A big data-based redundancy control method and system for brake-by-wire

Abstract

The application discloses a kind of redundancy control method and system based on line control brake, including obtaining the historical fault data of line control brake system and pre-processing as standard fault data after clustering analysis, extract fault component position, determine system weak position set;According to weak position set, construct initial fault propagation matrix, and calculate the system failure contribution degree of each fault component to update the matrix;Based on the updated matrix, use Monte Carlo simulation to form failure probability dataset;For the scenario with failure probability higher than the preset threshold, generate multiple candidate switching paths, and simulate the candidate switching paths through digital twin simulation to obtain the actual switching delay and success probability;According to the simulation results, calculate the comprehensive performance score of each path, and sort by score, when the operating load state of the fault component exceeds the preset threshold, select the path with the highest score for switching control.The application improves the redundancy control switching efficiency of the line control system under complex working conditions.

Description

Technical Field

[0001] This invention relates to the field of electronic control technology, and in particular to a redundant control method and system for linear braking based on big data. Background Technology

[0002] As the core of modern intelligent manufacturing, the reliability of industrial control systems directly affects production safety and efficiency. Braking-by-wire systems, a key module within industrial control systems and a core technology of electric vehicles, achieve precise braking control by replacing traditional mechanical connections with electronic signals, holding an irreplaceable position in the fields of autonomous driving and new energy vehicles.

[0003] Existing technology deploys a primary and backup dual controller, with the primary controller running in real time and the backup controller in standby mode. Sensors monitor parameters such as the load and communication latency of the primary controller. When a parameter exceeds a preset threshold, an alarm is triggered. For example, if the CPU load is greater than 90%, the system switches to the backup controller according to a preset path, using the fixed communication link.

[0004] However, static thresholds cannot adapt to dynamic operating conditions such as instantaneous load fluctuations, and are prone to false triggering or delayed switching; fixed paths do not consider real-time resource consumption, which may lead to extended switching time. Therefore, the existing technology has low efficiency in redundant control switching under complex operating conditions. Summary of the Invention

[0005] This invention provides a redundant control method and system for brake-by-wire based on big data, in order to solve the problem of low switching efficiency of redundant control under complex operating conditions in existing technologies.

[0006] Firstly, to address the aforementioned technical problems, this invention provides a redundant control method for line-controlled braking based on big data, comprising:

[0007] Historical fault data of the brake-by-wire system is acquired and preprocessed to obtain standard fault data;

[0008] Cluster analysis is performed on the standard fault data to extract the location of faulty components and determine the set of weak points in the system;

[0009] An initial fault propagation matrix is ​​constructed based on the set of weak points of the system. The system failure contribution of each faulty component is calculated and the initial fault propagation matrix is ​​updated to obtain the current fault propagation matrix.

[0010] Based on the current fault propagation matrix, the failure probability of each faulty component is calculated using Monte Carlo simulation to form a failure probability dataset.

[0011] Based on the failure probability dataset, multiple candidate switching paths are generated for each failure scenario of the faulty component. The candidate switching paths are simulated using digital twin simulation to obtain simulation results that include actual switching delay and actual success probability.

[0012] The overall performance score of each path is calculated based on the simulation results, and the candidate switching paths are sorted from high to low according to the overall performance score.

[0013] When the operating load of a faulty component exceeds the corresponding preset switching trigger threshold, the candidate switching path ranked first is selected for switching control.

[0014] Preferably, the step of acquiring historical fault data of the brake-by-wire system and preprocessing it to obtain standard fault data includes:

[0015] Historical fault data of the brake-by-wire system is obtained from sensor logs, controller records, and vehicle diagnostic systems.

[0016] The fault occurrence time, ambient temperature parameters, operating load status and fault triggering conditions are extracted from the historical fault data and integrated using a timestamp alignment method to generate initial fault data.

[0017] Linear interpolation was used to complete the missing fault occurrence time and ambient temperature parameters in the initial fault dataset, generating complete fault data.

[0018] The isolation forest algorithm is applied to the completed fault dataset to detect outliers in the fault occurrence time, the ambient temperature parameter, and the operating load status, and noise data exceeding the preset abnormal threshold is removed to generate denoised fault data.

[0019] The denoised fault dataset is sorted according to the fault occurrence time, and the ambient temperature parameter, the operating load status, and the fault triggering conditions are merged into a structured table to generate standard fault data.

[0020] Preferably, the step of performing cluster analysis on the standard fault data to extract the location of faulty components and determine the set of weak points in the system includes:

[0021] The frequency of failure and the scope of impact of each faulty component in the standard fault data are statistically analyzed.

[0022] Based on the fault impact range and combined with the preset impact range, cluster analysis is performed on all fault points to obtain multiple fault categories. The fault points with the highest fault occurrence frequency among all fault categories are extracted as the set of weak points of the system.

[0023] Preferably, the step of performing cluster analysis on all fault points based on the fault's impact range and in conjunction with a preset affected range to obtain multiple fault categories includes:

[0024] Set the minimum and step values ​​for the affected area;

[0025] Traverse all starting fault points and divide the fault points whose fault impact range is equal to the preset impact range into the same cluster;

[0026] The preset impact range is changed according to the step value, and all starting fault points are clustered again until all fault points with an impact range not less than the minimum preset impact range are clustered and divided to obtain multiple fault categories.

