A method for online identification of display faults of vehicle instrument panel

Abstract

The present application provides a method for online identification of display faults of a vehicle dashboard, which relates to the technical field of vehicle dashboards, including: obtaining N indicator lights of a vehicle dashboard, and real-time monitoring to obtain multi-dimensional working signal flows of the N indicator lights; performing feature extraction and dimensionality reduction screening to obtain N indicator light working key feature sets; calculating the degree of faults of the N indicator light working key feature sets; comparing and distinguishing the degree of faults of the N indicator lights to generate M indicator light fault parameters; performing fault data mining to obtain a dashboard fault database, and performing correlation analysis on the dashboard fault database; performing fault diagnosis analysis on the M indicator light fault parameters to determine vehicle dashboard fault diagnosis information. The present application can solve the technical problem in the prior art that it is impossible to accurately identify the mutual influence and propagation path between multiple indicator light faults, and achieve the technical effect of improving fault diagnosis accuracy and accelerating maintenance efficiency.

Description

Technical Field

[0001] The present application relates to the technical field of vehicle instrument panels, and in particular to an online identification method for display faults of a vehicle instrument panel. Background Art

[0002] With the development of modern automobile technology, the vehicle dashboard plays a vital role in monitoring the vehicle's operating status. The various indicator lights on the vehicle dashboard can reflect the vehicle's various performance indicators and working status in real time, helping the driver to find problems in time and take corresponding actions. However, with the increasing degree of automobile electrification, the dashboard indicator lights are diverse, and the types of faults involved are becoming increasingly complex.

[0003] At present, the existing fault diagnosis systems mostly rely on fixed fault databases and simple rule matching. Although they can provide some basic fault judgments to a certain extent, they are often unable to provide accurate diagnosis for various complex and dynamic fault situations in practical applications. For example, by analyzing the abnormal state of a single indicator light in the prior art, it may not be possible to identify the mutual influence and correlation between multiple indicator light failures. Due to the lack of in-depth analysis of the fault, the diagnosis results cannot effectively reflect the root cause of the fault, especially when the indicator light failures interact with each other or the fault generation chain is long, the prior art is prone to missed or misjudgment, affecting maintenance efficiency and accuracy.

[0004] In summary, the existing technology has technical problems such as the lack of in-depth analysis of faults and comprehensive consideration of the correlation between faults, which leads to the inability to accurately identify the mutual influence and propagation path between multiple indicator light faults, further affecting the accuracy of fault diagnosis and maintenance efficiency, and prone to missed or misjudgment in complex and dynamic fault scenarios, thereby increasing the difficulty of vehicle maintenance. Summary of the invention

[0005] The purpose of this application is to provide a method for online identification of display faults of a vehicle dashboard, so as to solve the technical problem in the prior art that, due to the lack of in-depth analysis of faults and comprehensive consideration of the correlation between faults, the mutual influence and propagation path between multiple indicator light faults cannot be accurately identified, which further affects the accuracy of fault diagnosis and maintenance efficiency, and is prone to missed judgments or misjudgments in complex and dynamic fault scenarios, thereby increasing the difficulty of vehicle maintenance.

[0006] In view of the above problems, the present application provides an online identification method for display faults of a vehicle instrument panel, comprising: obtaining N indicator lights of a vehicle instrument panel, and installing a sensor group at the rear end of the N indicator lights, and obtaining a multi-dimensional working signal flow of the N indicator lights through real-time monitoring of the sensor group; performing feature extraction and dimensionality reduction screening on the multi-dimensional working signal flow of the N indicator lights to obtain a set of key working features of the N indicator lights; determining an indicator light fault evaluation index set according to an instrument panel display application standard, and performing fault degree calculation on the key working feature set of the N indicator lights based on the indicator light fault evaluation index set to obtain the working fault degree of the N indicator lights; comparing and judging the working fault degree of the N indicator lights according to a preset fault degree judgment threshold to generate M indicator light fault parameters, wherein M≤N; performing fault data mining based on the N indicator lights to obtain an instrument panel fault database, performing correlation analysis on the instrument panel fault database, and constructing an instrument panel fault logic diagram; performing fault diagnosis analysis on the fault parameters of the M indicator lights using the instrument panel fault logic diagram to determine vehicle instrument panel fault diagnosis information.

