Control method and system for automobile electric brake

Abstract

The invention relates to the technical field of electric automobiles, and discloses a control method and system for an electric brake of an automobile, vibration and electromagnetic signals during motor operation are collected through a sensor array, and multi-dimensional analysis is carried out on rigid coupling state characteristics. Real-time monitoring and prediction of the rigid coupling state are realized by using a signal processing technology and a machine learning model. And when the state deviates from the normal range, a braking torque fluctuation correction coefficient is generated according to a prediction result, and driving parameters are dynamically optimized in combination with latest state sensing data. Stable braking torque output is achieved by adjusting an output instruction of the motor control unit. According to the method, operation feedback data under complex working conditions are continuously collected, a machine learning model is continuously optimized, and the state sensing and predicting capacity is improved, so that long-term optimization of the motor braking stability is achieved, and the problem of braking torque fluctuation caused by dynamic changes of the rigid coupling state is effectively solved.

Description

Technical Field

[0001] The present invention relates to the technical field of brake systems, and in particular to a control method and system for an automobile electric brake. Background Art

[0002] In the development of modern transportation, electric vehicles, as an important carrier of green energy, are crucial for performance optimization and safety assurance.

[0003] Current mainstream braking control methods often struggle to adapt to dynamic environments when dealing with complex operating conditions. This is particularly true when faced with varying road conditions or sudden load changes, where control strategies lack responsiveness and accuracy. These methods rely heavily on preset parameters and lack the ability to detect and adjust to subtle changes in operation. This can lead to unstable braking performance and potentially even pose safety risks.

[0004] Therefore, in the field of braking systems, how to ensure the smoothness and reliability of the braking process is directly related to vehicle safety and user experience, and has become one of the key directions of industry research. Summary of the Invention

[0005] The present invention provides a control method and system for an automobile electric brake, so as to achieve stability and reliability of a braking process.

[0006] In a first aspect, in order to solve the above technical problems, the present invention provides a control method for an automobile electric brake, comprising: Acquire vibration signals and electromagnetic signals when the motor is running to obtain a first feature set, obtain a second feature set based on the first feature set, classify and predict the rigid coupling state based on the second feature set, and determine whether the current state deviates from a normal range; If so, a fluctuation correction coefficient of the braking torque is generated to obtain a torque adjustment reference value, and a driving parameter is determined based on the torque adjustment reference value and the state perception data obtained in real time; Obtaining a first braking torque output value according to the driving parameters, determining whether the driving parameters need to be further optimized according to the first braking torque output value, and if so, obtaining a second braking torque output value and determining whether it meets a preset control accuracy; If not, then an improved prediction capability is obtained, and the driving parameters are cyclically obtained according to the improved prediction capability to determine a stable control result.

[0007] Preferably, the step of acquiring the vibration signal and the electromagnetic signal when the motor is running to obtain a first feature set, and acquiring the second feature set based on the first feature set includes: synchronously collecting vibration signals and electromagnetic signals while the motor is running, and obtaining the first feature set after filtering; If the interference component of the first feature set is lower than a preset component threshold, extracting time-frequency features, obtaining at least one relevant feature parameter, and determining a feature combination associated with the rigid coupling state; The feature combination is classified, and if the classification result matches the preset coupling standard, the second feature set is determined.

[0008] Preferably, the classifying and predicting the rigid coupling state according to the second feature set and judging whether the current state deviates from a normal range includes: Comparing the parameters in the second feature set with preset parameter thresholds one by one to obtain abnormal parameters and deviation values; Obtaining a change trend in different time periods according to the deviation value, extracting a fluctuation feature from the change trend, and obtaining a distribution status of the fluctuation feature; If the distribution condition exceeds a preset distribution threshold, the fluctuation characteristics are compared with pre-established standards to determine whether it belongs to an abnormal category; If so, the abnormal category is continuously monitored to determine the deviation.

[0009] Preferably, if so, generating a braking torque fluctuation correction coefficient to obtain a torque adjustment reference value includes: Acquire at least one dynamic fluctuation data related to the braking torque according to the deviation, and acquire a deviation value of the fluctuation according to a preset fluctuation threshold; If the deviation value of the fluctuation exceeds a preset deviation threshold, the dynamic fluctuation data is quantified, key fluctuation parameters are extracted from the quantification results, and the correction requirements of the key fluctuation parameters are determined; Generate a corresponding fluctuation correction coefficient according to the correction requirement, compare the correction coefficient with pre-established adjustment basis data, and determine the applicable range of the fluctuation correction coefficient; The fluctuation correction coefficient is converted into the torque adjustment reference value through the applicable range.

[0010] Preferably, determining the driving parameters according to the torque adjustment reference value and the state perception data acquired in real time includes: Fusing the torque adjustment reference value with the state perception data, extracting key state fluctuation features from the fusion result to obtain a basic data set; Dynamically calculating the basic data set, and if the key state fluctuation characteristics exceed a preset state fluctuation threshold, preliminarily correcting the driving parameters to determine a preliminary optimized value range; Obtaining the constraint conditions related to the response speed through the preliminary optimization value range and performing secondary adjustments, determining whether the adjusted value meets the preset optimization standard, and if so, obtaining the final optimization value; The driving parameters are determined according to the final optimized values ​​and in combination with the configuration scheme.

[0011] Preferably, obtaining the first braking torque output value according to the driving parameter includes: Obtain current state data from motor commands and determine preliminary adjustment basis based on stable control requirements; The driving parameters are calibrated for the first time using the preliminary adjustment basis, and if the fluctuation constraint characteristic exceeds a preset characteristic threshold range, an optimized parameter range is obtained; Performing a second calibration on the motor command according to the optimized parameter range to determine whether it meets the standard; If so, the final torque output is verified according to the adjusted braking torque value and the fluctuation constraint feature to determine the first braking torque output value.

[0012] Preferably, the determining whether further driving parameters are required based on the first braking torque output value, and if so, obtaining the second braking torque output value and determining whether it meets a preset control accuracy, includes: acquiring real-time operating data, and preliminarily organizing the real-time operating data according to a braking torque target to obtain an organized feedback data set; Evaluate the control accuracy using the organized feedback data set, and if the evaluation result deviates from a preset evaluation threshold, obtain a parameter set to be optimized; In combination with the real-time operating data and the requirements for working condition adaptation, the parameter set to be optimized is dynamically modified to obtain an adjusted parameter mapping table; The second braking torque output value is obtained through the adjusted parameter mapping table and the braking torque target, and it is determined whether it meets the preset control accuracy.