[0027] Preferably, the step of constructing an initial fault propagation matrix based on the set of weak points in the system, calculating the system failure contribution of each faulty component, updating the initial fault propagation matrix, and obtaining the current fault propagation matrix includes:

[0028] The fault points in the set of weak points are taken as the initial fault locations. The fault propagation path from the initial fault location to the final failure location is recursively analyzed according to the fault triggering conditions to obtain the propagation probability between each faulty component and construct the initial fault propagation matrix.

[0029] Calculate the system failure contribution of each fault point under different ambient temperature parameters and operating load conditions;

[0030] The initial fault propagation matrix is ​​updated based on the system failure contribution to obtain the current fault propagation matrix.

[0031] Preferably, the step of calculating the failure probability of each faulty component using Monte Carlo simulation based on the current fault propagation matrix to form a failure probability dataset includes:

[0032] Monte Carlo simulation is used to generate a sequence of random state parameters that include ambient temperature parameters and operating load conditions;

[0033] State parameters exceeding a preset state threshold are extracted from the random state parameter sequence, and the failure probability vector of each faulty component under different ambient temperature parameters and operating load conditions is determined by data sampling.

[0034] Based on the failure probability vector, the failure probability of each faulty component in the failure propagation path is calculated in conjunction with the current failure propagation matrix, and a failure probability dataset is generated.

[0035] Preferably, based on the failure probability dataset, multiple candidate switching paths are generated for each failure scenario of the faulty component, and digital twin simulation is used to simulate the candidate switching paths to obtain simulation results that include actual switching delay and actual success probability, including:

[0036] Extract failure scenarios with failure probabilities higher than a preset probability threshold from the failure probability dataset;

[0037] Based on preset switching delay and preset success probability, multiple candidate switching paths are generated for the failure scenario;

[0038] Based on the candidate switching paths, a digital twin simulation model is constructed, and the ambient temperature parameters and operating load status corresponding to each candidate path are loaded to simulate the switching process of each candidate path under complex operating conditions, so as to obtain simulation results including actual switching delay and actual success probability.

[0039] Preferably, the step of calculating the comprehensive performance score of each path based on the simulation results, and sorting the candidate switching paths from high to low according to the comprehensive performance score, includes:

[0040] Calculate the comprehensive performance score of each candidate handover path based on the actual handover latency and the actual success probability;

[0041] Candidate switching paths are sorted from high to low based on the comprehensive performance score.

[0042] Preferably, the formula for the comprehensive performance score is:

[0043]

[0044] in, A comprehensive performance score; This represents the actual probability of success. This represents the actual handover latency; The preset maximum switching delay; and These are preset weighting coefficients.

[0045] Secondly, the present invention provides a redundant control system for line-controlled braking based on big data, comprising:

[0046] The data acquisition and processing module is used to acquire historical fault data of the brake-by-wire system and preprocess it to obtain standard fault data.

[0047] The fault location extraction module is used to perform cluster analysis on the standard fault data, extract the location of the faulty component, and determine the set of weak points in the system.

[0048] The propagation matrix construction module is used to construct an initial fault propagation matrix based on the set of weak points of the system, calculate the system failure contribution of each faulty component and update the initial fault propagation matrix to obtain the current fault propagation matrix.

[0049] The failure probability calculation module is used to calculate the failure probability of each faulty component based on the current fault propagation matrix using Monte Carlo simulation, and form a failure probability dataset.

[0050] The switching path simulation module is used to generate multiple candidate switching paths for each failure scenario of the faulty component based on the failure probability dataset, and to simulate the candidate switching paths using digital twin simulation to obtain simulation results that include actual switching delay and actual success probability.

[0051] The path scoring calculation module is used to calculate the comprehensive performance score of each path based on the simulation results, and sort the candidate switching paths from high to low according to the comprehensive performance score;

[0052] The threshold-triggered switching module is used to select the first-ranked candidate switching path for switching control when the operating load of a faulty component exceeds the corresponding preset switching trigger threshold.

[0053] Thirdly, the present invention also provides an electronic device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor, when executing the computer program, implements the redundant control method for big data-based linear braking described in any one of the preceding claims.

[0054] Fourthly, the present invention also provides a computer-readable storage medium comprising a stored computer program, wherein, when the computer program is executed, it controls the device where the computer-readable storage medium is located to perform the redundant control method for big data-based linear braking described in any one of the above-described methods.

[0055] Compared with the prior art, the present invention has the following beneficial effects:

[0056] 1. This invention generates multiple candidate switching paths for each failure scenario of a faulty component and uses digital twin simulation to simulate these paths, obtaining simulation results that include actual switching delay and actual success probability. In this way, a comprehensive performance score for each path can be calculated based on the simulation results, and the candidate switching paths can be sorted from highest to lowest score. When a switch is required, the candidate switching path ranked first is selected for switching control. Unlike traditional fixed paths, this method fully considers the actual occupancy of current system resources (such as communication link bandwidth usage, controller processing capacity allocation, etc.) and selects the optimal path based on the performance indicators obtained from the simulation. For example, if at a certain moment the communication link on which the original fixed switching path depends is occupied by other high-priority tasks, leading to increased switching delay, this invention can select a backup path with less resource consumption and shorter switching delay based on the results of the digital twin simulation. This invention effectively avoids switching delay caused by fixed path resource occupancy issues, improving the efficiency and reliability of the switching process.

