Automobile brake energy recovery system integrated with ABS (Anti-lock Brake System)

Abstract

The invention relates to the technical field of brake energy recovery, and particularly discloses an ABS-integrated automobile brake energy recovery system, which comprises the following steps of: after a brake signal is triggered, acquiring and preprocessing wheel speed signal data of each tire to obtain a wheel speed data set, establishing a wheel speed fluctuation identification mechanism under a brake working condition, performing fluctuation type division on the wheel speed data set, and obtaining a wheel speed data set; if the fluctuation exists, based on the fluctuation type, layered self-adaptive adjustment is carried out through the regenerative braking torque, the regenerative braking torque output linearity difference is corrected, the corrected regenerative braking torque is obtained, a regenerative braking torque signal is transmitted to an ABS control system in real time, and dynamic cooperative matching is carried out; according to wheel speed fluctuation scenes under different braking conditions, a torque self-adaptive adjustment working condition library is established, and after a new braking signal is triggered, self-adaptive adjustment is carried out through the torque self-adaptive adjustment working condition library, so that the braking energy recovery efficiency is improved, and the safety of the braking process is guaranteed.

Description

Technical Field

[0001] This invention relates to the field of brake energy recovery technology, and more specifically to an automotive brake energy recovery system integrated with ABS. Background Technology

[0002] In the braking system of plug-in hybrid vehicles, the brake energy recovery technology integrating anti-lock braking system (ABS) is the core means to improve the vehicle's range and braking safety. In actual braking process, wheel speed signals are easily affected by factors such as sensor failure, sudden changes in road surface adhesion coefficient, uneven braking torque and tire grip failure, which not only reduces the efficiency of brake energy recovery, but also poses a safety hazard of insufficient braking stability.

[0003] In existing technologies, most plug-in hybrid vehicles use a single filtering algorithm to process wheel speed signals in their braking systems. This lacks differentiated torque adjustment strategies for different types of fluctuations, making it impossible to adapt to the dynamic demands under complex braking conditions. Furthermore, the matching of regenerative braking and ABS braking force often relies on a fixed ratio distribution, without combining wheel speed fluctuation characteristics with the optimal wheel slip ratio for dynamic optimization. Moreover, it lacks a working condition matching mechanism that can be quickly adapted. When facing complex braking scenarios, the braking force distribution response is lagging and lacks precision, making it difficult to achieve the optimal balance between energy recovery efficiency and braking safety. This fails to meet the technical requirements of plug-in hybrid vehicles for efficient and safe braking.

[0004] Therefore, the present invention provides an automotive braking energy recovery system integrating ABS. Summary of the Invention

[0005] The purpose of this invention is to provide an automotive braking energy recovery system with integrated ABS to solve the aforementioned background problems.

[0006] The objective of this invention can be achieved through the following technical solutions: An integrated ABS automotive brake energy recovery system includes: Identification and Classification Module: After the braking signal is triggered, the wheel speed signal data of each tire is acquired and preprocessed to obtain the wheel speed dataset. A wheel speed fluctuation identification mechanism under braking conditions is established, and the wheel speed dataset is classified into fluctuation types. Adjustment and correction module: If there are fluctuations in the wheel speed dataset, the regenerative braking torque is used for hierarchical adaptive adjustment based on the fluctuation type to correct the linearity difference of the regenerative braking torque output and obtain the corrected regenerative braking torque. Cooperative matching module: Based on the corrected regenerative braking torque, the regenerative braking torque signal is transmitted to the ABS control system in real time to perform dynamic cooperative matching between regenerative braking energy recovery and the ABS control system; Integrated construction modules: For wheel speed fluctuation identification, torque hierarchical correction, and braking force coordinated matching, a torque adaptive adjustment condition library is established for wheel speed fluctuation scenarios under different braking conditions. After a new braking signal is triggered, adaptive adjustment is performed through the torque adaptive adjustment condition library.

[0007] Furthermore, the wheel speed dataset is obtained as follows: After the braking system receives the braking signal triggered by the driver's operation, it uses the wheel speed sensors of each tire to collect the wheel speed signal data of each tire of the vehicle. The collected wheel speed signal data is preprocessed using a low-pass filtering algorithm and a linear interpolation method, and then standardized to obtain a standardized wheel speed dataset.

[0008] Furthermore, the wheel speed fluctuation identification mechanism is constructed as follows: Based on the wheel speed dataset, wheel speed fluctuation features are extracted using a sliding window. The duration and step size of the sliding window are set based on the sampling period of the ABS control system, and features are extracted for the wheel speed dataset within each sliding window. Based on the dynamic characteristics of different braking conditions, preset thresholds for judging wheel speed fluctuation under each braking condition are established. By comparing the core fluctuation features of the wheel speed dataset with the judgment threshold of the corresponding working conditions, wheel speed fluctuations are identified, and the wheel speed dataset is divided into normal wheel speed fluctuations and abnormal wheel speed fluctuations.

