization decision algorithm. Finally, the thermal management control parameters are applied to the vehicle's thermal management system to achieve precise control of the vehicle's thermal management system. This application generates dynamic thermal management control parameters by collecting and analyzing multi-source data and combining them with a multi-objective optimization decision-making algorithm. This overcomes the limitations of static and passive thermal management methods in existing technologies, achieving more refined and flexible thermal management control. It significantly improves the energy efficiency of new energy vehicles under various driving conditions and enhances the user's driving experience, thereby solving the technical problem of low reliability of vehicle thermal management methods in related technologies. Attached Figure Description

[0022] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0023] Figure 1 This is a flowchart of a vehicle thermal management method according to an embodiment of this application;

[0024] Figure 2This is a flowchart of a vehicle thermal management method according to an embodiment of this application;

[0025] Figure 3 This is a schematic diagram of a dynamic thermal management optimization system for energy flow in new energy vehicles based on big data driving behavior analysis.

[0026] Figure 4 This is a schematic diagram of a vehicle thermal management device according to an embodiment of this application. Detailed Implementation

[0027] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0028] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0029] According to an embodiment of this application, an embodiment of a vehicle thermal management method is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0030] Figure 1 This is a flowchart of a vehicle thermal management method according to an embodiment of this application, such as... Figure 1 As shown, the method includes the following steps:

[0031] Step S102: Obtain multi-source data collected by the vehicle, wherein the multi-source data includes at least: vehicle status data, environmental data and driving behavior data.

[0032] The aforementioned vehicles can refer to mobile vehicles used for transporting people or goods. Vehicle types may include, but are not limited to, pure electric vehicles, plug-in hybrid electric vehicles, and fuel cell vehicles, and have integrated battery, motor, and thermal management systems. The specific vehicle type needs to be determined based on actual driving conditions and is not limited here. The vehicle in this application can serve as the main body of the vehicle's thermal management method.

[0033] The aforementioned multi-source data can refer to diverse data types collected from different sensors, devices, and systems. Multi-source data encompasses information such as vehicle operating status, external environment, and driver operating habits, and can be used to provide a comprehensive operating environment context, supporting real-time adaptability and personalized adjustments to thermal management optimization algorithms.

[0034] The aforementioned vehicle status data refers to information reflecting the current operating and health status of the vehicle. Vehicle status data may include, but is not limited to, battery status, motor status, thermal management component status, and vehicle dynamic status. This data can be used to reflect the operating status of various vehicle components, ensuring that thermal management strategies respond to real-time demands and preventing overheating or overcooling.

[0035] The aforementioned environmental data can refer to the natural conditions and geographical information outside the vehicle. Environmental data may include, but is not limited to, meteorological conditions, geographical information, lighting conditions, and traffic environment. Specific environmental data needs to be determined based on the actual environment and the vehicle's thermal management requirements. Environmental data can be used to help the system predict and adapt to changes in external conditions and adjust thermal management strategies to cope with different climates and road conditions.

[0036] The aforementioned driving behavior data refers to data describing a driver's driving habits and style. This data may include, but is not limited to, acceleration and deceleration behavior, steering behavior, driving modes, and driving preferences. Specific driving behaviors need to be determined based on actual driving conditions. By analyzing this behavioral data, it is possible to predict heat load, adjust control strategies to improve energy efficiency, and adapt to different driving styles.

[0037] In one optional embodiment, a sensor network inside the vehicle and in the surrounding environment continuously collects multi-source data, including vehicle status data, environmental data, and driving behavior data. Vehicle status data includes battery temperature, voltage, and current status; motor operating temperature and speed information; and dynamic parameters of the air conditioning and cooling systems. Environmental data encompasses weather conditions (temperature, humidity, wind speed, and sunlight), geographic information (terrain, altitude), and real-time traffic flow. These external data have a significant impact on the vehicle's thermal management requirements and energy efficiency. Furthermore, through the vehicle bus, in-cabin sensors, and user interface, the system can accurately record driving behavior data, including acceleration mode, braking frequency, steering effort, and driving mode selection. This driving behavior data collectively reveals the driver's personal preferences and operating style, which is crucial for energy consumption and thermal load prediction. Through continuous acquisition and intelligent analysis of this multi-source data, the system can accurately identify current operating conditions, enabling personalized and proactive thermal management decisions, ensuring that new energy vehicles maintain high energy efficiency and passenger comfort under various driving environments.

[0038] Step S104: Perform cluster analysis on the multi-source data to determine the user's driving behavior profile.

[0039] The clustering analysis mentioned above can refer to an unsupervised learning method. The types of clustering analysis methods can include, but are not limited to, K-Means clustering, hierarchical clustering, density-based spatial clustering of applications with noise (DBSCAN), graph neural networks, etc. Clustering analysis can divide data objects in a dataset into several categories, making data objects in the same category as similar as possible, while data objects in different categories as dissimilar as possible, thereby discovering the inherent structure and patterns of the data.

[0040] The aforementioned user driving behavior profile refers to a personalized data model constructed by collecting and analyzing multi-dimensional data such as a user's driving habits, operational characteristics, and driving preferences. This model can describe and predict a user's driving style and behavioral patterns. This profile includes not only dynamic behavioral characteristics such as the driver's acceleration frequency, braking intensity, average speed, and steering habits, but also static information such as the user's preferences for temperature control and driving mode selection, thus forming a comprehensive view reflecting the driver's habits and needs. User driving behavior profiles can be used for personalized thermal management, improving energy efficiency, and more.

[0041] In one optional embodiment, the system first collects multi-source data, including but not limited to acceleration, braking intensity, steering angle, driving speed, ambient temperature, and humidity, from the vehicle's sensors, bus system, and environmental monitoring equipment. Then, advanced clustering algorithms, such as K-Means, DBSCAN, or graph neural networks, are used to process this data and identify different driving modes or behavior types. This process typically involves data preprocessing (cleaning, standardization), feature extraction, and clustering model training to ensure the accuracy and robustness of the clustering results. Finally, the system transforms the clustering results into easily understandable and applicable user driving behavior profiles, which include feature descriptions of different driving modes and related thermal management demand predictions. By establishing user driving behavior profiles, the vehicle can dynamically adjust thermal management system parameters based on the driver's actual behavior patterns, thereby reducing unnecessary energy consumption and improving overall efficiency while ensuring comfort.

[0042] Step S106: Based on the user's driving behavior profile and multi-objective optimization decision algorithm, generate the vehicle's thermal management control parameters.

[0043] The aforementioned multi-objective optimization decision-making algorithm refers to a mathematical tool that seeks a balance point to achieve a better overall solution when faced with multiple conflicting or incompatible objectives. In the field of thermal management, multi-objective optimization decision-making algorithms need to simultaneously consider multiple objectives such as energy consumption, temperature control, and component lifespan extension, each with its own importance and constraints. Such algorithms typically involve mathematical programming, machine learning models (such as neural networks and genetic algorithms), and decision theory, and can dynamically adjust based on real-time data and user profiles to adapt to constantly changing driving conditions and environments.