[0057] 2. This invention constructs a fault propagation matrix and uses Monte Carlo simulation to calculate the failure probability of each faulty component, forming a failure probability dataset. This means that the switching trigger condition is no longer a fixed, single threshold, but rather based on a comprehensive assessment of the failure probability of the faulty component. This method, based on big data analysis and simulation, can dynamically consider the changing failure probability of different components under various operating conditions in real time, thus effectively solving the problem of poor adaptability caused by static thresholds in existing technologies. For example, under different traffic conditions (such as long-term constant speed driving on highways and frequent start-stop operations on urban roads), the load conditions of the brake-by-wire system are different. This invention can flexibly trigger switching based on these dynamically changing load conditions and the failure probability data of the faulty components, avoiding false triggering or delayed switching due to fixed thresholds, and improving the accuracy and timeliness of switching control.

[0058] 3. This invention acquires historical fault data of a brake-by-wire system, preprocesses it to obtain standard fault data, and then performs cluster analysis on the standard fault data to extract the locations of faulty components and determine the set of weak points in the system. This process, through data mining and analysis techniques, deeply processes historical fault data, enabling precise location of components in the system prone to failure. Simultaneously, an initial fault propagation matrix is ​​constructed, and the system failure contribution of each faulty component is calculated to update the fault propagation matrix, resulting in the current fault propagation matrix. This helps to gain a deeper understanding of the scope and extent of the impact of faulty components on the entire system. Compared to existing technologies that may simply rely on partial sensor data to determine faults, the method of this invention is more systematic and comprehensive, mining commonalities from multiple fault samples to identify key weak points, providing a reliable basis for subsequent switching control strategies.

[0059] In summary, this invention can accurately locate the weak points of the system, comprehensively analyze the contribution of faulty components to system failure, optimize the redundancy control strategy of the entire brake-by-wire system, reduce the overall failure risk of the system, and improve the reliability and safety of the system. Attached Figure Description

[0060] Figure 1 This is a schematic flowchart of a redundant control method for line-controlled braking based on big data, provided in an embodiment of the present invention.

[0061] Figure 2 This is a schematic diagram of a redundant control system structure for big data-based linear braking provided in an embodiment of the present invention. Detailed Implementation

[0062] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0063] Reference Figure 1 The first embodiment of the present invention provides a schematic flowchart of a redundant control method for line-controlled braking based on big data, including the following steps:

[0064] S11: Obtain historical fault data of the brake-by-wire system and preprocess it to obtain standard fault data;

[0065] S12, perform cluster analysis on the standard fault data, extract the location of faulty components, and determine the set of weak points in the system;

[0066] S13, construct an initial fault propagation matrix based on the set of weak points of the system, calculate the system failure contribution of each faulty component and update the initial fault propagation matrix to obtain the current fault propagation matrix;

[0067] S14. Based on the current fault propagation matrix, the failure probability of each faulty component is calculated using Monte Carlo simulation to form a failure probability dataset.

[0068] S15. Based on the failure probability dataset, generate multiple candidate switching paths for each failure scenario of the faulty component, and simulate the candidate switching paths using digital twin simulation to obtain simulation results that include actual switching delay and actual success probability.

[0069] S16, Calculate the comprehensive performance score of each path based on the simulation results, and sort the candidate switching paths from high to low according to the comprehensive performance score;

[0070] S17, when the operating load of the faulty component exceeds the corresponding preset switching trigger threshold, the candidate switching path ranked first is selected for switching control.

[0071] In step S11, acquiring historical fault data of the brake-by-wire system and preprocessing it to obtain standard fault data includes:

[0072] Historical fault data of the brake-by-wire system is obtained from sensor logs, controller records, and vehicle diagnostic systems.

[0073] The fault occurrence time, ambient temperature parameters, operating load status and fault triggering conditions are extracted from the historical fault data and integrated using a timestamp alignment method to generate initial fault data.

[0074] Linear interpolation was used to complete the missing fault occurrence time and ambient temperature parameters in the initial fault dataset, generating complete fault data.

[0075] The isolation forest algorithm is applied to the completed fault dataset to detect outliers in the fault occurrence time, the ambient temperature parameter, and the operating load status, and noise data exceeding the preset abnormal threshold is removed to generate denoised fault data.

[0076] The denoised fault dataset is sorted according to the fault occurrence time, and the ambient temperature parameter, the operating load status, and the fault triggering conditions are merged into a structured table to generate standard fault data.

[0077] For example, historical fault data can be obtained from sensor logs, controller records, and the vehicle diagnostic system. The time of fault occurrence can be determined by comparing and analyzing the time of abnormal brake pressure changes recorded by the sensors, the execution command time recorded by the controller, and the fault codes returned by the diagnostic system. Then, the ambient temperature parameter corresponding to that time is obtained from the sensor records, the operating load status is obtained from the controller records, and the fault triggering conditions are obtained from the diagnostic system. Finally, these are integrated to form complete historical fault data containing the fault occurrence time, ambient temperature parameter, operating load status, and fault triggering conditions.