[0007] The technical solution provided in the present application has at least the following technical effects or advantages: by acquiring N indicator lights of a vehicle dashboard, and installing a sensor group at the rear end of the N indicator lights, the multi-dimensional working signal flow of the N indicator lights is acquired by real-time monitoring through the sensor group; feature extraction and dimension reduction screening are performed on the multi-dimensional working signal flow of the N indicator lights to obtain N indicator light working key feature sets; according to the dashboard display application standard, the indicator light fault evaluation index set is determined, and the fault degree of the N indicator light working key feature set is calculated based on the indicator light fault evaluation index set to obtain the N indicator light working fault degrees; the N indicator light working fault degrees are compared and judged according to a preset fault degree judgment threshold to generate M indicator light fault parameters, wherein M≤N; fault data mining is performed based on the N indicator lights to obtain a dashboard fault database, and the dashboard fault database is associated with the dashboard fault database to construct a dashboard fault logic diagram; the dashboard fault logic diagram is used to perform fault diagnosis analysis on the M indicator light fault parameters to determine the vehicle dashboard fault diagnosis information, that is, by achieving the technical goal of accurate diagnosis based on multi-indicator fault association rule mining and logic analysis, the technical effect of improving fault diagnosis accuracy and accelerating maintenance efficiency is achieved.

[0008] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented according to the contents of the specification, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are specifically cited below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in the present application or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0010] Figure 1 A schematic diagram of a flow chart of a method for online identification of display faults of a vehicle dashboard of the present application;

[0011] Figure 2 The present invention is a flow chart of obtaining a set of key working characteristics of N indicator lights in an online identification method for display faults of a vehicle instrument panel. DETAILED DESCRIPTION

[0012] This application provides a method for online identification of display faults on a vehicle dashboard, solving the technical problem that the prior art cannot accurately identify the mutual influence and propagation path between multiple indicator light faults due to the lack of in-depth analysis of faults and comprehensive consideration of the correlation between faults, further affecting the accuracy of fault diagnosis and maintenance efficiency, and easily misses or misjudges in complex and dynamic fault scenarios, thereby increasing the difficulty of vehicle maintenance. The technical goal of accurate diagnosis based on multi-indicator light fault association rule mining and logical analysis is achieved, achieving the technical effect of improving fault diagnosis accuracy and accelerating maintenance efficiency.

[0013] Below, the technical solutions in the present application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments of the present application. It should be understood that the present application is not limited to the example embodiments described herein. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application. It should also be noted that, for the convenience of description, only the parts related to the present application are shown in the accompanying drawings, rather than all of them.

[0014] Please see attached Figure 1The present application provides a method for online identification of display faults of a vehicle dashboard, which specifically comprises the following steps:

[0015] Step 1: Obtain N indicator lights on the vehicle dashboard, and install a sensor group at the rear end of the N indicator lights to obtain multi-dimensional working signal flows of the N indicator lights through real-time monitoring by the sensor group.

[0016] Specifically, obtaining N indicator lights on the vehicle dashboard means obtaining N indicator lights from the dashboard for monitoring. The indicator lights are used to display various working conditions of the vehicle, such as fuel level, temperature, speed, etc. Among them, N is a positive integer, which represents the number of all indicator lights. In order to obtain the working status of the indicator lights in real time, a sensor group is installed at the rear end of each indicator light. The sensor group is a system composed of multiple sensors that can obtain multiple signal sources at the same time. The function of the sensor group is to monitor and collect the multi-dimensional working signal flow of the indicator lights in real time, including data in multiple dimensions such as switch signals, brightness, frequency, etc., to help monitor and analyze the health status of the vehicle. For example, when the fuel level indicator light in the indicator light is on, the specific data of the fuel level change is obtained through the sensor group, and further analysis is performed on whether the fuel level is lower than a preset value, so as to give a warning or prompt.

[0017] Step 2: Perform feature extraction and dimensionality reduction screening on the multi-dimensional working signal streams of the N indicator lights to obtain a set of key working features of the N indicator lights.

[0018] Specifically, representative features are extracted from the signal stream of each indicator light, and key data points that can reflect the working status of the indicator light are identified, such as changes in the brightness of the indicator light, the frequency of the strobe light, etc., to help determine the current working status and health of the indicator light. Next, dimensionality reduction screening is performed to convert the multi-dimensional signal features into a small number of new features that contain the most important information, remove redundant information and noise, retain valuable features, reduce the complexity of the data, and improve the efficiency of subsequent processing. The key working feature set of N indicator lights is obtained, which can more accurately and concisely describe the status of the indicator light, facilitating subsequent analysis and decision-making.

