Dynamic load distribution method and device based on vehicle body controller, equipment and medium

Abstract

The invention relates to a dynamic load distribution method, device and equipment based on a vehicle body controller and a medium, and the method comprises the steps: collecting the mechanical load, electrical load and thermal load data of a vehicle body in real time through a distributed sensor network, and generating multi-source heterogeneous initial load data; performing standardization processing on the initial load data, and constructing a three-dimensional data cube type standard load data set; comprehensive characterization features are extracted through feature space mapping, and a fusion load feature set containing a threshold range is dynamically generated in combination with historical data; constructing a load correlation model based on the fusion features, and mapping to generate a steering and braking control parameter combination; and according to the real-time load threshold cross-border state, dynamically adjusting the task priority, allocating real-time calculation core resources to the chassis system, and generating an optimal allocation scheme. According to the method, the problems of multi-source load data fusion failure, resource allocation rigidity and safety task response delay in the prior art are solved, and dynamic load optimization of the vehicle body control system under complex working conditions is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle body dynamic load control, and in particular to a dynamic load distribution method, device, equipment and medium based on a vehicle body controller. Background Art

[0002] In modern intelligent vehicle systems, the body controller, as the core hub of the vehicle's electronic architecture, undertakes the critical task of coordinating the coordinated operation of the power system, chassis system, and various electronic devices. With the increasing complexity of on-board functional modules, the electrical loads, mechanical loads, and thermal loads generated during vehicle operation have shown multi-source heterogeneous characteristics. The ability to coordinate the real-time processing of these load data directly affects the vehicle's handling performance and safety performance. The load distribution method currently commonly used in the industry mainly relies on preset rules for static resource scheduling, which is difficult to cope with the dynamic fluctuations of the vehicle's load status under complex working conditions. Especially in extreme scenarios such as high-speed cornering and emergency braking, the mutual coupling effect of various load parameters often leads to delayed response of the control system and even causes abnormal execution of key safety functions.

[0003] Traditional load distribution schemes have significant limitations: First, there is a lack of effective integration of monitoring data from multi-dimensional parameters such as mechanical vibration, electrical fluctuations, and temperature changes, resulting in the control system being unable to accurately map load status to execution parameters. Second, the resource allocation mechanism is rigid. When there is a surge in steering power demand or a sudden change in brake energy recovery, the allocation strategy of processor core resources cannot be dynamically adjusted according to real-time load characteristics. Third, there is no quantitative basis for prioritizing critical safety tasks and non-real-time tasks, which can cause chassis control response delays when system resources are tight. These issues are particularly prominent in the high-power electronic and electrical architectures of new energy vehicles, manifesting as safety hazards such as steering system response timeouts and inaccurate regenerative braking torque distribution, seriously hindering further improvement of vehicle dynamic control performance. Summary of the Invention

[0004] Based on this, the purpose of the present invention is to provide a dynamic load distribution method, device, equipment and medium based on a vehicle body controller that can integrate multi-source load characteristics in real time and dynamically optimize processor resource allocation strategy.

[0005] The purpose of the present invention is achieved by the following scheme: In a first aspect, the present invention provides a method for dynamic load distribution based on a vehicle body controller, comprising the following steps: S1: Data is collected from various vehicle body components based on a distributed sensor network, including current and voltage data for electrical loads, force and acceleration data for mechanical loads, and temperature and heat flow data for thermal loads, to generate initial load data. S2: Preprocess the initial load data to eliminate the dimensional differences between mechanical load, electrical load, and thermal load, and generate a standard load data set; S3: Perform unified feature space mapping on the standard load data set, extract comprehensive characterization features of mechanical load, electrical load, and thermal load, and generate feature threshold ranges based on historical data distribution to generate a fused load feature set containing a fused feature vector and a preset threshold range; S4: Build a load association model of load characteristics and control parameters based on the fused load characteristic set, and input the fused characteristic vector into the load association model for parameter mapping to generate a mapping control parameter combination; S5: Dynamically adjust the load of the mapping control parameter combination based on the preset threshold range of the fused load feature set, adjust the task execution order and resource allocation ratio based on the preset load priority rules, and generate an optimized load distribution plan. The optimized load distribution plan is used to indicate the CPU core allocation strategy and control the body controller to make real-time parameter adjustments to the power system and chassis system.

[0006] In a second aspect, the present invention provides a dynamic load distribution device based on a vehicle body controller, the device being configured with the following modules: The data acquisition module is used to collect data from various vehicle body components based on a distributed sensor network, including current and voltage data of electrical loads, force and acceleration data of mechanical loads, and temperature and heat flow data of thermal loads, to generate initial load data. The data preprocessing module is used to preprocess the initial load data, eliminate the dimensional differences between mechanical load, electrical load and thermal load, and generate a standard load data set; The fusion load feature generation module is used to perform unified feature space mapping on the standard load data set, extract the comprehensive characterization features of mechanical load, electrical load and thermal load, and generate feature threshold ranges based on historical data distribution, thereby generating a fusion load feature set containing fusion feature vectors and preset threshold ranges; A load correlation model mapping module is used to construct a load correlation model of load characteristics and control parameters based on the fused load feature set, and input the fused feature vector into the load correlation model for parameter mapping processing to generate a mapping control parameter combination; The dynamic load optimization distribution module is used to dynamically adjust the load of the mapping control parameter combination based on the preset threshold range of the fused load feature set, adjust the task execution order and resource allocation ratio based on the preset load priority rules, and generate an optimized load distribution plan. The optimized load distribution plan is used to instruct the CPU core allocation strategy and control the body controller to make real-time parameter adjustments to the power system and chassis system.

[0007] In a third aspect, the present application provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, it implements any of the above-mentioned dynamic load distribution methods based on the vehicle body controller.

[0008] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which implements any of the above-mentioned dynamic load distribution methods based on a vehicle body controller when the computer program is executed by a processor.

[0009] In summary, For better understanding and implementation, the present invention is described in detail below with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 A schematic flow chart of a dynamic load distribution method based on a vehicle body controller provided in an embodiment of the present application; Figure 2 A schematic diagram of a process for generating an optimized load distribution solution provided in an embodiment of the present application; Figure 3 A schematic diagram of a process for updating a load association model provided in an embodiment of the present application; Figure 4 A schematic structural diagram of a dynamic load distribution device based on a vehicle body controller provided in another embodiment of the present application. DETAILED DESCRIPTION

[0011] To facilitate understanding of the present invention, the present invention will be described more fully below with reference to the accompanying drawings. The drawings illustrate preferred embodiments of the present invention. However, the present invention may be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided to provide a more thorough and comprehensive understanding of the disclosure.

[0012] Unless otherwise defined, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which this invention pertains. The terms used in this specification are for the purpose of describing specific embodiments only and are not intended to limit the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0013] In one embodiment, Figure 1 As shown, a dynamic load distribution method based on a vehicle body controller is provided. This embodiment uses the method applied to a terminal as an example for illustration. It is understood that the method can also be applied to a server, or to a system including a terminal and a server, and implemented through interaction between the terminal and the server. In this embodiment, the method includes the following steps: S1: Data is collected from various vehicle body components based on a distributed sensor network, including current and voltage data of electrical loads, force and acceleration data of mechanical loads, and temperature and heat flow data of thermal loads, to generate initial load data.

[0014] Specifically, within the vehicle's electronic architecture, a pre-deployed distributed sensor network covers key vehicle body components, such as the motors and batteries in the powertrain, the steering mechanism and brakes in the chassis, and various electronic devices. For electrical loads, high-precision current transformers and voltage sensors are used to collect real-time data on the current flowing through each electrical component and the voltage across it. The sampling frequency can be set based on actual needs, typically ranging from hundreds of hertz to several thousand hertz, to ensure that subtle changes in the electrical load are captured. For mechanical loads, appropriate force sensors and acceleration sensors are installed. Force sensors, for example, are installed at key connection points in the suspension system to measure the forces exerted on the vehicle from the road during driving. Acceleration sensors are located at key vehicle body and chassis components to obtain acceleration information. This data can reflect the dynamic loads borne by the vehicle's mechanical structure under different driving conditions. As for the thermal load, temperature sensors and heat flow sensors are configured. The temperature sensors are distributed around components that are prone to generating heat, such as motors and battery packs, to monitor their temperature changes in real time; the heat flow sensors are used to detect the rate of heat transfer between the components and the surrounding environment during operation, that is, the heat flux density.

