An intelligent fault detection and diagnosis method and system

Abstract

The present invention relates to the technical field of fault detection, and specifically to an intelligent fault detection and diagnosis method and system, which includes the following steps: Based on the operation data of the vehicle in normal and abnormal states, fault feature vectors are extracted item by item and classified, and then transmitted to the fault coding layer through encoding, input into the latent space, and the feature points are clustered to obtain preliminary fault identification features. In the present invention, by refining the fault feature extraction and expanding the sample analysis, the accuracy and efficiency of fault diagnosis are significantly improved. By analyzing the normal and abnormal state data, fine fault feature vectors are extracted, and clustering technology is used for preliminary identification. The sample set is further expanded through simulated samples to deepen the learning of fault modes and improve the prediction accuracy. Hierarchical fault detection is introduced to adjust the response according to the severity of the fault, and dynamic monitoring realizes early warning, effectively extending the equipment operation cycle, reducing the risk of downtime, saving maintenance costs, and enhancing the reliability of the system.

Description

Technical Field

[0001] The present invention relates to the field of technical fault detection, and particularly to an intelligent fault detection and diagnosis method and system. Background Art

[0002] The field of fault detection technology involves monitoring and analyzing the operating states of devices, systems or vehicles, aiming to identify abnormal phenomena and potential faults. This technology relies on technical means such as sensors, data acquisition and processing, signal analysis, pattern recognition, etc. By continuously monitoring the state data of machinery and equipment and extracting features, it identifies fault signs and judges the health status of the equipment. Fault detection can assist in predicting the fault time, improving the reliability of the system, extending the service life of the equipment, and reducing the maintenance cost, thereby avoiding economic losses and safety risks caused by unexpected equipment shutdowns.

[0003] Among them, the intelligent fault detection and diagnosis method is a technology that applies advanced artificial intelligence and machine learning algorithms in the field of fault detection, aiming to automatically identify and accurately diagnose the fault states of complex systems such as vehicles. By analyzing the massive data during the operation of the equipment, it uses algorithms to automatically learn fault patterns, locate the sources of potential faults and estimate the damages. Its main uses include enhancing the safety, stability and reliability of vehicle operation, reducing the occurrence of sudden faults, optimizing the maintenance process, achieving predictive maintenance, and thus providing more valuable technical support for vehicle management and maintenance.

[0004] The existing technologies rely on traditional sensor data and relatively static analysis models when dealing with fault detection, which limits their efficiency and agility in processing massive dynamic data. The lag and coarseness of fault detection result in the inability to accurately predict the specific time and type of faults, increasing the uncertainty and potential risks of the method. Without comprehensively utilizing machine learning and pattern recognition technologies, traditional methods are difficult to adapt to the rapidly changing equipment states, leading to problems such as inaccurate fault diagnosis and long fault response times in actual operations. This method that relies on preset fault models has poor effects when dealing with unknown or atypical faults, is difficult to adapt to the current fault patterns, thus ignoring incipient faults or abnormal phenomena, and increasing the operation and maintenance challenges and safety risks of the equipment. Summary of the Invention

[0005] The purpose of the present invention is to solve the deficiencies existing in the prior art, and to propose an intelligent fault detection and diagnosis method and system.

[0006] To achieve the above purpose, the present invention adopts the following technical solution: An intelligent fault detection and diagnosis method, comprising the following steps:

[0007] S1: Based on the operation data of the vehicle in normal and abnormal states, extract and classify the fault feature vectors item by item, transmit them through encoding to the fault coding layer, input them into the latent space, cluster the feature points, and obtain the preliminary fault identification features;

[0008] S2: Based on the preliminary fault identification features, through the fault simulation samples, extract the fault latent features and extended samples, analyze the distribution differences between the extended samples and the fault feature space, screen the extended samples associated with potential faults, generate a simulated fault sample set, verify the simulated fault sample set from multiple angles, and generate a sample extension structure for fault simulation;

[0009] S3: Based on the sample extension structure of the fault simulation, analyze the samples and abnormal density points, set the multi-level detection thresholds of the fault detection threshold layer, screen the fault features of different detection levels, establish a multi-level fault detection standard set, and according to the multi-level fault detection standard set, match the abnormal feature distribution of the detection level to obtain the hierarchical fault detection results;

[0010] S4: Based on the hierarchical fault detection results, call the real-time fault monitoring flow, dynamically monitor the signal features during vehicle operation, match the signal features with the detection standards, judge the vehicle state, embed the fault identifier into the monitoring configuration, and gradually update the identifier state to obtain the real-time fault detection and diagnosis scheme.

[0011] As a further solution of the present invention, the specific steps for obtaining the preliminary fault identification features are as follows:

[0012] S111: Based on the operation data of the vehicle in normal and abnormal states, extract the feature distributions of the vehicle in normal and abnormal states, call the operation parameters of speed, temperature, and vibration frequency and perform state encoding, and perform hierarchical classification processing on the state features to obtain a set of feature vectors;

[0013] S112: Perform information entropy analysis on the set of feature vectors. By calculating the information entropy value of each feature vector and eliminating the feature points with information entropy values lower than the set threshold, use the formula:

[0014]

[0015] Calculate the similarity value between feature vectors and establish a feature similarity matrix, where S represents the similarity value between feature vectors, a i is the feature weight coefficient, v i represents the value of the feature vector, g i is the feature difference parameter, b i is the additional adjustment parameter;

[0016] S113: According to the feature similarity matrix, apply clustering logic to group the feature points, calculate the clustering center features of each group, and generate preliminary fault identification features.