[0013] Preferably, if yes, obtaining improved predictive capability includes: If not, the operating status is acquired in real time, and the operating status is classified and processed to obtain a sorted status data set; Comparing the sorted state data set with a preset state threshold, and if the deviation exceeds an allowable range, obtaining a training data set; The training dataset is integrated into a pre-established prediction module to obtain the improved prediction capability.

[0014] Preferably, the cyclically acquiring the driving parameters according to the improved predictive capability to determine a stable control result comprises: determining a combination of driving parameters that need to be adjusted based on the improved predictive capability; Adjust the driving parameter combination that needs to be adjusted according to the optimization target to obtain adjusted parameters; The adjusted parameters are integrated into a pre-established operation status database, and the operation status database is regularly updated according to the stable control results.

[0015] In a second aspect, the present invention provides a control system for an automotive electric brake, comprising: A detection end is used to obtain vibration signals and electromagnetic signals when the motor is running to obtain a first feature set, and obtain a second feature set based on the first feature set. According to the second feature set, the rigid coupling state is classified and predicted to determine whether the current state deviates from a normal range; a processing end, configured to, if so, generate a fluctuation correction coefficient for the braking torque, obtain a torque adjustment reference value, and determine a driving parameter based on the torque adjustment reference value and the state perception data acquired in real time; obtain a first braking torque output value based on the driving parameter, determine whether the driving parameter needs to be further optimized based on the first braking torque output value, and if so, obtain a second braking torque output value and determine whether it meets a preset control accuracy; The optimization end is used to obtain an improved prediction capability if no, and cyclically obtain the driving parameters according to the improved prediction capability to determine a stable control result.

[0016] Compared with the prior art, the present invention has the following beneficial effects: (1) The present invention obtains a first feature set by acquiring vibration signals and electromagnetic signals when the motor is running, and obtains a second feature set based on the first feature set. The rigid coupling state is classified and predicted based on the second feature set to determine whether the current state deviates from the normal range. This comparison method helps to identify potential risks in advance and improve the accuracy of the prediction.

[0017] (2) In the present invention, if yes, a fluctuation correction coefficient of the braking torque is generated to obtain a torque adjustment reference value. The driving parameters are determined based on the torque adjustment reference value and the state perception data obtained in real time, thereby ensuring the reliability of the configuration content and providing a solid guarantee for subsequent operation.

[0018] (3) The present invention obtains a first braking torque output value based on the driving parameters, and determines whether the driving parameters need to be further optimized based on the first braking torque output value. If so, a second braking torque output value is obtained and determined whether it meets the preset control accuracy. This multi-dimensional analysis can enhance the reliability of parameter adjustment and ensure the stability of braking torque output under complex working conditions.

[0019] (4) In the present invention, if not, an improved prediction capability is obtained, and the driving parameters are cyclically obtained based on the improved prediction capability to determine a stable control result. This regular update mechanism ensures continuous improvement of the control effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 This is a flow chart of a control method for an automobile electric brake provided by an embodiment of the present invention; Figure 2 This is a system structure diagram for controlling an automobile electric brake provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0021] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0022] Reference Figure 1 , Figure 1 This is a flow chart of a control method for an automobile electric brake provided by an embodiment of the present invention, comprising the following steps: S1, obtaining a vibration signal and an electromagnetic signal when the motor is running to obtain a first feature set, and obtaining a second feature set based on the first feature set, classifying and predicting the rigid coupling state based on the second feature set to determine whether the current state deviates from a normal range; S2, if yes, generating a fluctuation correction coefficient of the braking torque to obtain a torque adjustment reference value, and determining the driving parameters according to the torque adjustment reference value and the state perception data obtained in real time; S3, obtaining a first braking torque output value according to the driving parameters, determining whether the driving parameters need to be further optimized according to the first braking torque output value, and if so, obtaining a second braking torque output value and determining whether it meets a preset control accuracy; S4: If not, obtain the improved prediction capability, and cyclically obtain the driving parameters according to the improved prediction capability to determine a stable control result.

[0023] In step S1, the vibration signal and electromagnetic signal of the motor during operation are obtained to obtain a first feature set, and a second feature set is obtained based on the first feature set. The rigid coupling state is classified and predicted based on the second feature set to determine whether the current state deviates from the normal range.

[0024] In the electric vehicle (EV) sector, rigid coupling refers to a drive system in which power and torque are directly transmitted between the motor and wheels via a mechanical connection without elastic elements or slip. In this state, the rotation of the motor rotor and the rotation of the wheels are strictly synchronized, with no buffering or flexible adjustment capabilities.

[0025] Preferably, the step of acquiring the vibration signal and the electromagnetic signal when the motor is running to obtain a first feature set, and acquiring the second feature set based on the first feature set includes: synchronously collecting vibration signals and electromagnetic signals while the motor is running, and obtaining the first feature set after filtering; If the interference component of the first feature set is lower than a preset component threshold, extracting time-frequency features, obtaining at least one relevant feature parameter, and determining a feature combination associated with the rigid coupling state; The feature combination is classified, and if the classification result matches the preset coupling standard, the second feature set is determined.

[0026] In the electric vehicle field, disturbance components and related characteristic parameters are key concepts for analyzing and optimizing system performance, especially in motor control, battery management, and vehicle dynamics. Disturbance components refer to undesired signals or disturbances that affect the normal operation of an electric vehicle system. Relevant characteristic parameters are key variables used to quantify system status or performance.

[0027] For example, when using a sensor array to synchronously collect vibration and electromagnetic signal data during motor operation, a specific industrial scenario can be imagined: during the operation of a large industrial motor, an acceleration sensor installed near the motor bearing collects vibration signals, while an electromagnetic sensor collects electromagnetic signals near the motor windings. These signals are named first vibration data and first electromagnetic data and stored. The acquisition of a preliminary signal feature set can be achieved by recording the amplitude changes of the vibration signal and the intensity fluctuations of the electromagnetic signal, such as when the peak value of the vibration signal reaches 2.5 mm / s and the intensity fluctuation of the electromagnetic signal is within 0.3 Tesla.

[0028] For signal denoising, a wavelet transform algorithm can be used to process the first vibration data and the first electromagnetic data, removing high-frequency noise interference and generating the second vibration data and the second electromagnetic data. Assuming that the noise component of the second vibration data is reduced to 0.1 mm / s after denoising, which is below the preset threshold of 0.2 mm / s, the data is deemed suitable for subsequent processing. This denoising process improves signal purity, providing a reliable foundation for subsequent feature extraction.