[0015] Each sensor collects the corresponding physical quantity in real time at a fixed sampling frequency. The sampling frequency is set according to different load types and vehicle operating conditions. For example, under dynamic conditions such as sudden acceleration and deceleration, the electrical, mechanical, and thermal loads change rapidly, so the sampling frequency is increased to 1000Hz to ensure data timeliness and integrity. Under steady-state conditions such as constant vehicle speed, the sampling frequency is set to 100Hz to reduce data processing pressure and power consumption. The collected data is transmitted to the body control module (BCM) via on-board network buses such as the CAN bus and FlexRay bus. The CAN bus has a data transmission rate of up to 1Mbps, and the FlexRay bus has a data transmission rate of up to 10Mbps, ensuring fast and stable data transmission, thereby generating initial load data containing raw data on electrical, mechanical, and thermal loads.

[0016] S2: Preprocess the initial load data to eliminate the dimensional differences between mechanical load, electrical load, and thermal load, and generate a standard load data set.

[0017] Specifically, because mechanical load, electrical load, and thermal load data are derived from different physical quantities, each with unique units and dimensions—for example, amperes for current, volts for voltage, newtons for force, meters per second squared for acceleration, degrees Celsius for temperature, and watts per square meter for heat flow—this diversity makes direct, comprehensive data analysis and processing difficult. To address this, the system applies a unified normalization operation to all data types. Specifically, the system determines a reasonable mapping range based on the historical statistical characteristics of each data type and accurately maps the raw data to this range using linear or nonlinear transformation functions. For example, for current data, the system converts it to a normalized value between 0 and 1, referencing the current fluctuation range of the vehicle's electrical system under normal operating conditions. Similarly, for mechanical load force and acceleration data, corresponding normalized values are assigned based on the mechanical performance indicators determined during the vehicle design phase. During this process, the system identifies and removes outliers that significantly deviate from the normal range. These outliers may be due to erroneous readings caused by sensor failure or temporary interference.

[0018] S3: Perform unified feature space mapping on the standard load data set, extract the comprehensive characterization features of mechanical load, electrical load and thermal load, and generate feature threshold ranges based on historical data distribution, thereby generating a fused load feature set containing a fused feature vector and a preset threshold range.

[0019] Specifically, for mechanical loads, the system can use short-time Fourier transform to extract its time-frequency domain characteristics, including the frequency distribution of force signals and the amplitude modulation characteristics of acceleration signals. For electrical loads, the time-domain statistical characteristics of current and voltage can be obtained through sliding window mean and variance calculations, and wavelet packet decomposition can be used to extract its frequency-domain energy distribution characteristics. Thermal load data is analyzed based on the heat conduction equation to extract physical characteristics such as temperature change rate and heat flux density gradient. The multi-dimensional original feature vectors extracted above are subjected to the principal component analysis dimensionality reduction algorithm to screen out key features that are highly sensitive to changes in load state and have strong correlation, forming the prototype of a fused feature vector. On this basis, the distribution characteristics of historical load data are integrated, and with the help of Gaussian mixture model cluster analysis, the characteristic threshold range of each load type under different working conditions is determined. For example, for the steering assist electrical load of new energy vehicles under high-speed conditions, combined with the current peak distribution in historical high-speed driving scenarios, the current threshold with a 95% confidence interval is set to [15A, 35A]. Correspondingly, the steering knuckle torque threshold on the mechanical load side is set to [800N, 1200N], and the temperature threshold of the motor controller on the thermal load side is set to [65°C, 85°C].

[0020] S4: A load association model of load characteristics and control parameters is constructed based on the fused load feature set, and the fused feature vector is input into the load association model for parameter mapping processing to generate a mapping control parameter combination.

[0021] Specifically, the load association model is used to establish a quantitative mapping relationship between the fused load characteristics and the body control parameters. The model structure adopts a deep neural network architecture. The input layer nodes correspond to the dimensions of the fused load feature vector, including the force frequency and acceleration amplitude of the mechanical load, the current mean and voltage variance of the electrical load, the temperature gradient of the thermal load and other key features; the hidden layer uses a combination of multi-layer nonlinear activation functions such as ReLU and sigmoid to explore the implicit association pattern between the load characteristics and the control parameters; the output layer nodes are mapped to the core control parameters of the vehicle power system and chassis system, such as the electronic throttle opening, the electric power steering assist ratio, the regenerative braking energy recovery ratio, etc.

[0022] During model training, the system uses a gradient descent optimization algorithm to adjust network weights based on a large amount of historical load data and corresponding actual control parameter samples, gradually converging the error loss function between the model's output mapping control parameter combination and the optimal control parameters under actual operating conditions to a minimum. When the fused feature vector collected in real time is input into the load association model, a fast forward propagation calculation using a deep neural network generates a mapping control parameter combination that precisely matches the current load state. This provides real-time, intelligent parameter decision support for vehicle dynamic control, effectively improving the vehicle's adaptive control performance under complex operating conditions.

[0023] S5: Dynamically adjust the load of the mapping control parameter combination based on the preset threshold range of the fused load feature set, adjust the task execution order and resource allocation ratio based on the preset load priority rules, and generate an optimized load distribution plan. The optimized load distribution plan is used to indicate the CPU core allocation strategy and control the body controller to make real-time parameter adjustments to the power system and chassis system.

[0024] Specifically, the system dynamically adjusts and optimizes the mapped control parameter combinations within the preset threshold ranges contained in the fused load signature set. Specifically, the system compares the currently mapped control parameters in real time with the threshold boundaries defined by preset load priority rules. For example, in an emergency braking scenario, if the brake system thermal load temperature exceeds a preset high-temperature threshold and the mechanical load value reported by the brake pressure sensor approaches the limit of the range, the highest-priority brake safety task assurance mechanism is triggered. The system then reorders the execution order of the powertrain and chassis tasks based on the preset load priority rules, prioritizing the energy supply and control response of the brake assist system and dynamically downgrading or suspending non-critical comfort electrical loads such as seat heating and window defoggers. Regarding resource allocation, the system dynamically adjusts the CPU core allocation strategy to allocate more computing core resources to tasks responsible for brake control and vehicle stability system operations. For example, 30% of the CPU core resources originally allocated to the in-vehicle entertainment system can be urgently allocated to the chassis control module to ensure the real-time and stability of critical safety control tasks. The optimized load distribution plan is output as a system command, directly instructing the body controller to make real-time parameter adjustments to the powertrain and chassis systems to adjust engine torque output, optimize motor speed, adjust suspension damping coefficient, and distribute braking torque, thereby achieving load balancing and performance optimization of the vehicle under complex dynamic conditions.

[0025] In summary, the embodiment of the present application provides a method for dynamic load distribution based on a vehicle body controller. In one embodiment, the present invention provides a method for dynamic load distribution based on a vehicle body controller, step S1, which specifically includes the following steps: S11: Monitor and process the stress distribution of vehicle body structural parts. Use strain sensors on the frame longitudinal beams to collect vibration spectrum data of key suspension nodes to generate a raw data set of mechanical loads.

[0026] Specifically, the system precisely deploys strain sensors on key structural components, such as the frame rails. These sensors sample vibration spectrum data at high frequencies at key suspension nodes, including structural vibration information caused by factors such as road surface unevenness, speed fluctuations, and uneven load distribution during driving. The sensors convert the acquired strain signals into electrical signals and transmit them via shielded cables to the signal acquisition module of the body controller to ensure signal stability and interference immunity during transmission. The system performs preliminary filtering on these raw strain signals to remove high-frequency noise and low-frequency drift, retaining the significant frequency components relevant to the vehicle's dynamic response. After filtering, the strain signals are further converted into stress values. Using material mechanics models and vehicle structural parameters, the stress distribution of structural components, such as the frame rails, under different operating conditions is calculated. This stress data not only reflects the mechanical load conditions of the vehicle during driving but also provides a foundation for subsequent mechanical load feature extraction and analysis.

