Battery current regulation optimization method and system based on hybrid drive

Abstract

The invention relates to the technical field of battery current adjustment, in particular to a battery current adjustment optimization method and system based on hybrid drive, and the system comprises an acquisition planning module, a collection screening module, a processing optimization module, an intelligent adjustment architecture building module and an adjustment strategy output module. Accurate data screening of the collecting and screening module and effective data processing of the processing and optimizing module provide high-quality data support for the intelligent adjusting framework building module. The intelligent adjustment architecture building module fuses the advantages of various algorithms and generates an accurate comprehensive adjustment strategy. The adjustment strategy output module ensures that the adjustment strategy can be timely and accurately applied to the vehicle battery management system, and real-time accurate regulation and control of the battery current are achieved. The energy utilization efficiency of the hybrid drive automobile is improved, the service life of a battery is prolonged, and the overall performance and driving experience of the automobile are enhanced.

Description

Technical Field

[0001] This invention relates to the field of battery current regulation technology, and in particular to a battery current regulation optimization method and system based on hybrid drive. Background Technology

[0002] With the rapid development of new energy vehicles, hybrid vehicles occupy an important position in the market due to their unique advantages. However, existing battery current regulation technologies have many shortcomings, which seriously restrict the further improvement of the performance of hybrid vehicles.

[0003] Traditional battery current regulation systems often only collect limited vehicle operating data, such as basic information like battery voltage and current. They lack comprehensive and systematic collection and analysis of multi-dimensional data on vehicle driving status, road conditions, and environmental factors. This results in an inaccurate and incomplete perception of the vehicle's operating status, failing to provide sufficient decision-making basis for current regulation. Moreover, most existing systems use fixed regulation rules and single algorithm models, making it difficult to adapt to complex and changing vehicle driving conditions and different user driving habits, resulting in poor regulation accuracy and adaptability.

[0004] Furthermore, traditional systems have limited data processing capabilities, making it impossible to effectively analyze and mine the large amounts of collected data. Data utilization is low, failing to fully leverage the value of the data to optimize current regulation strategies. Issues such as data lag and noise interference often exist, affecting the timeliness and accuracy of current regulation. In terms of system integration, existing battery current regulation systems lack sufficient coordination with other vehicle systems, failing to achieve efficient information sharing and collaborative control, resulting in overall vehicle performance falling short of optimal levels. Summary of the Invention

[0005] The purpose of this invention is to provide a battery current regulation optimization method and system based on hybrid drive, which aims to solve the problem of low regulation accuracy in existing battery current regulation optimization systems.

[0006] To achieve the above objectives, in a first aspect, the present invention provides a battery current regulation and optimization system based on hybrid drive, comprising a data acquisition and planning module, a data collection and filtering module, a processing and optimization module, an intelligent regulation architecture construction module, and a regulation strategy output module, wherein the data acquisition and planning module, the data collection and filtering module, the processing and optimization module, the intelligent regulation architecture construction module, and the regulation strategy output module are connected in sequence; The data acquisition planning module is used to select and deploy data acquisition terminals to form a data acquisition matrix, collect multi-dimensional information about the vehicle, and provide a data foundation for current regulation. The collection and filtering module is used to collect battery performance degradation cycle data as a benchmark, monitor the prototype vehicle's operating status, and filter data that meets the conditions to generate a set of original data sequences. The processing optimization module is used to process the raw data, remove noise, extract key features and patterns, optimize data quality, and generate high-quality data sequence clusters. The intelligent regulation architecture building module is used to select intelligent regulation algorithms, construct a composite regulation architecture including parallel regulation subunits and a central coordination main unit, and generate a comprehensive battery current regulation strategy. The adjustment strategy output module is used to collect real-time operating data, preprocess it, input it into the adjustment architecture for calculation, output the adjustment strategy, and send it to the vehicle battery management system.

[0007] The data acquisition planning module includes a data acquisition terminal selection unit, a sensor layout planning unit, and a prototype vehicle deployment unit. The data acquisition terminal selection unit is used to select a data acquisition terminal that is compatible with the target hybrid drive vehicle based on the vehicle's key performance indicators and sensor performance parameters, covering multi-dimensional data acquisition needs such as vehicle operating status, battery parameters, and environmental information. The sensor layout planning unit is responsible for determining the placement of the data acquisition terminal on the prototype vehicle and obtaining the optimal layout scheme through simulation and actual road test verification. The prototype deployment unit deploys data acquisition terminals on multiple prototype vehicles of the same type to form a data acquisition matrix.

[0008] The collection and screening module includes a performance decay cycle data acquisition unit, an operation data monitoring unit, and a data comparison and screening unit. The performance degradation cycle data acquisition unit is used to acquire performance degradation cycle data of the target vehicle battery. The operation data monitoring unit is used to monitor the prototype vehicle's operating status in real time and acquire multi-dimensional operation data such as driving speed, acceleration, driving mileage, road slope, road surface smoothness, battery voltage, current, and temperature. The data comparison and filtering unit is used to compare the running data with the reference benchmark, and adopts a dynamic threshold adjustment algorithm to dynamically adjust the filtering threshold according to the real-time battery status and historical data, and filters out the data that meets the conditions and stores it in the initial screening database to generate the original data sequence set.