[0017] As a further solution of the present invention, the steps for obtaining the simulated fault sample set are specifically as follows:

[0018] S211: Based on the preliminary fault identification features, call the feature parameters in the fault simulation samples for matching, respectively extract the potential feature values of the features in the fault simulation samples, determine the correlation between the potential features and the fault identification features by calculating the differences between the features, and generate a fault potential feature set;

[0019] S212: According to the feature parameters in the fault potential feature set, perform a position distribution analysis with the feature space of the extended samples, identify the differential distribution between the extended samples and the fault feature space, screen the extended sample set associated with the potential fault by setting a correlation threshold, and use the formula:

[0020]

[0021] Calculate the differential distribution value of the feature space to obtain the extended samples associated with the potential fault, where D represents the differential distribution value of the fault feature space, w is the weight coefficient, f and g respectively represent the feature values of the fault potential feature and the extended samples, k is the extended sample offset coefficient, h is the standard deviation of the extended sample features, and p is the offset adjustment parameter;

[0022] S213: According to the extended samples associated with the potential fault, screen the samples associated with the potential fault through the differential distribution value, integrate each associated sample parameter, and generate a simulated fault sample set.

[0023] As a further solution of the present invention, the steps for obtaining the sample extension structure of the fault simulation are specifically as follows:

[0024] S221: Based on each fault feature of the simulated fault sample set, analyze and extract the feature angle, value range, and distribution density, calculate the difference index and distribution weight of each feature, and generate a multi-angle analysis matrix of the fault features;

[0025] S222: Based on the feature difference index and weight value of the multi-angle analysis matrix of the fault features, calculate the influence score of the feature at the differential angle, and use the formula:

[0026]

[0027] Calculate the normalized influence value of the feature to obtain the verification result of the feature, where Y is the normalized influence value of the feature, m and n are the weight coefficients of the angle, x and y are the feature values under the differential angle, z is the differential adjustment parameter, and r and q are the additional adjustment parameters of the influence degree;

[0028] S223: Based on the verification result of the feature, call the screening result of the multi-angle analysis matrix, merge the feature sets with correlation degree, and generate a sample expansion structure for fault simulation.

[0029] As a further solution of the present invention, the obtaining steps of the multi-level fault detection standard set are specifically as follows:

[0030] S311: Based on the sample expansion structure for fault simulation, extract fault feature data from the sample data, calculate the difference value with the abnormal density point, normalize the calculation result, and generate a fault feature difference vector;

[0031] S312: Based on the fault feature difference vector, set the detection threshold layer of the multi-level fault detection standard set. For each feature data, through the formula:

[0032]

[0033] Calculate the fault feature difference value to generate a fault feature difference degree matrix, where Q represents the fault feature difference value, F i represents the i-th eigenvalue in the fault feature vector, T is the set detection threshold, W i represents the weight of the feature, and A i is the feature adjustment coefficient;

[0034] S313: According to the fault feature difference degree matrix, judge whether each feature difference degree is greater than the detection threshold, screen the fault features higher than the threshold, mark the features not exceeding the threshold as low-level features, and establish a multi-level fault detection standard set.

[0035] As a further solution of the present invention, the obtaining steps of the hierarchical fault detection result are specifically as follows:

[0036] S321: According to the abnormal feature distribution data of each level in the multi-level fault detection standard set, for each abnormal feature distribution, sequentially determine the matching detection level, and obtain the initial hierarchical fault detection feature by calculating the cumulative deviation amount of the abnormal feature distribution;

[0037] S322: Based on the initial hierarchical fault detection feature, calculate the cumulative deviation amount of the abnormal feature distribution, set the fault detection level threshold, judge the cumulative deviation amount, and use the formula:

[0038] E = |a·d x-b·d y +c·d z |

[0039] Generate the matching result of the hierarchical fault detection feature, where E represents the matching degree of the abnormal distribution deviation, d x represents the cumulative deviation amount of the initial hierarchical fault detection feature, d y is the set fault detection level threshold, d z is the difference value of the detection level cumulative value in the previous step result, and a, b, and c are the weight parameters of the cumulative deviation amount respectively;

[0040] S323: Analyze the matching result of the hierarchical fault detection feature, select the hierarchical feature whose abnormal distribution deviation exceeds the detection level threshold, and generate the hierarchical fault detection result by calling the matching situation and determination of the fault detection feature.

[0041] As a further solution of the present invention, the acquisition steps of the real-time fault detection and diagnosis solution are specifically as follows:

[0042] S411: Call the fault level parameter in the hierarchical fault detection result, extract the signal feature value during the vehicle operation according to the real-time fault monitoring flow, perform normalization processing on the collected signal data through dynamic acquisition of the signal feature, and calculate the deviation between the signal feature value and the standard value of the corresponding fault level to generate the real-time fault signal deviation result;

[0043] S412: According to the real-time fault signal deviation result, call the vehicle state judgment condition, analyze the monitoring threshold set for the deviation value, and use the formula:

[0044]

[0045] Calculate the standardized result of the signal deviation amount and determine whether it exceeds the threshold to generate the vehicle state matching result, where H is the signal deviation state value, δ s is the real-time signal deviation value, δ t is the deviation threshold in the fault level parameter, δ u is the initial deviation value of the hierarchical fault detection result, w a 、w b 、w c are the signal feature weight parameters;

[0046] S413: Dynamically monitor the vehicle state matching result, update the current fault identifier according to the vehicle state, gradually embed the fault identifier into the monitoring configuration, update and record the real-time state of the vehicle fault, and generate the real-time fault detection and diagnosis solution.

[0047] An intelligent fault detection and diagnosis system, which is used to execute the above-mentioned intelligent fault detection and diagnosis method, and the system includes:

[0048] The operating state analysis module separates various signals during operation based on the normal state and abnormal state operation data of the vehicle, extracts characteristic signals and fault parameters, performs differential operations on the correlation relationships between the signals, merges the fault characteristic values, combines the characteristics by category, and obtains a multi-dimensional fault characteristic signal set;

[0049] The fault coding analysis module maps the characteristic parameters of the signal set according to the requirements of the fault coding based on the multi-dimensional fault characteristic signal set, matches the characteristic signals with the corresponding fault characteristic points in the coding space, calculates the matching relationship between the characteristic values and the spatial distribution, divides the boundary characteristics of the differential faults, and obtains the fault aggregation recognition characteristics;

[0050] The simulated fault expansion module calls the original data samples based on the fault aggregation recognition characteristics, adds the samples that meet the fault characteristics, performs multi-dimensional distribution analysis on the matching samples, merges the expanded sample sets, and obtains a fault simulation characteristic structure;

[0051] The hierarchical fault judgment module sets multi-level fault judgment thresholds based on the fault simulation characteristic structure, compares the differential characteristics in the simulation characteristic structure with the thresholds of each level, sequentially screens the characteristic points of the differential levels, and merges them successively to generate a multi-level fault judgment structure;

[0052] The real-time fault monitoring module matches the real-time monitoring signal characteristics based on the multi-level fault judgment structure, analyzes the signal characteristics and the hierarchical boundaries, determines the matching situation of the fault levels, and embeds the fault state and the operation state signals into the configuration in real time to obtain a real-time fault detection and diagnosis solution.