[0044] The aforementioned thermal management control parameters refer to a series of numerical settings or operational commands used in the thermal management system to regulate and control the temperature of key vehicle components. These parameters may include, but are not limited to, battery temperature control parameters, motor temperature control parameters, air conditioning system parameters, coolant pump and fan control parameters, and system safety tolerance parameters. Specific thermal management control parameters need to be determined based on actual thermal management requirements. These parameters enable efficient heat dissipation and temperature control, meeting energy management and comfort requirements during driving.

[0045] In one optional embodiment, the system first identifies the driver's unique driving style based on the user's driving behavior profile, including rapid acceleration, rapid deceleration, average speed, and steering habits. These features are then transformed into key inputs for thermal load prediction. Subsequently, a multi-objective optimization decision algorithm, based on multiple objectives such as energy consumption, temperature control, and component lifespan extension, dynamically adjusts the weights of control parameters to generate in real time thermal management control parameters such as coolant flow rate, fan speed, air conditioning pre-cooling / pre-heating time, and coolant pump activation frequency. This ensures that the thermal management strategy achieves good results under different driving conditions. Generating thermal management control parameters through user driving behavior profiles and multi-objective optimization decision algorithms not only significantly improves the energy efficiency and driving safety of new energy vehicles but also provides drivers with a more personalized and intelligent driving experience.

[0046] Step S108: Perform thermal management on the vehicle based on thermal management control parameters.

[0047] In one optional embodiment, the operating states of various vehicle components are dynamically adjusted based on thermal management control parameters. For example, this includes regulating coolant flow, changing fan speed, turning the air conditioning system on or off, and adjusting the power of the heating pads to address changes in heat load caused by driving behavior. This allows the system to quickly respond to driver behavior patterns and external environmental conditions, preventing critical components from overheating or overcooling while improving energy efficiency and achieving effective thermal management. Intelligent thermal management of the vehicle based on thermal management control parameters enables a seamless link between data insights and practical operation, significantly improving vehicle energy efficiency, safety, and driving experience.

[0048] In this embodiment, firstly, multi-source data, including vehicle status data, environmental data, and driving behavior data, are acquired. Next, cluster analysis is performed on the multi-source data to determine the user's driving behavior profile. Further, based on the user's driving behavior profile and a multi-objective optimization decision algorithm, thermal management control parameters for the vehicle are generated. Finally, thermal management of the vehicle is implemented based on these parameters. This application first constructs a comprehensive data framework reflecting the actual driving environment and behavioral patterns by collecting multi-source data from all angles. Then, cluster analysis algorithms are used to process the collected multi-source data in depth to determine the user's driving behavior profile. This profile generation transforms abstract driving behavior into concrete and quantifiable features, laying the foundation for personalized thermal management strategies. Based on this, the most suitable thermal management control parameters for the current driving needs are determined using the user's driving behavior profile and a multi-objective optimization decision algorithm. Finally, the thermal management control parameters are applied to the vehicle's thermal management system to achieve precise control of the vehicle's thermal management system. This application generates dynamic thermal management control parameters by collecting and analyzing multi-source data and combining them with a multi-objective optimization decision-making algorithm. This overcomes the limitations of static and passive thermal management methods in existing technologies, achieving more refined and flexible thermal management control. It significantly improves the energy efficiency of new energy vehicles under various driving conditions and enhances the user's driving experience, thereby solving the technical problem of low reliability of vehicle thermal management methods in related technologies.

[0049] Optionally, cluster analysis is performed on multi-source data to determine the user's driving behavior profile, including: preprocessing the multi-source data to obtain preprocessed data; extracting features from the preprocessed data to obtain driving behavior indicator features; and performing cluster analysis on the driving behavior indicator features to obtain the user's driving behavior profile.

[0050] The aforementioned preprocessed data refers to the dataset after cleaning, formatting, and standardizing the original multi-source data. The purpose of preprocessing is to eliminate noise and outliers, standardize data formats, and resolve missing values, enabling subsequent feature extraction and analysis to be performed on high-quality data. Preprocessed data may include, but is not limited to, filtering of acceleration data, synchronization of data timestamps, calibration of spatial coordinates, and standardization of various sensor data. The specific preprocessed data needs to be determined based on actual requirements and is not limited here.

[0051] The aforementioned driving behavior indicators refer to specific quantitative indicators extracted from preprocessed data that reflect a driver's operating habits and style. These indicators may include, but are not limited to, characteristics such as the frequency of rapid acceleration, braking intensity, speed fluctuation index, steering volatility, and driving mode recognition results (e.g., determination of economical, balanced, or aggressive driving styles). Specific driving behavior indicators need to be determined based on the actual situation. These indicators can provide a basis for subsequent cluster analysis, helping the system understand the specific needs and habits of different drivers.

[0052] In one optional embodiment, the collected multi-source data is first preprocessed to obtain preprocessed data. This stage includes data cleaning, outlier detection, timestamp synchronization, and spatial coordinate alignment to ensure data accuracy and consistency, laying a solid foundation for subsequent analysis. Next, the preprocessed data is fed into a feature extraction stage. This stage uses machine learning techniques to extract quantifiable driving style indicators from data such as acceleration, braking, steering, and speed, including metrics like frequency of rapid acceleration, braking intensity, speed fluctuation index, and steering smoothness. Finally, through cluster analysis, these driving behavior indicators are summarized and organized to identify user groups with similar driving habits, forming user driving behavior profiles. This profile not only reflects each driver's personalized driving mode but also provides a basis for the thermal management system to dynamically adjust its strategies, enabling it to intelligently select the most suitable thermal management settings based on real-time driving behavior and environmental information.

[0053] Optionally, the multi-source data is preprocessed to obtain preprocessed data, including: filtering the acceleration data in the multi-source data to obtain filtered data; synchronizing the data timestamps of the filtered data to obtain synchronized data; and calibrating the spatial coordinates of the synchronized data to obtain preprocessed data.

[0054] The aforementioned acceleration data refers to information reflecting the acceleration of an object's motion. Acceleration data can include, but is not limited to, longitudinal acceleration (forward-backward direction), lateral acceleration (left-right direction), and vertical acceleration (up-down direction). Acceleration data in different directions is used to reflect the dynamic changes of a vehicle in different directions. Acceleration data can be used to reflect the acceleration and deceleration states of a vehicle during driving, and thus can be used to reflect the driver's behavior and road conditions.