[0009] The processing optimization module includes a filtering and noise reduction unit, a feature extraction and pattern mining unit, and a data quality assessment and optimization unit. The filtering and noise reduction processing unit is used to remove noise and interference components from the original data sequence set and improve data purity; The feature extraction and pattern mining unit is used to mine key features and temporal correlation patterns in the data and extract information valuable for battery current regulation. The data quality assessment and optimization unit performs quantitative evaluation of the processed data based on a preset data quality scoring system, iteratively optimizes the data, continuously improves the data quality, and ultimately generates a high-quality data sequence cluster.

[0010] The intelligent adjustment architecture building module includes an algorithm selection unit, a parallel adjustment building unit, and a central coordination main building unit. The algorithm selection unit is used to select a suitable algorithm from the multi-intelligence regulation algorithm library to provide an algorithmic basis for building the regulation architecture; The parallel adjustment building unit is used to build multiple parallel adjustment sub-units. Each sub-unit operates independently based on different algorithms and outputs its own battery current adjustment suggestions. The central coordination master unit is used to construct the central coordination master subunit. It is responsible for integrating the adjustment suggestions based on the real-time performance and historical data of the parallel subunits and using a dynamic weight allocation strategy to generate a comprehensive battery current adjustment strategy.

[0011] The adjustment strategy output module includes a real-time data acquisition unit, a preprocessing unit, a strategy calculation and output unit, and a strategy sending and control unit. The real-time data acquisition unit is used to collect real-time operating data of the target vehicle, providing a real-time basis for outputting precise adjustment strategies; The preprocessing unit is used to standardize and normalize the collected real-time operating data to make it meet the input requirements of the composite adjustment architecture. The strategy calculation output unit is used to input the preprocessed data into the composite regulation architecture for calculation and output a precise battery current regulation strategy. The strategy sending and control unit is used to send the adjustment strategy to the vehicle's battery management system to achieve real-time control of the battery current.

[0012] Secondly, a hybrid-driven battery current regulation optimization method, used in the hybrid-driven battery current regulation optimization system described in the first aspect, includes the following steps: By using the data acquisition planning module, data acquisition terminals are deployed on multiple prototype vehicles of the same type to form a data acquisition matrix and collect multi-dimensional data during vehicle operation. The battery performance degradation cycle data of the target vehicle is obtained by the collection and screening module and set as the data collection reference benchmark. The operating status of the prototype vehicle is monitored in real time, the operating data is compared with the reference benchmark, and the data that meets the conditions is filtered and stored in the initial screening database to generate the original data sequence set. The processing optimization module is used to perform noise reduction, feature extraction and optimization on the original data sequence set to generate a high-quality data sequence cluster. Select an appropriate algorithm from the multi-intelligence regulation algorithm library, construct a composite regulation architecture using the intelligent regulation architecture building module, import high-quality data sequence clusters for training and optimization, and complete the architecture construction. The system collects real-time operating data of the target vehicle, preprocesses it through the adjustment strategy output module, inputs it into the composite adjustment architecture, calculates and outputs the battery current adjustment strategy, and sends it to the vehicle battery management system.

[0013] This invention relates to a hybrid-drive-based battery current regulation and optimization system. Through a well-designed data acquisition and planning module, the system comprehensively acquires multi-dimensional vehicle information, providing a solid data foundation for subsequent current regulation. A data collection and filtering module precisely filters effective data, improving accuracy and representativeness. A processing and optimization module effectively removes data noise and extracts key features, further enhancing data quality. An intelligent regulation architecture module integrates the advantages of multiple algorithms to generate a comprehensive regulation strategy, enhancing the accuracy and adaptability of current regulation. A regulation strategy output module ensures that the regulation strategy can be applied to the vehicle's battery management system in a timely and accurate manner, achieving real-time and precise control of the battery current. These modules are connected sequentially and work collaboratively, significantly improving the energy efficiency of hybrid-drive vehicles, extending battery life, and enhancing overall vehicle performance and driving experience, thereby solving the problem of low regulation accuracy in existing battery current regulation and optimization systems. Attached Figure Description

[0014] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0015] Figure 1 This is a schematic diagram of the battery current regulation and optimization system based on hybrid drive provided by the present invention.

[0016] Figure 2 This is a schematic diagram of the data acquisition and planning module.

[0017] Figure 3 This is a schematic diagram of the collection and filtering module.

[0018] Figure 4 This is a schematic diagram of the optimization module.

[0019] Figure 5 This is a schematic diagram of the intelligent adjustment architecture building module.

[0020] Figure 6 This is a schematic diagram of the adjustment strategy output module.

[0021] Figure 7 This is a flowchart of the battery current regulation optimization method based on hybrid drive provided by the present invention.

[0022] In the diagram: 1-Acquisition planning module, 2-Collection and filtering module, 3-Processing and optimization module, 4-Intelligent adjustment architecture construction module, 5-Adjustment strategy output module, 11-Data acquisition terminal selection unit, 12-Sensor layout planning unit, 13-Prototype deployment unit, 21-Performance decay cycle data acquisition unit, 22-Operational data monitoring unit, 23-Data comparison and filtering unit, 31-Filtering and noise reduction processing unit, 32-Feature extraction and pattern mining unit, 33-Data quality assessment and optimization unit, 41-Algorithm selection unit, 42-Parallel adjustment construction unit, 43-Central coordination main construction unit, 51-Real-time data acquisition unit, 52-Preprocessing unit, 53-Strategy calculation and output unit, 54-Strategy transmission and control unit. Detailed Implementation

[0023] Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain the present invention, and should not be construed as limiting the present invention.