[0053] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0054] In the present invention, by introducing a refined fault feature extraction and extended sample analysis method, the accuracy and efficiency of fault diagnosis are significantly improved. By analyzing the normal and abnormal state data, more detailed fault feature vectors are extracted, and based on the vectors, clustering technology is used to initially identify faults. Further, the sample set is expanded through fault simulation samples, which not only optimizes the process of fault diagnosis but also enhances the learning depth of fault modes, making the prediction of potential faults more accurate. The introduction of hierarchical fault detection allows the method to respond according to the severity of the faults, ensuring that the monitoring parameters can be dynamically adjusted during real-time monitoring, realizing early fault warning, effectively extending the operation cycle of the equipment, reducing the risk of unexpected shutdowns, thereby saving maintenance costs and enhancing the reliability of the overall method. Description of the Drawings

[0055] Figure 1 It is a schematic diagram of the working process of the present invention;

[0056] Figure 2 It is a flowchart of the preliminary fault identification features in the present invention;

[0057] Figure 3 It is a flowchart of the simulated fault sample set in the present invention;

[0058] Figure 4 It is a flowchart of the sample expansion structure of fault simulation in the present invention;

[0059] Figure 5 It is a flowchart of the multi-level fault detection standard set in the present invention;

[0060] Figure 6 It is a flowchart of the hierarchical fault detection result in the present invention;

[0061] Figure 7 It is a flowchart of the real-time fault detection and diagnosis scheme in the present invention. Detailed implementation manners

[0062] In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0063] In the description of the present invention, it should be understood that the orientation or positional relationship indicated by the terms "length", "width", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. is based on the orientation or positional relationship shown in the drawings, and is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the present invention. In addition, in the description of the present invention, the meaning of "a plurality" is two or more, unless otherwise specifically defined.

[0064] Embodiment 1

[0065] Please refer to Figure 1 , the present invention provides a technical solution: an intelligent fault detection and diagnosis method, including the following steps:

[0066] S1: Based on the operation data of the vehicle in normal and abnormal states, extract and classify the fault feature vectors item by item, transfer them to the fault coding layer through encoding, input them into the latent space, cluster the feature points, and obtain the preliminary fault identification features;

[0067] S2: Based on the preliminary fault identification features, through the fault simulation samples, extract the potential fault features and extended samples, analyze the difference in the spatial distribution of the extended samples and the fault features, screen the extended samples associated with the potential faults, generate a simulated fault sample set, conduct multi-angle verification on the simulated fault sample set, and generate a sample extension structure for fault simulation;

[0068] S3: Based on the sample extension structure for fault simulation, analyze the samples and the abnormal density points. By setting multi-level detection thresholds for the fault detection threshold layer, screen the fault features with different detection levels, establish a multi-level fault detection standard set. According to the multi-level fault detection standard set, match the abnormal feature distribution of the detection level to obtain the hierarchical fault detection results;

[0069] S4: Based on the hierarchical fault detection results, call the real-time fault monitoring stream, dynamically monitor the signal features during vehicle operation, match the signal features with the detection standards, judge the vehicle state, embed the fault identifier into the monitoring configuration, and gradually update the identifier state to obtain a real-time fault detection and diagnosis solution.

[0070] The preliminary fault identification features include fault category identifiers, density distribution features, and clustering feature points. The sample extension structure for fault simulation includes simulated fault samples, potential fault types, and extended sample sets. The multi-level fault detection standard set includes detection threshold levels, fault feature classification, and abnormal distribution standards. The hierarchical fault detection results include fault detection levels, abnormal identifier types, and distribution detection results. The real-time fault monitoring stream includes dynamic signal data, detection matching standards, and real-time monitoring identifiers. The real-time fault detection and diagnosis solution includes identification status updates, monitoring configuration adjustments, and fault identifier statuses.

[0071] Please refer to Figure 2 , and the specific steps for obtaining the preliminary fault identification features are as follows:

[0072] S111: Based on the operation data of the vehicle in normal and abnormal states, extract the feature distributions of the vehicle in normal and abnormal states, call the operation parameters of speed, temperature, and vibration frequency and conduct state encoding, and perform hierarchical classification processing on the state features to obtain a set of feature vectors;

[0073] Based on the vehicle operation data, extract the operation characteristic data of the vehicle in normal and abnormal states. First, preliminarily classify the collected parameters such as speed, temperature, vibration frequency, etc. in sequence. By comparing the value ranges, fluctuation frequencies, etc. of the parameters in different states, identify the distribution ranges of each state characteristic. After obtaining the distribution ranges, perform state marking on each parameter to generate a state marking matrix to facilitate the analysis of the boundary relationships between different states. Further, call the feature screening rules to compare different parameters one by one, eliminate duplicate or highly interfering feature items, and perform hierarchical processing on the feature distribution results of various states through state coding. During the hierarchical process, uniformly manage the features in the state coding set to ensure that accurate feature distribution data can be called during subsequent fault feature classification, and obtain a preliminary feature vector set.