[0009] Furthermore, the classification method for fluctuation types is as follows: The extracted feature data are first normalized and then summed to obtain the wheel speed fluctuation determination coefficient. If the wheel speed fluctuation judgment coefficient is greater than or equal to the wheel speed fluctuation judgment threshold, the corresponding wheel speed dataset will be marked as abnormal wheel speed fluctuation. If the wheel speed fluctuation judgment coefficient is less than the wheel speed fluctuation judgment threshold, the corresponding wheel speed dataset is marked as normal wheel speed fluctuation. Based on the wheel speed dataset labeled as abnormal wheel speed fluctuations, the fluctuation types are classified. The fluctuation causes and dynamic characteristics are used as the classification criteria. Combined with the differences in braking conditions, a feature matching classification system for abnormal wheel speed fluctuations is constructed, which classifies abnormal wheel speed fluctuations into four typical types: sensor failure type fluctuations, road adhesion coefficient mutation type fluctuations, braking torque unevenness type fluctuations, and tire grip failure type fluctuations. We constructed feature maps for various types of abnormal fluctuations, matched the feature data in the wheel speed dataset with the feature maps of each type of fluctuation, and determined the fluctuation type of the wheel speed dataset.

[0010] Furthermore, the adjustment method of hierarchical adaptive adjustment is as follows: To address sensor-failure-induced fluctuations, a torque output stabilization strategy is adopted to reduce the adjustment amplitude and step size of the regenerative braking torque, fix the torque output slope, and avoid sudden torque changes caused by sensor signal jumps. To address the abrupt fluctuations in road surface adhesion coefficient, a stepped torque adjustment strategy is adopted. Based on the real-time changes in wheel speed fluctuation amplitude, the regenerative braking torque output value is dynamically adjusted, and the torque is increased or decreased step by step according to a preset step size. To address uneven fluctuations in braking torque, a wheel-end differentiated torque compensation strategy is adopted. Based on the wheel speed deviation of the tires on the same side or diagonally opposite sides, the regenerative braking torque of the corresponding wheel end is precisely compensated and adjusted. To address tire grip failure-related fluctuations, a torque limiting control strategy is employed to significantly reduce the regenerative braking torque output or temporarily cut off the regenerative braking torque output, thereby quickly switching the vehicle's braking force core to the mechanical braking system.

[0011] Furthermore, the method for obtaining the linearity difference in regenerative braking torque output is as follows: Obtain the actual linear output curves of regenerative braking torque under different fluctuation types, compare and analyze the actual linear output curves with the corresponding theoretical linear output curves, and extract the deviation data between the actual linear output curves and the theoretical linear output curves.

[0012] Furthermore, the revised method for obtaining regenerative braking torque is as follows: By using the least squares method to fit the deviation data to obtain a linear deviation compensation function for torque output that is adapted to the current braking condition and fluctuation type; Substitute the regenerative braking torque value after stratified adjustment into the compensation function to calculate the torque compensation amount. Then, compensate and correct the torque after stratified adjustment to obtain the corrected regenerative braking torque.

[0013] Furthermore, the dynamic coordination and matching method between regenerative braking energy recovery and the ABS control system is as follows: Using the corrected regenerative braking torque as the core input signal, data extraction is performed on the corrected regenerative braking torque signal; The extracted key information is fused with data collected by the system, such as wheel slip ratio, brake line pressure, and vehicle speed, to construct a basic dataset. Based on the current braking conditions and wheel speed fluctuation type, the ABS control system presets the braking force distribution ratio range between regenerative braking and mechanical braking. With the optimal wheel slip ratio as the core control index, the ABS control system extracts multi-dimensional parameters from the basic dataset and dynamically adjusts the braking force distribution ratio in real time.

[0014] Furthermore, the method for establishing the torque adaptive adjustment operating condition library is as follows: A three-level architecture of braking condition classification, wheel speed fluctuation scenario clustering, and adjustment strategy matching is adopted to construct a torque adaptive adjustment condition library; Level 1 is the main classification of braking conditions, which is divided into three main categories: emergency braking, conventional braking, and low-speed braking. The second level is the clustering of wheel speed fluctuation scenarios. Under each main category of braking conditions, based on the divided wheel speed fluctuation types and incorporating influencing factors such as road surface type, the clusters are classified into typical wheel speed fluctuation scenarios. Level 3 is for precise matching of adjustment strategies, which uniquely matches the corresponding regenerative braking torque adjustment scheme, torque output linearity correction parameters, and ABS braking force coordination matching strategy for each typical wheel speed fluctuation scenario, forming a standardized scenario-strategy matching system. Core feature data for each typical scenario is collected, and after standardization and normalization of the collected feature data, it is input into the three-level architecture to build a torque adaptive adjustment condition library.