[0024] Please see Figures 1 to 6 In a first aspect, the present invention provides a battery current regulation and optimization system based on hybrid drive, including a data acquisition and planning module 1, a data collection and filtering module 2, a processing and optimization module 3, an intelligent regulation architecture construction module 4, and a regulation strategy output module 5, wherein the data acquisition and planning module 1, the data collection and filtering module 2, the processing and optimization module 3, the intelligent regulation architecture construction module 4, and the regulation strategy output module 5 are connected in sequence. The data acquisition planning module 1 is used to select and deploy data acquisition terminals to form a data acquisition matrix, collect multi-dimensional information about the vehicle, and provide a data basis for current regulation. The collection and screening module 2 is used to collect battery performance degradation cycle data as a benchmark, monitor the prototype vehicle's operating status, and filter data that meets the conditions to generate an original data sequence set. The processing optimization module 3 is used to process the raw data, remove noise, extract key features and patterns, optimize data quality, and generate high-quality data sequence clusters. The intelligent regulation architecture building module 4 is used to select an intelligent regulation algorithm, construct a composite regulation architecture including a parallel regulation subunit and a central coordination main unit, and generate a comprehensive battery current regulation strategy. The adjustment strategy output module 5 is used to collect real-time operating data, preprocess it, input it into the adjustment architecture for calculation, output the adjustment strategy, and send it to the vehicle battery management system.

[0025] In this embodiment, the invention, through the rational layout of the data acquisition and planning module 1, can comprehensively acquire multi-dimensional information about the vehicle, providing a solid data foundation for subsequent current regulation; the data collection and filtering module 2 accurately filters effective data, improving the accuracy and representativeness of the data; the processing and optimization module 3 effectively removes data noise and extracts key features, further improving data quality; the intelligent regulation architecture building module 4 integrates the advantages of multiple algorithms to generate a comprehensive regulation strategy, enhancing the accuracy and adaptability of current regulation; and the regulation strategy output module 5 ensures that the regulation strategy can be applied to the vehicle battery management system in a timely and accurate manner, achieving real-time and precise control of the battery current. These modules are connected sequentially and work together to greatly improve the energy utilization efficiency of hybrid vehicles, extend battery life, and enhance the overall performance and driving experience of the vehicle, thereby solving the problem of low regulation accuracy in existing battery current regulation optimization systems.

[0026] Furthermore, the data acquisition planning module 1 includes a data acquisition terminal selection unit 11, a sensor layout planning unit 12, and a prototype vehicle deployment unit 13; The data acquisition terminal selection unit 11 is used to select a data acquisition terminal that is compatible with the target hybrid drive vehicle based on the vehicle's key performance indicators and sensor performance parameters, covering multi-dimensional data acquisition needs such as vehicle operating status, battery parameters, and environmental information. The sensor layout planning unit 12 is responsible for determining the placement of the data acquisition terminal on the prototype vehicle and obtaining the optimal layout scheme through simulation and actual road test verification. The prototype deployment unit 13 deploys data acquisition terminals on multiple prototype vehicles of the same type to form a data acquisition matrix.

[0027] In this embodiment, the data acquisition terminal selection unit 11 selects data acquisition terminals suitable for the target hybrid drive vehicle based on the vehicle's key performance indicators and sensor performance parameters. Optimization algorithms such as genetic algorithms are used to select a combination of terminals that can comprehensively cover multi-dimensional data acquisition needs, including vehicle operating status, battery parameters, and environmental information. The sensor layout planning unit 12 determines the placement of the data acquisition terminals on the prototype vehicle. Using a multi-objective optimization algorithm, NSGA-II, with optimization objectives such as data acquisition accuracy, minimization of signal interference between sensors, and physical feasibility of the layout, combined with a vehicle structure model and simulation tools, the sensor layout is iteratively optimized to obtain the optimal layout scheme. The prototype vehicle deployment unit 13 deploys data acquisition terminals on multiple prototype vehicles of the same type, forming a data acquisition matrix. Following the planned layout scheme, standardized installation processes and calibration procedures are adopted to ensure accurate installation and normal operation of each terminal, providing a stable and reliable foundation for subsequent data acquisition. The data acquisition terminal selection unit 11 establishes a database containing multiple sensors, encodes different sensor combinations using a genetic algorithm, and performs multi-generational evolutionary selection with the comprehensiveness and accuracy of data acquisition as the fitness function to ultimately determine the optimal sensor combination, thereby achieving effective acquisition of multi-dimensional vehicle information. Multiple potential sensor placement locations are divided on the vehicle model. The sensor layout planning unit 12 uses the NSGA-II algorithm to evaluate and filter the layout of different locations, ultimately obtaining the Pareto optimal layout scheme under various constraints. The final layout is determined after verification through actual road testing.