[0019] Step three: according to the instrument panel display application standard, determine the indicator light fault evaluation index set, calculate the fault degree of the N indicator light working key feature sets based on the indicator light fault evaluation index set, and obtain the N indicator light working fault degrees.

[0020] Specifically, the instrument panel display application standard refers to the display requirements and performance specifications of the instrument panel in different scenarios, including the color, brightness, response speed and flashing frequency of the indicator light, as a benchmark for measuring the working status of the indicator light. According to the instrument panel display application standard, the indicator light fault evaluation index set is determined, and specific indicators are selected to determine whether the indicator light is faulty, such as numerical response compliance, color brightness compliance and flashing frequency compliance, which respectively represent whether the timeliness of the signal change, the accuracy of the display brightness and the flashing frequency of the indicator light are consistent with the standard when it is working.

[0021] Based on each indicator of the indicator light fault evaluation indicator set, the fault degree of the N indicator light working key feature sets is calculated. By comparing the data in the working feature set with the evaluation criteria, the fault degree of each indicator light can be quantified, and the working fault degree of the N indicator lights can be obtained to describe the overall working status of the indicator light. The working fault degree of the indicator light indicates that the greater the fault risk of the indicator light, the smaller it is.

[0022] Step 4: Compare and judge the working fault degrees of the N indicator lights according to a preset fault degree judgment threshold, and generate M indicator light fault parameters, where M≤N.

[0023] Specifically, the preset fault degree discrimination threshold is the set judgment standard. The setting of the preset fault degree discrimination threshold can be based on experimental data or industry standards, and can be obtained by technical personnel in this field through customized settings according to actual conditions. According to the preset fault degree discrimination threshold, when analyzing the indicator light failure, the working fault degrees of N indicator lights are compared and judged, and the fault degree of each indicator light is compared with the above preset threshold one by one to determine whether it exceeds the threshold range, which is used to distinguish whether the indicator light has an abnormal working state, and M indicator lights that are identified as faulty are screened out to generate M indicator light failure parameters. M is the number of faulty indicator lights obtained after actual comparison. Since only some indicator lights may exceed the threshold, the value of M is less than or equal to the total number N, indicating that under any circumstances, the number of faulty indicator lights will not exceed the total number of indicator lights.

[0024] Step 5: Perform fault data mining based on the N indicator lights to obtain a dashboard fault database, perform correlation analysis on the dashboard fault database, and construct a dashboard fault logic diagram.

[0025] Specifically, fault data mining is performed based on N indicator lights to obtain the rules and patterns of faults and reveal the commonalities between different indicator light faults. For example, when the brightness of a certain indicator light is abnormal, it is usually accompanied by abnormal numerical response. Through mining, the potential association of indicator light faults can be found, and a dashboard fault database containing all indicator light fault records can be obtained. The logical relationship between indicator light faults can be established, thereby helping to identify potential fault propagation paths. The fault relationship obtained through association analysis is presented in a graphical way to construct a dashboard fault logic diagram.

[0026] Step 6: Use the instrument panel fault logic diagram to perform fault diagnosis analysis on the M indicator light fault parameters to determine vehicle instrument panel fault diagnosis information.

[0027] Specifically, the fault logic diagram of the instrument panel is used to perform fault diagnosis and analysis on the fault parameters of M indicator lights, and the root cause or propagation path of the fault is inferred based on the correlation in the diagram to further diagnose and analyze the fault parameters of the M indicator lights that have been identified. For example, if the flashing frequency of a certain indicator light is abnormal, and according to the fault logic diagram, the fault of this indicator light may be related to the abnormal numerical response of another indicator light, then it is inferred that the two faults may be caused by the same circuit fault. Through the diagnosis and analysis of the fault parameters, the final fault diagnosis conclusion is obtained, and the vehicle instrument panel fault diagnosis information is determined, including the specific fault location, type and possible cause, to provide a basis for maintenance and treatment.

[0028] The method for online identification of display faults of a vehicle instrument panel can achieve the technical goal of accurate diagnosis based on multi-indicator light fault association rule mining and logic analysis, and achieve the technical effect of improving fault diagnosis accuracy and accelerating maintenance efficiency.