[0055] The filtered data mentioned above refers to clean data obtained by applying filtering algorithms to the original acceleration data to remove noise, high-frequency interference, and other unwanted signals. Commonly used filtering techniques include, but are not limited to, Hamming window function filtering, infinite impulse response filtering, finite impulse response filtering, Butterworth filtering, Chebyshev filtering, and elliptic filtering. The specific filtering technique needs to be determined based on actual usage requirements. Filtering eliminates unnecessary interference, improves the signal-to-noise ratio of the data, makes subsequent feature extraction more accurate, and reduces the possibility of misjudgment.

[0056] The aforementioned data timestamps refer to a set of time information appended to data records, indicating the specific moment the data was collected or the event occurred. Data timestamp types can include, but are not limited to, absolute timestamps and relative timestamps relative to a specific event. The specific data timestamp type needs to be determined based on the actual situation. Data timestamps allow for correct alignment of data on the timeline, ensuring the correlation and consistency between data.

[0057] The aforementioned synchronized data refers to the result obtained after time alignment of data with different timestamps. Synchronized data can accurately reflect the vehicle's state at a specific moment, which is crucial for analyzing vehicle dynamics and driving behavior. Only after data synchronization can effective time series analysis and spatial positioning analysis be performed.

[0058] The aforementioned spatial coordinate calibration refers to the calibration of spatial position data from different sensors during multi-source data preprocessing to eliminate coordinate system differences caused by sensor installation positions, angles, or vehicle motion. Spatial coordinate calibration can be achieved through various techniques, including but not limited to using fusion algorithms of Inertial Measurement Units (IMUs) and Global Positioning Systems (GPS), or using environmental references provided by LiDAR or visual sensors for spatial positioning calibration. The specific spatial coordinate calibration technique needs to be determined based on actual requirements. Spatial coordinate calibration ensures that all data is in the same coordinate system, providing accurate position information and direction of motion, thereby improving data usability and analytical accuracy.

[0059] In one optional embodiment, acceleration data is first selected as the key analysis object from multi-source vehicle data. This acceleration data covers acceleration and deceleration events during vehicle operation and is an important signal for understanding driving style and habits. However, raw acceleration data often carries noise and high-frequency interference, which can affect the accuracy of subsequent analysis. Therefore, the first step is to filter the acceleration data. Here, filtering techniques such as the Hamming window function are used to effectively remove noise from the data, resulting in cleaner filtered data. Subsequently, considering that different sensors in the vehicle may have different time bases, which may affect the consistency and correlation of the data, the timestamps of the filtered data are synchronized. By analyzing and correcting the delay between sensors, the time base of all acceleration data is ensured to be unified. The resulting synchronized data can accurately reflect the true time series of various events during driving, avoiding analytical bias caused by time differences. Finally, given that acceleration data may be affected by changes in vehicle attitude during vehicle movement, the synchronized data is calibrated in spatial coordinates. This process ensures that the coordinate representation of the acceleration data in three-dimensional space matches the actual motion state of the vehicle, eliminating coordinate system inconsistencies caused by vehicle tilting or turning, thus obtaining preprocessed data that more accurately reflects vehicle dynamics. Through this series of preprocessing steps, not only is the usability and effectiveness of the raw acceleration data improved, but a solid data foundation is also laid for building accurate driving behavior profiles and intelligent thermal management strategies, further promoting the economic efficiency and reliability of new energy vehicles.

[0060] Optionally, feature extraction is performed on the preprocessed data to obtain driving behavior index features, including: obtaining a rapid acceleration frequency feature based on the acceleration change rate in the preprocessed data, wherein the rapid acceleration frequency index feature is used to characterize the number of times the acceleration change rate is greater than a change rate threshold within a preset time period; obtaining a braking attack intensity feature based on the deceleration and braking frequency in the preprocessed data, wherein the braking attack intensity index feature is used to characterize the braking force or braking depth of the brake pedal; obtaining a speed fluctuation index feature based on the vehicle speed in the preprocessed data, wherein the speed fluctuation index feature is used to characterize the frequency and amplitude of vehicle speed changes; obtaining a steering variability feature based on the steering wheel angle in the preprocessed data, wherein the steering variability feature is used to characterize the change rate of the steering wheel angle within a preset time period; and obtaining driving behavior index features based on the rapid acceleration frequency feature, braking attack intensity feature, speed fluctuation index feature, and steering variability feature.

[0061] The aforementioned rate of change of acceleration refers to the change in acceleration per unit time. This rate of change can be further subdivided into longitudinal, lateral, and vertical acceleration rates, corresponding to acceleration changes in the vehicle's forward, lateral, and vertical directions, respectively. The specific rate of change needs to be determined based on actual conditions. The rate of change of acceleration reflects the speed change during vehicle acceleration or deceleration. Monitoring this rate helps identify whether the driver tends to accelerate rapidly, thus providing a basis for the thermal management system to dynamically adjust parameters and ensure that critical components do not overheat under aggressive driving modes.

[0062] The aforementioned rapid acceleration frequency characteristic can refer to the characteristic that characterizes the driver's tendency to accelerate rapidly by statistically analyzing the number of times the rate of change of acceleration exceeds a set threshold. The rapid acceleration frequency characteristic can help the system understand the driver's preferences, especially in situations of frequent starts in urban areas or overtaking at high speeds. This is beneficial for the thermal management system to prepare response strategies in advance and extend the service life of the power battery and electric motor.

[0063] The aforementioned rate of change threshold can refer to a standard value set when calculating the frequency characteristics of rapid acceleration. Only when the rate of change of acceleration exceeds this threshold is it considered a rapid acceleration event. The selection of this threshold directly affects the analysis results. The type of rate of change threshold can include, but is not limited to, a fixed rate of change threshold, or a dynamic rate of change threshold that is dynamically adjusted based on factors such as vehicle status and ambient temperature. The specific rate of change threshold needs to be determined based on the accuracy of the judgment. Setting a reasonable rate of change threshold can filter out minor acceleration events in daily driving and focus on rapid acceleration events that truly have a significant impact on the vehicle's thermal load, thereby accurately identifying key areas for thermal management improvement.

[0064] The aforementioned deceleration refers to the rate at which the vehicle's speed decreases during deceleration, and the aforementioned number of braking strokes refers to the total number of times the driver presses the brake pedal. These two indicators together depict the characteristics of the driver's braking behavior. Statistical analysis of deceleration and braking strokes can be used to reveal whether the driver tends towards sudden stops or gradual braking.

[0065] The aforementioned braking aggression intensity characteristic refers to a combination of deceleration and braking frequency, used to measure the force and frequency of braking by the driver. It is an important indicator for assessing whether a driver's driving style is aggressive. Braking aggression intensity characteristics can help the system identify whether the driver frequently engages in aggressive braking, aiding in the maintenance planning of the braking system and the adjustment and formulation of cooling requirements in the thermal management system.