[0074] S112: Conduct information entropy analysis on the feature vector set. By calculating the information entropy values of each feature vector and eliminating the feature points with information entropy values lower than the set threshold, use the formula:

[0075]

[0076] Calculate the similarity values between feature vectors and establish a feature similarity matrix. Among them, S represents the similarity values between feature vectors, a i is the feature weight coefficient, v i represents the value of the feature vector, g i is the feature difference degree parameter, b i is the additional adjustment parameter;

[0077] The advantage of the formula is that through the combination of the feature weight coefficient, the feature difference degree parameter, and the additional adjustment parameter, the calculation of the similarity between feature points is more flexible and accurate, so as to more precisely judge the similarity relationship between different features in high-dimensional data;

[0078] Set the weight coefficient a i = 0.5, the feature vector value v i = 10, the feature difference degree parameter g i = 2, the additional adjustment parameter b i = 1.5, substitute the values into the formula to calculate the feature similarity value;

[0079] First, calculate the numerator part:

[0080]

[0081] Next, calculate the denominator part:

[0082]

[0083] Substitute the numerator and denominator into the formula to obtain the similarity value:

[0084]

[0085] The result shows that the similarity value between the feature vectors is 0.97, which is close to 1, indicating a high degree of similarity between the feature points. This value can be used for further analysis of the feature clustering results and provide a reference for the feature center set in the subsequent steps.

[0086] S113: According to the feature similarity matrix, apply clustering logic to group the feature points, calculate the clustering center features of each group, and generate preliminary fault identification features;

[0087] Based on the feature similarity matrix, apply similarity clustering logic for feature classification. By calculating the similarity values between each feature point in the feature similarity matrix, feature points with a high degree of similarity are grouped into the same class according to a certain similarity threshold to form a preliminary clustering grouping. Then, a representative central feature point is calculated within each clustering grouping. The selection of the central feature point is determined by calculating the average similarity between each feature point in the grouping and the feature point, and the feature point with the highest average value is selected as the clustering center. After the clustering centers of each group are selected, record the central feature points of each group, form a feature center set with each clustering center feature point, and mark the features in the set as preliminary fault identification features.

[0088] Please refer to Figure 3 , and the specific steps for obtaining the simulated fault sample set are as follows:

[0089] S211: Based on the preliminary fault identification features, call the feature parameters in the fault simulation samples for matching, extract the potential feature values of the features in the fault simulation samples respectively, determine the correlation between the potential features and the fault identification features by calculating the differences between the features, and generate a fault potential feature set;

[0090] Based on the preliminary fault identification features, extract the parameters in the fault simulation samples, compare the feature parameters of the fault simulation samples with the preliminary fault identification features item by item, and judge the correlation between the features by calculating the numerical differences between each feature value. First, extract each feature data contained in the simulation sample, call the potential feature values of the associated fault features, compare the statistical features such as the value range, standard deviation, and variance of each fault feature, screen out the fault features associated with the preliminary identification features from them, establish a preliminary fault feature set. After extracting the potential fault feature set, standardize the parameters in the fault sample to ensure that all parameters in the feature set have the same dimension and scale. After completing the standardization, generate a fault potential feature set to ensure the accuracy of feature comparison and call in the subsequent analysis process.

[0091] S212: According to the characteristic parameters in the potential fault feature set, conduct a position distribution analysis with the feature space of the extended samples, identify the differential distribution between the extended samples and the fault feature space, and screen the set of extended samples associated with potential faults by setting an association threshold. Use the formula:

[0092]

[0093] Calculate the differential distribution value of the feature space to obtain the extended samples associated with potential faults. Among them, D represents the differential distribution value of the fault feature space, w is the weight coefficient, f and g respectively represent the feature values of potential faults and extended samples, k is the extended sample offset coefficient, h is the standard deviation of the extended sample features, and p is the offset adjustment parameter;

[0094] The advantage of the formula is that by introducing the combination of the weight coefficient, offset coefficient, and adjustment parameter, the contribution of different features can be adjusted more flexibly in the calculation of the similarity degree in the feature space, so as to more accurately reflect the relationship between potential fault features and extended samples;

[0095] Set the weight coefficient w = 0.8, the potential fault feature value f = 15, the extended sample feature value g = 12, the offset coefficient k = 0.5, the standard deviation h of the extended sample features = 4, and the adjustment parameter p = 1.0, and substitute them into the formula to calculate the differential distribution value;

[0096] First, calculate the square of the differential value of the first term:

[0097] (w·(f - g)) 2 =(0.8·(15 - 12)) 2 =(0.8·3) 2 =2.4 2 =5.76

[0098] Next, calculate the square of the offset value of the second term:

[0099] (k·h) 2 =(0.5·4) 2 =2 2 =4

[0100] Then, substitute the values into the formula to calculate the final differential distribution value:

[0101]

[0102] The result shows that the differential distribution value is 3.28, indicating that the difference degree between the potential fault features and the extended samples is small, which can be used to further screen the extended samples associated with the potential fault features and serve as a component of the simulated fault sample set.

[0103] S213: According to the extended samples associated with potential faults, screen the samples associated with potential faults through the differential distribution values, integrate each associated sample parameter, and generate a simulated fault sample set;

[0104] On the basis of completing the differential distribution analysis of the difference between the potential fault feature set and the extended sample feature space, by setting the threshold of the differential distribution to screen the extended samples with higher correlation, first analyze each parameter in the obtained extended sample set, compare the previously calculated differential distribution value with the threshold, and mark the samples with differential distribution values less than or equal to the set threshold as relevant samples. In this way, gradually classify various associated samples into the finally generated simulated fault sample set, so that the set can comprehensively cover the diversity of potential fault features.

[0105] Please refer to Figure 4 , and the specific steps for obtaining the sample expansion structure of fault simulation are as follows:

[0106] S221: Based on each fault feature of the simulated fault sample set, analyze and extract the feature angle, value range, and distribution density, calculate the difference index and distribution weight of each feature, and generate a multi-angle analysis matrix of fault features;

[0107] Based on each fault feature of the simulated fault sample set, analyze and extract the angle, value range, and distribution density of the feature item by item. First, calculate the statistical attributes of each feature parameter, including the mean, variance, and extreme values, etc., to describe the performance of each fault feature at different angles, and further generate its difference index. Then, assign an appropriate weight to different angles of each feature. The assignment of the weight depends on the change range of the feature at different angles and the proportion it occupies in the overall feature set. By comparing the feature performance at different angles through this process, judge which features are more representative at specific angles, form a complete multi-angle analysis matrix of features, and use this matrix to screen and judge the relevance and difference of each feature, which can provide data support for the subsequent steps and ensure the diversity and relevance of features at different angles.