[0029] During the feature extraction phase, the root mean square value and peak factor of the secondary vibration data can be extracted through time domain analysis, while the primary frequency component and harmonic components can be extracted through frequency domain analysis. For the secondary electromagnetic data, the frequency distribution and amplitude change rate of the electromagnetic signal can be extracted. Given the dynamic variability of the rigid coupling state, it is assumed that the extracted relevant feature parameters include the primary frequency of the vibration signal being 50 Hz, which is related to the motor speed, and the harmonic components of the electromagnetic signal being related to the tightness of the coupling state. This combination of features can reflect the stability of the coupling state; for example, a primary frequency deviation may indicate a loose coupling.

[0030] If the classification results show that a feature combination matches the preset coupling state criteria by at least 90%, then the feature combination is considered capable of characterizing the dynamic characteristics of the rigid coupling state, ultimately forming a feature dataset. This classification process offers the advantage of quickly identifying whether the coupling state is normal, providing data support for subsequent fault diagnosis and maintenance.

[0031] For example, analyzing the dynamic variability of rigid coupling from multiple dimensions can combine the time and frequency domain characteristics of the vibration and electromagnetic signals to verify each other. For example, when the vibration signal shows an abnormal peak, the electromagnetic signal also shows harmonic anomalies. Combining these two can more accurately determine whether there is a problem with the coupling state. The advantage of this multi-dimensional analysis is that it improves the accuracy and reliability of diagnosis and avoids misjudgment based on a single signal.

[0032] Automated scripts can be set up to regularly update data to ensure real-time signal characteristics. This approach effectively supports long-term monitoring, promptly identifying potential coupling status issues and extending motor life.

[0033] Preferably, the classifying and predicting the rigid coupling state according to the second feature set and judging whether the current state deviates from a normal range includes: Comparing the parameters in the second feature set with preset parameter thresholds one by one to obtain abnormal parameters and deviation values; Obtaining a change trend in different time periods according to the deviation value, extracting a fluctuation feature from the change trend, and obtaining a distribution status of the fluctuation feature; If the distribution condition exceeds a preset distribution threshold, the fluctuation characteristics are compared with pre-established standards to determine whether it belongs to an abnormal category; If so, the abnormal category is continuously monitored to determine the deviation.

[0034] For example, consider an industrial motor operating scenario. For monitoring the rigid coupling state, assume the parameter set includes the dominant frequency of the vibration signal and the rate of change of the electromagnetic signal's amplitude. Comparing these parameters against preset thresholds reveals that the normal range for the dominant frequency of the vibration signal is 55-65 Hz, while the actual detected value reaches 70 Hz, significantly exceeding the range and being identified as an outlier. The degree of deviation is assessed by calculating the difference in the values ​​of the excess, and the deviation reaches 5 Hz, which is considered a high deviation.

[0035] For example, in the time series analysis of outliers, we track the changing trends of the outliers in the main frequency over different time periods. Suppose that over three consecutive hours of monitoring, the main frequency gradually increases from 70 Hz to 72 Hz, showing a trend of continuous deviation. Extracting fluctuation characteristics from this data reveals that the fluctuation frequency exhibits irregular jumps, with the distribution exceeding the preset stable range. This distribution indicates potential instability in the coupling state, providing important evidence for subsequent analysis.

[0036] For example, if a fluctuation characteristic exceeds a threshold, classification is used to categorize the characteristic. Assuming the fluctuation characteristic is classified as normal, slightly abnormal, and severely abnormal, combined with pre-established criteria, if the main frequency continuously deviates and fluctuates irregularly, it is classified as severely abnormal. This classification helps quickly identify the severity of the problem and provides guidance for subsequent handling.

[0037] Continuous monitoring of abnormalities collects real-time data on severe abnormal conditions. For example, key indicators extracted from the monitoring data are the deviation of the vibration signal's main frequency and the number of sudden amplitude changes in the electromagnetic signal. The main frequency deviation is found to be stable above 5 Hz, and the number of sudden amplitude changes exceeds 10 per hour, far exceeding the normal range. Based on these indicators, the current deviation is determined to be at a high risk level. This monitoring method can promptly identify deteriorating trends and provide data support for maintenance decisions.

[0038] When analyzing the deviation and fluctuation characteristics of outliers from multiple perspectives, it's helpful to combine the time and frequency domains. Time domain analysis reveals a periodic increase in the deviation of the main frequency, while frequency domain analysis reveals a significant increase in harmonic components. These two factors reinforce each other and point to the possibility of mechanical looseness in the coupling state. This multi-dimensional analysis can enhance the comprehensiveness of state identification.

[0039] When achieving state prediction, historical data can be used as an aid when comparing classified features with standards. For example, if similar combinations of frequency deviation and amplitude mutations in historical data often correspond to bearing wear issues, and the current state closely matches this, a similar risk may be predicted. This comparison method helps identify potential risks in advance and improves prediction accuracy.

[0040] In step S2, if yes, a fluctuation correction coefficient of the braking torque is generated to obtain a torque adjustment reference value, and a driving parameter is determined according to the torque adjustment reference value and the state perception data obtained in real time.

[0041] In electric vehicle (EV) braking control systems, the braking torque fluctuation correction factor and the torque adjustment reference value are key parameters for optimizing braking performance and ensuring safety and comfort. The braking torque fluctuation correction factor is a dynamic adjustment parameter used to quantify and compensate for random fluctuations in braking torque (such as motor torque pulsation, mechanical transmission backlash, road surface excitation, etc.). The torque adjustment reference value serves as the reference anchor point for the braking control system, representing the ideal braking torque under the current operating conditions (theoretical value in the absence of interference).

[0042] State awareness data and drive parameters are two key pieces of information used for real-time monitoring and decision-making, and system performance control, respectively. State awareness data (SAD) is vehicle dynamics information acquired by the system through sensors and in real time. It is used to accurately determine current operating conditions and provide input for control strategies. Drive parameters are adjustable variables that control the operation of the motor and transmission system and directly influence power output characteristics.

[0043] Preferably, if so, generating a braking torque fluctuation correction coefficient to obtain a torque adjustment reference value includes: Acquire at least one dynamic fluctuation data related to the braking torque according to the deviation, and acquire a deviation value of the fluctuation according to a preset fluctuation threshold; If the deviation value of the fluctuation exceeds a preset deviation threshold, the dynamic fluctuation data is quantified, key fluctuation parameters are extracted from the quantification results, and the correction requirements of the key fluctuation parameters are determined; Generate a corresponding fluctuation correction coefficient according to the correction requirement, compare the correction coefficient with pre-established adjustment basis data, and determine the applicable range of the fluctuation correction coefficient; The fluctuation correction coefficient is converted into the torque adjustment reference value through the applicable range.