[0029] For further information, please see the attached Figure 2 The present application also includes: filtering, cleaning and amplifying and enhancing the N indicator light multi-dimensional working signal streams respectively to obtain N available indicator light multi-dimensional working signal streams; extracting related features of the N available indicator light multi-dimensional working signal streams according to the indicator light application scenarios to obtain N indicator light related working feature sets; decentralizing the N indicator light related working feature sets respectively to obtain N covariance matrices, and performing principal component solution selection based on the N covariance matrices to obtain N feature principal component information; projecting and reducing the dimension of the N indicator light related working feature sets based on the N feature principal component information to obtain the N indicator light working key feature sets.

[0030] Specifically, the multi-dimensional working signal streams of N indicator lights are filtered, cleaned, and amplified and enhanced to effectively improve the reliability of the data and obtain N usable multi-dimensional working signal streams of indicator lights. Filtering and cleaning are used to remove noise or irrelevant information in the signal. Filtering refers to the use of specific algorithms or tools to retain the important parts of the signal, such as low-pass filters to remove high-frequency interference signals; cleaning is to remove outliers or invalid data. Next, the process of amplification and enhancement is used to increase the strength or resolution of the signal to ensure that the data quality is more suitable for subsequent processing.

[0031] By analyzing the correlation in multiple scenarios, the correlation features of the multi-dimensional working signal streams of N available indicator lights are extracted according to the indicator light application scenarios. The relevant features in the pre-processed signal streams are extracted for the use of different indicator lights in actual scenarios, which are used for subsequent modeling and analysis to help accurately describe the relationship between signals. The application scenarios of indicator lights refer to their functions in different situations. For example, there may be a correlation between brake lights and vehicle speed signals, while fuel level lights may be related to mileage.

[0032] Decentralize the N indicator light associated work feature sets respectively, subtract the average value from each indicator light associated work feature value, make the center point of the data distribution zero, and eliminate the influence of the difference in the magnitude of the feature value on the analysis. Next, calculate the covariance matrix of each indicator light associated work feature set. The covariance matrix is ​​a mathematical tool to describe the relationship between different features, in which each element represents the covariance size between two features. Then, principal component analysis is a mathematical dimensionality reduction method used to find the most important direction of change in the data. The covariance matrix is ​​solved and selected for the principal components. By selecting the principal components, the most important information in the data is retained, while reducing redundancy, and the principal component information of N features is obtained.

[0033] Based on the information of N characteristic principal components, the projected dimension reduction and screening of N indicator light-related working feature sets can be performed to retain important features and eliminate minor features, which can further simplify the data dimension, reduce the computational complexity, and ensure that the retained information is sufficient to describe the key characteristics of the indicator light. For example, a feature set originally contained ten dimensions, but after dimensionality reduction, only the two most significant features were retained, representing brightness changes and frequency fluctuations, which not only reduced the data size, but also improved the efficiency of subsequent analysis.

[0034] Through filtering and cleaning, amplification and enhancement, correlation feature extraction, decentralization, covariance matrix calculation, principal component analysis and dimensionality reduction screening, the multi-dimensional signals of N indicator lights are converted into a key working feature set with clear structure and concentrated information, which not only improves the data analysis efficiency, but also enhances the ability to accurately judge the vehicle status.

[0035] Furthermore, the present application also includes: arranging the N indicator light working key feature sets according to timing information to obtain N indicator light working timing feature sets; according to the indicator light fault evaluation index set, the indicator light fault evaluation index set includes numerical response compliance, color brightness compliance and flashing frequency compliance; evaluating and calculating the N indicator light working timing feature sets respectively according to the indicator light fault evaluation index set to obtain N indicator light evaluation index compliance sets; performing fault degree evaluation on the N indicator light working key feature sets based on the N indicator light evaluation index compliance sets to obtain N indicator light working fault degrees.

[0036] Specifically, the key feature sets of N indicator lights are arranged according to the timing information. The timing information refers to the changes of the indicator light signal on the time axis. The timing arrangement is to be able to capture the trend of the indicator light state change and obtain its evolution over time. For example, the brightness of the indicator light gradually increases in a certain period of time and then decreases in another period of time, which helps to determine whether the indicator light has abnormal working behavior. The working timing feature set of N indicator lights is formed, which contains the key status information of the indicator light at different time points.