[0066] The aforementioned speed fluctuation index characteristic can refer to an indicator that assesses the stability of a driver's driving style by statistically analyzing the frequency and amplitude of vehicle speed changes. The speed fluctuation index characteristic can be calculated by calculating the standard deviation of speed, the difference between maximum and minimum speeds, or the integral of the rate of speed change. The specific method chosen depends on the accuracy requirements of the thermal management system. The speed fluctuation index characteristic can indirectly reflect the workload and frequency of the powertrain system during vehicle operation.

[0067] The aforementioned steering variability characteristics refer to features used to characterize the stability and effort of a driver's actions when turning or changing lanes. Steering variability can be calculated based on the average or peak value of the rate of change of steering wheel angle over a continuous period, or it can be a combination of the frequency and amplitude of steering operations. The specific calculation method needs to be determined according to actual needs. Steering variability characteristics can help the system adjust its cooling strategy in a timely manner.

[0068] In one optional embodiment, firstly, by acquiring the rate of change of acceleration in the preprocessed data, the frequency of the driver's rapid acceleration within a preset time period is determined. This feature reveals whether the driver tends to accelerate rapidly, which is particularly important for predicting energy consumption and thermal load. Next, the deceleration and braking frequency in the preprocessed data are analyzed to measure the intensity of braking aggression, i.e., the force and depth of the driver's braking. This is another key indicator for assessing the aggressiveness of driving style and has a direct impact on the health monitoring and thermal management of the braking system. Further, focusing on vehicle speed fluctuations, the frequency and amplitude of speed changes are determined by calculating the speed fluctuation index. This not only reflects the stability of driving habits but also indicates fluctuations in energy demand, which is crucial for adjusting battery and motor thermal management strategies. Finally, the steering variability feature is determined based on the rate of change of steering wheel angle within a preset time period. The steering variability feature describes the driver's steering adjustment habits and is equally important for assessing vehicle handling and steering system thermal load. By combining the extraction of the above features, a rich framework of driving behavior indicators is constructed, covering multiple dimensions such as the frequency of rapid acceleration, braking aggression intensity, speed fluctuation index, and steering variability. This framework not only accurately captures the driver's subtle operating habits but also provides comprehensive data support for the thermal management system's intelligent decision-making, enabling real-time strategy adjustments. Through these steps, the thermal management system can deeply understand the driver's driving habits and road condition changes, thereby achieving refined energy flow management and significantly improving vehicle economy and reliability.

[0069] Optionally, based on user driving behavior profiles and multi-objective optimization decision-making algorithms, vehicle thermal management control parameters are generated, including: generating initial control parameters of the vehicle based on user driving behavior profiles and multi-objective optimization decision-making algorithms; constructing a performance degradation model of the vehicle based on user driving behavior profiles, wherein the performance degradation model is used to characterize the correlation between user driving behavior profiles and vehicle performance degradation; and adjusting the initial control parameters based on vehicle historical data and the performance degradation model to obtain thermal management control parameters.

[0070] The aforementioned initial control parameters refer to the preliminary calculated operating parameters of the thermal management system. These parameters may include, but are not limited to, water pump power, fan speed, and refrigerant flow rate. Specific initial control parameters need to be determined based on the user's driving behavior profile and a multi-objective optimization decision-making algorithm. These initial control parameters serve as the starting point for the thermal management system's operation, determining the system's initial response strategy in response to different driving behaviors. This is a crucial step in ensuring normal vehicle operation and efficient energy utilization.

[0071] The aforementioned performance degradation model can be considered a mathematical model used to quantify the long-term impact of driving behavior on vehicle performance. This model can be based on various theories, including but not limited to physical models, statistical models, and machine learning models. Physical models consider the effects of physical processes such as material fatigue and wear, statistical models focus on summarizing degradation patterns from historical data, while machine learning models use techniques such as deep learning to uncover nonlinear correlations and make more accurate long-term predictions. Performance degradation models can help thermal management systems plan ahead, reducing component damage caused by overuse and extending vehicle lifespan. Furthermore, they can adjust control parameters in a timely manner based on changes in user driving behavior and vehicle aging, maintaining the system in good condition during long-term operation.

[0072] The aforementioned vehicle history data can refer to all information accumulated during the vehicle's previous use. This information may include, but is not limited to, driving behavior records, thermal management system operation logs, maintenance history, and environmental condition data. The specific vehicle history data needs to be determined based on actual requirements. Vehicle history data can not only be used to verify the accuracy of performance degradation models but also provide rich learning samples during model training, helping the system better understand the trend of vehicle performance changes over time and the impact of driving behavior on this trend.

[0073] In one optional embodiment, firstly, based on driving behavior profiles and multi-objective optimization decision-making algorithms, initial control parameters for the vehicle are generated. This step is equivalent to the "intelligent initialization" of the thermal management system. It pre-sets optimal parameters such as water pump speed, fan power, and compressor operating status based on the driver's driving mode (e.g., economy, balanced, aggressive) and real-time environmental conditions, ensuring the system achieves a good thermal management configuration upon startup. Subsequently, based on the user's driving behavior profile, a performance degradation model for the vehicle is constructed. This model reveals the impact path of driving habits on the performance degradation of key systems (such as the battery, motor, and thermal management system). Through big data mining and model iteration, the potential degradation trends of various components under specific driving modes can be predicted. For example, the temperature rise of the battery under aggressive driving mode may lead to an accelerated decline in its efficiency and lifespan. With this predictive model, the thermal management system can be proactively maintained, avoiding sudden failures caused by performance degradation, and providing a scientific basis for long-term vehicle health management. Finally, to make the thermal management system more closely match the actual state of the vehicle, the generated initial control parameters are dynamically adjusted based on historical vehicle data. Historical data encompasses the vehicle's past operating records, maintenance information, and changes in environmental conditions. Through analysis using performance degradation models, the effectiveness of control parameters can be reassessed based on this information, with fine-tuning as necessary to achieve personalized and real-time thermal management strategies. This process iterates repeatedly until the most suitable thermal management control parameters for the current vehicle state and driving mode are found. This process not only allows for immediate adjustments to the thermal management strategy based on driving behavior but also predicts long-term changes in vehicle performance, achieving an organic unity between immediate response and proactive maintenance.

[0074] Optionally, a user driving behavior profile is used to characterize driving behavior patterns. Based on the user driving behavior profile and a multi-objective optimization decision algorithm, initial control parameters for the vehicle are generated, including: predicting the heat load demand of the user driving behavior profile using a preset association model to obtain the target heat load demand, wherein the preset association model is used to characterize the quantitative relationship between driving behavior and thermal management demand; constructing the objective function of the multi-objective optimization decision algorithm based on the classification type of driving behavior patterns and the target heat load demand; constructing constraints based on battery temperature, pump control signal, and total power; and solving the objective function using a parameter mapping algorithm and constraints to obtain the initial control parameters.