[0028] Furthermore, the collection and screening module 2 includes a performance decay cycle data acquisition unit 21, an operation data monitoring unit 22, and a data comparison and screening unit 23; The performance degradation cycle data acquisition unit 21 is used to acquire the performance degradation cycle data of the target vehicle battery. The operation data monitoring unit 22 is used to monitor the prototype vehicle's operating status in real time and acquire multi-dimensional operation data such as driving speed, acceleration, driving mileage, road slope, road surface smoothness, battery voltage, current, and temperature. The data comparison and filtering unit 23 is used to compare the running data with the reference benchmark, and adopts a dynamic threshold adjustment algorithm to dynamically adjust the filtering threshold according to the real-time battery status and historical data, and filters out the data that meets the conditions and stores it in the initial screening database to generate the original data sequence set.

[0029] In this embodiment, the performance degradation cycle data acquisition unit 21 employs sensors and data acquisition chips from a high-precision battery management system (BMS) to collect real-time performance degradation cycle data of the target vehicle's battery. This data includes key indicators such as the number of charge-discharge cycles, capacity degradation curve, and internal resistance changes. The battery's equivalent circuit model and capacity estimation algorithm, combined with the time-integration method and Kalman filtering algorithm, can be used to accurately model and acquire data on the battery performance degradation cycle. The operation data monitoring unit 22 monitors the prototype vehicle's operating status in real-time through the vehicle bus system (CAN bus) and a distributed sensor network. A sensor fusion algorithm is used to fuse multi-dimensional operating data from different sensors, including driving speed, acceleration, mileage, road slope, road surface smoothness, battery voltage, current, and temperature, improving data accuracy and reliability. A Kalman filtering algorithm is used to fuse speed data from different wheel speed sensors to obtain accurate vehicle speed information; a neural network algorithm is used to process road slope and road surface smoothness data from multiple sensors, enabling accurate identification and quantification of road conditions. When comparing the running data with the reference benchmark, the data comparison and screening unit 23 employs a dynamic threshold adjustment algorithm. Based on the online learning algorithm in machine learning, the Adaptive Resonance Theory (ART) network dynamically adjusts the screening threshold according to the real-time battery status and historical data. Real-time monitoring of battery voltage, current, and other data is compared with the expected values ​​calculated based on the battery model. When the deviation exceeds the threshold dynamically adjusted based on historical data and the current battery status, the data is deemed abnormal and filtered out. This process selects data that meets the criteria and stores it in the initial screening database, generating an original data sequence set.

[0030] Furthermore, the processing optimization module 3 includes a filtering and noise reduction processing unit 31, a feature extraction and pattern mining unit 32, and a data quality assessment and optimization unit 33; The filtering and noise reduction processing unit 31 is used to remove noise and interference components from the original data sequence set and improve data purity; The feature extraction and pattern mining unit 32 is used to mine key features and temporal correlation patterns in the data and extract information valuable for battery current regulation. The data quality assessment and optimization unit 33 performs quantitative assessment of the processed data based on a preset data quality scoring system, iteratively optimizes the data, continuously improves the data quality, and finally generates a high-quality data sequence cluster.

[0031] In this embodiment, the filtering and noise reduction processing unit 31 uses an adaptive filtering and noise reduction algorithm to remove noise and interference components from the original data sequence set. A wavelet transform algorithm is employed, and by selecting appropriate wavelet basis functions and decomposition scales, the data is decomposed and reconstructed at multiple scales to effectively filter out noise. Wavelet decomposition is performed on the acquired vehicle acceleration signal to separate signals in different frequency bands. After removing high-frequency noise components, the signal is reconstructed to obtain a clean acceleration signal, improving data purity. The feature extraction and pattern mining unit 32 uses intelligent feature extraction technology to mine key features and temporal correlation patterns in the data. A convolutional neural network (CNN) from deep learning algorithms is used to extract features from multi-dimensional data, automatically learning spatial and temporal features. Time-series data during vehicle operation (such as battery current and voltage) is used as input to the CNN. Through feature extraction operations in convolutional and pooling layers, key features valuable for battery current regulation are mined, such as characteristic waveforms during battery charging and discharging, and current change patterns under vehicle acceleration and deceleration conditions. Temporal correlation patterns are extracted from the data to provide a basis for subsequent regulation strategy generation. The data quality assessment and optimization unit 33 performs a quantitative evaluation of the processed data based on a preset data quality scoring system. Using a fuzzy comprehensive evaluation algorithm, multiple quality indicators such as accuracy, completeness, and consistency are used as evaluation factors to construct a fuzzy evaluation matrix and comprehensively score the data quality. Based on the scoring results, iterative optimization methods such as particle swarm optimization (PSO) are employed to adjust the parameters of the data processing algorithm, continuously improving data quality and ultimately generating a high-quality data sequence cluster.

[0032] Furthermore, the intelligent adjustment architecture building module 4 includes an algorithm selection unit 41, a parallel adjustment building unit 42, and a central coordination main building unit 43; The algorithm selection unit 41 is used to select a suitable algorithm from the multi-intelligence regulation algorithm library to provide an algorithmic basis for building the regulation architecture; The parallel adjustment building unit 42 is used to build multiple parallel adjustment sub-units. Each sub-unit operates independently based on different algorithms and outputs its own battery current adjustment suggestions. The central coordination master construction unit 43 is used to construct the central coordination master sub-unit, which is responsible for integrating the adjustment suggestions based on the real-time performance and historical data of the parallel sub-units and adopting a dynamic weight allocation strategy to generate a comprehensive battery current adjustment strategy.