[0027] S12: Perform real-time parameter acquisition and processing on the vehicle body's high-voltage electrical system, obtain the battery management system's discharge current ripple and motor controller's phase voltage fluctuation data through the CAN bus, and generate the electrical load original data set.

[0028] Specifically, the system leverages the efficient data transmission capabilities of the CAN bus to acquire real-time data on the battery management system's discharge current ripple and the motor controller's phase voltage fluctuations. The battery management system monitors the battery pack's discharge current via high-precision current sensors. The amplitude and frequency of the ripple current reflect the battery's dynamic characteristics during discharge, including internal chemical reaction rates, contact resistance variations, and external load fluctuations. The motor controller, in turn, monitors phase voltage fluctuations in real time via voltage sensors. This fluctuation data provides crucial information on the motor's operating status, the switching characteristics of the controller's power module, and the dynamic response of the vehicle's powertrain.

[0029] The system digitizes the collected current ripple and phase voltage fluctuation data, converting the analog signals into digital signals via an analog-to-digital converter. It then performs data packaging and frame synchronization to ensure reliable data transmission on the CAN bus. After receiving this data, the system uses signal processing algorithms to perform preliminary analysis, extracting key characteristic parameters such as the amplitude and frequency of the current ripple and the amplitude and phase of the phase voltage fluctuation. These parameters not only reflect the operating status of the high-voltage electrical system but are also closely related to the vehicle's power output, energy recovery, and electrical load distribution.

[0030] S13: Perform temperature field scanning on the heat-sensitive areas of the power battery pack. Use the on-board infrared thermal imager to capture the surface temperature gradient distribution of the motor controller and the heat flux density changes of the power module to generate the original heat load data set.

[0031] Specifically, the infrared thermal imager captures infrared radiation from key heat source areas, such as the motor controller surface and power modules, accurately reflecting the temperature distribution and variations in these areas. The system calibrates and calibrates the onboard infrared thermal imager to ensure the accuracy and reliability of its measurement results. During vehicle operation, the infrared thermal imager continuously scans heat-sensitive areas of the power battery pack at set intervals, capturing temperature gradients and changes in heat flux density. The temperature gradient reflects the direction and rate of heat transfer within the battery pack, while changes in heat flux density indicate the amount of heat generated during battery charging and discharging, as well as the efficiency of the heat dissipation system. The system digitizes the captured infrared thermal image data and uses image processing algorithms to extract key characteristic parameters of the temperature field, such as the highest temperature point, temperature change rate, and heat flux density distribution. These parameters not only provide a visual representation of the battery pack's thermal state but also provide important guidance for optimizing the thermal management system.

[0032] During processing, the system performs noise filtering and image enhancement on the infrared thermal image data to improve data quality and readability. Ultimately, the system integrates this processed thermal load data into a raw thermal load dataset, which details the thermal state of the power battery pack under different operating conditions. This dataset provides critical thermal load information for subsequent thermal load feature extraction and the development of dynamic load distribution strategies.

[0033] S14: Based on the precision time protocol, the mechanical load raw data set, the electrical load raw data set, and the thermal load raw data set are temporally and spatially aligned. The time bases of the mechanical load data, the electrical load data, and the thermal load data are unified through a timestamp calibration algorithm to generate initial load data.

[0034] Specifically, to enable the integration and analysis of mechanical, electrical, and thermal load data, the system uses a precision time protocol to perform spatiotemporal alignment of the raw data sets for each load. The raw data sets for mechanical, electrical, and thermal loads originate from different sensor networks and monitoring systems, and these data may exhibit certain deviations in acquisition time and spatial location.

[0035] Specifically, the system uses a precision time protocol and high-precision clocks to synchronize and calibrate each data collection node. During the time alignment process, the system assigns a unique timestamp to each data collection node, ensuring that each data point accurately corresponds to a specific time point. Furthermore, the system employs a timestamp calibration algorithm to align the time bases of different load data. This algorithm analyzes the time offset and frequency differences between different data sets to calculate calibration parameters. It then interpolates or samples the data to align the time bases of mechanical, electrical, and thermal load data to the same time scale. After the spatiotemporal alignment process, the system generates initial load data. This dataset not only preserves the integrity and accuracy of the original data but also ensures temporal and spatial consistency across different load types. This spatiotemporal alignment lays a solid foundation for subsequent multi-source load data fusion and comprehensive analysis, enabling the system to more accurately assess the overall vehicle load status under complex operating conditions and providing a more reliable basis for the formulation of dynamic load distribution strategies.

[0036] In one embodiment, the present invention provides a method for dynamic load distribution based on a vehicle body controller, step S2, which specifically includes the following steps: S21: performing amplitude normalization processing on the mechanical load data of the initial load data, scaling the suspension stress data to a range of 0-1 according to the upper limit of the material yield strength, and generating normalized mechanical load data.

[0037] Specifically, the system analyzes the suspension stress data, compares the stress data of the suspension system with the upper limit of the material yield strength, and scales the stress data to the range of 0-1 according to a specific mathematical proportional relationship. In this process, the system accurately determines the parameters required for normalization based on the theoretical basis of material mechanics and the physical properties of the materials used in the vehicle suspension system. For example, for suspension components made of high-strength alloy steel, the upper limit of its yield strength can reach XXXMPa. The system divides the actually measured suspension stress value by this upper limit to obtain the normalized stress value. This processing method not only eliminates the data inconsistency caused by differences in size and material of different suspension components, but also enables the normalized mechanical load data to reflect the actual load conditions of the suspension system on a unified scale.

[0038] S22: Performing dynamic range compression processing on the electrical load data of the initial load data, normalizing the current data according to the maximum discharge current of the battery, scaling the voltage data according to the rated voltage of the drive motor, and generating standard electrical load data.

[0039] Specifically, the system monitors the current ripple data provided by the battery management system in real time, compares this data with the maximum discharge current value of the battery, and maps the current value to a standardized range through proportional scaling. At the same time, the system performs similar processing on the phase voltage fluctuation data of the motor controller, normalizing the voltage data according to the rated voltage of the drive motor to ensure that the voltage data can also be analyzed on the same scale. This dynamic range compression processing can not only effectively reduce the fluctuation amplitude of the electrical load data, but also retain key feature information in the data, such as the current ripple frequency and voltage fluctuation amplitude. During the processing process, the system will verify the normalized data in real time to ensure its accuracy and reliability. The processed standard electrical load data can not only reflect the electrical load status of the battery and motor under different working conditions, but also provide standardized data support for the subsequent electrical load feature extraction and the formulation of dynamic load distribution strategies.

[0040] S23: Performing temperature rise characteristic conversion processing on the thermal load data of the initial load data, calculating the relative heat flux density based on the maximum tolerance temperature of the power module, and generating relative heat flux density data.

[0041] Specifically, the system uses an on-board infrared thermal imager to capture temperature gradient distribution and heat flux density change data in key heat source areas, such as the power battery pack and motor controller. This data contains information about the vehicle's thermal status, such as the temperature change rate and heat flux density distribution. The system compares this heat load data with the maximum temperature tolerance of the power module and calculates the relative heat flux density using a mathematical model. This process not only eliminates the temperature dimension differences between different heat source areas, but also intuitively reflects the relative relationship between the current thermal load status and the power module's extreme performance.

[0042] During processing, the system calibrates and verifies relative heat flux data in real time to ensure its accuracy and consistency. The generated relative heat flux data not only reflects the thermal state of key components such as the power battery pack and motor controller, but also provides an important basis for optimizing the thermal management system. This approach effectively prevents battery overheating and thermal runaway risks, ensuring the safe and stable operation of new energy vehicle power battery systems. It also provides critical thermal load information for subsequent thermal load feature extraction and the development of dynamic load distribution strategies.

[0043] S24: Perform matrix reconstruction on the normalized mechanical load data, standard electrical load data, and relative heat flux density data, integrate the stress data, current and voltage data, and heat flux data into a three-dimensional data cube in time series, and generate a standard load data set. The standard load data set is used to indicate the real-time load status of each component of the vehicle body.