[0075] The aforementioned driving behavior patterns refer to types of driving habits identified from driving behavior characteristics. These patterns can include, but are not limited to, economical, balanced, aggressive, and specific patterns that vary with driving environment (e.g., city, highway, mountain roads) and time (e.g., rush hour, night). Specific driving behavior patterns need to be determined based on actual user driving behavior profiles. Identifying driving behavior patterns can help the thermal management system better understand driver operating habits, providing crucial information for heat load demand forecasting.

[0076] The aforementioned pre-defined correlation model can refer to a mathematical model used to quantify and predict the relationship between driving behavior patterns and thermal management needs. This model can be based on physical laws, such as thermodynamic models of batteries and motors, or it can be a data-driven model based on machine learning, such as a neural network model, used to capture complex nonlinear relationships. The specific pre-defined correlation model needs to be determined based on actual needs. The pre-defined correlation model can serve as the core of thermal demand forecasting, ensuring the scientific rigor and targeted nature of strategy adjustments.

[0077] The aforementioned target heat load demand refers to the amount of cooling or heating power that the thermal management system needs to execute to maintain the thermal balance of the vehicle operation, as predicted based on driving behavior patterns. Target heat load demand can be categorized into battery heat load, motor heat load, and air conditioning system heat load, etc. The specific target heat load demand needs to be determined based on the vehicle's configuration and the structure of the thermal management system. The prediction of the target heat load demand provides a quantitative basis for the formulation of thermal management control strategies, ensuring that strategy adjustments accurately match real-time operating conditions and the driver's operating habits.

[0078] The aforementioned objective function can refer to the objective function of a multi-objective optimization decision-making algorithm constructed based on the classification type of driving behavior patterns and the target thermal load demand. This objective function integrates three indicators: energy consumption, temperature control, and lifespan. By setting different weights, it balances the efficiency of the thermal management system with the health status of vehicle components. The objective function can be a linear or nonlinear objective function; the specific objective function needs to be determined based on the control requirements of the thermal management system and the performance characteristics of the vehicle components. The objective function can serve as the core of the multi-objective optimization algorithm, guiding the algorithm to find a balance point among multiple objectives such as minimizing energy consumption, optimizing temperature control, and maximizing component lifespan.

[0079] The battery temperature mentioned above may refer to the current actual temperature of the battery. The pump control signals mentioned above include the operating status commands of the water pump and the fan. The total power mentioned above is the total cooling or heating power that the thermal management system needs to output under specific operating conditions.

[0080] The aforementioned constraints refer to conditions used to ensure the thermal management system operates within a safe and feasible range. These constraints can be determined based on key parameters such as battery temperature, pump control signals, and total power. Constraints can serve as boundaries for multi-objective optimization decision-making algorithms, preventing other parameters from exceeding safe limits when the thermal management strategy pursues a particular objective, thus ensuring the stability and safety of the system.

[0081] The aforementioned parameter mapping algorithm refers to the process of transforming the optimized objective function solution into specific thermal management control parameters (such as pump control signals, fan control signals, etc.). The parameter mapping algorithm can serve as a bridge connecting optimization decisions and practical applications. It transforms abstract optimization results into executable instructions for the thermal management system and is a key step in strategy implementation.

[0082] In one optional embodiment, firstly, a pre-defined correlation model is used to perform in-depth analysis of the user's driving behavior profile to predict the upcoming thermal load demand. In this step, the pre-defined correlation model can establish a quantitative relationship between driving behavior characteristics (such as rapid acceleration, smooth driving, etc.) and vehicle thermal management requirements (such as cooling requirements, temperature control requirements, etc.), providing accurate data support for subsequent strategy adjustments.

[0083] Subsequently, based on the classification of driving behavior patterns and the predicted target heat load demand, a multi-objective optimization decision algorithm objective function is constructed. The objective function comprehensively considers three key indicators: energy consumption, temperature control, and lifespan. By dynamically adjusting the weights of these indicators, it ensures that the thermal management strategy can meet immediate heat load demands while also considering the long-term operating efficiency of the vehicle and the health status of key components. This multi-objective optimization method avoids the limitations of single-indicator control and achieves dual optimization of energy utilization efficiency and system health maintenance.

[0084] While ensuring the optimization objective of the objective function is achieved, constraints were constructed based on key parameters such as battery temperature, pump control signals, and total power. These constraints ensure that the operating parameters of the thermal management system do not exceed safe boundaries during the optimization process; for example, the battery temperature will not be too high, pump control will not cause abnormal system pressure, and the adjustment of total power will not exceed the available resource range. By setting these constraints, a protective barrier is built for the optimization decision algorithm, ensuring that the strategy is implemented both efficiently and safely.

[0085] Finally, using a parameter mapping algorithm and considering the set constraints, the objective function is solved to obtain the initial control parameters most suitable for the current driving behavior and vehicle state. The parameter mapping algorithm transforms the optimized decision results into executable thermal management commands, such as water pump speed and fan speed, ensuring that the optimized strategy can be directly applied to the actual thermal management system.

[0086] This process not only enables dynamic adjustment of thermal management control but also automatically updates and optimizes parameters as driving behavior changes and vehicle performance degrades, ensuring the continuous effectiveness of the thermal management strategy. Through these steps, intelligent optimization of the thermal management control strategy is achieved, from driving behavior characteristics to thermal management demand prediction. This process not only improves the responsiveness and control accuracy of the thermal management system but also ensures long-term adaptive adjustment and cloud-based collaborative optimization by introducing performance degradation compensation and reinforcement learning strategies. This significantly improves the economy and reliability of new energy vehicles under different driving modes.

[0087] Optionally, based on the user's driving behavior profile, a vehicle performance degradation model is constructed, including: tracking the user's driving behavior profile based on a hidden Markov model to determine the changing trend of the user's driving behavior profile; constructing a battery capacity degradation model and a cooling system efficiency degradation model based on the changing trend of the user's driving behavior profile; and obtaining the performance degradation model based on the battery capacity degradation model and the cooling system efficiency degradation model.

[0088] The aforementioned Hidden Markov Model (HMM) can refer to a statistical model. HMMs can include, but are not limited to, first-order and second-order models, as well as models using different types of observation probability (such as Gaussian mixture models, multinomial distributions, etc.). The specific HMM model needs to be determined based on actual needs. HMMs can capture the patterns of driving behavior evolution over time, even if these changes are not immediately apparent. By identifying shifts in driving habits, the model can predict future driving behaviors, which is crucial for developing adaptive thermal management strategies. Using HMMs, the uncertainty of driving behavior can be modeled, its dynamic changes predicted, and the control parameters of the thermal management system adjusted in advance to cope with sudden changes in driving style.