[0037] The indicator light fault assessment index set includes numerical response compliance, color brightness compliance, and flashing frequency compliance. The fault assessment index set refers to the standard used to evaluate whether the indicator light is in normal working condition. The numerical response compliance measures the consistency between the indicator light signal strength and the preset value, the color brightness compliance evaluates whether the color or brightness of the indicator light meets the set standard, and the flashing frequency compliance is used to detect whether the flashing frequency of the indicator light meets the expectation. The indicators help determine whether the indicator light has a fault and ensure that it is in normal working condition.

[0038] According to the indicator light fault evaluation index set, the N indicator light working timing feature sets are evaluated and calculated respectively. By calculating the compliance of the indicator light's numerical response, brightness, and strobe frequency at each timing point with the standard, the N indicator light evaluation index compliance set is obtained, and then quantified whether the indicator light is within the normal working range.

[0039] Based on the compliance set of N indicator light evaluation indicators, the fault degree of N indicator light working key feature sets is evaluated. Through weighted summation or other mathematical methods, the various evaluation indicators are combined to obtain the working fault degree of N indicator lights, and it is judged whether there are serious problems in the operation of the indicator lights.

[0040] By arranging the working characteristics of N indicator lights in time sequence and comparing them with the fault assessment index set, the working fault degree of each indicator light is obtained, which can accurately reflect whether the indicator light is faulty or about to fail, help to find and repair the problem in time, and thus ensure the safe operation of the vehicle.

[0041] Furthermore, the present application also includes: constructing N indicator light working characteristic matrices according to the N indicator light evaluation index compliance sets: ;in, is the working characteristic matrix of the nth indicator light, T is the number of timing characteristics, R T is the numerical response conformity under the Tth time series, L T is the color brightness conformity at the Tth time sequence, F T is the flashing frequency compliance under the Tth timing; the N indicator light working characteristic matrices are respectively forward-processed and standardized to obtain N standard indicator light working characteristic matrices; a fault degree evaluation function is constructed, and the fault degree of the N standard indicator light working characteristic matrices is evaluated based on the fault degree evaluation function to obtain the working fault degree of the N indicator lights.

[0042] Specifically, based on the compliance set of N indicator light evaluation indicators, an indicator light working characteristic matrix of each indicator light in different time series is constructed, where each row of the matrix represents different timing information, and each column represents a specific evaluation indicator (such as numerical response, brightness, frequency, etc.), which is used to reflect the performance of the indicator light in different time periods and help analyze the working status of the indicator light.

[0043] Where Rn is the nth indicator light working characteristic matrix, each matrix contains multiple time series evaluation information. T is the number of time series characteristics, that is, the number of time points involved in the analysis process. T is the numerical response conformity under the Tth time sequence. T is the color brightness conformity at the Tth time sequence. T is the flicker frequency compliance under the Tth timing. T , L T and F T They represent whether the numerical response, color brightness and flashing frequency of the indicator light meet the preset standards in each timing sequence.

[0044] The N indicator light working characteristic matrices are processed by forward and standardization respectively. Forward processing is to adjust the data to a certain range or direction to make it easier to analyze later. Standardization is to convert the characteristic values ​​of different scales or magnitudes into a unified standard so that they can be compared and processed at the same scale. After forward and standardization processing, the working characteristic matrices of all indicator lights will be within a unified standard range, which is convenient for analysis and evaluation, and N standard indicator light working characteristic matrices are obtained.

[0045] A fault degree evaluation function is constructed, and the fault degree of N standard indicator light working characteristic matrices is evaluated based on the fault degree evaluation function. The fault degree evaluation function is a mathematical model used to calculate the fault degree of the indicator light based on the standard matrix of the indicator light working characteristics, and then obtain the working fault degree of N indicator lights. The fault degree evaluation function evaluates whether the indicator light is faulty based on factors such as the indicator light's compliance and working state fluctuations.

[0046] Furthermore, the present application also includes: the fault degree evaluation function is specifically: ,in ;

[0047] ;

[0048] ;in, is the fault degree score of the nth indicator light, is the fault degree score of the i-th time series feature, is the decision weight of the i-th time series feature, , and They are the weights of numerical response compliance, color brightness compliance, and flashing frequency compliance, which can be determined according to the degree of influence on the indicator light failure. , and are the maximum values ​​of the elements in each matrix column, , and are the minimum values ​​of the elements of each matrix column, , and It is the numerical response conformity, color brightness conformity, and flicker frequency conformity under the i-th timing.