[0044] Specifically, the system creates a three-dimensional data cube structure, with time series as the first dimension, used to record data changes over time; different types of load data as the second dimension, used to distinguish the characteristics of various loads; and the third dimension is used to store the specific values of each type of load data at each time point. Through matrix operations and data filling algorithms, the system sequentially fills the normalized mechanical load data, standard electrical load data, and relative heat flux density data into the three-dimensional data cube, ensuring that each data point accurately reflects its actual value at a specific time and under a specific load type. The generated standard load data set not only integrates the real-time load status of various vehicle components, but also provides efficient data structure support for subsequent data analysis and mining in the form of a three-dimensional data cube.

[0045] In one embodiment, S3 of a dynamic load distribution method based on a vehicle body controller provided by the present invention specifically includes the following steps: S31: Perform multimodal feature extraction on the standard load data set, calculate the covariance matrix of the mechanical vibration frequency domain energy distribution, the electrical load time domain ripple coefficient, and the thermal gradient spatial change rate, and generate a feature correlation matrix.

[0046] Specifically, the system uses the Fast Fourier Transform (FFT) algorithm to transform the mechanical vibration signal into the frequency domain and calculate the frequency-domain energy distribution of the mechanical vibration. The system decomposes the vibration signal into multiple frequency components and calculates the energy contribution of each frequency component, thereby obtaining an energy distribution that reflects the energy transfer and dissipation characteristics of the mechanical system at different frequencies. For electrical load data, the system uses the wavelet transform to decompose and reconstruct the time-domain signal and calculate the time-domain ripple coefficient of the electrical load. Through multi-scale analysis, the wavelet transform can capture subtle fluctuations and variations in the electrical load over time. The ripple coefficient quantifies the amplitude and frequency of these fluctuations, characterizing the stability and smoothness of current and voltage in the electrical system. When processing thermal load data, the system uses a spatial difference algorithm to calculate the covariance matrix of the spatial rate of change of the thermal gradient. In the vehicle thermal field model, the system performs spatial difference calculations on temperature data at key heat sources, such as the power module, and their surrounding areas to obtain thermal gradient vectors. The covariance matrix is then solved, revealing the spatial trends and correlations of the thermal load and reflecting the heat transfer and distribution patterns among different components.

[0047] S32: Perform principal component screening on the feature correlation matrix, retain the principal component directions with eigenvalues greater than 1 based on the Kaiser criterion, calculate the feature subspace with cumulative variance contribution rate greater than 95%, and generate a dimensionality reduction feature projection matrix.

[0048] Specifically, based on the Kaiser criterion, the system automatically screens out the principal component directions with eigenvalues greater than 1. These principal component directions represent the main information and change trends contained in the data, and can effectively remove the influence of noise and redundant information. The system further calculates the cumulative variance contribution rate of the selected principal component to ensure that it is greater than 95%. This process determines the final characteristic subspace by gradually accumulating the variance contribution of the principal component until the preset 95% threshold is reached. The determination of the characteristic subspace not only retains the variation information of the original data to the greatest extent, but also significantly reduces the data dimension, thereby improving the efficiency and accuracy of subsequent data processing. The system generates a dimensionality reduction feature projection matrix based on this characteristic subspace.

[0049] S33: Perform feature space mapping processing on the initial load data, project the suspension vibration spectrum, battery current ripple, and power module temperature gradient into a unified feature space based on the dimensionality reduction feature projection matrix, calculate the similarity between the feature vectors, and generate a fused feature vector.

[0050] Specifically, the system takes multi-source heterogeneous data, such as suspension vibration spectra, battery current ripple, and power module temperature gradients, as input. It then performs a linear transformation using a dimensionality-reduction feature projection matrix, projecting this raw feature data into a unified feature space. This process converts data of varying physical quantities and dimensions into eigenvectors within the same feature space, enabling comparison and analysis of previously dispersed and heterogeneous load data within a common mathematical framework.

[0051] The system then calculates the similarity between the projected feature vectors using the cosine similarity metric. This similarity is calculated by taking the ratio of the dot product of two feature vectors to the product of their moduli to obtain a similarity index. This step not only reveals the inherent correlations and similarities between different load characteristics but also provides a quantitative basis for subsequent feature fusion. Finally, based on the similarity calculation results, the system fuses similar features to generate a fused feature vector. This fused feature vector integrates key characteristic information from mechanical, electrical, and thermal loads, forming a comprehensive representation of the vehicle's overall load status.

[0052] S34: Retrieve the historical load feature set stored in the historical operation process of the vehicle from the on-board storage unit as a historical data sample, perform dynamic distribution analysis and processing on the historical load feature set and the fused feature vector based on the ISO 26262 functional safety standard, update the feature distribution boundary of the fused feature vector, and determine the preset threshold range of the current feature vector with three times the standard deviation to generate a fused load feature set.

[0053] Specifically, the system performs rigorous dynamic distribution analysis on historical load signatures and fused feature vectors based on the ISO 26262 functional safety standard. The ISO 26262 standard provides functional safety guidelines for automotive electrical / electronic systems, ensuring that the system fully considers safety and reliability requirements when processing data. During the dynamic distribution analysis, the system uses Gaussian mixture models and kernel density estimation to model and fit the distribution characteristics of the historical load signatures and fused feature vectors. Gaussian mixture models capture the multimodal distribution characteristics of the data, while kernel density estimation provides a non-parametric probability density estimate. Combined, these two methods enable the system to accurately depict the dynamic distribution of load signatures. Based on the analysis results, the system updates the feature distribution boundaries of the fused feature vector. This process uses statistical methods to re-determine the reasonable fluctuation range of the features, eliminate the influence of outliers, and ensure that the feature distribution boundaries more closely align with actual operating conditions. Finally, the system determines the preset threshold range for the current feature vector using three times the standard deviation. The three times standard deviation principle, based on the characteristics of the normal distribution, ensures that under normal operation, feature vector values have a high probability of falling within this range, thus providing a scientific basis for anomaly detection.

[0054] In one embodiment, S4 of a dynamic load distribution method based on a vehicle body controller provided by the present invention specifically includes the following steps: S41: Perform nonlinear modeling on the feature vectors and control parameters of the fused load feature set, construct a support vector regression architecture based on the radial basis kernel function, set regularization parameters to constrain the model complexity, and build an initial association model between the load features and the control parameters.

[0055] Specifically, the system constructs a support vector regression (SVR) architecture based on the radial basis kernel function to capture the nonlinear relationship between load characteristics and control parameters. Specifically, in the SVR model, the radial basis function (RBF) serves as the kernel function, mapping the original feature vector into a high-dimensional space. This makes it easier to find a linear hyperplane to fit the data in this high-dimensional space, thereby effectively modeling complex nonlinear relationships. The system sets the regularization parameter C and the kernel parameter γ to constrain model complexity and prevent overfitting. The regularization parameter C controls the balance between the model's fit to the training data and its generalization ability. A larger C value causes the model to prioritize fitting the training data, potentially leading to overfitting; a smaller C value results in better generalization but potentially underfitting. The kernel parameter γ determines the width of the RBF, affecting the model's sensitivity to local data features. By adjusting these two parameters, the system constructs an initial model of the relationship between load characteristics and control parameters.

[0056] S42: Grid search optimization is performed on the model hyperparameters of the initial association model. The prediction accuracy of different (C, γ) combinations is evaluated using the cross-validation method in the parameter space. The parameter configuration with the minimum mean square error is selected to generate the optimized association model.

[0057] Specifically, the system divides the possible value ranges of C and γ into multiple discrete points, forming a grid-like parameter combination space. For example, C may take the value , γ may take the value of For each (C, γ) combination, the system uses cross-validation to evaluate its prediction accuracy. Cross-validation divides the training dataset into k subsets, alternately using k-1 subsets as the training set and the remaining subset as the validation set. The model is trained and validated k times, and the average of the k validation results is used as the performance indicator for that parameter combination.