[0089] The aforementioned trend in user driving behavior profiles refers to identifying the evolution of driving behavior characteristics over time through long-term observation and analysis of user driving data. This trend reflects natural changes in user driving habits, helping the thermal management system predict future thermal management needs, especially when driving behavior may cause significant changes in thermal load. By tracking this trend, the thermal management system can adjust its strategies in a timely manner to cope with changes in driving style, maintain vehicle performance stability, and extend component lifespan.

[0090] The aforementioned battery capacity degradation model refers to a mathematical model used to simulate and predict the degradation of battery capacity over time. This model can help thermal management systems understand the battery's health status, predict its degradation rate under different driving modes, and thus adjust cooling strategies to slow down the degradation process, extend battery life, and maintain high battery performance.

[0091] The aforementioned cooling system efficiency degradation model refers to a mathematical model used to describe the decline in vehicle cooling system efficiency over time. This model can predict the potential performance degradation of the cooling system during long-term use, allowing for proactive increases in cooling power or adjustments to cooling strategies to maintain the effectiveness and efficiency of the entire thermal management system.

[0092] In one optional embodiment, firstly, a Hidden Markov Model (HMM) is used to continuously track and analyze the user's driving behavior. The HMM can capture the implicit trends in driving habits over time, even if these changes are not intuitive to external observers. By learning from past driving data, the HMM establishes a dynamic behavioral pattern map, revealing the long-term evolution of user driving behavior and laying a solid foundation for subsequent performance prediction. Next, based on the changing trends in the driving behavior profile revealed by the HMM, battery capacity degradation models and cooling system efficiency degradation models are constructed. These two models utilize big data analytics and machine learning techniques to quantify the characteristics of driving behavior and establish a mathematical relationship with the degradation of battery capacity and cooling system efficiency. Through this series of analyses, it is possible to predict how the battery and cooling system will gradually lose their optimal performance over time under different driving modes. The battery capacity degradation model considers the impact of high-intensity operations such as rapid acceleration and frequent braking on battery life, while the cooling system efficiency degradation model focuses on how changes in heat load caused by driving behavior accelerate the aging of cooling components. Finally, the two key models are integrated into a complete performance degradation prediction system—the performance degradation model. This model comprehensively considers the combined effects of driving behavior on the battery and cooling system in the long term, providing comprehensive performance degradation predictions. Based on these predictions, the thermal management system can intelligently adjust its control strategies, such as dynamically optimizing battery thermal management parameters and adjusting the workload of the cooling system in a timely manner, to mitigate performance degradation and improve the overall energy efficiency and reliability of the vehicle.

[0093] Optionally, the initial control parameters are adjusted based on vehicle historical data and a performance degradation model to obtain thermal management control parameters, including: compensating the initial control parameters based on the performance degradation model to obtain compensated control parameters; and optimizing the compensated control parameters using vehicle historical data to obtain thermal management control parameters.

[0094] The aforementioned compensation control parameters refer to a set of control parameters that are dynamically adjusted and compensated based on the preliminary control parameters and the predicted battery degradation state and cooling system efficiency decline by the performance degradation model. These compensation control parameters can be used to compensate for the effects of performance degradation by predictively adjusting the thermal management strategy, thereby maintaining the long-term effectiveness of the thermal management system and the economy and reliability of vehicle operation.

[0095] In one optional embodiment, firstly, the initial control parameters derived from a multi-objective optimization decision algorithm are dynamically compensated based on a performance degradation model. The performance degradation model integrates long-term trends in driving behavior and its impact on battery capacity and cooling system efficiency. The resulting compensation control parameters can proactively adjust the control strategy to compensate for performance losses when battery and cooling system performance deteriorates. For example, if the battery capacity degradation model predicts a decline in battery health, the compensation control parameters will moderately increase cooling efficiency or adjust the battery charging strategy to reduce additional heat generation and thus slow down battery aging. Secondly, the compensation control parameters are further adjusted and improved using a large amount of historical vehicle data to obtain the final thermal management control parameters. This optimization process, based on deep learning and big data analysis, can uncover hidden patterns in historical data and continuously adjust the precision of the control parameters, making the thermal management system's response more closely aligned with actual operating conditions and driving behavior. Through continuous optimization, the thermal management control parameters can more accurately match the driver's preferences and the vehicle's actual needs, ensuring optimal thermal management performance under all circumstances. The above process, through dynamic compensation control parameters, can effectively cope with the natural performance degradation of the battery and cooling system over time, ensuring that the thermal management system maintains a high-efficiency state for a long time.

[0096] In one alternative embodiment, Figure 2 This is a flowchart of a vehicle thermal management method according to an embodiment of this application, such as... Figure 2 As shown, the method includes the following steps:

[0097] Step S202: Preprocess the collected multi-source data and extract data features, and perform cluster analysis on users based on clustering algorithms to construct driving behavior profiles for different users.

[0098] Step S204: Construct the objective function for multi-objective optimization, set the weight dynamic adjustment vector piecewise function and constraints, and generate control parameters in real time to achieve multi-objective optimization decision-making.

[0099] Step S206: Based on big data mining and reinforcement learning, ensure that the thermal management system adapts to changes in user habits and vehicle performance degradation trends, and achieve decision-making strategy optimization and collaborative updates.

[0100] In the aforementioned thermal management method, firstly, based on the preprocessing and feature extraction of multi-source data, combined with clustering algorithms, the driving behavior patterns of different users can be accurately depicted, making the thermal management strategy more closely aligned with actual driving needs and reducing unnecessary energy consumption and wear. Secondly, the objective function of multi-objective optimization, combined with dynamic weight adjustment and real-time generation of control parameters, ensures a delicate balance between energy consumption, battery temperature control, and system lifespan, improving the real-time response capability and decision-making flexibility of the thermal management system. Finally, through big data mining and reinforcement learning techniques, the thermal management strategy can self-improve and update as user habits evolve and vehicle status changes, maintaining long-term high efficiency and avoiding a decline in thermal management effectiveness due to performance degradation. In summary, this method effectively solves the problems of insufficient perception of driving behavior, static strategy optimization, and poor adaptability to system performance degradation in traditional thermal management solutions, significantly enhancing the intelligence, reliability, and economy of the thermal management system for new energy vehicles.

[0101] According to an embodiment of this application, a vehicle thermal management system is provided. It should be noted that this device can be used to execute the aforementioned vehicle thermal management method. The specific implementation method and preferred application scenarios are the same as those in the above embodiment, and will not be repeated here.

[0102] The system includes: a data acquisition module for acquiring multi-source data collected from the vehicle, wherein the multi-source data includes at least: vehicle status data, environmental data, and driving behavior data; an analysis module for performing cluster analysis on the multi-source data to determine the user's driving behavior profile; a generation module for generating vehicle thermal management control parameters based on the user's driving behavior profile and a multi-objective optimization decision algorithm; and a control module for performing thermal management on the vehicle based on the thermal management control parameters.