[0033] In this embodiment, the algorithm selection unit 41 selects a suitable algorithm from a multi-dimensional intelligent regulation algorithm library. Based on the characteristics and requirements of the hybrid drive vehicle battery current regulation problem, and considering the advantages and applicability of different algorithms, an expert system and machine learning algorithm are combined for algorithm selection. By establishing an algorithm library containing various intelligent algorithms such as neural network algorithms, fuzzy control algorithms, and genetic algorithms, the expert system is used to evaluate and screen the performance characteristics and applicable scenarios of different algorithms. Simultaneously, machine learning algorithms are combined to predict and optimize the performance of the algorithms, providing the optimal algorithm combination for constructing the regulation architecture. The parallel regulation construction unit 42 constructs multiple parallel regulation sub-units, each operating independently based on a different algorithm. One sub-unit uses a neural network algorithm to predict and regulate the battery current. By training the neural network model to learn the battery's charging and discharging characteristics, it predicts the battery's current demand under different operating conditions and outputs corresponding regulation suggestions. Another sub-unit uses a fuzzy control algorithm to perform real-time regulation and control of the battery current according to the fuzzy rules of battery current regulation. A third sub-unit uses a genetic algorithm to optimize the battery current regulation strategy to improve the accuracy and adaptability of the regulation. Each sub-unit operates independently without interfering with each other, and each outputs its own battery current regulation suggestions. The central coordination master unit 43 constructs a central coordination master subunit, responsible for fusing adjustment suggestions based on the real-time performance of parallel subunits and historical data, using a dynamic weight allocation strategy. Ensemble learning algorithms such as AdaBoost are employed to weight and fuse the outputs of each parallel subunit. Based on the accuracy and stability of each subunit in historical data processing, its weight coefficients are dynamically adjusted, and the adjustment suggestions from the neural network subunit, fuzzy control subunit, and genetic algorithm subunit are weighted and summed to generate a comprehensive battery current adjustment strategy, improving the reliability and accuracy of the adjustment strategy.

[0034] Furthermore, the adjustment strategy output module 5 includes a real-time data acquisition unit 51, a preprocessing unit 52, a strategy calculation output unit 53, and a strategy sending and control unit 54; The real-time data acquisition unit 51 is used to collect real-time operating data of the target vehicle, providing a real-time basis for outputting precise adjustment strategies; The preprocessing unit 52 is used to standardize and normalize the collected real-time operating data to make it meet the input requirements of the composite adjustment architecture. The strategy calculation output unit 53 is used to input the preprocessed data into the composite regulation architecture for calculation and output a precise battery current regulation strategy. The strategy sending and control unit 54 is used to send the control strategy to the vehicle's battery management system to realize real-time control of the battery current.

[0035] In this embodiment, the real-time data acquisition unit 51 employs a high-speed, high-precision data acquisition card and sensors to collect real-time operating data of the target vehicle. Utilizing the vehicle bus system and wireless sensor network, it acquires information such as vehicle speed, acceleration, and battery status in real time, ensuring that the collected data accurately reflects the vehicle's real-time operating status and providing a real-time basis for outputting precise adjustment strategies. The preprocessing unit 52 standardizes and normalizes the collected real-time operating data. Using the Z-Score normalization algorithm or the Min-Max normalization algorithm, data of different types and dimensions are converted to a unified numerical range, conforming to the input requirements of the composite adjustment architecture. The collected vehicle speed, battery current, and other data are standardized to eliminate dimensional differences between data, improve data consistency and comparability, and provide accurate data support for subsequent adjustment strategy calculations. The strategy calculation output unit 53 inputs the preprocessed data into the composite adjustment architecture for calculation. The composite adjustment architecture comprehensively utilizes the aforementioned intelligent algorithms to analyze and process the input data, outputting a precise battery current adjustment strategy. Based on real-time and historical data, a neural network algorithm is used to predict the battery current change trend. This is combined with a fuzzy control algorithm to adjust the current in real time. Simultaneously, the optimal battery current regulation strategy is generated by referencing the optimized adjustment parameters obtained from a genetic algorithm, and this strategy is sent to the vehicle's battery management system. The strategy sending and control unit 54 employs a reliable communication protocol and data transmission technology to send the regulation strategy to the vehicle's battery management system. Through vehicle communication protocols such as CAN bus or LIN bus, the regulation strategy is sent to the battery management system in a specified data format and transmission rate, ensuring that the regulation strategy can be applied to battery current regulation in a timely and accurate manner, achieving real-time and precise control of the battery current, and improving the energy utilization efficiency and battery performance of hybrid vehicles.