[0058] The system selects the parameter configuration with the smallest mean squared error (MSE) as the optimal hyperparameter combination. This ensures that the initial correlation model achieves optimal predictive performance within the parameter space, generating an optimized correlation model. This optimized model more accurately predicts vehicle control parameters such as power steering response time and regenerative braking torque distribution.

[0059] S43: Perform incremental learning processing on the training sample set of the optimized association model, implement online iterative update of model parameters on the domain controller platform of the AUTOSAR architecture, use the exponential decay forgetting factor mechanism to eliminate outdated data, and generate the final load association model.

[0060] Specifically, after generating the optimized correlation model, the system uses an online iterative update method. When new fused load feature set sample data is acquired, it is incorporated into the training sample set, triggering an update of the model parameters. To prevent information loss caused by the complete replacement of old data by new data, the system introduces an exponentially decaying forgetting factor mechanism. The forgetting factor is a parameter between 0 and 1, typically set to a value close to 1, such as 0.9 or 0.99.

[0061] When model parameters are updated, the weight of old data gradually decays according to the exponential power of the forgetting factor. For example, assuming the current number of iterations is t and the forgetting factor is α, then at the tth iteration, the weight of the i-th historical data sample is α^{ti}. In this way, the impact of outdated data will gradually weaken over time, but it will not be completely discarded. In this way, the system achieves a smooth transition and dynamic update of model parameters, generating the final load correlation model. The load correlation model can not only predict vehicle control parameters, but also has the ability to adapt to changes in the vehicle's dynamic operating conditions in real time. This ensures that the system always accurately evaluates and effectively controls the vehicle's load status based on the latest data and models throughout the vehicle's life cycle, improving the vehicle's performance and safety level in different operating stages.

[0062] S44: performing control parameter mapping processing on the fused feature vector of the fused load feature set, predicting the response time of the power steering system and the torque distribution ratio of the regenerative braking system through the load association model, and generating a mapping control parameter combination.

[0063] Specifically, the system inputs the fused feature vector into an optimized and incrementally learned load-dependent model. Leveraging the model's nonlinear mapping capabilities, it predicts key control parameters such as the response time of the power steering system and the torque split ratio of the regenerative braking system. The system implements model predictions through algorithms such as matrix multiplication and kernel function calculations. During the prediction process, the system dynamically adjusts the model's internal parameters based on the current fused feature vector to adapt to changing load conditions.

[0064] For example, when a vehicle is cornering at high speed, the system can predict the appropriate steering assist response time based on the current mechanical and thermal load characteristics, ensuring vehicle stability and safety. Similarly, for the regenerative braking system, the system can predict the optimal torque distribution ratio based on the real-time status of the electrical and thermal loads to improve the efficiency of brake energy recovery. These predictions generate a mapping control parameter combination that guides the body controller in real-time adjustments to the powertrain and chassis systems.

[0065] In one embodiment, Figure 2 As shown, S5 of a dynamic load distribution method based on a vehicle body controller provided by the present invention specifically includes the following steps: S51: Perform real-time out-of-bounds detection on the preset threshold range of the fusion load feature set. When the suspension vibration energy characteristic value exceeds the preset threshold upper limit, recalculate the characteristic offset. Call the sliding time window algorithm to confirm whether the duration of the abnormality exceeds the preset warning threshold, and generate an abnormality detection report containing the abnormality trigger signal and characteristic offset data.

[0066] Specifically, the system implements a real-time cross-border detection mechanism for the preset threshold range of the fused load feature set. Taking the suspension vibration energy characteristic value as an example, the system sets the upper threshold limit according to the ISO 26262 functional safety standard. When the suspension vibration energy characteristic value is detected to have exceeded the preset threshold, the abnormal response program is immediately initiated. Among them, the ISO 26262 standard specifies the functional safety requirements for automotive electrical / electronic systems. ASIL-D is the highest level, requiring the system to have extremely high safety and reliability. Preferably, the feature offset is recalculated using the following formula:

[0067] in, represents the feature offset, is the current suspension vibration energy characteristic value, The upper limit of the preset threshold. By calculating the offset, the system accurately quantifies the degree to which the feature value exceeds the threshold.

[0068] Then, the system calls the sliding time window algorithm to evaluate the duration of the abnormal state. In the time series data, the data is intercepted with the window length L and the step size S. For each data point in the window , calculate the abnormal duration T:

[0069] The system will calculate the duration of the abnormality T and preset warning thresholds For comparison. T > , an anomaly detection report is generated. The report includes the anomaly trigger signal and feature offset data. The anomaly trigger signal is represented in binary form, 1 indicates an anomaly occurs, and 0 indicates normal.

[0070] S52: Reorders safety-critical tasks based on anomaly detection reports. Based on the ISO 26262 ASIL-D level, the electronic stability program control task is placed at the top of the real-time execution queue and the electric power steering task is placed in the second-priority queue, generating a reordered task execution sequence.

[0071] Specifically, the system re-arranges safety-critical tasks based on the ISO 26262 ASIL-D level for abnormal detection reports. In the real-time execution queue, the electronic stability program control task is placed first, with a priority weight set to The system then sets the priority weight of the electric power steering task to The task execution sequence is reordered based on priority weights, and the execution order of tasks is calculated using a scheduling algorithm. Task scheduling follows the priority queue rules, with tasks with higher priority weights being executed first. Specifically, the system assigns a priority weight to each task in the task queue; the higher the weight, the higher the priority of the task in the queue. The high priority of the electronic stability program control task ensures dynamic stability in abnormal situations, while the lower priority of the electric power steering task ensures timely response to steering operations.

[0072] S53: Dynamically partitions multi-core processor resources based on the reordered task execution sequence, allocating computing resources of the ARM Cortex-R52 real-time core to the chassis control system and backend resources of the Cortex-A78 performance core to the in-vehicle infotainment system, generating a core-task binding mapping table.

[0073] Specifically, the Cortex-R52 processor, with its excellent real-time performance and functional safety, is suitable for chassis control systems that require high-precision, low-latency control tasks, such as electronic stability programs and braking systems, to ensure that these critical tasks can be completed within strict time constraints. At the same time, the system allocates the background resources of the Cortex-A78 performance core to the in-vehicle infotainment system. The Cortex-A78 processor is good at handling complex computing tasks and providing efficient performance. It is suitable for infotainment systems with relatively low real-time requirements but high processing power requirements, such as multimedia playback and navigation functions. Specifically, the system uses the following formula to calculate the computing resources required for each task:

[0074] in, represents the amount of computing resources required for task i, represents the execution time of task i, represents the deadline for task i. By calculating the resource requirements for each task, the system allocates computing resources from the ARM Cortex-R52 real-time core to the chassis control system, ensuring it can handle critical control tasks such as electronic stability program control and electric power steering in real time. Simultaneously, the in-vehicle infotainment system is allocated background resources from the Cortex-A78 performance core to ensure its normal operation without affecting critical tasks.

[0075] For in-vehicle infotainment systems, the system uses a dynamic partitioning algorithm to generate a core-task binding mapping table based on task priorities and resource requirements, tailored to energy efficiency requirements. This mapping table details the binding relationship between each task and processor core, as well as the resource allocation for each core. For example, it records the binding relationship between the chassis control system task and the ARM Cortex-R52 real-time core, along with the processor clock cycles and memory resources allocated to that task. This approach enables refined management of multi-core processor resources, improves resource utilization, and ensures the real-time execution of critical tasks and the smooth operation of non-critical tasks.

[0076] S54: Perform instruction encoding processing on the core-task binding mapping table, use the AUTOSAR COM module to encapsulate the task execution sequence, core binding relationship and resource quota parameters, and generate an optimized load distribution plan including a time trigger mechanism.

[0077] Specifically, the system generates an optimized load distribution plan including a time-triggered mechanism by defining task priorities, allocating time slices, and setting resource limits. The specific algorithm formula is as follows:

[0078] in, represents the total scheduling weight, represents the priority of task i, represents the weight of task i. This formula is used to calculate the total weight of task scheduling, ensuring timely execution of high-priority tasks. The system dynamically adjusts the execution order and resource allocation of tasks based on their priority and weight. Furthermore, the system employs a time-triggered mechanism, setting task execution cycles and deadlines to ensure that tasks are completed within the specified timeframe. This optimized load distribution scheme rationally allocates processor resources, ensuring stable vehicle operation under various operating conditions and improving the system's real-time performance and reliability.