[0103] Figure 3 This is a schematic diagram of a dynamic thermal management optimization system for energy flow in new energy vehicles based on big data driving behavior analysis, such as... Figure 3 As shown, the system includes: a cloud database, a multi-source data acquisition module, a big data analysis module, a dynamic optimization decision-making module, and an execution feedback module.

[0104] The system includes a cloud database for integrating and storing historical trip data, user driving habit models, operational geographic information, and environmental data; a multi-source data acquisition module that uses a high-precision sensing, edge computing, and cloud-based collaborative architecture to collect and process multi-source data in real time; a big data analysis module for identifying driving behavior patterns and performing behavior-heat load coupled predictive analysis; a dynamic optimization decision-making module for making multi-objective optimization decisions based on driving behavior pattern classification; and an execution feedback module for using deep reinforcement learning to optimize decision-making strategies and achieve collaborative updates between the cloud and edge.

[0105] More specifically, the multi-source data acquisition module includes an onboard sensor network, a vehicle bus data source, and a cloud-based collaborative unit connected to a cloud database. The onboard sensor network is equipped with multiple sensor modules to collect key thermal management parameters in real time. The vehicle bus data source includes a CAN bus and an onboard Ethernet. Driving behavior data is acquired via the CAN bus and transmitted via the onboard Ethernet while parameters are adjusted synchronously. The big data analysis module includes a driving behavior recognition engine and an energy flow demand prediction module. The driving behavior recognition engine includes a behavior feature extraction unit, a pattern classifier, and an abnormal behavior detection unit. The dynamic optimization decision-making module includes a strategy decision-making engine, a multi-objective optimizer, and a system collaborative controller. The strategy decision-making engine includes a rule base decision-maker, a learning optimizer, and an emergency intervention unit. The execution feedback module includes an actuator cluster and a sensor feedback network. This system architecture can quickly adjust the thermal management strategy based on the driver's unique behavior patterns, ensuring optimal energy flow configuration under various driving conditions. Simultaneously, it continuously improves the decision-making strategy through a deep reinforcement learning mechanism, addressing long-term changes in user habits and the natural degradation of vehicle performance, significantly enhancing the system's adaptability and long-term stability.

[0106] Optionally, the generation module includes: a decision module for generating initial control parameters of the vehicle based on the user's driving behavior profile and a multi-objective optimization decision algorithm; and a feedback module for constructing a performance degradation model of the vehicle based on the user's driving behavior profile, adjusting the initial control parameters based on historical vehicle data and the performance degradation model to obtain thermal management control parameters, wherein the performance degradation model is used to characterize the correlation between the user's driving behavior profile and the vehicle's performance degradation.

[0107] According to an embodiment of this application, a device for vehicle thermal management is provided. It should be noted that this device can be used to execute the aforementioned vehicle thermal management method. The specific implementation method and preferred application scenarios are the same as those in the above embodiment, and will not be repeated here.

[0108] Figure 4 This is a schematic diagram of a vehicle thermal management device according to an embodiment of this application, such as... Figure 4 As shown, the device includes the following: acquisition module 402, analysis module 404, generation module 406, and control module 408.

[0109] The acquisition module 402 is used to acquire multi-source data collected by the vehicle, wherein the multi-source data includes at least: vehicle status data, environmental data and driving behavior data; the analysis module 404 is used to perform cluster analysis on the multi-source data to determine the user driving behavior profile; the generation module 406 is used to generate the vehicle's thermal management control parameters based on the user driving behavior profile and a multi-objective optimization decision algorithm; and the control module 408 is used to perform thermal management on the vehicle based on the thermal management control parameters.

[0110] Optionally, the analysis module is used to preprocess multi-source data to obtain preprocessed data; extract features from the preprocessed data to obtain driving behavior indicator features; and perform cluster analysis on the driving behavior indicator features to obtain a user driving behavior profile.

[0111] Optionally, the analysis module is also used to filter the acceleration data in the multi-source data to obtain filtered data; to synchronize the data timestamps of the filtered data to obtain synchronized data; and to calibrate the spatial coordinates of the synchronized data to obtain preprocessed data.

[0112] Optionally, the analysis module is also used to obtain the rapid acceleration frequency feature based on the acceleration change rate in the preprocessed data, wherein the rapid acceleration frequency index feature is used to characterize the number of times the acceleration change rate is greater than the change rate threshold within a preset time period; to obtain the braking attack intensity feature based on the deceleration and braking frequency in the preprocessed data, wherein the braking attack intensity index feature is used to characterize the braking force or braking depth of the brake pedal; to obtain the speed fluctuation index feature based on the vehicle speed in the preprocessed data, wherein the speed fluctuation index feature is used to characterize the frequency and amplitude of vehicle speed changes; to obtain the steering variability feature based on the steering wheel angle in the preprocessed data, wherein the steering variability feature is used to characterize the rate of change of the steering wheel angle within a preset time period; and to obtain the driving behavior index feature based on the rapid acceleration frequency feature, braking attack intensity feature, speed fluctuation index feature, and steering variability feature.

[0113] Optionally, the generation module is used to generate initial control parameters for the vehicle based on the user's driving behavior profile and a multi-objective optimization decision algorithm; to construct a performance degradation model for the vehicle based on the user's driving behavior profile, wherein the performance degradation model is used to characterize the correlation between the user's driving behavior profile and the vehicle's performance degradation; and to adjust the initial control parameters based on the vehicle's historical data and the performance degradation model to obtain thermal management control parameters.

[0114] Optionally, the user driving behavior profile is used to characterize driving behavior patterns; the generation module is also used to predict the heat load demand of the user driving behavior profile using a preset association model to obtain the target heat load demand, wherein the preset association model is used to characterize the quantitative relationship between driving behavior and heat management demand; based on the classification type of driving behavior pattern and the target heat load demand, the objective function of a multi-objective optimization decision algorithm is constructed; based on battery temperature, pump control signal and total power, constraints are constructed; the objective function is solved using a parameter mapping algorithm and constraints to obtain the initial control parameters.

[0115] Optionally, the generation module is also used to track the user's driving behavior profile based on the Hidden Markov Model and determine the changing trend of the user's driving behavior profile; based on the changing trend of the user's driving behavior profile, construct the battery capacity decay model and the cooling system efficiency decay model respectively; and based on the battery capacity decay model and the cooling system efficiency decay model, obtain the performance decay model.

[0116] Optionally, the generation module is also used to compensate the initial control parameters based on the performance degradation model to obtain compensated control parameters; and to optimize the compensated control parameters using historical vehicle data to obtain thermal management control parameters.

[0117] Embodiments of this application also provide a vehicle, including: a memory storing an executable program; and a processor for running the program, wherein the program executes the methods described in various embodiments of this application when it runs.