[0015] Furthermore, the adaptive adjustment method is as follows: Each typical scenario includes braking conditions, fluctuation types, road surface types, etc., which are then integrated to construct a feature retrieval code. When a vehicle triggers a new braking signal, the wheel speed fluctuation recognition mechanism collects and extracts the core parameters and wheel speed fluctuation feature data of the current braking condition in real time, and generates a feature retrieval code to be matched. The feature retrieval code to be matched is compared with the feature retrieval codes of each typical scenario in the torque adaptive adjustment condition database. The K-nearest neighbor algorithm is used to calculate the feature similarity between the feature retrieval code to be matched and the feature retrieval codes of each typical scenario in the torque adaptive adjustment condition database. The typical scenario with the highest similarity is selected as the matching result. Based on the matching results, the corresponding regenerative braking torque hierarchical adjustment strategy, linearity correction parameters and ABS braking force coordination matching scheme are extracted from the torque adaptive adjustment condition library and directly used as the adjustment parameters for the current braking scenario.

[0016] The beneficial effects of this invention are as follows: 1. This invention acquires multi-dimensional wheel speed signals after triggering the braking signal, and obtains a standardized wheel speed dataset through filtering, interpolation, and standardization. It establishes a three-tiered wheel speed fluctuation recognition mechanism—feature extraction, threshold determination, and fluctuation identification—to identify normal and abnormal wheel speed fluctuations. Based on the fluctuation type and braking condition, it constructs a hierarchical adaptive adjustment system for regenerative braking torque. Differentiated torque adjustment strategies are adopted for different fluctuation types, and the torque output linearity is corrected by fitting a deviation compensation function using the least squares method. Ultimately, this provides accurate torque input for the coordinated matching of regenerative braking and the ABS system, effectively solving the problems of torque misjudgment and poor output linearity caused by wheel speed fluctuations, laying the foundation for improving braking energy recovery efficiency and ensuring braking safety.

[0017] 2. Based on the obtained corrected regenerative braking torque, this invention transmits the torque signal to the ABS control system in real time and integrates multi-dimensional vehicle operation data. It dynamically matches the braking force distribution ratio with the optimal wheel slip ratio as the core, simultaneously matches the braking pressure adjustment strategy, and dynamically optimizes the collaborative scheme based on the vehicle's real-time status. At the same time, it establishes a torque adaptive adjustment condition library and achieves rapid adaptation to typical scenarios through feature retrieval code similarity matching. This realizes dynamic collaboration between regenerative braking energy recovery and the ABS control system, improving adaptability and reliability to braking scenarios, further ensuring braking safety and improving energy recovery efficiency. Attached Figure Description

[0018] The invention will now be further described with reference to the accompanying drawings.

[0019] Figure 1 This is a flowchart of an integrated ABS automotive braking energy recovery system according to an embodiment of the present invention; Figure 2 This is a logic diagram of an integrated ABS automotive braking energy recovery system according to an embodiment of the present invention. Detailed Implementation

[0020] To make the technical means, creative features, objectives and effects of this invention easier to understand, the invention will be further described below in conjunction with specific embodiments.