[0108] S222: Based on the feature difference index and weight value of the multi-angle analysis matrix of fault features, calculate the influence score of the feature at different differential angles, using the formula:

[0109]

[0110] Calculate the normalized influence value of the feature to obtain the verification result of the feature. Among them, Y is the normalized influence value of the feature, m and n are the weight coefficients of the angle, x and y are the feature values at different differential angles, z is the differential adjustment parameter, and r and q are the additional adjustment parameters of the influence degree;

[0111] The advantage of the formula is that by using different weights and adjustment parameters, the differential calculation between different features can be precisely controlled in multi-angle analysis, and the feature adaptation degree can be flexibly adjusted among features with significant differences;

[0112] Set the weight coefficients m = 0.6 and n = 0.4, the differential feature value x = 18, the feature value y of another angle = 12, the adjustment parameter z = 5, the influence degree additional parameters r = 2 and q = 3, and substitute the values into the formula to calculate the multi-angle normalization influence value;

[0113] First, calculate the numerator part:

[0114] |m·(x - y)| = |0.6·(18 - 12)| = |0.6·6| = 3.6

[0115]

[0116] The total of the numerator is: 3.6 + 0.896 = 4.496;

[0117] Next, calculate the denominator part:

[0118] r + q = 2 + 3 = 5

[0119] Finally, obtain the normalization influence value:

[0120]

[0121] This result indicates that the normalization influence value is 0.899, which means that in the multi-angle verification process, the comprehensive influence value of this feature is relatively high, meeting the requirements of multi-angle analysis, and can be used for the comprehensive analysis of multi-angle verification scores in the follow-up to support the multi-angle verification and expansion of fault features.

[0122] S223: Based on the verification results of features, call the screening results of the multi-angle analysis matrix, merge the feature sets with correlation, and generate a sample expansion structure for fault simulation;

[0123] Based on the comprehensive results of multi-angle verification scores, compare the screening results of each feature in the obtained multi-angle analysis matrix, determine the correlation between features through the comparison process, classify and merge the features with higher correlation, and at the same time verify the merged feature set to ensure that there is no loss or deviation of feature information in the multi-angle verification process. Mark the feature set after multi-angle verification as highly correlated features, further integrate the feature set with the highest correlation, and form a sample expansion structure for fault simulation. Based on this feature expansion structure, multiple fault features can be organically combined to cover more potential fault features, ensuring that the sample expansion structure for fault simulation has a more comprehensive fault feature coverage ability.

[0124] Please refer toFigure 5 , the steps for obtaining the multi-level fault detection standard set are specifically as follows:

[0125] S311: Based on the sample expansion structure for fault simulation, extract the fault feature data from the sample data, calculate the difference values with the abnormal density points, normalize the calculation results, and generate the fault feature difference vector;

[0126] Based on the sample expansion structure for fault simulation, extract the fault feature data, compare each sample data item by item with the abnormal density points, calculate the distance differences between each sample point and the abnormal density points, statistically analyze the difference distribution between the samples and the abnormal points, then normalize all the calculation results to ensure comparison of different feature data on a unified scale, and then convert the normalized data into independent numerical values for each fault feature through the feature decomposition method to obtain a series of standardized feature values, generating a fault feature difference vector containing various fault feature components. The difference values of different features can be clearly distinguished for subsequent calculations, providing specific input data support for subsequent multi-level detection.

[0127] S312: Based on the fault feature difference vector, set the detection threshold layer of the multi-level fault detection standard set. For each feature data, through the formula:

[0128]

[0129] calculate the fault feature difference value, and generate the fault feature difference degree matrix. Among them, Q represents the fault feature difference value, F i represents the i-th eigenvalue in the fault feature vector, T is the set detection threshold, W i represents the weight of the feature, and A i is the feature adjustment coefficient;

[0130] The advantage of the formula is that by adding the feature adjustment coefficient A i and the feature weight W i , the calculation result is more adaptable, capable of self-adjusting the difference degree calculation under different detection conditions, and improving the accuracy of detection;

[0131] Let F i = 100 be the eigenvalue, T = 80 be the detection threshold, the feature adjustment coefficient A i = 1.2 is calculated through the previous feature distribution analysis, and the feature weight W i = 0.5 is set based on the feature importance weight distribution;

[0132] Substitute each value into the formula for calculation:

[0133]

[0134] First, calculate the numerator part:

[0135] |100 - 80|·1.2 = 20·1.2 = 24

[0136] Then, calculate the denominator part:

[0137]

[0138] Combine the numerator and denominator:

[0139]

[0140] The result shows that the calculated difference degree Q = 19.6 indicates a relatively high difference level of this feature under the current threshold, which can be used as the basis for screening high - level fault features.

[0141] S313: According to the fault feature difference degree matrix, determine whether the difference degree of each feature is greater than the detection threshold, screen the fault features higher than the threshold, mark the features not exceeding the threshold as low - level features, and establish a multi - level fault detection standard set;

[0142] According to each data point in the fault feature difference degree matrix, compare the difference degree of each feature with the detection threshold. If the difference degree of a certain feature is greater than the set threshold, mark this feature as a high - level fault feature and add it to the high - level fault feature set. Continue to loop and judge the size of the difference degree values in each feature data, screen all the data that meet the high - level feature criteria. If the difference degree does not exceed the threshold, classify this feature as a low - level feature and maintain its original state. Through this classification method, the classification and categorization of fault features are realized, so that each feature is arranged in an orderly manner according to its detection level, forming a multi - level fault detection standard set.

[0143] Please refer to Figure 6 , and the steps for obtaining the hierarchical fault detection results are specifically as follows:

[0144] S321: According to the abnormal feature distribution data of each level in the multi - level fault detection standard set, for each abnormal feature distribution, sequentially determine the matching detection level, and obtain the initial hierarchical fault detection features by calculating the cumulative deviation amount of the abnormal feature distribution;

[0145] According to the abnormal feature distribution data of each level in the multi - level fault detection standard set, sequentially retrieve the abnormal features of each classification, perform feature quantization processing on the detection content of each level, convert the abnormal feature data in the classification into numerical data, and perform comparison operations one by one according to the abnormal data interval of each level of the standard set. Determine the interval through the set detection threshold, calculate the numerical deviation amount of the distribution feature, record the cumulative deviation value of each abnormal feature, so as to determine the matching degree of the abnormal features at each classification level, and match the cumulative deviation value with the set detection standards of each level to obtain the initial hierarchical fault detection features.