[0036] Please see Figure 7 Secondly, a hybrid-driven battery current regulation optimization method, used in the hybrid-driven battery current regulation optimization system described in the first aspect, includes the following steps: S1 uses the data acquisition planning module 1 to deploy data acquisition terminals on multiple prototype vehicles of the same type to form a data acquisition matrix and collect multi-dimensional data during vehicle operation. Specifically, upon receiving the task, the first step is to conduct an in-depth analysis of key information regarding the hybrid vehicle's model, powertrain, and battery type to determine the key data types that need to be collected, including but not limited to multi-dimensional data such as vehicle speed, acceleration, battery voltage, current, temperature, and engine speed. Based on these data requirements, sensors suitable for different types of data acquisition are selected from the sensor database, such as high-precision wheel speed sensors, current sensors, voltage sensors, and temperature sensors. After completing the sensor selection, the sensor layout planning phase begins. A 3D model of the hybrid vehicle is created, and multiple potential sensor installation locations are marked on the model. Taking the vehicle speed sensor as an example, it is considered to be installed near the hubs of the front and rear wheels to ensure direct measurement of wheel speed, thereby accurately calculating vehicle speed. For battery-related sensors, such as voltage and current sensors, they are planned to be installed near the battery pack, close to the battery's output ports, to accurately acquire battery voltage and current signals. The selected sensors are then installed on multiple prototype vehicles of the same type to form a data acquisition matrix. During installation, standardized installation procedures are strictly followed to ensure that the installation position and angle of each sensor meet design requirements. For wheel speed sensors, high-precision positioning tools are used to ensure a precise perpendicular distance between them and the wheel's rotation axis, guaranteeing measurement accuracy. After installation, all sensors undergo rigorous calibration and standardization. Professional calibration equipment, such as a high-precision speed standard source, is used to calibrate the wheel speed sensors, and high-precision current and voltage sources are used to calibrate the battery sensors, ensuring the accuracy of the collected data.

[0037] S2 obtains the battery performance degradation cycle data of the target vehicle through the collection and screening module 2, sets it as the data collection reference benchmark, monitors the operating status of the prototype vehicle in real time, compares the operating data with the reference benchmark, filters the data that meets the conditions and stores it in the initial screening database, and generates the original data sequence set. Specifically, collecting battery performance degradation cycle data is a crucial step. A high-precision Battery Management System (BMS) is employed, equipped with advanced sensors and data acquisition chips, capable of real-time monitoring of key battery indicators during charge-discharge cycles, such as capacity changes, internal resistance increases, and self-discharge rate. To accurately capture the characteristics of battery performance degradation, a capacity estimation algorithm based on the battery's equivalent circuit model was developed, combining the ampere-hour integral method and Kalman filtering algorithm to estimate battery capacity in real time. During data collection, information such as the number of charge-discharge cycles, capacity changes in each cycle, and battery usage time are recorded simultaneously. As data accumulates, a battery performance degradation database is constructed, storing a large amount of battery performance degradation data under different operating conditions, including battery model, manufacturer, usage environment, and corresponding performance degradation curves. Through in-depth analysis of the database, regression analysis algorithms from machine learning are used to fit a general model for battery performance degradation. During real-time monitoring of the prototype vehicle's operating status, multi-dimensional operational data of the vehicle is comprehensively collected using vehicle bus systems (such as CAN bus) and distributed sensor networks. Sensor fusion algorithms are used to fuse data from different sensors. For vehicle speed data, a Kalman filter algorithm is used to fuse data from multiple wheel speed sensors, effectively filtering out noise interference to obtain accurate vehicle speed information. To filter out valuable data, a dynamic threshold adjustment algorithm is employed. Based on a battery performance degradation model and considering the vehicle's real-time operating conditions, the threshold for data filtering is dynamically calculated. When the battery is discharging and the vehicle is accelerating, the battery voltage threshold is adjusted in real-time according to factors such as the battery's used capacity, current discharge current, and battery temperature. This dynamic adjustment mechanism accurately filters out data that conforms to the characteristics of the battery performance degradation cycle.

[0038] S3 uses the processing optimization module 3 to perform noise reduction, feature extraction and optimization on the original data sequence set to generate a high-quality data sequence cluster. Specifically, the original data sequence often contains a large amount of noise and interference, requiring various data processing techniques for purification and optimization. First, wavelet transform algorithms are used to denoise the original data. Taking vehicle acceleration signals as an example, by selecting appropriate wavelet basis functions and decomposition scales, the signal is decomposed into multi-scale components. During decomposition, the signal is divided into sub-band signals of different frequency bands. Then, noise components in the high-frequency sub-band signals are thresholded. Finally, the processed sub-band signals are reconstructed using inverse wavelet transform to obtain a clean acceleration signal. After denoising, intelligent feature extraction techniques are used to mine key features and temporal correlation patterns in the data. Convolutional Neural Networks (CNNs) algorithms from deep learning are used to extract features from multi-dimensional data. The denoised vehicle operation data, such as battery current, voltage, temperature, and vehicle speed, are input into the CNN model as time-series data samples. The convolutional layers in the model slide their kernels across the data, automatically learning local spatial features, while the pooling layers downsample these features to further highlight the main features. After multiple convolution and pooling operations, the final output is a key feature vector characterizing the vehicle's operating status and battery performance. To further optimize data quality, the processed data is quantitatively evaluated according to a pre-defined data quality scoring system. A fuzzy comprehensive evaluation algorithm is used, employing multiple quality indicators such as data accuracy, completeness, and consistency as evaluation factors. A fuzzy evaluation matrix is ​​constructed, where rows represent different evaluation factors and columns represent individual data samples. The scores of each data sample under different evaluation factors are fuzzily transformed to obtain the comprehensive evaluation result. Based on the evaluation result, the particle swarm optimization (PSO) algorithm is used to optimize and adjust the parameters in the data processing process. This includes adjusting the decomposition scale and threshold processing parameters in the wavelet transform, or optimizing the convolution kernel size and pooling method in the CNN model. Through this iterative optimization process, a high-quality data sequence cluster is ultimately generated.