[0021] Example 1 Please see Figure 1 - Figure 2 As shown in the embodiment of the present invention, an integrated ABS automotive brake energy recovery system addresses the problem in integrated ABS automotive brake energy recovery systems where wheel speed signal fluctuations lead to misjudgment of regenerative braking torque output and poor linearity of torque output, thereby interfering with the dynamic matching of ABS braking force. The invention identifies wheel speed fluctuation types and adjusts regenerative braking torque in layers to determine torque correction parameters that adapt to the fluctuation type and braking condition. It dynamically matches the braking force distribution ratio according to the optimal slip ratio and utilizes a torque adaptive adjustment condition library to achieve rapid scenario adaptation, ultimately improving brake energy recovery efficiency and ensuring braking safety. Specifically, it includes the following modules: Identification and Classification Module: After the braking signal is triggered, the wheel speed signal data of each tire is acquired and preprocessed to obtain the wheel speed dataset. A wheel speed fluctuation identification mechanism under braking conditions is established, and the wheel speed dataset is classified into fluctuation types. Specifically, after the braking system receives the braking signal triggered by the driver's operation (including foot brake signal, handbrake signal and emergency brake signal), it uses wheel speed sensors (such as Hall effect wheel speed sensors or electromagnetic wheel speed sensors) to collect wheel speed signal data of each tire of the vehicle. The collected parameters include the instantaneous wheel speed of each tire, the wheel speed change rate, the wheel speed difference between adjacent tires, and the duration of continuous wheel speed change. The collection frequency is consistent with the sampling frequency of the ABS control system. The collected wheel speed signal data is preprocessed using a low-pass filtering algorithm and a linear interpolation method, and then standardized to obtain a standardized wheel speed dataset. A wheel speed fluctuation identification mechanism under braking conditions is established. Based on the classification of braking conditions (emergency braking, normal braking, and low-speed braking), and combined with the dynamic characteristics of the wheel speed dataset, a three-level identification architecture of feature extraction, threshold determination, and fluctuation identification is constructed. Based on the wheel speed dataset, wheel speed fluctuation features are extracted using a sliding window. Taking the sampling period of the ABS control system as a basis, the duration and step size of the sliding window are set. Features are extracted for the wheel speed dataset within each sliding window, including wheel speed fluctuation amplitude, fluctuation frequency, wheel speed deviation (the difference in wheel speed between each tire), and wheel speed change rate fluctuation coefficient. Among them, wheel speed fluctuation amplitude reflects the maximum difference between the wheel speed and the average wheel speed, fluctuation frequency reflects the number of wheel speed fluctuations per unit time, wheel speed deviation reflects the synchronicity of the wheel speeds of each tire, and wheel speed change rate fluctuation coefficient reflects the stability of the wheel speed change rate. Based on the dynamic characteristics of different braking conditions, preset thresholds for judging wheel speed fluctuation under each braking condition are established. By comparing the core fluctuation features of the wheel speed dataset with the judgment threshold of the corresponding working conditions, wheel speed fluctuations are identified, and the wheel speed dataset is divided into normal wheel speed fluctuations and abnormal wheel speed fluctuations. The extracted feature data are first normalized and then summed to obtain the wheel speed fluctuation determination coefficient. Compare the wheel speed fluctuation judgment coefficient with the wheel speed fluctuation judgment threshold; If the wheel speed fluctuation judgment coefficient is greater than or equal to the wheel speed fluctuation judgment threshold, the corresponding wheel speed dataset will be marked as abnormal wheel speed fluctuation. If the wheel speed fluctuation judgment coefficient is less than the wheel speed fluctuation judgment threshold, the corresponding wheel speed dataset is marked as normal wheel speed fluctuation. Based on the wheel speed dataset labeled as abnormal wheel speed fluctuations, the fluctuation types are classified. The fluctuation causes and dynamic characteristics are used as the classification criteria. Combined with the differences in braking conditions, a feature matching classification system for abnormal wheel speed fluctuations is constructed, which classifies abnormal wheel speed fluctuations into four typical types: sensor failure type fluctuations, road adhesion coefficient mutation type fluctuations, braking torque unevenness type fluctuations, and tire grip failure type fluctuations. Build feature maps of various types of abnormal fluctuations, and mark the core feature threshold ranges corresponding to different types of fluctuations. For example, sensor failure type fluctuations correspond to irregular fluctuation frequency, wheel speed values ​​jump and wheel speed change rate fluctuation coefficient exceeds the normal range; road adhesion coefficient abrupt fluctuations correspond to a sudden increase in wheel speed fluctuation amplitude and the difference in wheel speed between adjacent tires exceeds the threshold in a short period of time; uneven braking torque type fluctuations correspond to continuous excessive wheel speed deviation of tires on the same side or diagonally and regular differences in fluctuation amplitude; tire grip failure type fluctuations correspond to wheel speed fluctuation amplitude reaching extreme values ​​and wheel speed continuous change duration exceeding the working condition set value. The feature data in the wheel speed dataset is matched with the feature maps of various types of fluctuations to determine the fluctuation type of the wheel speed dataset. It should be noted that the function of this module is to collect multi-dimensional wheel speed parameters through wheel speed sensors, and obtain a standardized wheel speed dataset through filtering, interpolation and standardization. The core features of wheel speed fluctuations are extracted using a sliding window. Combined with preset thresholds for different braking conditions, normal and abnormal fluctuations are determined. Then, based on the causes and dynamic characteristics of the fluctuations, abnormal fluctuations are classified into four typical types through a feature matching classification system. A feature map is built to achieve accurate matching of fluctuation types. By capturing the dynamic change pattern of wheel speed during braking, the effective identification of wheel speed fluctuations and accurate classification of abnormal types are achieved. This provides an accurate basis for subsequent hierarchical adaptive adjustment of regenerative braking torque based on fluctuation type and correction of torque output linearity issues. Adjustment and correction module: If there are fluctuations in the wheel speed dataset, the regenerative braking torque is used for hierarchical adaptive adjustment based on the fluctuation type to correct the linearity difference of the regenerative braking torque output and obtain the corrected regenerative braking torque. Specifically, based on the wheel speed fluctuation identification results, the type of wheel speed fluctuation and the corresponding braking condition are confirmed. Combining the core characteristics of different fluctuation types and the dynamic requirements of braking conditions, a two-dimensional regenerative braking torque hierarchical adaptive adjustment system of fluctuation type and braking condition is established. Different regenerative braking torque adjustment benchmark parameters are preset for different fluctuation types, including torque adjustment range, adjustment step size, torque output slope and adjustment response time. The benchmark parameter settings are adapted to the torque adjustment logic of the ABS control system. To address sensor-failure-induced fluctuations, a torque output stabilization strategy is adopted to reduce the adjustment amplitude and step size of the regenerative braking torque, fix the torque output slope, and avoid sudden torque changes caused by sensor signal jumps. To address the abrupt fluctuations in road surface adhesion coefficient, a stepped torque adjustment strategy is adopted. Based on the real-time changes in wheel speed fluctuation amplitude, the regenerative braking torque output value is dynamically adjusted, and the torque is increased or decreased step by step according to a preset step size. To address uneven fluctuations in braking torque, a wheel-end differentiated torque compensation strategy is adopted. Based on the wheel speed deviation of the tires on the same side or diagonally opposite sides, the regenerative braking torque of the corresponding wheel end is precisely compensated and adjusted. To address tire grip failure-related fluctuations, a torque limiting control strategy is used to significantly reduce the regenerative braking torque output or temporarily cut off the regenerative braking torque output, quickly switching the vehicle's braking force core to the mechanical braking system. At the same time, in conjunction with the anti-lock braking adjustment strategy of the ABS control system, the vehicle's braking stability is prioritized. Based on the adaptive adjustment of regenerative braking torque, the actual linear output curve of regenerative braking torque under different fluctuation types is obtained. The actual linear output curve is compared and analyzed with the corresponding theoretical linear output curve, and the deviation data between the actual linear output curve and the theoretical linear output curve is extracted. By using the least squares method to fit the deviation data to obtain a linear deviation compensation function for torque output that is adapted to the current braking condition and fluctuation type; Substitute the regenerative braking torque value after stratified adjustment into the compensation