[0146] S322: Calculate the cumulative deviation of the abnormal feature distribution based on the initial hierarchical fault detection feature, set the fault detection level threshold, and judge the cumulative deviation using the formula:

[0147] E = |a·d x -b·d y +c·d z |

[0148] Generate the matching result of the hierarchical fault detection feature, where E represents the matching degree of the abnormal distribution deviation, d x represents the cumulative deviation of the initial hierarchical fault detection feature, d y is the set fault detection level threshold, d z is the difference value of the detection level cumulative value in the previous step result, and a, b, and c are the weight parameters of the cumulative deviation respectively;

[0149] The advantage of the formula is that it introduces the weight parameters of the cumulative deviation, so that the detection result can more accurately reflect the actual influence degree of the abnormal feature, and enhances the refinement and accuracy of the hierarchical feature matching result;

[0150] The parameter d x is the cumulative deviation obtained from the initial hierarchical fault detection feature, and is specifically calculated through the cumulative value of the aforementioned detection threshold deviation;

[0151] The parameter d y is the set detection level threshold, obtained based on the mean value of the upper and lower limits of the hierarchical threshold in the multi-level fault detection standard set;

[0152] The parameter d z is the difference in the cumulative values of each detection level, representing the numerical difference range of each level of detection feature, and is calculated based on the cumulative deviation value of the abnormal feature under the aforementioned classification;

[0153] The setting basis of the weight parameters a, b, and c is the weight influence of each parameter on the influence degree of the abnormal feature, and it fluctuates with the upper and lower limit thresholds of the classification standard;

[0154] Set d x = 15, d y = 10, d z = 5, set a = 1.2, b = 0.8, c = 1.5, then the formula calculation process is as follows:

[0155] Calculate a·d x = 1.2 × 15 = 18;

[0156] Calculate b·d y = 0.8 × 10 = 8;

[0157] Calculate c·d z = 1.5 × 5 = 7.5;

[0158] The result of the formula E = |18 - 8 + 7.5| = |17.5| = 17.5;

[0159] This result indicates that the matching degree of the abnormal distribution deviation is 17.5, which has a high degree of conformity with the deviation interval of the fault detection level, and generates the matching result of the hierarchical fault detection feature.

[0160] S323: Analyze the matching result of the hierarchical fault detection feature, select the hierarchical feature whose abnormal distribution deviation exceeds the detection level threshold, and generate the hierarchical fault detection result by invoking the matching situation and determination of the fault detection feature;

[0161] According to the abnormal distribution situation of the foregoing detection levels, further analyze the matching result in the second step. For the fault feature whose abnormal distribution deviation exceeds the detection level threshold in each level, record the final matching situation of this hierarchical feature, perform a composite operation on the result parameter and the initial hierarchical fault detection feature, and call the fault level threshold for further hierarchical processing according to the increasing trend of the deviation degree of the matching parameter. Generate the hierarchical fault detection result by recording the final matching situation of each level.

[0162] Please refer to Figure 7 , and the specific steps for obtaining the real-time fault detection and diagnosis solution are as follows:

[0163] S411: Invoke the fault level parameter in the hierarchical fault detection result, extract the signal feature value during the vehicle operation according to the real-time fault monitoring stream, perform normalization processing on the collected signal data through dynamic acquisition of the signal feature, calculate the deviation between the signal feature value and the standard value of the corresponding fault level, and generate the real-time fault signal deviation result;

[0164] Call the fault level parameters in the hierarchical fault detection results, sort out item by item the signal feature data collected in real time during the vehicle operation, perform normalization processing on each signal feature in turn, map the signal feature value range to a standard scale uniformly, so that different source signals have consistency. Through this normalization process, a standardized representation of each signal feature is generated to ensure that the signal features are within a unified range for subsequent operations. Match the normalized signal feature data according to the standard values of each level of fault level, calculate item by item the difference between the actual signal feature value and the standard fault detection value. By gradually calculating the deviation values of each signal feature, a deviation amount set of signal features is formed. Comprehensively consider the differences in the deviation value set, determine whether it exceeds the set range of the fault level parameters, accumulate the signal deviations that meet the conditions to form a real-time fault signal deviation result, and this deviation result will be used to further determine the matching situation between the vehicle operation state and the fault level.

[0165] S412: According to the real-time fault signal deviation result, call the vehicle state judgment conditions, analyze by monitoring the threshold value set for the deviation value, and use the formula:

[0166]

[0167] Calculate the standardized result of the signal deviation amount and determine whether it exceeds the threshold value to generate a vehicle state matching result, where H is the signal deviation state value, δ s is the real-time signal deviation value, δ t is the deviation threshold value in the fault level parameters, δ u is the initial deviation value of the hierarchical fault detection result, w a 、w b 、w c are the signal feature weight parameters;

[0168] The advantage of the formula is that by adjusting multiple signal weights, the detection result is more flexible, can adapt to the actual influence of different signals, and helps to more carefully reflect the current state of the vehicle;

[0169] This formula is used to calculate the signal deviation state value H;

[0170] δ s is the real-time signal deviation value, which is the difference between the current running signal of the vehicle and the fault standard value, obtained by collecting real-time signal data on site and comparing it with the preset standard value;

[0171] δ t represents the deviation threshold value in the set fault level parameters, which is the tolerance deviation range obtained through data analysis and actual monitoring;

[0172] δ u ​The initial deviation value for hierarchical fault detection is obtained through processing hierarchical detection data;

[0173] Weight parameter w a 、w b 、w c Perform weighted processing on each signal deviation term respectively, and the weight setting is dynamically adjusted according to the actual operating status;

[0174] Substitute the numerical calculation example: Let δ s =12, δ t =8, δ u =6, and set w a =1.1, w b =0.9, w c =1.0;

[0175] The calculation process is as follows:

[0176] Calculate w a ·δ s =1.1×12=13.2;

[0177] Calculate w b ·δ t =0.9×8=7.2;

[0178] Calculate w c ·δ u =1.0×6=6;

[0179] Substitute into the formula:

[0180] The result shows that the signal deviation status value is 9.33. This value indicates that the current signal deviation is within the detection range, and a compliant vehicle state matching result is generated.