[0039] S4 selects an appropriate algorithm from the multi-intelligence regulation algorithm library, uses the intelligent regulation architecture building module 4 to build a composite regulation architecture, imports high-quality data sequence clusters for training and optimization, and completes the architecture building. Specifically, this paper first conducts an in-depth analysis of the battery current regulation problem in hybrid vehicles, clarifying that its key challenge lies in adapting to complex and ever-changing vehicle operating conditions and current regulation requirements under different battery states. To address these issues, multiple candidate algorithms, including neural networks, fuzzy control, and genetic algorithms, are selected from a multivariate intelligent regulation algorithm library. To determine the optimal algorithm combination, an evaluation model containing multiple intelligent regulation algorithms is constructed. This model uses the algorithm's prediction accuracy, convergence speed, and adaptability as evaluation indicators, and tests and scores each algorithm by simulating battery current regulation scenarios under different operating conditions. After selecting the algorithms, a composite regulation architecture is constructed using intelligent regulation architecture module 4. The architecture contains multiple parallel regulation sub-units, each running a different intelligent regulation algorithm. One sub-unit uses a neural network algorithm, training a neural network model to learn the battery's charging and discharging characteristics and the vehicle's operating patterns to predict the optimal current regulation strategy for the battery under different operating conditions. Another sub-unit uses a fuzzy control algorithm, adjusting and controlling the battery current in real time based on a fuzzy rule base for battery current regulation. To achieve collaborative work among the sub-units, a central coordinating master unit is constructed. This unit employs the AdaBoost algorithm to fuse the outputs of each parallel sub-unit. During the fusion process, the central coordinating unit dynamically allocates weights based on the performance of each sub-unit in historical data processing, and performs a weighted summation of the adjustment strategies of the neural network sub-unit, fuzzy control sub-unit, etc., to generate a comprehensive adjustment strategy. By continuously inputting high-quality data sequence clusters for training and optimization, the model's prediction accuracy and adaptability are continuously improved, ultimately completing the construction of the intelligent adjustment architecture.

[0040] S5 collects real-time operating data of the target vehicle, which is then preprocessed by the adjustment strategy output module 5 and input into the composite adjustment architecture to calculate and output the battery current adjustment strategy, which is then sent to the vehicle battery management system.

[0041] Specifically, a vehicle bus system and wireless sensor network are used to collect real-time operating data of the target vehicle. This data includes key information such as the vehicle's real-time speed, acceleration, battery voltage, current, temperature, and engine speed. To ensure the accuracy and reliability of the data, the collected data undergoes rigorous verification and completion processing. Data interpolation algorithms are used to reasonably estimate missing data points, and historical data is combined to correct abnormal data. The preprocessed real-time operating data is input into the composite regulation architecture. Each parallel regulation subunit in the architecture analyzes and processes the data according to its own algorithm characteristics. The neural network subunit extracts and analyzes features from the real-time data using a trained model to predict the optimal battery current regulation strategy under the current operating conditions. The fuzzy control subunit matches the real-time data with corresponding rules in the fuzzy rule base and quickly outputs current regulation suggestions. These regulation suggestions are sent to the central coordination master unit, where they undergo weighted fusion processing to generate the final comprehensive regulation strategy. Finally, the battery current regulation strategy is sent to the vehicle battery management system. Through the vehicle bus system, the regulation strategy is transmitted to the battery management system in a specified data format and transmission rate. The battery management system adjusts the battery's charging and discharging current in real time based on the received adjustment strategy, achieving precise control of the battery current. When the vehicle accelerates, the system increases the battery discharging current according to the adjustment strategy to ensure that the vehicle receives sufficient power output; when the vehicle decelerates or brakes, the system adjusts the battery charging current to effectively recover braking energy and improve energy utilization efficiency.

[0042] The above-disclosed embodiments are merely preferred embodiments of the battery current regulation optimization method and system based on hybrid drive of the present invention. Of course, they should not be construed as limiting the scope of the present invention. Those skilled in the art can understand that all or part of the processes of the above embodiments can be implemented, and equivalent changes made in accordance with the claims of the present invention are still within the scope of the invention.

Claims

1. A battery current regulation optimization system based on hybrid drive, characterized in that, It includes a data acquisition and planning module, a data collection and filtering module, a processing and optimization module, an intelligent adjustment architecture construction module, and an adjustment strategy output module, wherein the data acquisition and planning module, the data collection and filtering module, the processing and optimization module, the intelligent adjustment architecture construction module, and the adjustment strategy output module are connected in sequence; The data acquisition planning module is used to select and deploy data acquisition terminals to form a data acquisition matrix, collect multi-dimensional information about the vehicle, and provide a data foundation for current regulation. The collection and filtering module is used to collect battery performance degradation cycle data as a benchmark, monitor the prototype vehicle's operating status, and filter data that meets the conditions to generate a set of original data sequences. The processing optimization module is used to process the raw data, remove noise, extract key features and patterns, optimize data quality, and generate high-quality data sequence clusters. The intelligent regulation architecture building module is used to select intelligent regulation algorithms, construct a composite regulation architecture including parallel regulation sub-units and a central coordination main unit, and generate a comprehensive battery current regulation strategy. The adjustment strategy output module is used to collect real-time operating data, preprocess it, input it into the adjustment architecture for calculation, output the adjustment strategy, and send it to the vehicle battery management system.