[0049] Specifically, the fault degree evaluation function is used to comprehensively analyze the characteristic data of each indicator light and obtain its fault degree. Wherein, Bn is the fault degree score of the nth indicator light, which is used to quantify the overall fault degree of a certain indicator light. is the fault degree score of the ith timing feature, indicating the specific fault score of the nth indicator light under the ith timing. The greater the fault degree, the greater the fault degree score, which means that the distance between the conformity of the ith timing feature and the maximum value of the elements of each matrix column number and the distance from the minimum value of the elements of each matrix column number is greater, and the degree of conformity with the normal operation of the indicator light is lower, and vice versa. is the decision weight of the i-th time series feature, which indicates the importance of different time series features when calculating the fault degree. The weight distribution is determined according to the impact of the feature on the indicator light fault.

[0050] , and The weights of the numerical response compliance, color brightness compliance, and flashing frequency compliance can be determined according to the degree of influence on the indicator light failure, indicating the importance distribution of the three main indicators in the evaluation process. The determination of the weight depends on the influence of each compliance in the actual scenario.

[0051] , and are the maximum values ​​of the elements in each matrix column, , and They are the minimum values ​​of the elements in each matrix column, indicating that in order to normalize the data, the range of the elements in each column of the matrix is ​​determined so that the values ​​can be standardized to the same scale.

[0052] , and It is the numerical response compliance, color brightness compliance, and flashing frequency compliance at the i-th timing, indicating the specific compliance degree of different indicators at each timing point, which is calculated according to the actual working status of the indicator light.

[0053] Furthermore, the present application also includes: assigning identifiers to the N indicator lights to obtain N indicator light identifiers; performing association mapping on the instrument panel fault database according to the N indicator light identifiers to obtain N indicator light fault association data sets; using an association rule mining algorithm to perform rule mining and fault identification on the N indicator light fault association data sets to obtain an indicator light fault rule pattern set; using the N indicator light identifiers as nodes, performing cascade analysis on the indicator light fault rule pattern set to construct the instrument panel fault logic diagram.

[0054] Specifically, identifiers are assigned to the N indicator lights, and each indicator light is assigned a unique number or label to obtain N indicator light identifiers so that the indicator lights can be distinguished and identified in subsequent operations. The indicator light identifier may be a combination of numbers, letters, or other symbols, which is used to indicate the identity of the indicator light and helps to manage and track the status of each indicator light.

[0055] The instrument panel fault database is associated and mapped according to the N indicator light identifiers to obtain N indicator light fault association data sets, and the working status of each indicator light and the fault information are matched one by one to obtain the fault information data set of each indicator light. For example, the fault data of indicator light one may include color brightness non-compliant, slow numerical response, etc., while the fault data of indicator light two may include excessive flashing frequency, etc., to ensure that no abnormal situation of any indicator light is missed.

[0056] The association rule mining algorithm is used to perform rule mining and fault identification on N indicator light fault association data sets, analyze the inherent rules in the indicator light fault data set, identify the common patterns and rules of indicator light failures, and obtain the indicator light failure rule pattern set. The association rule mining algorithm extracts effective rules from the commonalities in a large amount of data, such as certain faults often appear on certain specific indicator lights at the same time, or appear frequently in certain time periods.

[0057] Taking N indicator light identifiers as nodes, the indicator light fault rule pattern set is cascaded and analyzed to build a dashboard fault logic diagram. The mutual impact of faults and the fault propagation path between different indicator lights are obtained. For example, the failure of a certain indicator light may cause problems in other indicators. The propagation of the fault is predicted, so that measures can be taken in advance to repair it.

[0058] Furthermore, the present application also includes: using an association rule mining algorithm to perform fault association rule mining on the N indicator light fault association data sets to obtain multiple indicator light fault association rules; setting a rule confidence threshold according to rule mining requirements and indicator light fault data characteristics; performing rule filtering on the multiple indicator light fault association rules based on the rule confidence threshold to obtain an available indicator light fault association rule set; performing fault identification on the N indicator light fault association data sets based on the available indicator light fault association rule set to obtain an indicator light fault rule pattern set.