[0044] Among them, in the braking control system of electric vehicles (EVs), braking torque, key fluctuation parameters and fluctuation correction coefficient are three closely related core concepts, which jointly affect the smoothness, safety and energy recovery efficiency of braking.

[0045] Braking torque is the rotational resistance applied to the wheels by the braking system (including friction and regenerative braking) to slow or stop the vehicle. Key Fluctuation Parameters (CFPs) refer to the physical quantities that cause periodic or random fluctuations in braking torque during braking, directly impacting driving comfort and control accuracy. The Fluctuation Correction Coefficient (FCC) is a dynamic adjustment factor used to compensate for braking torque fluctuations, resulting in smoother output.

[0046] For example, when analyzing dynamic fluctuation data related to braking torque, one can start with data acquisition and characteristic analysis. Braking torque is a critical parameter in industrial equipment operation, and its fluctuations often reflect system stability. Imagine, for example, an industrial motor operating scenario where sensors collect dynamic fluctuation data on braking torque in real time and discover that the fluctuation amplitude exhibits irregular changes over short periods of time. This characteristic may indicate potential operational issues.

[0047] For example, a data comparison scenario can be used to match the characteristics of dynamic fluctuations with a preset threshold range. Suppose the preset braking torque fluctuation threshold range is plus or minus 2 Nm, but the actual collected fluctuation data reaches plus or minus 3.5 Nm, clearly exceeding the range. Calculating the deviation value reveals a value of 1.5 mm, indicating abnormal fluctuations that require further attention. This comparison method helps quickly locate problematic parameters.

[0048] For example, when fluctuation data exceeds a threshold, using fluctuation analysis for quantitative processing can focus on the frequency and amplitude of the fluctuations. It was found that the frequency of the fluctuations experienced five irregular jumps per minute, and the amplitude remained consistently high. The key fluctuation parameters extracted were the number of frequency jumps and the peak amplitude. Based on this, it was determined that the correction required was high, requiring timely adjustments to prevent further system instability.

[0049] To determine the fluctuation correction factor and its applicable range, an initial correction factor (assuming it's 0.85) is generated. This factor is then compared with pre-established adjustment data to determine if it's applicable within the current motor load range. This comparison ensures the correction factor's relevance and reliability.

[0050] For example, when converting a correction factor into a torque adjustment reference value, this step can be accomplished using numerical conversion. Assuming that a correction factor of 0.85 corresponds to a reference value of 2.2 Nm, this reference value serves as the basis for subsequent adjustments. This conversion method provides a clear reference standard for torque adjustment.

[0051] For example, when adjusting braking torque fluctuations, extracting specific adjustment parameters from a baseline value can determine the adjustment method as a gradual reduction in torque output. For example, based on a baseline value of 2.2 N·m, the adjustment parameter is set to decrease by 0.2 N·m per hour until the fluctuation returns to a normal range. This application effectively smooths fluctuations and ensures stable equipment operation.

[0052] Analyzing the entire process of fluctuation adjustments from multiple perspectives can be supported by combining historical data with real-time monitoring data. For example, if similar fluctuations in historical data often indicate overload issues, and current real-time data also indicates excessive load, the combination of these two confirms the need for adjustment. This multi-dimensional analysis enhances comprehensive decision-making and helps reduce the risk of misjudgment.

[0053] Preferably, determining the driving parameters according to the torque adjustment reference value and the state perception data acquired in real time includes: Fusing the torque adjustment reference value with the state perception data, extracting key state fluctuation features from the fusion result to obtain a basic data set; Dynamically calculating the basic data set, and if the key state fluctuation characteristics exceed a preset state fluctuation threshold, preliminarily correcting the driving parameters to determine a preliminary optimized value range; Obtaining the constraint conditions related to the response speed through the preliminary optimization value range and performing secondary adjustments, determining whether the adjusted value meets the preset optimization standard, and if so, obtaining the final optimization value; The driving parameters are determined according to the final optimized values ​​and in combination with the configuration scheme.

[0054] In the electric vehicle (EV) sector, key state fluctuation characteristics refer to periodic or random variations in state quantities that affect vehicle performance, safety, or comfort. These characteristics typically manifest as signal noise, mechanical vibration, or control deviation. These characteristics are crucial for system optimization and fault diagnosis.

[0055] For example, when integrating torque baseline data with real-time monitoring status perception information, a starting point is data fusion. This unified processing of multi-source data ensures comprehensive information. For example, in an industrial motor operating scenario, the torque baseline data is 2.2 N·m, while real-time monitoring reveals a fluctuation range of plus or minus 3 N·m. Combining these two factors, the fluctuation characteristic can be extracted as an excess of 1 N·m. This fusion approach helps more accurately understand system status and provides a reliable basis for subsequent adjustments.

[0056] Combining dynamic calculations of the basic data set with changes in monitoring frequency allows analysis of fluctuation characteristics. For example, if the monitoring frequency is 10 times per minute and the state fluctuation characteristic exceeds the preset threshold range of ±2 N·m, reaching ±3 N·m, preliminary corrections to the drive parameters are made, with the initial optimized value range determined to be 1.8 to 2.0 N·m. This dynamic calculation method enables timely response to fluctuations and ensures real-time parameter adjustments.

[0057] Based on the initial optimization range, constraints related to response speed are obtained and secondary adjustments are performed. Assuming the response speed adjustment time does not exceed 5 seconds, and the initial optimization range is 1.8 to 2.0 Nm, after analysis, it is adjusted to 1.9 Nm to ensure compliance with the preset standard. This secondary adjustment method can balance the requirements of speed and accuracy, improving the rationality of parameter configuration.

[0058] For example, parameter conversion can be used to determine the final optimized value and translate it into the required configuration. For example, if the final optimized value is 1.9 Nm, combined with the stability requirements of the configuration, the conversion result extracts an applicable parameter of 1.85 Nm, which serves as the final drive parameter configuration. This conversion ensures that the parameters match the actual application scenario, contributing to smooth system operation.

[0059] Analyzing the entire process of state fluctuation feature extraction and parameter adjustment from multiple perspectives can be mutually supported by combining historical fluctuation data with current monitoring data. For example, if similar fluctuations exceeding the specified range in historical data are often associated with sudden load changes, and current monitoring also shows an increase in load, these two data points can confirm the necessity of adjustment. This multi-dimensional analysis approach can improve the comprehensiveness of decision-making and reduce the risk of bias in parameter configuration.