2. The battery current regulation and optimization system based on hybrid drive as described in claim 1, characterized in that, The data acquisition planning module includes a data acquisition terminal selection unit, a sensor layout planning unit, and a prototype vehicle deployment unit. The data acquisition terminal selection unit is used to select a data acquisition terminal that is compatible with the target hybrid drive vehicle based on the vehicle's key performance indicators and sensor performance parameters, covering multi-dimensional data acquisition needs such as vehicle operating status, battery parameters, and environmental information. The sensor layout planning unit is responsible for determining the placement of the data acquisition terminal on the prototype vehicle and obtaining the optimal layout scheme through simulation and actual road test verification. The prototype deployment unit deploys data acquisition terminals on multiple prototype vehicles of the same type to form a data acquisition matrix.

3. The battery current regulation and optimization system based on hybrid drive as described in claim 1, characterized in that, The collection and screening module includes a performance decay cycle data acquisition unit, an operation data monitoring unit, and a data comparison and screening unit. The performance degradation cycle data acquisition unit is used to acquire performance degradation cycle data of the target vehicle battery. The operation data monitoring unit is used to monitor the prototype vehicle's operating status in real time and acquire multi-dimensional operation data such as driving speed, acceleration, driving mileage, road slope, road surface smoothness, battery voltage, current, and temperature. The data comparison and filtering unit is used to compare the running data with the reference benchmark, and adopts a dynamic threshold adjustment algorithm to dynamically adjust the filtering threshold according to the real-time battery status and historical data, and filters out the data that meets the conditions and stores it in the initial screening database to generate the original data sequence set.

4. The battery current regulation and optimization system based on hybrid drive as described in claim 1, characterized in that, The processing optimization module includes a filtering and noise reduction unit, a feature extraction and pattern mining unit, and a data quality assessment and optimization unit. The filtering and noise reduction processing unit is used to remove noise and interference components from the original data sequence set and improve data purity; The feature extraction and pattern mining unit is used to mine key features and temporal correlation patterns in the data and extract information valuable for battery current regulation. The data quality assessment and optimization unit quantitatively evaluates the processed data based on a preset data quality scoring system, iterates and optimizes the data in a loop, continuously improves the data quality, and finally generates a high-quality data sequence cluster.

5. The battery current regulation and optimization system based on hybrid drive as described in claim 1, characterized in that, The intelligent adjustment architecture building module includes an algorithm selection unit, a parallel adjustment building unit, and a central coordination main building unit. The algorithm selection unit is used to select a suitable algorithm from the multi-intelligence regulation algorithm library to provide an algorithmic basis for building the regulation architecture; The parallel adjustment building unit is used to build multiple parallel adjustment sub-units. Each sub-unit operates independently based on different algorithms and outputs its own battery current adjustment suggestions. The central coordination master unit is used to construct the central coordination master subunit. It is responsible for integrating the adjustment suggestions based on the real-time performance and historical data of the parallel subunits and using a dynamic weight allocation strategy to generate a comprehensive battery current adjustment strategy.

6. The battery current regulation and optimization system based on hybrid drive as described in claim 1, characterized in that, The adjustment strategy output module includes a real-time data acquisition unit, a preprocessing unit, a strategy calculation and output unit, and a strategy transmission and control unit. The real-time data acquisition unit is used to collect real-time operating data of the target vehicle, providing a real-time basis for outputting precise adjustment strategies; The preprocessing unit is used to standardize and normalize the collected real-time operating data to make it meet the input requirements of the composite adjustment architecture. The strategy calculation output unit is used to input the preprocessed data into the composite regulation architecture for calculation and output a precise battery current regulation strategy. The strategy sending and control unit is used to send the adjustment strategy to the vehicle's battery management system to achieve real-time control of the battery current.

7. A battery current regulation optimization method based on hybrid drive, used in the battery current regulation optimization system based on hybrid drive as described in any one of claims 1-6, characterized in that, Includes the following steps: By using the data acquisition planning module, data acquisition terminals are deployed on multiple prototype vehicles of the same type to form a data acquisition matrix and collect multi-dimensional data during vehicle operation. The battery performance degradation cycle data of the target vehicle is obtained by the collection and screening module and set as the data collection reference benchmark. The operating status of the prototype vehicle is monitored in real time, the operating data is compared with the reference benchmark, and the data that meets the conditions is filtered and stored in the initial screening database to generate the original data sequence set. The processing optimization module is used to perform noise reduction, feature extraction and optimization on the original data sequence set to generate a high-quality data sequence cluster. Select an appropriate algorithm from the multi-intelligence regulation algorithm library, construct a composite regulation architecture using the intelligent regulation architecture building module, import high-quality data sequence clusters for training and optimization, and complete the architecture construction. The system collects real-time operating data of the target vehicle, preprocesses it through the adjustment strategy output module, and then inputs it into the composite adjustment architecture. It calculates and outputs the battery current adjustment strategy and sends it to the vehicle battery management system to achieve real-time control of the battery current.

Images