[0078] Preferably, timestamp alignment is used to integrate data from different sources along a unified timeline. This is achieved by extracting the timestamp corresponding to each data point in the historical fault data; aligning the data from different sources according to their timestamps; and integrating the aligned data into a complete initial fault data set. Specifically, the timestamp of the fault code that marks the formal record of the fault can be used as a benchmark to search for data in the sensor and controller records that has a time error within ±1 second.

[0079] For example, if a vehicle experiences brake pedal failure at 10:00:00 on March 1, 2025, the sensor shows an ambient temperature of 25°C, the controller records a load of 70%, and the diagnostic system marks the trigger condition as "insufficient hydraulic pump pressure," these are integrated into a single initial fault data set by aligning the timestamps.

[0080] In one possible implementation, assuming 1000 fault records are collected, 50 fault occurrence times are missing, and 80 ambient temperatures are missing, resulting in missing rates of 5% and 8%, respectively. If the preset threshold is 3%, data completion is required. Linear interpolation is used to complete the time and temperature records, calculating the missing values ​​based on the preceding and following time points. For example, if the temperatures at 10:00 and 10:02 are 25°C and 27°C, respectively, the missing temperature at 10:01 can be interpolated to 26°C.

[0081] Specifically, the Isolation Forest algorithm is used to detect outliers and is suitable for high-dimensional data analysis. The algorithm identifies points that deviate from the normal distribution by randomly splitting the data; for example, a record showing a fault time of 3:00 AM, a temperature of -50°C, and a load of 200% is clearly abnormal. Isolation Forest marks these as noise and removes them, making the dataset more accurate and reducing the risk of misjudgment.

[0082] Preferably, the generated structured table includes columns: failure time "2025-03-01 10:00:00", failure type "pedal failure", temperature "25°C", load "70%", and trigger condition "insufficient pressure". This format facilitates the analysis of failure trends.

[0083] In step S12, the cluster analysis of the standard fault data to extract the location of faulty components and determine the set of weak points in the system includes:

[0084] The frequency of failure and the scope of impact of each faulty component in the standard fault data are statistically analyzed.

[0085] Based on the fault impact range and combined with the preset impact range, cluster analysis is performed on all fault points to obtain multiple fault categories. The fault points with the highest fault occurrence frequency among all fault categories are extracted as the set of weak points of the system.

[0086] Preferably, the method of calculating the failure frequency of each faulty component can be achieved by extracting the component failure type field from all fault records, such as hydraulic pump failure or controller failure, grouping by component type, counting the number of failures for each component, and dividing the number of failures by the total running time of the statistical period to obtain the failure frequency.

[0087] The scope of impact of a fault refers to the number of other components affected by a faulty component. For example, if the frequency of a brake master cylinder failure is 5 times per hour, and the brake wheel cylinder and brake pedal also fail or experience performance degradation after the brake master cylinder fails, affecting a total of 2 components, then the scope of impact of the fault is 2.

[0088] It should be noted that cluster analysis is performed by comparing the consistency between the fault impact range of each fault point and the preset impact range to obtain multiple fault categories. For each fault category, the fault occurrence frequency of all fault points is compared, and the fault point with the highest fault occurrence frequency is selected and integrated into a set as the set of weak points of the system.

[0089] In one implementation, the step of performing cluster analysis on all fault points based on the fault impact range and in conjunction with a preset spillover range to obtain multiple fault categories, including:

[0090] Set the minimum and step values ​​for the affected area;

[0091] Traverse all starting fault points and divide the fault points whose fault impact range is equal to the preset impact range into the same cluster;

[0092] The preset impact range is changed according to the step value, and all starting fault points are clustered again until all fault points with an impact range not less than the minimum preset impact range are clustered and divided to obtain multiple fault categories.

[0093] The minimum value of the affected range can be set to 2, and the step value can be set to 1, or it can be set according to the analysis of historical fault data. The affected range of most fault points is concentrated between 2 and 5, so a step value of 1 can make the classification of fault points more refined.

[0094] Preferably, an empty cluster is created based on a preset impact range to store fault points within that impact range; each starting fault point is checked one by one to obtain its fault impact range, i.e., the number of impacts; the number of impacts of the current fault point is compared with the preset impact range, and if they are equal, the fault point is added to the cluster to form a fault category.

[0095] Preferably, after one traversal, the preset impact threshold is increased according to the preset step value, and the previous step is repeated until all fault points with an impact range of not less than 2 have been checked and classified, forming multiple fault categories.

[0096] By following the steps above, we can accurately locate the weak points in the system, provide a basis for taking targeted redundancy control measures, improve control effectiveness, and clearly classify faults. This facilitates the development of personalized redundancy control strategies for different types of faults, thereby enhancing system reliability.

[0097] In step S13, the process of constructing an initial fault propagation matrix based on the set of weak points in the system, calculating the system failure contribution of each faulty component, updating the initial fault propagation matrix, and obtaining the current fault propagation matrix includes:

[0098] The fault points in the set of weak points are taken as the initial fault locations. The fault propagation path from the initial fault location to the final failure location is recursively analyzed according to the fault triggering conditions to obtain the propagation probability between each faulty component and construct the initial fault propagation matrix.