function to calculate the torque compensation amount. Then, compensate and correct the torque after stratified adjustment to obtain the corrected regenerative braking torque. It should be noted that the regenerative braking torque value is calculated collaboratively from the braking conditions, brake pedal travel and real-time vehicle driving parameters, and is adjusted by a hierarchical adaptive adjustment strategy corresponding to the fluctuation type to obtain the adjusted regenerative braking torque value. It should be noted that the purpose of this step is to construct a two-dimensional adjustment system based on the fluctuation type and braking condition. For four typical fluctuation types, such as sensor failure and sudden road surface changes, different strategies such as torque stabilization and step adjustment are adopted to achieve precise hierarchical control of regenerative braking torque. Combined with the least squares method to fit the deviation compensation function between the actual output curve and the theoretical curve, the torque after hierarchical adjustment is corrected a second time, so as to provide accurate torque input for subsequent coordinated control with the ABS system. Example 2 like Figure 1 - Figure 2 As shown in the embodiment of the present invention, an automotive braking energy recovery system integrating ABS includes the following modules: Cooperative matching module: Based on the corrected regenerative braking torque, the regenerative braking torque signal is transmitted to the ABS control system in real time to perform dynamic cooperative matching between regenerative braking energy recovery and the ABS control system; Specifically, the corrected regenerative braking torque is used as the core input signal. Data is extracted from the corrected regenerative braking torque signal to extract key information such as torque output value, torque adjustment level, torque change rate, and linear correction accuracy. The encoded torque signal is transmitted in real time to the core control unit (ECU) of the ABS control system via the vehicle CAN bus, and the transmission frequency is consistent with the sampling frequency of the ABS control system. The extracted key information is fused with data collected by the system, such as wheel slip ratio, brake line pressure, and vehicle speed, to construct a basic dataset. Based on the current braking conditions (emergency braking, conventional braking, low-speed braking) and wheel speed fluctuation type, the braking force distribution ratio range of regenerative braking and mechanical braking is preset. With the optimal wheel slip ratio as the core control index, the ABS control system extracts multi-dimensional parameters from the basic dataset and dynamically adjusts the braking force distribution ratio in real time. It should be noted that the braking force distribution ratio range is preset based on the results of real vehicle test calibration to ensure that it adapts to the actual needs of different braking conditions and wheel speed fluctuation scenarios. For example, when the corrected regenerative braking torque can meet the current braking demand, and the wheel slip ratio is in the optimal braking range with no risk of lock-up, the proportion of regenerative braking force is maximized to fully release the energy recovery potential. When the wheel speed fluctuation is severely abnormal (such as tire grip failure type fluctuation) or the wheel slip ratio is close to the lock-up threshold, the proportion of regenerative braking force is quickly reduced or even cleared to zero, and the mechanical braking system undertakes the main braking force to prioritize braking safety. In response to the dynamic changes in the regenerative braking torque after correction, the ABS control system matches the corresponding braking pressure adjustment strategy in real time. When the regenerative braking torque increases, the mechanical braking line pressure at the corresponding wheel end is reduced simultaneously. When the regenerative braking torque decreases, the mechanical braking line pressure at the corresponding wheel end is increased simultaneously. At the same time, based on the real-time driving status of the vehicle (road adhesion coefficient, vehicle speed), the braking force coordination matching strategy is dynamically optimized. The real-time estimated value of the road adhesion coefficient is introduced to adaptively correct the braking force distribution ratio and braking pressure adjustment parameters. It should be noted that the purpose of this step is to use the corrected regenerative braking torque as the core input, and introduce it into the ABS control system through data extraction and signal transmission. It is then integrated with the multi-dimensional data collected by the vehicle itself to construct a basic dataset. The braking force distribution ratio range is preset based on the braking conditions and wheel speed fluctuation types. The braking force distribution is dynamically adjusted with the optimal wheel slip ratio as an indicator. At the same time, the braking pressure adjustment strategy is matched in real time according to the changes in regenerative braking torque, and the collaborative matching strategy is dynamically optimized based on the real-time driving status of the vehicle. This achieves accurate, dynamic, and safe collaborative matching between regenerative braking energy recovery and the ABS control system, improving braking energy recovery efficiency and ensuring braking safety. Integrated construction modules: For wheel speed fluctuation identification, torque hierarchical correction, and braking force coordinated matching, a torque adaptive adjustment condition library is established for wheel speed fluctuation scenarios under different braking conditions. After a new braking signal is triggered, adaptive adjustment is performed through the torque adaptive adjustment condition library. Specifically, with wheel speed fluctuation identification method, regenerative braking torque hierarchical adaptive adjustment strategy, and dynamic collaborative matching mechanism between regenerative braking and ABS braking system as core technical support, and combined with typical wheel speed fluctuation scenarios under different vehicle braking conditions, a standardized, searchable, self-learning, and iterative torque adaptive adjustment condition library is constructed. Among them, a three-level architecture of braking condition classification, wheel speed fluctuation scenario clustering, and adjustment strategy matching is adopted to build a torque adaptive adjustment condition library; Level 1 is the main classification of braking conditions, which is divided into three main categories: emergency braking, conventional braking, and low-speed braking. The second level is wheel speed fluctuation scenario clustering. Under each braking condition main category, based on the divided wheel speed fluctuation type (normal wheel speed fluctuation type, sensor failure type, road adhesion coefficient sudden change type, braking torque uneven type, tire grip failure type), and incorporating influencing factors such as road surface type (dry hard road surface, wet and slippery road surface, icy and snowy road surface, soft road surface), it is clustered into typical wheel speed fluctuation scenarios. Level 3 is for precise matching of adjustment strategies. For each typical wheel speed fluctuation scenario, a unique regenerative braking torque adjustment scheme (adjustment level, adjustment range, adjustment rate), torque output linearity correction parameters (compensation function, PID adjustment parameters), and ABS braking force coordination matching strategy (braking force distribution ratio range, braking pressure adjustment standard) are matched to form a standardized scenario-strategy matching system. Core feature data for each typical scenario are collected, including braking condition parameters, wheel speed fluctuation characteristic indicators, optimal torque adjustment parameters, braking force coordination matching ratio, energy recovery efficiency, braking stability evaluation indicators, etc. After standardizing and normalizing the collected feature data, it is input into the three-level architecture to build a torque adaptive adjustment condition library. Each typical scenario includes braking conditions, fluctuation types, road surface types, etc., which are then integrated to construct a feature retrieval code. When a vehicle triggers a new braking signal, the wheel speed fluctuation recognition mechanism collects and extracts the core parameters and wheel speed fluctuation feature data of the current braking condition in real time, and generates a feature retrieval code to be matched. The feature retrieval code to be matched is compared with the feature retrieval codes of each typical scenario in the torque adaptive adjustment condition database. The K-nearest neighbor algorithm is used to calculate the feature similarity between the feature retrieval code to be matched and the feature retrieval codes of each typical scenario in the torque adaptive adjustment condition database. The typical scenario with the highest similarity is selected as the matching result. Based on the matching results, the corresponding regenerative braking torque hierarchical adjustment strategy, linearity correction parameters and ABS braking force coordination matching scheme are extracted from the torque adaptive adjustment condition library and directly used as the adjustment parameters for the current braking scenario. It should be noted that the purpose of this step is to construct a three-level architecture torque adaptive adjustment condition library, which integrates typical wheel speed fluctuation scenarios and corresponding adjustment strategies under different braking conditions, by constructing a braking condition classification, wheel speed fluctuation scenario clustering, and adjustment strategy matching. When the vehicle triggers a new braking signal, it can quickly and accurately match the characteristics of the current braking scenario, extract the optimal adjustment parameters, and realize adaptive adjustment of regenerative braking torque, linearity correction, and dynamic coordination and matching with the ABS braking system. This effectively improves braking energy recovery efficiency and ensures braking safety, further enhancing the adaptability and reliability of the system.