[0060] When determining parameter update requirements and finalizing the configuration, focus on the applicable range of the adjusted value. Assuming the adjusted value of 1.85 Nm passes verification and meets the stability requirements within the load range, this approach ensures the reliability of the configuration and provides a solid foundation for subsequent operation.

[0061] In step S3, a first braking torque output value is obtained according to the driving parameters, and it is determined whether the driving parameters need to be further optimized according to the first braking torque output value. If so, a second braking torque output value is obtained and it is determined whether it meets the preset control accuracy.

[0062] In the electric vehicle (EV) sector, braking torque output refers to the instantaneous braking torque actually applied to the wheels by the braking system (including regenerative braking and friction braking). It is a core control parameter when the vehicle is decelerating or stopping. Essentially, it is the physical quantity ultimately executed by the system after calculation based on driver demand, vehicle status, and control strategies.

[0063] Preferably, obtaining the first braking torque output value according to the driving parameter includes: Obtain current state data from motor commands and determine preliminary adjustment basis based on stable control requirements; The driving parameters are calibrated for the first time using the preliminary adjustment basis, and if the fluctuation constraint characteristic exceeds a preset characteristic threshold range, an optimized parameter range is obtained; Performing a second calibration on the motor command according to the optimized parameter range to determine whether it meets the standard; If so, the final torque output is verified according to the adjusted braking torque value and the fluctuation constraint feature to determine the first braking torque output value.

[0064] In the electric vehicle (EV) sector, stability control requirements, fluctuation constraint characteristics, and torque output are the three core concepts of braking and drive systems. Together, they ensure vehicle safety, ride comfort, and energy efficiency. Stability control requirements refer to the control objectives for maintaining dynamic stability under complex operating conditions (such as low-adhesion roads and rapid acceleration and deceleration). Fluctuation constraint characteristics limit the non-ideal fluctuations in braking or driving torque, aiming to improve comfort and durability. Torque output refers to the effective torque actually applied to the wheels by the drive motor or braking system.

[0065] For example, when obtaining current state data from a motor command, one can first focus on the torque output information and operating status feedback contained in the command. Imagine an industrial motor control scenario where the command indicates a target torque of 2.5 Nm, while the current state data reflects an actual output of 2.3 Nm, indicating a certain deviation. Using data integration, the command data and feedback information can be processed uniformly, extracting a deviation feature of 0.2 mm as a preliminary basis for adjustment. This approach helps clarify the direction of adjustment.

[0066] When combining the initial adjustment criteria with dynamic monitoring results for correction, the drive parameters can be modified. For example, suppose dynamic monitoring shows a torque fluctuation range of plus or minus 0.3 N·m, while the preset threshold is plus or minus 0.2 N·m, exceeding the constraint. In this case, by analyzing the fluctuation characteristics, the drive parameters are adjusted from the initial value of 2.3 N·m to 2.4 N·m, forming an optimized parameter range. This adjustment method promptly responds to fluctuations and ensures that the parameters are closer to the target value.

[0067] For example, when performing secondary calibration based on the optimized parameter range, numerical conversion can be used to process the motor command. Assume the optimized range is 2.4 to 2.5 Nm, and the command optimization requires that the braking torque adjustment be updated synchronously with the state feedback. The command parameter is calibrated to 2.45 Nm, and the braking torque is adjusted based on real-time feedback to ensure that it meets the preset standard. This secondary calibration method can improve the match between the command and actual operation.

[0068] Data verification can be used to verify the final torque output and determine the control configuration. Assuming the adjusted braking torque is 2.45 N·m, combined with the real-time calibration constraints, the output fluctuation must not exceed plus or minus 0.1 N·m. Analysis of the current output data confirms that the fluctuation range is plus or minus 0.08 N·m, meeting the target requirement. The control configuration is ultimately determined to be 2.45 N·m. This verification method ensures the stability of the torque output.

[0069] For example, from integrating state feedback information to determining the final configuration, the rationality of the adjustment process can be analyzed from multiple perspectives. Suppose historical data indicates that similar torque deviations are often caused by load changes, and current feedback also shows a slight increase in load. By corroborating historical and real-time data, the necessity of adjusting to 2.45 N·m can be confirmed. Simultaneously, combined with analysis of fluctuation constraint characteristics, the adaptability of the adjusted parameters under different loads can be verified. This multi-dimensional analysis approach helps improve configuration reliability.

[0070] To achieve the braking torque target, we can focus on the synergy between dynamic monitoring and parameter correction. Assuming a monitoring frequency of eight times per minute, if fluctuations exceed the threshold, we can promptly adjust the parameters to ensure that the torque output approaches the target. This collaborative processing approach effectively addresses unexpected changes during operation and ensures effective control.

[0071] Preferably, the determining whether further driving parameters are required based on the first braking torque output value, and if so, obtaining the second braking torque output value and determining whether it meets a preset control accuracy, includes: Acquiring real-time operating data, and preliminarily organizing the real-time operating data according to a braking torque target to obtain an organized feedback data set; Evaluate the control accuracy using the collated feedback data set, and if the evaluation result deviates from a preset evaluation threshold, obtain a parameter set to be optimized; In combination with the real-time operating data and the requirements for working condition adaptation, the parameter set to be optimized is dynamically modified to obtain an adjusted parameter mapping table; The second braking torque output value is obtained through the adjusted parameter mapping table and the braking torque target, and it is determined whether it meets the preset control accuracy.

[0072] Among them, in the field of electric vehicles (EVs), the braking torque target value (Braking Torque Target) refers to the ideal braking torque value calculated by the vehicle control system based on the driver's braking request (such as pedal opening), current driving status and safety strategy. It is the benchmark instruction for the braking system to execute control.

[0073] For example, data collection is crucial for processing operational feedback under complex operating conditions. Consider an industrial motor control scenario where real-time data, such as speed, current, and temperature, needs to be acquired from multiple sensor nodes. During initial compilation, this data can be aligned by timestamp to form a feedback dataset containing multi-dimensional information. This approach facilitates rapid identification of anomalies during subsequent analysis. For example, a sudden drop in speed data within a certain time period may be associated with unstable braking torque output. The compiled dataset can intuitively reflect this correlation, providing a foundation for subsequent evaluation.

[0074] When evaluating control accuracy, data comparison can effectively identify deviations. For example, suppose the preset braking torque target is 2.6 Nm, but the real-time feedback dataset shows the actual output is 2.4 Nm, a deviation of 0.2 Nm, exceeding the preset threshold of 0.1 Nm. In this case, by comparing and analyzing the source of the deviation, it may be discovered that insufficient output is caused by a sudden load change, triggering the parameter adjustment process. This evaluation method can identify problems promptly and ensure the control system's responsiveness to deviations.