[0099] Calculate the system failure contribution of each fault point under different ambient temperature parameters and operating load conditions;

[0100] The initial fault propagation matrix is ​​updated based on the system failure contribution to obtain the current fault propagation matrix.

[0101] It is worth noting that the construction of the initial fault propagation matrix can be achieved through the following steps: taking each fault point in the set of weak points as the initial fault location, analyzing its fault propagation path under different fault triggering conditions, and determining the probability and conditions of the fault propagating from the initial location to the next component based on historical fault data; starting from the initial fault location, recursively analyzing the possible propagation paths of the fault until the final failure location is determined; statistically determining the propagation probability between each pair of adjacent faulty components based on historical fault data; creating a matrix where the rows and columns of the matrix represent the various components of the system, and the matrix elements represent the probability of the fault propagating from the row component to the column component, and filling the matrix elements according to the propagation probability to obtain the initial fault propagation matrix.

[0102] The fault triggering conditions may include factors such as excessively high ambient temperature or excessive operating load, or other triggering conditions obtained from the analysis of historical fault data.

[0103] It should be noted that the system failure contribution can be calculated using the following formula;

[0104]

[0105] in, For temperature ,load Below, components The probability weight of a failure leading to the overall failure of the system; For temperature ,load Below, components Failure leads to component The probability of failure; Assess the severity of component failure on system functionality; This represents the number of fault points.

[0106] Preferably, ABS failure can be set. =1.0, sensor failure =0.3.

[0107] In the above embodiments, components are determined based on historical fault data and system analysis. Failure can cause other components to fail. Probability of failure This reflects the correlation between component failures, and the components The probability of a failure causing other components to fail is multiplied by the severity score of the corresponding component failure, and the summation is applied to all components to obtain the system failure contribution. This comprehensively considers the correlation between component failures and the severity of the failure consequences, and can fully reflect the impact of component failures on the system. By using operating parameters such as temperature and load, the calculation results are made to better fit actual operating conditions, which helps to identify key weak components under specific operating conditions.

[0108] For each element in the initial fault propagation matrix Multiply by the corresponding system failure contribution Update faults from components propagation to components The probability is used to obtain the current fault propagation matrix. This updated fault propagation matrix can more accurately reflect the propagation characteristics of faults within the system under actual operating conditions.

[0109] In step S14, the failure probability of each faulty component is calculated using Monte Carlo simulation based on the current fault propagation matrix to form a failure probability dataset, including:

[0110] Monte Carlo simulation is used to generate a sequence of random state parameters that include ambient temperature parameters and operating load conditions;

[0111] State parameters exceeding a preset state threshold are extracted from the random state parameter sequence, and the failure probability vector of each faulty component under different ambient temperature parameters and operating load conditions is determined by data sampling.

[0112] Based on the failure probability vector, the failure probability of each faulty component in the failure propagation path is calculated in conjunction with the current failure propagation matrix, and a failure probability dataset is generated.

[0113] In one possible implementation, the Monte Carlo simulation generates a sequence of random state parameters by setting a reasonable range of ambient temperature and operating load; for example, the reasonable temperature range of the system operating environment is defined as [-20, 80] degrees Celsius, and the operating load range is [0%, 100%].

[0114] It should be noted that the temperature distribution follows a normal distribution with an average temperature of 25 degrees Celsius and a standard deviation of 15 degrees Celsius; the load distribution follows a truncated normal distribution or a mixed distribution, for example, the load on urban roads is concentrated between 30% and 70%, and the load on highways is concentrated between 70% and 100%.

[0115] Set state thresholds for ambient temperature and operating load. For example, the normal range for ambient temperature is -10 to 70 degrees Celsius, and the normal range for operating load is 0% to 70%. Traverse the random state parameter sequence and filter out state parameters that exceed the preset normal range. For each state parameter that exceeds the threshold, sample the corresponding failure situation from historical fault data to form a failure probability vector corresponding to the state parameter and the failure situation.

[0116] The failure probability vector is multiplied by the fault propagation matrix to calculate the comprehensive failure probability of each faulty component in the fault propagation path. The calculated comprehensive failure probabilities of each faulty component are then compiled into a dataset to obtain a dataset containing the failure probability of each faulty component under different ambient temperatures and operating load conditions.

[0117] It is worth noting that industrial control systems operate under complex and diverse conditions, and the risk of component failure varies greatly under different environments. For example, in the high-temperature environment where electric vehicles operate for a long time, the accuracy of sensors is easily affected and may fail. This step, by analyzing state parameters that exceed the threshold, can accurately predict the risk of component failure under high-risk conditions, thereby improving the reliability and safety of the system.

[0118] In step S15, based on the failure probability dataset, multiple candidate switching paths are generated for each failure scenario of the faulty component. Digital twin simulation is then used to simulate these candidate switching paths, yielding simulation results that include actual switching delay and actual success probability, including:

[0119] Extract failure scenarios with failure probabilities higher than a preset probability threshold from the failure probability dataset;

[0120] Based on preset switching delay and preset success probability, multiple candidate switching paths are generated for the failure scenario;

[0121] Based on the candidate switching paths, a digital twin simulation model is constructed, and the ambient temperature parameters and operating load status corresponding to each candidate path are loaded to simulate the switching process of each candidate path under complex operating conditions, so as to obtain simulation results including actual switching delay and actual success probability.