[0079] Specifically, the system encapsulates task execution sequences, core binding relationships, and resource quota parameters into instruction packets, which are then transmitted and executed via the AUTOSAR COM module. For example, for the electronic stability program control task, the system sets its highest priority, allocates a larger time slice and resource quota, and uses a time-triggered mechanism to ensure timely execution within each control cycle.

[0080] In real-world vehicle operating environments, this computer-based dynamic load distribution method can significantly improve vehicle performance and safety. For example, in new energy vehicles, during high-speed cornering maneuvers, the system monitors the vibration energy characteristics of the suspension system in real time, enabling timely detection of potential roll risks. Upon detecting an anomaly, the system immediately prioritizes ESP and EPS tasks according to pre-set safety policies and allocates high-performance real-time processor core resources. This rapid system response effectively prevents vehicle skidding, improves driving stability, and ensures passenger safety. Furthermore, by rationally allocating computing resources, the system ensures that non-critical tasks, such as the in-vehicle infotainment system, can operate normally without compromising vehicle safety, enhancing the user experience. This intelligent dynamic load distribution method not only optimizes the resource utilization efficiency of the vehicle's electronic systems but also enhances the vehicle's adaptability and reliability in complex operating conditions, providing important technical support for the development of intelligent vehicles.

[0081] In one embodiment, the present invention provides a method for dynamic load distribution based on a vehicle body controller, further comprising the following steps: S61: Collect the response delay and control accuracy data of the vehicle body actuator after executing the optimized load distribution plan through the vehicle bus, calculate the deviation between the actual value and the expected value of the system collaborative efficiency index, and generate an execution status deviation report.

[0082] Specifically, the system utilizes on-board communication networks such as the CAN bus or FlexRay bus to obtain real-time response timestamps and control accuracy parameters for various actuators (such as the power steering motor, brake controller, and suspension adjuster). The power steering motor's response latency is defined as the time interval between receiving a control command and the actual start of steering force adjustment. Control accuracy is measured by the difference between the actual steering angle and the target steering angle. For the suspension adjuster, response latency is the time difference between receiving a command and the start of suspension height or damping adjustment. Control accuracy is determined by the deviation between the actual suspension parameters and the target parameters.

[0083] The system calculates the deviation between the actual value and the expected value of the system collaborative efficiency index. The collaborative efficiency index is a parameter that comprehensively evaluates the collaborative effect between various vehicle systems. Its calculation formula is:

[0084] in, is the collaborative efficiency indicator, and They are the actuator control accuracy weight and response delay weight, respectively, and both are pre-set according to the importance of the actuator in the vehicle dynamic performance. To score the control accuracy, the actuator is scored based on how close its actual control accuracy is to the ideal accuracy. The delay penalty factor increases with response latency, negatively impacting collaborative efficiency. The system compares the calculated actual collaborative efficiency index with the expected value based on historical best performance or factory-set standards, generating an execution status deviation report. This report details each actuator's response latency, control accuracy data, and the overall collaborative efficiency index deviation.

[0085] S62: Perform incremental learning processing on the execution state deviation report and the fused load feature set, use the stochastic gradient descent algorithm to calculate the parameter update amount of the load association model, and generate model parameter adjustment suggestions.

[0086] Specifically, the system performs incremental learning on the execution state deviation report and the fused load feature set, uses the stochastic gradient descent algorithm to calculate the parameter update amount of the load association model, and generates model parameter adjustment suggestions. Specifically, the system uses the deviation amount in the execution state deviation report and the feature vector in the fused load feature set as input to construct a loss function to measure the difference between the model's predicted value and the actual value. The loss function is defined as:

[0087] in, is the actual value, is the predicted value, and n is the number of samples. The system minimizes the loss function using the stochastic gradient descent algorithm to calculate the update amount of the model parameters. The core idea of the stochastic gradient descent algorithm is to randomly select a sample, calculate the gradient of the sample, and use the gradient to update the model parameters. The parameter update formula is:

[0088] in, are model parameters, is the learning rate, is the gradient of the loss function with respect to the model parameters. The system updates the model parameters through multiple iterations until the loss function converges to a smaller value. Finally, the system generates a model parameter adjustment recommendation that details the amount and direction of the update.

[0089] S63: The load association model and the model parameter adjustment suggestion are updated online, and the model weight parameters are adjusted in real time by applying the parameter update amount during vehicle operation to generate an updated load association model.

[0090] Specifically, the system reads the parameter updates from the model parameter adjustment recommendations and adjusts the weight parameters of the load-related model accordingly. The updated model better reflects the vehicle's actual operating state, improving prediction accuracy. The system applies these parameter updates in real time during vehicle operation, ensuring the model remains optimal through an online update mechanism. The updated load-related model not only improves system coordination efficiency but also enhances its adaptability to dynamic changes, providing more precise support for intelligent vehicle control.

[0091] After each update, the system verifies the model in real time to ensure that the updated model accurately predicts the actuator's response latency and control accuracy. This verification process is performed using a reserved validation dataset, calculating the prediction error of the updated model on the validation dataset. If the prediction error is within an acceptable range, the updated model is accepted; otherwise, the system rolls back to the previous model parameters and recalculates the update.

[0092] Through this online update mechanism, the system ensures that the load association model is always in optimal condition, accurately reflecting the vehicle's actual operating conditions, thereby improving the accuracy and efficiency of the vehicle's dynamic load distribution. In the actual operating environment of new energy vehicles, this real-time update mechanism can significantly enhance vehicle performance and safety. For example, under complex road conditions, the system can quickly adjust the load distribution strategy based on the real-time updated model, optimizing the vehicle's handling performance and energy efficiency. Over long-term use, the system's adaptive capabilities ensure that the vehicle maintains optimal operating condition, reducing maintenance costs and improving the user experience.

[0093] It should be understood that, although the various steps in the flowcharts involved in the various embodiments described above are displayed in sequence according to the instructions of the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and these steps can be executed in other orders. Moreover, at least a portion of the steps in the flowcharts involved in the various embodiments described above can include multiple steps or multiple stages, and these steps or stages are not necessarily executed and completed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a portion of steps or stages in other steps.

[0094] Based on the same inventive concept, embodiments of the present application also provide a vehicle body controller-based dynamic load distribution device for implementing the aforementioned vehicle body controller-based dynamic load distribution method. The solution provided by this device is similar to the solution described in the aforementioned method. Therefore, the specific limitations of one or more of the following embodiments of the vehicle body controller-based dynamic load distribution device can be found in the aforementioned limitations of the vehicle body controller-based dynamic load distribution method and will not be further elaborated here.

[0095] Preferably, if Figure 4 As shown, the present invention provides a dynamic load distribution device 700 based on a vehicle body controller, which is configured with the following modules: The data acquisition module 710 is used to collect data from various vehicle body components based on a distributed sensor network, including current and voltage data of electrical loads, force and acceleration data of mechanical loads, and temperature and heat flow data of thermal loads, to generate initial load data. The data preprocessing module 720 is used to preprocess the initial load data, eliminate the dimensional differences between mechanical load, electrical load and thermal load, and generate a standard load data set; The fused load feature generation module 730 is used to perform unified feature space mapping on the standard load data set, extract comprehensive characterization features of mechanical load, electrical load, and thermal load, and generate feature threshold ranges based on historical data distribution to generate a fused load feature set containing a fused feature vector and a preset threshold range; The load correlation model mapping module 740 is used to construct a load correlation model of load characteristics and control parameters based on the fused load characteristic set, and input the fused characteristic vector into the load correlation model for parameter mapping processing to generate a mapping control parameter combination; The dynamic load optimization distribution module 750 is used to dynamically adjust the load of the mapping control parameter combination based on the preset threshold range of the fused load feature set, adjust the task execution order and resource allocation ratio based on the preset load priority rules, and generate an optimized load distribution plan. The optimized load distribution plan is used to indicate the CPU core allocation strategy and control the body controller to make real-time parameter adjustments to the power system and chassis system.