[0075] After determining the set of parameters to be optimized, it's crucial to dynamically adjust them based on real-time monitoring data. Suppose monitoring data indicates that the load under the current operating conditions has increased, and the braking torque output needs to be increased to 2.5 Nm. The original parameters can be mapped to the load variation characteristics to create an adjusted parameter mapping table. This approach allows for flexible parameter adjustments based on changing operating conditions, ensuring system adaptability. For example, the mapping table may contain optimal parameter values ​​under different loads, providing a reference for subsequent optimization.

[0076] For example, when processing status update data, the focus is on synchronously updating the braking torque output value. Suppose the adjusted parameter map indicates the output value should be 2.5 N·m, but the current status update data reflects the actual value as 2.45 N·m. Through conversion, the output command can be fine-tuned to 2.5 N·m, and compliance with control accuracy standards can be determined in real time, such as whether the fluctuation range is within 0.05 N·m. This synchronous processing ensures high consistency between the command and actual operation, improving system stability.

[0077] Analyzing the above process from multiple perspectives can further verify its rationality. Assume that historical operating data shows that the braking torque often needs to be increased by 0.1 to 0.2 Nm when the load increases, and current real-time monitoring also shows a similar trend. Through the mutual verification of historical and real-time data, the necessity of adjusting to 2.5 Nm is confirmed. At the same time, combined with the analysis of feedback characteristics under different operating conditions, the adaptability of the adjusted parameters is verified. This multi-dimensional analysis can enhance the reliability of parameter adjustment and ensure the stability of braking torque output under complex operating conditions.

[0078] In step S4 , if not, then the improved prediction capability is obtained, and the driving parameters are cyclically obtained according to the improved prediction capability to determine a stable control result.

[0079] Preferably, if not, obtaining improved predictive capabilities includes: If not, the operating status is acquired in real time, and the operating status is classified and processed to obtain a sorted status data set; Comparing the sorted state data set with a preset state threshold, and if the deviation exceeds an allowable range, obtaining a training data set; The training dataset is integrated into a pre-established prediction module to obtain the improved prediction capability.

[0080] For example, in an industrial motor control scenario, the topic of obtaining real-time operating status information from multiple sensor nodes can be analyzed from the perspective of comprehensive data collection. Sensor nodes may be located in different parts of the motor, monitoring key indicators such as speed, voltage, and vibration. For example, during a particular operation, the speed sensor displays a value of 1500 rpm, the voltage sensor displays a value of 220 volts, and the vibration sensor displays a value of 0.5 mm / s. This data is recorded with a unified timestamp to form a preliminary set of status information. This multi-dimensional collection method provides a comprehensive basis for subsequent analysis.

[0081] For example, when control accuracy falls below the preset threshold, data organization and classification of operating status information can be discussed from a data structuring perspective. For example, suppose the preset speed threshold is 1550 rpm, but the actual speed data collected is 1500 rpm, significantly below the standard. Grouping information such as speed, voltage, and vibration by category creates a clear status data set, facilitating subsequent comparison and analysis. This classification process helps quickly identify problematic indicators.

[0082] In the step of matching the collated status data set with the preset threshold, we can use the perspective of deviation identification as an example. For example, suppose the vibration index's tolerance is 0.2 mm / s, but the actual data is 0.5 mm / s, exceeding the range. This anomaly is flagged, and a deviation report is generated, triggering the data update process. This approach allows for the timely identification of potential problems and provides guidance for subsequent adjustments.

[0083] The integration of the training dataset into the prediction module can be analyzed from the perspective of data integration practicality. Suppose the training dataset contains the motor's operating records under different loads over the past month. The data importer organizes the data by chronological order and operating condition type before entering the prediction module. This integration ensures that the prediction module can learn from historical data and improve its state awareness capabilities.

[0084] For example, feature extraction from a dataset through batch processing can be explained from the perspective of feature screening. Suppose the speed trend and vibration frequency in the dataset are extracted as key features. The relationship between these features and the motor's operating status is analyzed. For example, a decrease in speed may be related to an increase in load. This feature extraction method lays the foundation for subsequent adjustments to the prediction logic.

[0085] Dynamically adjusting the operating status prediction logic can be done from the perspective of parameter adaptation. For example, if the prediction module detects an increase in load, it will adjust the control parameters from their default values ​​to a configuration more suitable for the higher load. A new mapping table will be generated to ensure that the prediction logic matches the actual operating conditions. This dynamic adjustment method can improve the system's adaptability.

[0086] For example, during real-time monitoring, synchronously updating parameter configurations and determining whether they meet accuracy requirements can be implemented from the perspective of real-time response. Suppose the adjusted parameter configuration requires the speed to be controlled within 1550 rpm, and the real-time monitoring data is 1548 rpm, which meets the requirement. The synchronous update mechanism records this result and continuously monitors subsequent data to ensure stable accuracy. This real-time update method effectively ensures system operational reliability.

[0087] Preferably, the cyclically acquiring the driving parameters according to the improved predictive capability to determine a stable control result comprises: determining a combination of driving parameters that need to be adjusted based on the improved predictive capability; Adjust the driving parameter combination that needs to be adjusted according to the optimization target to obtain adjusted parameters; The adjusted parameters are integrated into a pre-established operation status database, and the operation status database is regularly updated according to the stable control results.

[0088] For example, in industrial motor control scenarios, real-time monitoring of rigid-coupling operating status can be achieved by acquiring data from multiple sensor nodes and conducting analysis based on state perception requirements. When processing operating status, data acquisition categorizes different types of data, such as temperature, torque, and speed, to form a state information set. This classification facilitates subsequent accurate comparison. For example, if torque data is 800 Nm during a given monitoring session, while other indicators such as speed of 1480 rpm and temperature of 45°C are correlated via timestamps, forming a multi-dimensional state information set that lays the foundation for subsequent analysis.

[0089] Matching the preset threshold can be explained from the perspective of deviation identification. Suppose the preset torque threshold is 850 Nm, with an allowable deviation range of plus or minus 20 Nm. However, the actual monitored value is 800 Nm, significantly below the standard range. This anomaly is flagged, a deviation report is generated, and the parameter configuration process is triggered. This approach allows for rapid identification of problematic indicators and provides a basis for subsequent adjustments.

[0090] When determining the drive parameter combination that requires adjustment, consider parameter screening. For example, if the deviation report reveals insufficient torque, possibly due to a low drive current setting, the system will filter the current parameters and associated control frequencies that require adjustment, creating a parameter combination. This screening process ensures targeted adjustments. For example, if the original current setting is 10 amps, it may need to be increased to 12 amps after adjustment to meet torque requirements.