[0122] This involves setting a failure probability threshold to distinguish between high-risk and low-risk failure scenarios. For example, setting the threshold to 0.1 means that scenarios with a failure probability higher than 0.1 are considered high-risk scenarios. The failure probability dataset is traversed, and the failure probability of each record is checked one by one. Failure scenarios with a failure probability higher than 0.1 are then selected.

[0123] It should be noted that the preset switchover latency and preset success probability are set as baseline values ​​based on system design requirements and historical data. For example, the preset switchover latency can be 50 milliseconds or 60 milliseconds or even shorter, and the preset success probability can be 90% or 95% or even higher. For each high-risk failure scenario, multiple possible switchover paths are generated based on different redundancy designs, each corresponding to a different backup component or a different switchover strategy. The theoretical expected switchover latency and success probability of the generated different paths are statistically analyzed.

[0124] Preferably, an accurate digital twin simulation model is constructed based on the actual design and parameters of the brake-by-wire system. This model includes all key components of the system, control logic, and fault propagation mechanisms. For each candidate switching path, the corresponding ambient temperature parameters and operating load status are obtained from the failure probability dataset and loaded into the simulation model. In the simulation model, the switching process of each candidate switching path under complex operating conditions is simulated, and the actual switching delay and success probability are recorded.

[0125] In step S16, the step of calculating the comprehensive performance score of each path based on the simulation results and sorting the candidate switching paths from high to low according to the comprehensive performance score includes:

[0126] Calculate the comprehensive performance score of each candidate handover path based on the actual handover latency and the actual success probability;

[0127] Candidate switching paths are sorted from high to low based on the comprehensive performance score.

[0128] Preferably, the comprehensive performance score of each candidate handover path calculated based on the actual handover delay and the actual success probability can be obtained by the following formula:

[0129]

[0130] in, A comprehensive performance score; This represents the actual probability of success. This represents the actual handover latency; The preset maximum switching delay; and These are preset weighting coefficients.

[0131] It should be noted that, and It can be configured according to needs. For example, if you need to switch to a path with a higher success rate, you can... Set it to 0.6 or 0.7 or even higher; for switching paths that require shorter switching times, you can set it to... Setting it to 0.4 or 0.5 is recommended, considering the actual application scenarios of brake-by-wire systems. The value is not less than 0.5.

[0132] This embodiment combines the success probability of switching and the latency, and uses weighted coefficients to comprehensively evaluate the performance of the switching path. For example, in the brake-by-wire system of an electric vehicle, it is necessary to ensure both the reliability of braking control (i.e., high success probability) and timely response (i.e., short latency). This comprehensive evaluation method of the formula meets practical needs and allows for flexible setting of weighted coefficients according to specific requirements, enabling the formula to adapt to different application scenarios and priority requirements.

[0133] Preferably, based on the calculated comprehensive performance score of each path, the paths are sorted from highest to lowest according to the comprehensive performance score. For candidate switching paths with the same score, a preset auxiliary sorting rule is used for further sorting, including prioritizing the path with the shortest switching latency or the lowest resource utilization.

[0134] In step S17, when the operating load state of the faulty component is detected to exceed the corresponding preset switching trigger threshold, the candidate switching path ranked first is selected for switching control.

[0135] The preset switching trigger threshold is set differently for different components. This requires a detailed analysis of each key component, including its function, failure mode, historical failure data, and operating environment; the basis for setting the threshold is determined based on the component's failure characteristics, system reliability requirements, and historical operating data; and different switching trigger thresholds are set for each component, which can be based on parameters such as operating load, temperature, and pressure.

[0136] In one possible implementation, the system monitors the operating status of each component in real time and compares it with a preset switching trigger threshold. When the operating status of a component exceeds the corresponding preset switching trigger threshold, the system triggers a switching mechanism and selects the candidate switching path ranked first for switching control.

[0137] For example, when the operating load of the brake pedal sensor exceeds 85%, signal delay or distortion may occur; when the brake master cylinder pressure exceeds 120 bar, it may indicate abnormal system pressure; when the brake wheel cylinder pressure exceeds 110 bar, it may indicate brake wheel cylinder failure; when the brake disc temperature exceeds 200°C, it may cause brake disc thermal fade. Therefore, the above data can be set as the preset switching trigger thresholds for the corresponding components.

[0138] In summary, this invention acquires historical fault data of a brake-by-wire system and preprocesses it into standard fault data. Then, it performs cluster analysis on this data to extract the locations of faulty components and determine the set of weak points in the system. Next, it constructs an initial fault propagation matrix based on the weak point set and calculates the system failure contribution of each faulty component to update the matrix. Based on the updated matrix, it uses Monte Carlo simulation to generate a failure probability dataset. Subsequently, it generates multiple candidate switching paths for scenarios where the failure probability exceeds a threshold, and simulates these paths using digital twin simulation to obtain the actual switching delay and success probability. Finally, it calculates the comprehensive performance score of each path based on the simulation results, sorts them by score, and selects the path with the highest score for switching control when the operating load of the faulty component exceeds a preset threshold. This process improves the efficiency of redundant control switching under complex operating conditions.

[0139] Reference Figure 2 The second embodiment of the present invention provides a redundant control system for wired braking based on big data, comprising:

[0140] The data acquisition and processing module is used to acquire historical fault data of the brake-by-wire system and preprocess it to obtain standard fault data.