[0022] The technical solution of this invention is as follows: This invention acquires multi-dimensional wheel speed signals after the braking signal is triggered, and obtains a standardized wheel speed dataset through filtering, interpolation, and standardization. It establishes a three-tiered wheel speed fluctuation recognition mechanism—feature extraction, threshold determination, and fluctuation identification—to identify normal and abnormal wheel speed fluctuations. Based on the fluctuation type and braking condition, a hierarchical adaptive adjustment system for regenerative braking torque is constructed. Differentiated torque adjustment strategies are adopted for different fluctuation types, and the torque output linearity is corrected by fitting a deviation compensation function using the least squares method. Ultimately, this provides accurate torque input for the coordinated matching of regenerative braking and the ABS system, effectively solving the problems of torque misjudgment and poor output linearity caused by wheel speed fluctuations, and improving the efficiency of braking performance. This lays the foundation for improving regenerative braking efficiency and ensuring braking safety. Based on the obtained corrected regenerative braking torque, the torque signal is transmitted to the ABS control system in real time and integrated with multi-dimensional vehicle operation data. The braking force distribution ratio is dynamically matched with the optimal wheel slip ratio as the core, and the braking pressure adjustment strategy is matched synchronously. The collaborative scheme is dynamically optimized based on the real-time vehicle status. At the same time, a torque adaptive adjustment condition library is established, and rapid adaptation to typical scenarios is achieved through feature retrieval code similarity matching. This realizes the dynamic collaboration between regenerative braking energy recovery and the ABS control system, improves the adaptability and reliability of braking scenarios, further ensures braking safety and improves energy recovery efficiency.