[0118] Embodiments of this application also provide a computer-readable storage medium including a stored executable program, wherein, when the executable program is running, it controls the device where the computer-readable storage medium is located to perform the methods of various embodiments of this application.

[0119] Embodiments of this application also provide a computer program product, including a computer program that, when executed by a processor, implements the methods of various embodiments of this application.

[0120] Embodiments of this application also provide a computer program product, including a non-volatile computer-readable storage medium for storing a computer program that, when executed by a processor, implements the methods in various embodiments of this application.

[0121] Embodiments of this application also provide a computer program that, when executed by a processor, implements the methods described in the various embodiments of this application.

[0122] In the above embodiments of this application, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0123] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units can be a logical functional division, and in actual implementation, there may be other division methods. For instance, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling, direct coupling, or communication connection may be through some interfaces; the indirect coupling or communication connection between units or modules may be electrical or other forms.

[0124] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0125] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0126] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard drive, magnetic disk, or optical disk.

[0127] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A thermal management method for a vehicle, characterized in that, include: Acquire multi-source data collected from the vehicle, wherein the multi-source data includes at least: vehicle status data, environmental data, and driving behavior data; Cluster analysis is performed on the multi-source data to determine the user's driving behavior profile; Based on the user driving behavior profile and multi-objective optimization decision algorithm, the thermal management control parameters of the vehicle are generated; Thermal management of the vehicle is performed based on the aforementioned thermal management control parameters.

2. The vehicle thermal management method according to claim 1, characterized in that, Cluster analysis is performed on the multi-source data to determine user driving behavior profiles, including: The multi-source data is preprocessed to obtain preprocessed data; Feature extraction is performed on the preprocessed data to obtain driving behavior index features; Cluster analysis is performed on the driving behavior indicator features to obtain the user driving behavior profile.

3. The vehicle thermal management method according to claim 2, characterized in that, The multi-source data is preprocessed to obtain preprocessed data, including: The acceleration data in the multi-source data is filtered to obtain filtered data. The timestamps of the filtered data are synchronized to obtain synchronized data; The synchronized data is calibrated in spatial coordinates to obtain the preprocessed data.

4. The vehicle thermal management method according to claim 2, characterized in that, Feature extraction is performed on the preprocessed data to obtain driving behavior indicator features, including: Based on the acceleration change rate in the preprocessed data, a rapid acceleration frequency feature is obtained, wherein the rapid acceleration frequency index feature is used to characterize the number of times the acceleration change rate is greater than the change rate threshold within a preset time period. Based on the deceleration and braking frequency in the preprocessed data, braking attack intensity characteristics are obtained, wherein the braking attack intensity index characteristics are used to characterize the braking force or braking depth of the brake pedal. Based on the vehicle speed in the preprocessed data, a speed fluctuation index feature is obtained, wherein the speed fluctuation index feature is used to characterize the frequency and amplitude of the vehicle speed change; Based on the steering wheel angle in the preprocessed data, a steering variability feature is obtained, wherein the steering variability feature is used to characterize the rate of change of the steering wheel angle within a preset time period; The driving behavior index features are obtained based on the rapid acceleration frequency features, the braking attack intensity features, the speed fluctuation index features, and the steering fluctuation features.

5. The vehicle thermal management method according to claim 1, characterized in that, Based on the user driving behavior profile and multi-objective optimization decision algorithm, the thermal management control parameters of the vehicle are generated, including: Based on the user driving behavior profile and the multi-objective optimization decision algorithm, the initial control parameters of the vehicle are generated; Based on the user driving behavior profile, a performance degradation model for the vehicle is constructed, wherein the performance degradation model is used to characterize the correlation between the user driving behavior profile and the performance degradation of the vehicle. The initial control parameters are adjusted based on vehicle historical data and the performance degradation model to obtain the thermal management control parameters.

6. The vehicle thermal management method according to claim 5, characterized in that, The user driving behavior profile is used to characterize driving behavior patterns; Based on the user driving behavior profile and multi-objective optimization decision algorithm, the initial control parameters of the vehicle are generated, including: The user's driving behavior profile is used to predict the heat load demand to obtain the target heat load demand. The preset association model is used to characterize the quantitative relationship between driving behavior and heat management demand. Based on the classification type of the driving behavior pattern and the target heat load demand, the objective function of the multi-objective optimization decision algorithm is constructed. Constraints are constructed based on battery temperature, pump control signal, and total power. The objective function is solved using a parameter mapping algorithm and the constraints to obtain the initial control parameters.

7. The vehicle thermal management method according to claim 5, characterized in that, Based on the user's driving behavior profile, a performance degradation model for the vehicle is constructed, including: The user's driving behavior profile is tracked based on a hidden Markov model to determine the changing trend of the user's driving behavior profile. Based on the changing trends of the user driving behavior profile, a battery capacity decay model and a cooling system efficiency decay model are constructed respectively. Based on the battery capacity decay model and the cooling system efficiency decay model, the performance decay model is obtained.

8. The vehicle thermal management method according to claim 5, characterized in that, The initial control parameters are adjusted based on the vehicle historical data and the performance degradation model to obtain the thermal management control parameters, including: The initial control parameters are compensated based on the performance degradation model to obtain the compensated control parameters; The compensation control parameters are optimized using the vehicle's historical data to obtain the thermal management control parameters.

9. A thermal management system for a vehicle, characterized in that, include: The acquisition module is used to acquire multi-source data collected by the vehicle, wherein the multi-source data includes at least: vehicle status data, environmental data, and driving behavior data; The analysis module is used to perform cluster analysis on the multi-source data to determine the user's driving behavior profile; The generation module is used to generate the vehicle's thermal management control parameters based on the user's driving behavior profile and the multi-objective optimization decision algorithm. A control module is used to perform thermal management on the vehicle based on the thermal management control parameters.

10. The vehicle thermal management system according to claim 9, characterized in that, The generation module includes: The decision module is used to generate the initial control parameters of the vehicle based on the user driving behavior profile and the multi-objective optimization decision algorithm. The feedback module is used to construct a performance degradation model of the vehicle based on the user driving behavior profile, and to adjust the initial control parameters based on the vehicle's historical data and the performance degradation model to obtain the thermal management control parameters. The performance degradation model is used to characterize the correlation between the user driving behavior profile and the performance degradation of the vehicle.

11. A vehicle, characterized in that, include: Memory, which stores executable programs; A processor for running the program, wherein the program, when running, performs the method according to any one of claims 1 to 8.

12. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored executable program, wherein, when the executable program is executed, it controls the device on which the storage medium is located to perform the method according to any one of claims 1 to 8.

13. A computer program product, characterized in that, Includes a computer program that, when executed by a processor, implements the method according to any one of claims 1 to 8.

Images