[0091] For example, we can analyze this from the perspective of adaptability. Assuming the adjusted current parameter is 12 amps, the parameter mapping table will generate a new mapping relationship table based on the current load and operating environment, ensuring that the control logic matches the actual operating conditions. This dynamic update method can adapt to different operating scenarios. If the load increases, the mapping table will further fine-tune the frequency parameter to ensure braking stability.

[0092] An example of real-time performance can be used to illustrate the synchronous refresh of parameter configurations during real-time monitoring. For example, suppose the adjusted parameters require torque control between 830 and 870 Nm. The real-time monitoring data is 845 Nm, which meets expectations. The synchronous refresh mechanism records this result and continuously monitors subsequent data fluctuations to ensure stability. This approach enables timely response to operational changes.

[0093] Assuming the adjusted parameter data includes current of 12 amps and torque of 845 Nm, store it chronologically and by operating condition type. This integrated approach facilitates subsequent tracing and analysis, providing data support for long-term control effectiveness.

[0094] For example, when regularly updating data to determine whether continuous optimization is being met, consider long-term planning. For example, assuming three months of accumulated operating data, the system will regularly analyze fluctuations in indicators like torque and speed. If a metric deviates from expectations for a long period of time, a new round of parameter optimization will be triggered. This regular update mechanism ensures continuous improvement in control effectiveness.

[0095] In summary, the present invention discloses a torque motor control method for electric vehicle braking, which collects vibration and electromagnetic signals of the motor during operation through a sensor array, and performs multi-dimensional analysis on the characteristics of the rigid coupling state. Signal processing technology and machine learning models are used to achieve real-time monitoring and prediction of the rigid coupling state. When the state deviates from the normal range, a braking torque fluctuation correction coefficient is generated according to the prediction results, and the drive parameters are dynamically optimized in combination with the latest state perception data. Stable braking torque output is achieved by adjusting the output instructions of the motor control unit. The present invention also continuously collects operating feedback data under complex working conditions, continuously optimizes the machine learning model, and improves state perception and prediction capabilities, thereby achieving long-term optimization of motor braking stability and effectively solving the problem of braking torque fluctuation caused by dynamic changes in the rigid coupling state.

[0096] Reference Figure 2 , an embodiment of the present invention provides a system structure diagram for controlling an automobile electric brake, including: The detection terminal 201 is used to obtain vibration signals and electromagnetic signals when the motor is running to obtain a first feature set, and obtain a second feature set based on the first feature set. Based on the second feature set, the rigid coupling state is classified and predicted to determine whether the current state deviates from the normal range; The processing end 202 is configured to, if yes, generate a fluctuation correction coefficient for the braking torque, obtain a torque adjustment reference value, and determine a driving parameter based on the torque adjustment reference value and the state perception data acquired in real time; obtain a first braking torque output value based on the driving parameter, determine whether the driving parameter needs to be further optimized based on the first braking torque output value, and if yes, obtain a second braking torque output value and determine whether it meets a preset control accuracy; The optimization end 203 is configured to obtain an improved prediction capability if no, and cyclically obtain the driving parameters according to the improved prediction capability to determine a stable control result.

[0097] It should be noted that the control system for an automobile electric brake provided in an embodiment of the present invention is used to execute all the process steps of the control method for an automobile electric brake in the above embodiment. The working principles and beneficial effects of the two correspond one to one, and thus will not be described in detail.

[0098] The embodiment of the present invention further provides a terminal device. The terminal device includes: a processor, a memory, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps in the above-mentioned control method for an automobile electric brake are implemented, for example: Figure 1 Alternatively, the processor implements the functions of the modules / units in the above-mentioned system embodiments when executing the computer program.

[0099] Exemplarily, the computer program may be divided into one or more modules / units, which are stored in the memory and executed by the processor to implement the present invention. The one or more modules / units may be a series of computer program instruction segments capable of performing specific functions, and the instruction segments are used to describe the execution process of the computer program in the terminal device.

[0100] The terminal device may be a computing device such as a desktop computer, laptop, PDA, or smart tablet. The terminal device may include, but is not limited to, a processor and memory. Those skilled in the art will appreciate that the aforementioned components are merely examples of terminal devices and do not constitute a limitation of the terminal device. The terminal device may include more or fewer components than those described above, or a combination of certain components or different components. For example, the terminal device may also include input / output devices, network access devices, buses, and the like.

[0101] The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor. The processor is the control center of the terminal device and connects various parts of the entire terminal device using various interfaces and lines.

[0102] The memory can be used to store the computer programs and / or modules. The processor implements the various functions of the terminal device by running or executing the computer programs and / or modules stored in the memory and accessing the data stored in the memory. The memory may primarily include a program storage area and a data storage area. The program storage area may store an operating system and at least one application required for a function (such as a sound playback function or an image playback function); the data storage area may store data generated based on the use of the mobile phone (such as audio data, a phone book, etc.). Furthermore, the memory may include high-speed random access memory and non-volatile memory, such as a hard disk, internal memory, a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, at least one disk storage device, a flash memory device, or other volatile solid-state storage device.

[0103] If the module / unit integrated into the terminal device is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention can implement all or part of the process steps in the above-mentioned method embodiments by using a computer program to instruct the relevant hardware. The computer program can be stored in a computer-readable storage medium. When executed by a processor, the computer program can implement the steps of each of the above-mentioned method embodiments. The computer program includes computer program code, which can be in source code form, object code form, executable file, or some intermediate form. The computer-readable medium can include: any entity or system capable of carrying the computer program code, recording medium, USB flash drive, mobile hard drive, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal, and software distribution medium. It should be noted that the content of the computer-readable medium can be appropriately increased or decreased based on the requirements of legislation and patent practice in a jurisdiction. For example, in some jurisdictions, based on legislation and patent practice, computer-readable media does not include electric carrier signals and telecommunication signals.

[0104] It should be noted that the system embodiment described above is merely illustrative, wherein the units described as separate components may or may not be physically separate, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed across multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the present embodiment. In addition, in the drawings of the system embodiment provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, which can be specifically implemented as one or more communication buses or signal lines. A person of ordinary skill in the art can understand and implement the present invention without inventive work.

[0105] The specific embodiments described above further illustrate the objectives, technical solutions, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. In particular, it should be noted that any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included within the scope of protection of the present invention for those skilled in the art.