[0141] The fault location extraction module is used to perform cluster analysis on the standard fault data, extract the location of the faulty component, and determine the set of weak points in the system.

[0142] The propagation matrix construction module is used to construct an initial fault propagation matrix based on the set of weak points of the system, calculate the system failure contribution of each faulty component and update the initial fault propagation matrix to obtain the current fault propagation matrix.

[0143] The failure probability calculation module is used to calculate the failure probability of each faulty component based on the current fault propagation matrix using Monte Carlo simulation, and form a failure probability dataset.

[0144] The switching path simulation module is used to generate multiple candidate switching paths for each failure scenario of the faulty component based on the failure probability dataset, and to simulate the candidate switching paths using digital twin simulation to obtain simulation results that include actual switching delay and actual success probability.

[0145] The path scoring calculation module is used to calculate the comprehensive performance score of each path based on the simulation results, and sort the candidate switching paths from high to low according to the comprehensive performance score;

[0146] The threshold-triggered switching module is used to select the first-ranked candidate switching path for switching control when the operating load of a faulty component exceeds the corresponding preset switching trigger threshold.

[0147] It should be noted that the redundant control system for brake-by-wire based on big data provided in this embodiment of the invention is used to execute all the process steps of the redundant control method for brake-by-wire based on big data in the above embodiment. The working principles and beneficial effects of the two are one-to-one, so they will not be described again.

[0148] This invention also provides an electronic device. The electronic device includes a processor, a memory, and a computer program stored in the memory and executable on the processor, such as a digital twin simulation program. When the processor executes the computer program, it implements the steps in the aforementioned embodiments of the redundant control method for linear braking based on big data, for example... Figure 1 The step S11 shown. Alternatively, when the processor executes the computer program, it implements the functions of each module / unit in the above-described device embodiments, such as the path scoring calculation module.

[0149] For example, the computer program may be divided into one or more modules / units, which are stored in the memory and executed by the processor to complete the present invention. The one or more modules / units may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of the computer program in the electronic device.

[0150] The electronic device may be a desktop computer, laptop, handheld computer, or smart tablet, etc. The electronic device may include, but is not limited to, a processor and memory. Those skilled in the art will understand that the above components are merely examples of electronic devices and do not constitute a limitation on the electronic device. It may include more or fewer components than described above, or combine certain components, or different components. For example, the electronic device may also include input / output devices, network access devices, buses, etc.

[0151] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the electronic device, connecting all parts of the electronic device via various interfaces and lines.

[0152] The memory can be used to store the computer programs and / or modules. The processor implements various functions of the electronic device by running or executing the computer programs and / or modules stored in the memory and by calling data stored in the memory. The memory may mainly include a program storage area and a data storage area. The program storage area may store the operating system, at least one application program required for a function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created according to the use of the mobile phone (such as audio data, phonebook, etc.). In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as hard disk, memory, plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0153] Wherein, if the modules / units integrated in the electronic device are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. It should be noted that the content included in the computer-readable medium can be appropriately added or removed according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electrical carrier signals and telecommunication signals.

[0154] It should be noted that the device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Furthermore, in the accompanying drawings of the device embodiments provided by this invention, the connection relationships between modules indicate that they have communication connections, which can be specifically implemented as one or more communication buses or signal lines. Those skilled in the art can understand and implement this without any creative effort.

[0155] The specific embodiments described above further illustrate the purpose, technical solution, 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 redundant control method for brake-by-wire based on big data, characterized in that, include: Historical fault data of the brake-by-wire system is acquired and preprocessed to obtain standard fault data; Cluster analysis is performed on the standard fault data to extract the location of faulty components and determine the set of weak points in the system; An initial fault propagation matrix is ​​constructed based on the set of weak points of the system. The system failure contribution of each faulty component is calculated and the initial fault propagation matrix is ​​updated to obtain the current fault propagation matrix. Based on the current fault propagation matrix, the failure probability of each faulty component is calculated using Monte Carlo simulation to form a failure probability dataset. Based on the failure probability dataset, multiple candidate switching paths are generated for each failure scenario of the faulty component. The candidate switching paths are simulated using digital twin simulation to obtain simulation results that include actual switching delay and actual success probability. The overall performance score of each path is calculated based on the simulation results, and the candidate switching paths are sorted from high to low according to the overall performance score. When the operating load of a faulty component exceeds the corresponding preset switching trigger threshold, the candidate switching path ranked first is selected for switching control. The process of constructing an initial fault propagation matrix based on the set of weak points in the system, calculating the system failure contribution of each faulty component, updating the initial fault propagation matrix, and obtaining the current fault propagation matrix includes: The fault points in the set of weak points are taken as the initial fault locations. The fault propagation path from the initial fault location to the final failure location is recursively analyzed according to the fault triggering conditions to obtain the propagation probability between each faulty component and construct the initial fault propagation matrix. Calculate the system failure contribution of each fault point under different ambient temperature parameters and operating load conditions; The initial fault propagation matrix is ​​updated based on the system failure contribution to obtain the current fault propagation matrix.