[0181] S413: Dynamically monitor the vehicle state matching result, update the current fault flag according to the vehicle state, gradually embed the fault flag into the monitoring configuration, update and record the real-time state of vehicle faults, and generate a real-time fault detection and diagnosis scheme;

[0182] Based on the vehicle state matching result, further use the dynamic monitoring mechanism to gradually embed the vehicle fault flag, call the previously generated real-time signal deviation status value result, embed the qualified fault flag information into the monitoring configuration system, update the fault flag in the configuration system for specific signal deviation status values, and update each time according to the current data of the signal deviation state during the vehicle operation process. Embed the completed fault flags one by one, gradually increase the vehicle's fault monitoring status information, embed the currently detected fault status result for recording, and continuously generate a real-time fault detection and diagnosis scheme.

[0183] The intelligent fault detection and diagnosis system is used to execute the above-mentioned intelligent fault detection and diagnosis method. The system includes:

[0184] The operating state analysis module separates various signals during the operation process based on the normal state and abnormal state operation data of the vehicle, extracts characteristic signals and fault parameters, performs differential operations on the correlation relationships between the signals, merges the fault characteristic values, combines the characteristics by category, and obtains a multi-dimensional fault characteristic signal set;

[0185] The fault coding analysis module maps the characteristic parameters of the signal set according to the requirements of the fault coding based on the multi-dimensional fault characteristic signal set, matches the characteristic signals with the corresponding fault characteristic points in the coding space, calculates the matching relationship between the characteristic values and the spatial distribution, divides the boundary characteristics of the differential faults, and obtains the fault aggregation recognition characteristics;

[0186] The simulated fault expansion module calls the original data samples based on the fault aggregation recognition characteristics, adds the samples that meet the fault characteristics, performs multi-dimensional distribution analysis on the conforming samples, merges the expanded sample sets, and obtains the fault simulation characteristic structure;

[0187] The hierarchical fault judgment module sets multi-level fault judgment thresholds based on the fault simulation characteristic structure, compares the differential characteristics in the simulation characteristic structure with the thresholds of each level, sequentially screens the characteristic points of the differential levels, and merges them successively to generate a multi-level fault judgment structure;

[0188] The real-time fault monitoring module matches the real-time monitoring signal characteristics based on the multi-level fault judgment structure, analyzes the signal characteristics and the hierarchical boundaries, determines the matching situation of the fault levels, and embeds the fault state and the operation state signals into the configuration in real time to obtain the real-time fault detection and diagnosis solution.

[0189] The above is only the preferred embodiment of the present invention, and it does not limit the present invention in other forms. Any person skilled in the art may use the disclosed technical content to make changes or modifications into equivalent embodiments with equivalent changes and apply them to other fields. However, as long as it does not depart from the technical content of the technical solution of the present invention, any simple modification, equivalent change, and modification made to the above embodiments based on the technical essence of the present invention still belong to the protection scope of the technical solution of the present invention.

Claims

1. An intelligent fault detection and diagnosis method, characterized in that: The following steps are involved: Based on the vehicle's normal and abnormal operating data, the fault feature vectors are extracted and classified item by item, transmitted to the fault encoding layer through encoding, input into the latent space, and the feature points are clustered to obtain preliminary fault identification features; Based on the preliminary fault identification features, extracting fault potential features and extended samples through fault simulation samples, analyzing the spatial distribution differences between extended samples and fault features, screening extended samples associated with potential faults, generating a simulated fault sample set, verifying the simulated fault sample set from multiple angles, and generating a sample extension structure for fault simulation; The steps for obtaining the sample extension structure of the fault simulation are specifically as follows: Based on each fault feature of the simulated fault sample set, the feature angle, value range and distribution density are analyzed and extracted, the difference index and distribution weight of each feature are calculated, and a multi-angle analysis matrix of the fault features is generated; Based on the feature difference index and weight value of the fault feature multi-angle analysis matrix, the impact score of the feature under the differentiation angle is calculated using the formula: Calculate the normalized influence value of the feature and obtain the verification result of the feature, where: is the normalized influence value of the feature, and is the angle weight coefficient, and is the eigenvalue under the differentiation angle, Adjust the parameters for the difference, and is an additional adjustment parameter for the influence; Based on the verification results of the features, the screening results of the multi-angle analysis matrix are called, the feature sets of the correlation are merged, and the sample extension structure of the fault simulation is generated; Based on the sample extension structure of the fault simulation, the samples and abnormal density points are analyzed, and the fault features of the differentiated detection levels are screened by setting the multi-level detection threshold of the fault detection threshold layer, and a multi-level fault detection standard set is established. According to the multi-level fault detection standard set, the abnormal feature distribution of the detection level is matched to obtain the graded fault detection result; Based on the hierarchical fault detection results, the real-time fault monitoring flow is called to dynamically monitor the signal characteristics of the vehicle in operation, match the signal characteristics with the detection standards, judge the vehicle status, embed the fault identification into the monitoring configuration, gradually update the identification status, and obtain real-time fault detection and diagnosis solutions.

2. The intelligent fault detection and diagnosis method according to claim 1, characterized in that: The steps for obtaining the preliminary fault identification features are specifically as follows: Based on the normal and abnormal state operation data of the vehicle, the characteristic distribution of the vehicle in the normal and abnormal state is extracted, the operation parameters of speed, temperature, and vibration frequency are called and state-encoded, the state characteristics are hierarchically classified and processed, and a feature vector set is obtained; The information entropy analysis is performed on the feature vector set, by calculating the information entropy value of each feature vector and eliminating feature points whose information entropy value is lower than the set threshold, using the formula: Calculate the similarity values ​​between feature vectors and establish a feature homogeneity matrix, where: represents the similarity value between feature vectors, is the feature weight coefficient, represents the value of the eigenvector, is the feature difference parameter, For additional adjustment parameters; According to the feature homogeneity matrix, clustering logic is applied to group the feature points, and the cluster center features of each group are calculated to generate preliminary fault identification features.