[0096] Preferably, the data acquisition module 710 provided in this application is configured with the following units: The mechanical load data acquisition unit is used to monitor and process the stress distribution of the vehicle body structure. It collects the vibration spectrum data of the key nodes of the suspension through the strain sensors of the frame longitudinal beams to generate the original data set of mechanical load. The electrical load data acquisition unit is used to collect and process real-time parameters of the vehicle body's high-voltage electrical system. It obtains the battery management system's discharge current ripple and motor controller phase voltage fluctuation data through the CAN bus to generate the electrical load raw data set. The thermal load data acquisition unit is used to scan and process the temperature field of the heat-sensitive areas of the power battery pack. It uses an on-board infrared thermal imager to capture the surface temperature gradient distribution of the motor controller and the heat flux density changes of the power module to generate a thermal load raw data set. The multi-load data spatiotemporal alignment unit is used to perform spatiotemporal alignment on the mechanical load original data set, the electrical load original data set, and the thermal load original data set based on the precision time protocol. The time base of the mechanical load data, the electrical load data, and the thermal load data is unified through the timestamp calibration algorithm to generate the initial load data.

[0097] Preferably, the data preprocessing module 720 provided in this application is configured with the following units: a mechanical load amplitude normalization unit, for performing amplitude normalization processing on the mechanical load data of the initial load data, scaling the suspension stress data to a range of 0-1 according to the upper limit of the material yield strength, and generating normalized mechanical load data; An electrical load dynamic compression unit is used to perform dynamic range compression processing on the electrical load data of the initial load data, normalize the current data according to the maximum discharge current of the battery, and scale the voltage data according to the rated voltage of the drive motor to generate standard electrical load data; The thermal load temperature rise conversion unit is used to convert the thermal load data of the initial load data into temperature rise characteristics, calculate the relative heat flux density based on the maximum tolerance temperature of the power module, and generate relative heat flux density data; The multi-load data matrix reconstruction unit is used to perform matrix reconstruction processing on normalized mechanical load data, standard electrical load data and relative heat flux density data, integrate stress data, current and voltage data and heat flux data into a three-dimensional data cube in time series, and generate a standard load data set. The standard load data set is used to indicate the real-time load status of each component of the vehicle body.

[0098] Preferably, the fused load signature generation module 730 provided in this application is configured with the following units: Multimodal feature correlation analysis unit: used to extract multimodal features from standard load data sets, calculate the covariance matrix of mechanical vibration frequency domain energy distribution, electrical load time domain ripple coefficient and thermal gradient spatial change rate, and generate a feature correlation matrix; Principal component dimension reduction projection unit: used to perform principal component screening on the feature correlation matrix, retain the principal component direction with eigenvalue greater than 1 based on the Kaiser criterion, calculate the feature subspace with cumulative variance contribution rate greater than 95%, and generate the dimension reduction feature projection matrix; Unified feature space fusion unit: This unit is used to perform feature space mapping on the initial load data. It projects the suspension vibration spectrum, battery current ripple, and power module temperature gradient into a unified feature space based on the dimension reduction feature projection matrix. It calculates the similarity between feature vectors and generates a fused feature vector. Historical data threshold modeling unit: used to retrieve the historical load feature set stored in the vehicle's historical operation process from the on-board storage unit as a historical data sample, perform dynamic distribution analysis and processing on the historical load feature set and fused feature vector based on the ISO 26262 functional safety standard, update the feature distribution boundary of the fused feature vector, and determine the preset threshold range of the current feature vector with three times the standard deviation to generate a fused load feature set.

[0099] Preferably, the load association model mapping module 740 provided in this application is configured with the following units: The nonlinear modeling construction unit is used to perform nonlinear modeling processing on the feature vectors and control parameters of the fused load feature set, build a support vector regression architecture based on the radial basis kernel function, set regularization parameters to constrain the model complexity, and build an initial correlation model between the load features and the control parameters; The hyperparameter grid optimization unit is used to perform grid search optimization on the model hyperparameters of the initial correlation model, use cross-validation in the parameter space to evaluate the prediction accuracy of different (C, γ) combinations, select the parameter configuration with the minimum mean square error, and generate the optimized correlation model; The incremental learning update unit is used to perform incremental learning on the training sample set of the optimized association model. It implements online iterative updates of model parameters on the AUTOSAR architecture domain controller platform, uses an exponentially decaying forgetting factor mechanism to eliminate outdated data, and generates the final load association model. The control parameter mapping unit is used to perform control parameter mapping processing on the fusion feature vector of the fusion load feature set, predict the response time of the steering power system and the torque distribution ratio of the regenerative braking system through the load association model, and generate a mapping control parameter combination.

[0100] Preferably, the dynamic load optimization distribution module 750 provided in this application is configured with the following units: The load anomaly detection unit is used to perform real-time cross-border detection processing on the preset threshold range of the fused load feature set. When the suspension vibration energy characteristic value exceeds the preset upper threshold, the characteristic offset is recalculated, and a sliding time window algorithm is used to confirm whether the duration of the anomaly exceeds the preset warning threshold. The anomaly detection report containing the anomaly trigger signal and characteristic offset data is generated; The safety task reordering unit reorders safety-critical tasks based on anomaly detection reports. Based on the ISO 26262 ASIL-D level, it places the electronic stability program control task at the top of the real-time execution queue and the electric power steering task in the next priority queue, generating a reordered task execution sequence. The processor resource partitioning unit is used to dynamically partition the multi-core processor resources based on the reordered task execution sequence, allocating the computing resources of the ARM Cortex-R52 real-time core to the chassis control system and the background resources of the Cortex-A78 performance core to the in-vehicle infotainment system, and generating a core-task binding mapping table; The load distribution encoding unit is used to perform instruction encoding processing on the core-task binding mapping table, and uses the AUTOSAR COM module to encapsulate the task execution sequence, core binding relationship and resource quota parameters to generate an optimized load distribution plan with a time trigger mechanism.

[0101] In one embodiment, the vehicle body controller-based dynamic load distribution device 700 provided in this application is further configured with the following units: The execution status monitoring unit is used to collect the response delay and control accuracy data of the body actuators after the optimized load distribution plan is executed through the vehicle bus, calculate the deviation between the actual value and the expected value of the system coordination efficiency index, and generate an execution status deviation report; The incremental learning optimization unit is used to perform incremental learning processing on the execution state deviation report and the fused load feature set, using the stochastic gradient descent algorithm to calculate the parameter update amount of the load association model and generate model parameter adjustment suggestions; The online model updating unit is used to update the load association model and the model parameter adjustment suggestions online, and to adjust the model weight parameters by applying the parameter update amount in real time during the vehicle operation to generate an updated load association model.

[0102] In one embodiment, the present application further provides a computer device including a memory and a processor, wherein the memory stores a computer program, and the processor implements the above-mentioned dynamic load distribution method based on the vehicle body controller when executing the computer program.

[0103] In one embodiment, the present application further provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the above-mentioned dynamic load distribution method based on the vehicle body controller is implemented.

[0104] In the description of this specification, the reference terms "one embodiment," "some embodiments," "example," "specific example," or "some examples" mean that the specific features, structures, materials, or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. Moreover, the specific features, structures, materials, or characteristics described may be combined in any appropriate manner in any one or more embodiments or examples. In addition, those skilled in the art may combine and integrate different embodiments or examples described in this specification, as well as features of different embodiments or examples, unless they are mutually inconsistent.

[0105] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to the partial description of the method embodiments. The device embodiments described above are merely illustrative, wherein the components described as separate parts may or may not be physically separated, and the parts displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the disclosed solution. A person of ordinary skill in the art can understand and implement it without expending creative work.