2. The redundant control method for line-controlled braking based on big data according to claim 1, characterized in that, The acquisition of historical fault data of the brake-by-wire system and the preprocessing to obtain standard fault data include: Historical fault data of the brake-by-wire system is obtained from sensor logs, controller records, and vehicle diagnostic systems. The fault occurrence time, ambient temperature parameters, operating load status and fault triggering conditions are extracted from the historical fault data and integrated using a timestamp alignment method to generate initial fault data. Linear interpolation was used to complete the missing fault occurrence time and ambient temperature parameters in the initial fault dataset, generating complete fault data. The isolation forest algorithm is applied to the completed fault dataset to detect outliers in the fault occurrence time, the ambient temperature parameter, and the operating load status, and noise data exceeding the preset abnormal threshold is removed to generate denoised fault data. The denoised fault dataset is sorted according to the fault occurrence time, and the ambient temperature parameter, the operating load status, and the fault triggering conditions are merged into a structured table to generate standard fault data.

3. The redundant control method for line-controlled braking based on big data according to claim 2, characterized in that, The process of performing cluster analysis on the standard fault data to extract the locations of faulty components and determine the set of weak points in the system includes: The frequency of failure and the scope of impact of each faulty component in the standard fault data are statistically analyzed. Based on the fault impact range and combined with the preset impact range, cluster analysis is performed on all fault points to obtain multiple fault categories. The fault points with the highest fault occurrence frequency among all fault categories are extracted as the set of weak points of the system.

4. The redundant control method for line-controlled braking based on big data according to claim 3, characterized in that, The method involves clustering all fault points based on the fault's impact range and a preset affected range to obtain multiple fault categories, including: Set the minimum and step values ​​for the affected area; Traverse all starting fault points and divide the fault points whose fault impact range is equal to the preset impact range into the same cluster; The preset impact range is changed according to the step value, and all starting fault points are clustered again until all fault points with an impact range not less than the minimum preset impact range are clustered and divided to obtain multiple fault categories.

5. The redundant control method for line-controlled braking based on big data according to claim 1, characterized in that, Based on the current fault propagation matrix, the failure probability of each faulty component is calculated using Monte Carlo simulation to form a failure probability dataset, including: Monte Carlo simulation is used to generate a sequence of random state parameters that include ambient temperature parameters and operating load conditions; State parameters exceeding a preset state threshold are extracted from the random state parameter sequence, and the failure probability vector of each faulty component under different ambient temperature parameters and operating load conditions is determined by data sampling. Based on the failure probability vector, the failure probability of each faulty component in the failure propagation path is calculated in conjunction with the current failure propagation matrix, and a failure probability dataset is generated.

6. The redundant control method for line-controlled braking based on big data according to claim 1, characterized in that, Based on the failure probability dataset, multiple candidate switching paths are generated for each failure scenario of the faulty component. Digital twin simulation is then used to simulate these candidate switching paths, yielding simulation results that include actual switching delay and actual success probability, including: Extract failure scenarios with failure probabilities higher than a preset probability threshold from the failure probability dataset; Based on preset switching delay and preset success probability, multiple candidate switching paths are generated for the failure scenario; Based on the candidate switching paths, a digital twin simulation model is constructed, and the ambient temperature parameters and operating load status corresponding to each candidate path are loaded to simulate the switching process of each candidate path under complex operating conditions, so as to obtain simulation results including actual switching delay and actual success probability.

7. The redundant control method for line-controlled braking based on big data according to claim 1, characterized in that, The step of calculating the comprehensive performance score of each path based on the simulation results, and sorting the candidate switching paths from high to low according to the comprehensive performance score, includes: Calculate the comprehensive performance score of each candidate handover path based on the actual handover latency and the actual success probability; Candidate switching paths are sorted from high to low based on the comprehensive performance score.

8. The redundant control method for line-controlled braking based on big data according to claim 6, characterized in that, The formula for the comprehensive performance score is: in, A comprehensive performance score; This represents the actual probability of success. This represents the actual handover latency; The preset maximum switching delay; and These are preset weighting coefficients.

9. A redundant control system for brake-by-wire based on big data, used to implement the redundant control method for brake-by-wire based on big data as described in any one of claims 1 to 8, characterized in that, include: The data acquisition and processing module is used to acquire historical fault data of the brake-by-wire system and preprocess it to obtain standard fault data. The fault location extraction module is used to perform cluster analysis on the standard fault data, extract the location of the faulty component, and determine the set of weak points in the system. The propagation matrix construction module is used to construct an initial fault propagation matrix based on the set of weak points of the system, calculate the system failure contribution of each faulty component and update the initial fault propagation matrix to obtain the current fault propagation matrix. The failure probability calculation module is used to calculate the failure probability of each faulty component based on the current fault propagation matrix using Monte Carlo simulation, and form a failure probability dataset. The switching path simulation module is used to generate multiple candidate switching paths for each failure scenario of the faulty component based on the failure probability dataset, and to simulate the candidate switching paths using digital twin simulation to obtain simulation results that include actual switching delay and actual success probability. The path scoring calculation module is used to calculate the comprehensive performance score of each path based on the simulation results, and sort the candidate switching paths from high to low according to the comprehensive performance score; The threshold-triggered switching module is used to select the first-ranked candidate switching path for switching control when the operating load of a faulty component exceeds the corresponding preset switching trigger threshold.

Images