[0023] The foregoing has provided a detailed description of one embodiment of the present invention, but this description is merely a preferred embodiment and should not be construed as limiting the scope of the invention. All equivalent variations and modifications made within the scope of the present invention should still fall within the scope of the present invention.

Claims

1. A vehicle braking energy recovery system integrating ABS, characterized in that: Includes the following modules: Identification and Classification Module: After the braking signal is triggered, the wheel speed signal data of each tire is acquired and preprocessed to obtain the wheel speed dataset. A wheel speed fluctuation identification mechanism under braking conditions is established, and the wheel speed dataset is classified into fluctuation types. Adjustment and correction module: If there are fluctuations in the wheel speed dataset, the regenerative braking torque is used for hierarchical adaptive adjustment based on the fluctuation type to correct the linearity difference of the regenerative braking torque output and obtain the corrected regenerative braking torque. Cooperative matching module: Based on the corrected regenerative braking torque, the regenerative braking torque signal is transmitted to the ABS control system in real time to perform dynamic cooperative matching between regenerative braking energy recovery and the ABS control system; Integrated construction modules: For wheel speed fluctuation identification, torque hierarchical correction, and braking force coordinated matching, a torque adaptive adjustment condition library is established for wheel speed fluctuation scenarios under different braking conditions. After a new braking signal is triggered, adaptive adjustment is performed through the torque adaptive adjustment condition library.

2. The automotive braking energy recovery system integrating ABS according to claim 1, characterized in that: The wheel speed dataset is obtained as follows: After the braking system receives the braking signal triggered by the driver's operation, it uses the wheel speed sensors of each tire to collect the wheel speed signal data of each tire of the vehicle. The collected wheel speed signal data is preprocessed using a low-pass filtering algorithm and a linear interpolation method, and then standardized to obtain a standardized wheel speed dataset.

3. The automotive braking energy recovery system integrating ABS according to claim 2, characterized in that: The wheel speed fluctuation recognition mechanism is constructed as follows: Based on the wheel speed dataset, wheel speed fluctuation features are extracted using a sliding window. The duration and step size of the sliding window are set based on the sampling period of the ABS control system, and features are extracted for the wheel speed dataset within each sliding window. Based on the dynamic characteristics of different braking conditions, preset thresholds for judging wheel speed fluctuation under each braking condition are established. By comparing the core fluctuation features of the wheel speed dataset with the judgment threshold of the corresponding working conditions, wheel speed fluctuations are identified, and the wheel speed dataset is divided into normal wheel speed fluctuations and abnormal wheel speed fluctuations.

4. The automotive braking energy recovery system integrating ABS according to claim 3, characterized in that: The classification method for wave types is as follows: The extracted feature data are first normalized and then summed to obtain the wheel speed fluctuation determination coefficient. If the wheel speed fluctuation judgment coefficient is greater than or equal to the wheel speed fluctuation judgment threshold, the corresponding wheel speed dataset will be marked as abnormal wheel speed fluctuation. If the wheel speed fluctuation judgment coefficient is less than the wheel speed fluctuation judgment threshold, the corresponding wheel speed dataset is marked as normal wheel speed fluctuation. Based on the wheel speed dataset labeled as abnormal wheel speed fluctuations, the fluctuation types are classified. The fluctuation causes and dynamic characteristics are used as the classification criteria. Combined with the differences in braking conditions, a feature matching classification system for abnormal wheel speed fluctuations is constructed, which classifies abnormal wheel speed fluctuations into four typical types: sensor failure type fluctuations, road adhesion coefficient mutation type fluctuations, braking torque unevenness type fluctuations, and tire grip failure type fluctuations. We constructed feature maps for various types of abnormal fluctuations, matched the feature data in the wheel speed dataset with the feature maps of each type of fluctuation, and determined the fluctuation type of the wheel speed dataset.