[0106] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any person skilled in the art can easily conceive of various modifications or substitutions within the technical scope disclosed in this application, and such modifications or substitutions should be included within the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

Claims

1. A dynamic load distribution method based on a vehicle body controller, characterized in that: The following steps are involved: S1: Data is collected from various vehicle body components based on a distributed sensor network, including current and voltage data for electrical loads, force and acceleration data for mechanical loads, and temperature and heat flow data for thermal loads, to generate initial load data. S2: Preprocessing the initial load data to eliminate dimensional differences among mechanical load, electrical load, and thermal load, and generating a standard load data set; S3: performing unified feature space mapping on the standard load data set, extracting comprehensive characterization features of mechanical load, electrical load, and thermal load, and generating feature threshold ranges based on historical data distribution, thereby generating a fused load feature set containing a fused feature vector and a preset threshold range; S4: constructing a load association model of load characteristics and control parameters based on the fused load characteristic set, and inputting the fused characteristic vector into the load association model for parameter mapping processing to generate a mapping control parameter combination; S5: Dynamically adjust the load of the mapping control parameter combination based on the preset threshold range of the fused load feature set, adjust the task execution order and resource allocation ratio based on the preset load priority rules, and generate an optimized load distribution plan. The optimized load distribution plan is used to indicate the CPU core allocation strategy and control the body controller to make real-time parameter adjustments to the power system and chassis system.

2. The method according to claim 1, characterized in that Said S1 comprises: S11: Monitors and processes stress distribution of vehicle body structural components. Vibration spectrum data of key suspension nodes is collected through strain sensors on the frame rails to generate a raw data set of mechanical loads. S12: Performs real-time parameter acquisition and processing on the vehicle body's high-voltage electrical system, obtains battery management system discharge current ripple and motor controller phase voltage fluctuation data via the CAN bus, and generates an electrical load raw data set; S13: Perform temperature field scanning on the heat-sensitive areas of the power battery pack. Use the on-board infrared thermal imager to capture the surface temperature gradient distribution of the motor controller and the heat flux density changes of the power module to generate a raw heat load data set. S14: performing spatiotemporal alignment on the mechanical load original data set, the electrical load original data set, and the thermal load original data set based on a precision time protocol, unifying the time bases of the mechanical load data, the electrical load data, and the thermal load data through a timestamp calibration algorithm, and generating initial load data.

3. The method according to claim 1, characterized in that The S2 includes: S21: performing amplitude normalization processing on the mechanical load data of the initial load data, scaling the suspension stress data to a range of 0-1 according to the upper limit of the material yield strength, and generating normalized mechanical load data; S22: performing dynamic range compression processing on the electrical load data of the initial load data, normalizing the current data according to the maximum discharge current of the battery, and scaling the voltage data according to the rated voltage of the drive motor to generate standard electrical load data; S23: performing temperature rise characteristic conversion processing on the thermal load data of the initial load data, calculating the relative heat flux density based on the maximum tolerance temperature of the power module, and generating relative heat flux density data; S24: Matrix reconstruction is performed on the normalized mechanical load data, the standard electrical load data, and the relative heat flux density data, and the stress data, the current and voltage data, and the heat flux data are integrated into a three-dimensional data cube in time series to generate a standard load data set, which is used to indicate the real-time load status of each component of the vehicle body.

4. The method according to claim 1, wherein The S3 includes: S31: performing multimodal feature extraction processing on the standard load data set, calculating the covariance matrix of the mechanical vibration frequency domain energy distribution, the electrical load time domain ripple coefficient, and the thermal gradient spatial change rate, and generating a feature correlation matrix; S32: performing principal component screening on the feature correlation matrix, retaining principal component directions with eigenvalues greater than 1 based on the Kaiser criterion, calculating feature subspaces with cumulative variance contribution rates greater than 95%, and generating a dimensionality-reduced feature projection matrix; S33: performing feature space mapping processing on the initial load data, projecting the suspension vibration spectrum, battery current ripple, and power module temperature gradient into a unified feature space based on the dimension reduction feature projection matrix, calculating the similarity between feature vectors, and generating a fused feature vector; S34: Retrieve the historical load feature set stored in the historical operation process of the vehicle from the on-board storage unit as a historical data sample, perform dynamic distribution analysis and processing on the historical load feature set and the fused feature vector based on the ISO 26262 functional safety standard, update the feature distribution boundary of the fused feature vector, and determine the preset threshold range of the current feature vector with three times the standard deviation to generate a fused load feature set.

5. The method according to claim 1, wherein The S4 includes: S41: performing nonlinear modeling processing on the feature vectors and control parameters of the fused load feature set, constructing a support vector regression architecture based on a radial basis kernel function, setting a regularization parameter to constrain the model complexity, and constructing an initial association model between the load features and the control parameters; S42: performing grid search optimization processing on the model hyperparameters of the initial association model, using a cross-validation method in the parameter space to evaluate the prediction accuracy of different (C, γ) combinations, selecting the parameter configuration with the minimum mean square error, and generating an optimized association model; S43: performing incremental learning processing on the training sample set of the optimized association model, implementing online iterative update of model parameters on the domain controller platform of the AUTOSAR architecture, using an exponential decay forgetting factor mechanism to eliminate obsolete data, and generating a final load association model; S44: performing control parameter mapping processing on the fused feature vector of the fused load feature set, predicting the response time of the power steering system and the torque distribution ratio of the regenerative braking system through the load association model, and generating a mapping control parameter combination.

6. The method according to claim 1, characterized in that The S5 includes: S51: Performing real-time out-of-bounds detection on a preset threshold range of the fused load feature set, recalculating a feature offset when a suspension vibration energy feature value exceeds a preset upper threshold, calling a sliding time window algorithm to determine whether the duration of the abnormality exceeds a preset warning threshold, and generating an abnormality detection report containing an abnormality trigger signal and feature offset data; S52: Rearrange the safety-critical tasks based on the anomaly detection report, place the electronic stability program control task at the top of the real-time execution queue based on the ISO 26262 ASIL-D level, and place the electric power steering task in a lower priority queue, thereby generating a reordered task execution sequence. S53: Dynamically partitioning the multi-core processor resources based on the reordered task execution sequence, allocating computing resources of the ARM Cortex-R52 real-time core to the chassis control system and backend resources of the Cortex-A78 performance core to the in-vehicle infotainment system, and generating a core-task binding mapping table; S54: performing instruction encoding processing on the core-task binding mapping table, using the AUTOSAR COM module to encapsulate the task execution sequence, core binding relationship and resource quota parameters, and generating an optimized load distribution solution including a time trigger mechanism.

7. The method according to any one of claims 1 to 6, characterized in that After S5, the following is also included: S61: Collecting response delay and control accuracy data of the vehicle body actuators after executing the optimized load distribution solution through the vehicle bus, calculating the deviation between the actual value and the expected value of the system coordination efficiency index, and generating an execution status deviation report; S62: performing incremental learning processing on the execution state deviation report and the fused load feature set, calculating parameter update amounts of the load association model using a stochastic gradient descent algorithm, and generating model parameter adjustment suggestions; S63: performing online updating on the load association model and the model parameter adjustment suggestion, and adjusting the model weight parameters by applying the parameter update amount in real time during vehicle operation to generate an updated load association model.

8. A dynamic load distribution device based on a vehicle body controller, characterized in that: The device comprises: The data acquisition module is used to collect data from various vehicle body components based on a distributed sensor network, including current and voltage data of electrical loads, force and acceleration data of mechanical loads, and temperature and heat flow data of thermal loads, to generate initial load data. a data preprocessing module, configured to preprocess the initial load data, eliminate dimensional differences among mechanical load, electrical load, and thermal load, and generate a standard load data set; a fused load feature generation module, configured to perform unified feature space mapping on the standard load data set, extract comprehensive characterization features of mechanical load, electrical load, and thermal load, generate feature threshold ranges based on historical data distribution, and generate a fused load feature set containing a fused feature vector and a preset threshold range; a load association model mapping module, configured to construct a load association model of load characteristics and control parameters based on the fused load feature set, and input the fused feature vector into the load association model for parameter mapping processing to generate a mapping control parameter combination; A dynamic load optimization distribution module is used to dynamically adjust the load of the mapping control parameter combination based on the preset threshold range of the fused load feature set, adjust the task execution order and resource allocation ratio based on the preset load priority rules, and generate an optimized load distribution plan. The optimized load distribution plan is used to indicate the CPU core allocation strategy and control the body controller to make real-time parameter adjustments to the power system and chassis system.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the method according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images