3. The intelligent fault detection and diagnosis method according to claim 2, characterized in that: The steps of obtaining the simulated fault sample set are specifically as follows: Based on the preliminary fault identification features, feature parameters in the fault simulation samples are called for matching, potential feature values ​​of features in the fault simulation samples are extracted respectively, and the correlation between the potential features and the fault identification features is determined by calculating the differences between the features, so as to generate a potential fault feature set; According to the characteristic parameters in the potential fault feature set, the position distribution analysis is performed with the feature space of the extended sample to identify the difference distribution between the extended sample and the fault feature space, and the extended sample set associated with the potential fault is screened by setting the associated threshold, using the formula: Calculate the difference distribution value of the feature space to obtain the extended sample of potential fault associations, where represents the difference distribution value of the fault feature space, is the weight coefficient, and Represent the potential features of the fault and the feature values ​​in the extended sample, respectively. is the extended sample offset factor, is the standard deviation of the extended sample characteristics, Adjust parameters for offset; According to the extended samples associated with the potential fault, samples associated with the potential fault are screened by using difference distribution values, and each associated sample parameter is integrated to generate a simulated fault sample set.

4. The intelligent fault detection and diagnosis method according to claim 1, characterized in that: The steps for obtaining the multi-level fault detection standard set are specifically as follows: Based on the sample extension structure of the fault simulation, fault feature data is extracted from the sample data, difference values ​​are calculated with abnormal density points, and the calculation results are normalized to generate a fault feature difference vector; Based on the fault feature difference vector, the detection threshold layer of the multi-level fault detection standard set is set, and for each feature data, the formula is used: Calculate the fault feature difference value and generate the fault feature difference matrix, where: represents the fault feature difference value, represents the first eigenvalues, is the detection threshold set, represents the weight of the feature, is the characteristic adjustment coefficient; According to the fault feature difference matrix, it is determined whether each feature difference is greater than a detection threshold, fault features above the threshold are screened, features that do not exceed the threshold are marked as low-level features, and a multi-level fault detection standard set is established.

5. The intelligent fault detection and diagnosis method according to claim 4, characterized in that: The steps for obtaining the hierarchical fault detection results are specifically as follows: According to the abnormal feature distribution data of each level in the multi-level fault detection standard set, for each level of abnormal feature distribution, the matching detection level is determined in turn, and the initial graded fault detection feature is obtained by calculating the cumulative deviation of the abnormal feature distribution; Based on the initial hierarchical fault detection features, the cumulative deviation of the abnormal feature distribution is calculated, the fault detection level threshold is set, and the cumulative deviation is judged using the formula: Generate matching results of hierarchical fault detection features, where Indicates the matching degree of abnormal distribution deviation, represents the cumulative deviation of the initial classification fault detection features, is the set fault detection level threshold, is the difference in the cumulative value of the detection level in the previous step. , , are the weight parameters of the cumulative deviation respectively; The matching results of the hierarchical fault detection features are analyzed, hierarchical features whose abnormal distribution deviation exceeds the detection level threshold are selected, and hierarchical fault detection results are generated by calling the matching conditions and judgments of the fault detection features.

6. The intelligent fault detection and diagnosis method according to claim 5, characterized in that: The steps for obtaining the real-time fault detection and diagnosis solution are specifically as follows: Calling the fault level parameter in the hierarchical fault detection result, extracting the signal characteristic value in the vehicle operation process according to the real-time fault monitoring flow, performing normalization processing on the collected signal data by dynamically collecting the signal characteristics, calculating the deviation between the signal characteristic value and the standard value of the corresponding fault level, and generating a real-time fault signal deviation result; According to the real-time fault signal deviation result, the vehicle status judgment condition is called, and the monitoring threshold set by the deviation value is analyzed, and the formula is used: Calculate the normalized result of the signal deviation and determine whether it exceeds the threshold, and generate the vehicle state matching result, where: is the signal deviation state value, is the real-time signal deviation value, is the deviation threshold in the fault level parameter, is the initial deviation value of the hierarchical fault detection result, , , is the signal feature weight parameter; The vehicle status matching result is dynamically monitored, the current fault identification is updated according to the vehicle status, the fault identification is gradually embedded into the monitoring configuration, the real-time status of the vehicle fault is updated and recorded, and a real-time fault detection and diagnosis solution is generated.

7. An intelligent fault detection and diagnosis system, characterized in that: According to the intelligent fault detection and diagnosis method according to any one of claims 1 to 6, the system comprises: The running status analysis module separates various signals in the running process based on the normal and abnormal running data of the vehicle, extracts characteristic signals and fault parameters, performs difference calculation on the correlation between signals, merges the fault characteristic values, and combines the characteristics by category to obtain a multi-dimensional fault characteristic signal set; The fault coding analysis module maps the characteristic parameters of the signal set according to the requirements of fault coding based on the multi-dimensional fault characteristic signal set, matches the characteristic signal with the corresponding fault characteristic point in the coding space, calculates the matching relationship between the characteristic value and the spatial distribution, divides the boundary characteristics of the differentiated faults, and obtains the fault aggregation identification characteristics; The simulated fault expansion module calls the original data samples based on the fault aggregation identification features, adds samples that meet the fault features, performs multi-dimensional distribution analysis on the samples that meet the features, merges and expands the sample set, and obtains the fault simulation feature structure; The hierarchical fault judgment module sets multi-level fault judgment thresholds based on the fault simulation feature structure, compares the differentiated features in the simulation feature structure with the thresholds of each level, selects the feature points of the differentiated levels in turn, merges them one by one, and generates a multi-level fault judgment structure; The real-time fault monitoring module matches the real-time monitoring signal characteristics based on the multi-level fault judgment structure, analyzes the signal characteristics with the classification boundaries, determines the matching of the fault levels, embeds the fault status and operating status signals into the configuration in real time, and obtains a real-time fault detection and diagnosis solution.

Images