5. The automotive braking energy recovery system integrating ABS according to claim 1, characterized in that: The adjustment method of hierarchical adaptive adjustment is as follows: To address sensor-failure-induced fluctuations, a torque output stabilization strategy is adopted to reduce the adjustment amplitude and step size of the regenerative braking torque, fix the torque output slope, and avoid sudden torque changes caused by sensor signal jumps. To address the abrupt fluctuations in road surface adhesion coefficient, a stepped torque adjustment strategy is adopted. Based on the real-time changes in wheel speed fluctuation amplitude, the regenerative braking torque output value is dynamically adjusted, and the torque is increased or decreased step by step according to a preset step size. To address uneven fluctuations in braking torque, a wheel-end differentiated torque compensation strategy is adopted. Based on the wheel speed deviation of the tires on the same side or diagonally opposite sides, the regenerative braking torque of the corresponding wheel end is precisely compensated and adjusted. To address tire grip failure-related fluctuations, a torque limiting control strategy is employed to significantly reduce the regenerative braking torque output or temporarily cut off the regenerative braking torque output, thereby quickly switching the vehicle's braking force core to the mechanical braking system.

6. The automotive braking energy recovery system integrating ABS according to claim 5, characterized in that: The method for obtaining the linearity difference in regenerative braking torque output is as follows: Obtain the actual linear output curves of regenerative braking torque under different fluctuation types, compare and analyze the actual linear output curves with the corresponding theoretical linear output curves, and extract the deviation data between the actual linear output curves and the theoretical linear output curves.

7. The automotive braking energy recovery system integrating ABS according to claim 6, characterized in that: The revised method for obtaining regenerative braking torque is as follows: By using the least squares method to fit the deviation data to obtain a linear deviation compensation function for torque output that is adapted to the current braking condition and fluctuation type; Substitute the regenerative braking torque value after stratified adjustment into the compensation function to calculate the torque compensation amount. Then, compensate and correct the torque after stratified adjustment to obtain the corrected regenerative braking torque.

8. The automotive braking energy recovery system integrating ABS according to claim 7, characterized in that: The dynamic coordination and matching method between regenerative braking energy recovery and the ABS control system is as follows: Using the corrected regenerative braking torque as the core input signal, data extraction is performed on the corrected regenerative braking torque signal; The extracted key information is fused with data collected by the system, such as wheel slip ratio, brake line pressure, and vehicle speed, to construct a basic dataset. Based on the current braking conditions and wheel speed fluctuation type, the ABS control system presets the braking force distribution ratio range between regenerative braking and mechanical braking. With the optimal wheel slip ratio as the core control index, the ABS control system extracts multi-dimensional parameters from the basic dataset and dynamically adjusts the braking force distribution ratio in real time.

9. The automotive braking energy recovery system integrating ABS according to claim 1, characterized in that: The method for establishing the torque adaptive adjustment operating condition library is as follows: A three-level architecture of braking condition classification, wheel speed fluctuation scenario clustering, and adjustment strategy matching is adopted to construct a torque adaptive adjustment condition library; Level 1 is the main classification of braking conditions, which is divided into three main categories: emergency braking, conventional braking, and low-speed braking. The second level is the clustering of wheel speed fluctuation scenarios. Under each main category of braking conditions, based on the divided wheel speed fluctuation types and incorporating influencing factors such as road surface type, the clusters are classified into typical wheel speed fluctuation scenarios. Level 3 is for precise matching of adjustment strategies, which uniquely matches the corresponding regenerative braking torque adjustment scheme, torque output linearity correction parameters, and ABS braking force coordination matching strategy for each typical wheel speed fluctuation scenario, forming a standardized scenario-strategy matching system. Core feature data for each typical scenario is collected, and after standardization and normalization of the collected feature data, it is input into the three-level architecture to build a torque adaptive adjustment condition library.

10. The automotive braking energy recovery system integrating ABS according to claim 1, characterized in that: The adaptive adjustment method is as follows: Each typical scenario includes braking conditions, fluctuation types, road surface types, etc., which are then integrated to construct a feature retrieval code. When a vehicle triggers a new braking signal, the wheel speed fluctuation recognition mechanism collects and extracts the core parameters and wheel speed fluctuation feature data of the current braking condition in real time, and generates a feature retrieval code to be matched. The feature retrieval code to be matched is compared with the feature retrieval codes of each typical scenario in the torque adaptive adjustment condition database. The K-nearest neighbor algorithm is used to calculate the feature similarity between the feature retrieval code to be matched and the feature retrieval codes of each typical scenario in the torque adaptive adjustment condition database. The typical scenario with the highest similarity is selected as the matching result. Based on the matching results, the corresponding regenerative braking torque hierarchical adjustment strategy, linearity correction parameters and ABS braking force coordination matching scheme are extracted from the torque adaptive adjustment condition library and directly used as the adjustment parameters for the current braking scenario.

Images