Claims

1. A control method for an automobile electric brake, characterized in that: The method comprises: Acquire vibration signals and electromagnetic signals when the motor is running to obtain a first feature set, obtain a second feature set based on the first feature set, classify and predict the rigid coupling state based on the second feature set, and determine whether the current state deviates from a normal range; If so, a fluctuation correction coefficient of the braking torque is generated to obtain a torque adjustment reference value, and a driving parameter is determined based on the torque adjustment reference value and the state perception data obtained in real time; Obtaining a first braking torque output value according to the driving parameters, determining whether the driving parameters need to be further optimized according to the first braking torque output value, and if so, obtaining a second braking torque output value and determining whether it meets a preset control accuracy; If not, then an improved prediction capability is obtained, and the driving parameters are cyclically obtained according to the improved prediction capability to determine a stable control result.

2. The control method for an automobile electric brake according to claim 1, characterized in that: The step of obtaining a vibration signal and an electromagnetic signal when the motor is running to obtain a first feature set, and obtaining a second feature set based on the first feature set, includes: synchronously collecting vibration signals and electromagnetic signals while the motor is running, and obtaining the first feature set after filtering; If the interference component of the first feature set is lower than a preset component threshold, extracting time-frequency features, obtaining at least one relevant feature parameter, and determining a feature combination associated with the rigid coupling state; The feature combination is classified, and if the classification result matches the preset coupling standard, the second feature set is determined.

3. The control method for an automobile electric brake according to claim 1, characterized in that: The classifying and predicting the rigid coupling state according to the second feature set and judging whether the current state deviates from a normal range includes: Comparing the parameters in the second feature set with preset parameter thresholds one by one to obtain abnormal parameters and deviation values; Obtaining a change trend in different time periods according to the deviation value, extracting a fluctuation feature from the change trend, and obtaining a distribution status of the fluctuation feature; If the distribution condition exceeds a preset distribution threshold, the fluctuation characteristics are compared with pre-established standards to determine whether it belongs to an abnormal category; If so, the abnormal category is continuously monitored to determine the deviation.

4. The control method for an automobile electric brake according to claim 3, characterized in that: If so, then generating a braking torque fluctuation correction coefficient to obtain a torque adjustment reference value includes: Acquire at least one dynamic fluctuation data related to the braking torque according to the deviation, and acquire a deviation value of the fluctuation according to a preset fluctuation threshold; If the deviation value of the fluctuation exceeds a preset deviation threshold, the dynamic fluctuation data is quantified, key fluctuation parameters are extracted from the quantification results, and the correction requirements of the key fluctuation parameters are determined; Generate a corresponding fluctuation correction coefficient according to the correction requirement, compare the correction coefficient with pre-established adjustment basis data, and determine the applicable range of the fluctuation correction coefficient; The fluctuation correction coefficient is converted into the torque adjustment reference value through the applicable range.

5. The control method for an automobile electric brake according to claim 1, characterized in that: The determining of the driving parameters according to the torque adjustment reference value and the state perception data acquired in real time includes: Fusing the torque adjustment reference value with the state perception data, extracting key state fluctuation features from the fusion result to obtain a basic data set; Dynamically calculating the basic data set, and if the key state fluctuation characteristics exceed a preset state fluctuation threshold, preliminarily correcting the driving parameters to determine a preliminary optimized value range; Obtaining the constraint conditions related to the response speed through the preliminary optimization value range and performing secondary adjustments, determining whether the adjusted value meets the preset optimization standard, and if so, obtaining the final optimization value; The driving parameters are determined according to the final optimized values ​​and in combination with the configuration scheme.

6. The control method for an automobile electric brake according to claim 1, characterized in that: The obtaining of the first braking torque output value according to the driving parameter includes: Obtain current state data from motor commands and determine preliminary adjustment basis based on stable control requirements; The driving parameters are calibrated for the first time using the preliminary adjustment basis, and if the fluctuation constraint characteristic exceeds a preset characteristic threshold range, an optimized parameter range is obtained; Performing a second calibration on the motor command according to the optimized parameter range to determine whether it meets the standard; If so, the final torque output is verified according to the adjusted braking torque value and the fluctuation constraint feature to determine the first braking torque output value.

7. The control method for an automobile electric brake according to claim 6, characterized in that: The determining whether further driving parameters are required based on the first braking torque output value, and if so, obtaining the second braking torque output value and determining whether it meets the preset control accuracy, includes: acquiring real-time operating data, and preliminarily organizing the real-time operating data according to a braking torque target to obtain an organized feedback data set; Evaluate the control accuracy using the organized feedback data set, and if the evaluation result deviates from a preset evaluation threshold, obtain a parameter set to be optimized; In combination with the real-time operating data and the requirements for working condition adaptation, the parameter set to be optimized is dynamically modified to obtain an adjusted parameter mapping table; The second braking torque output value is obtained through the adjusted parameter mapping table and the braking torque target, and it is determined whether it meets the preset control accuracy.

8. The control method for an automobile electric brake according to claim 1, characterized in that: If not, obtain improved predictive capabilities, including: If not, the operating status is acquired in real time, and the operating status is classified and processed to obtain a sorted status data set; Comparing the sorted state data set with a preset state threshold, and if the deviation exceeds an allowable range, obtaining a training data set; The training dataset is integrated into a pre-established prediction module to obtain the improved prediction capability.

9. The control method for an automobile electric brake according to claim 8, characterized in that: The step of cyclically acquiring the driving parameters according to the improved predictive capability to determine a stable control result includes: determining a combination of driving parameters that need to be adjusted based on the improved predictive capability; Adjust the driving parameter combination that needs to be adjusted according to the optimization target to obtain adjusted parameters; The adjusted parameters are integrated into a pre-established operation status database, and the operation status database is regularly updated according to the stable control results.

10. A control system for an automobile electric brake, characterized in that: include: a detection end, configured to obtain a vibration signal and an electromagnetic signal when the motor is running, obtain a first feature set, obtain a second feature set based on the first feature set, classify and predict the rigid coupling state based on the second feature set, and determine whether the current state deviates from a normal range; a processing end, configured to, if so, generate a fluctuation correction coefficient for the braking torque, obtain a torque adjustment reference value, and determine a driving parameter based on the torque adjustment reference value and the state perception data acquired in real time; obtain a first braking torque output value based on the driving parameter, determine whether the driving parameter needs to be further optimized based on the first braking torque output value, and if so, obtain a second braking torque output value and determine whether it meets a preset control accuracy; The optimization end is used to obtain an improved prediction capability if no, and cyclically obtain the driving parameters according to the improved prediction capability to determine a stable control result.

Images