[0059] Specifically, association rule mining algorithms (such as Apriori and FP-Growth) are used to mine fault association rules of N indicator light fault association data sets, extract the rules and patterns existing in the indicator light fault data sets, and then find the association between indicator light faults, and then identify potential fault association patterns, thereby providing a basis for subsequent fault prediction and troubleshooting, including the failure of a single indicator light, the joint failure of multiple indicator lights, or failures under specific conditions. For example, when the color brightness of the indicator light does not meet the standard, it may cause similar failures in other indicator lights.

[0060] According to the rule mining requirements and the characteristics of indicator light fault data, the rule confidence threshold is set to judge the reliability of the rule. Confidence refers to the probability of the rule occurring. If the confidence of the rule is high, it means that the possibility of the rule occurring is high, and vice versa. By setting the rule confidence threshold, you can avoid filtering out invalid rules with too low confidence.

[0061] Based on the rule confidence threshold, multiple indicator light fault association rules are filtered to screen out the fault association rules that meet the standards and have higher confidence, remove unreliable rules, retain the fault modes that are more likely to occur in actual situations, and obtain a set of available indicator light fault association rules to ensure that the obtained fault rules are more reliable and accurate.

[0062] Through the available indicator light fault association rule set, fault identification is performed on N indicator light fault association data sets, and the indicator lights that may fail are identified. The potential failure modes can be effectively identified, and the indicator light failure rule pattern set is obtained. An effective failure mode rule set is generated, and then the manifestation and law of the indicator light failure are obtained, providing a basis for subsequent maintenance and improvement.

[0063] Furthermore, the present application also includes: matching the N indicator light fault association data sets based on the available indicator light fault association rule set, and recording the association rule indicator light fault data sets; performing fault identification classification and association relationship integration on the association rule indicator light fault data sets to obtain the indicator light fault rule pattern set.

[0064] Specifically, the N indicator light fault association data sets are matched one by one using the screened indicator light fault association rule set with high confidence, and the conditions in the rules are compared with the characteristics of the indicator light fault data one by one to determine whether the N indicator light fault association data sets meet the rules. When the match is successful, the relevant data is recorded to form an association rule indicator light fault data record set.

[0065] The association rule indicator light fault data set is used for fault identification classification and association relationship integration. By analyzing the matched indicator light fault data record set, the faults are divided into different categories according to specific standards, the types of faults are identified, and the types of faults are integrated according to the association relationships described in the rules. The potential relationship between fault categories is clearly expressed. For example, it is found that an indicator light with abnormal frequency may also be accompanied by abnormal brightness, thus forming a new association pattern. Finally, a pattern set containing all classified faults and association relationships is output.

[0066] In summary, the method for online identification of display faults of a vehicle instrument panel provided by the present application has the following technical effects: by acquiring N indicator lights of a vehicle instrument panel, and installing a sensor group at the rear end of the N indicator lights, the multi-dimensional working signal stream of the N indicator lights is acquired through real-time monitoring by the sensor group; feature extraction and dimensionality reduction screening are performed on the multi-dimensional working signal stream of the N indicator lights to obtain a set of key working features of the N indicator lights; according to the instrument panel display application standard, an indicator light fault evaluation index set is determined, and the fault degree of the key working feature set of the N indicator lights is calculated based on the indicator light fault evaluation index set to obtain the working fault degree of the N indicator lights; The working fault degrees of the N indicator lights are compared and judged according to a preset fault degree judgment threshold, and M indicator light fault parameters are generated, wherein M≤N; fault data mining is performed based on the N indicator lights to obtain an instrument panel fault database, and correlation analysis is performed on the instrument panel fault database to construct an instrument panel fault logic diagram; fault diagnosis analysis is performed on the M indicator light fault parameters using the instrument panel fault logic diagram to determine vehicle instrument panel fault diagnosis information, that is, by achieving the technical goal of accurate diagnosis based on multi-indicator light fault association rule mining and logic analysis, the technical effect of improving fault diagnosis accuracy and accelerating maintenance efficiency is achieved.

[0067] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

[0068] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application belong to the scope of the present application and its equivalent technology, the present application is also intended to include these modifications and variations.

Claims

1. A method for online identification of display faults of a vehicle instrument panel, characterized in that: Methods include: Obtain N indicator lights on the vehicle dashboard, and install a sensor group at the rear end of the N indicator lights to obtain multi-dimensional working signal flows of the N indicator lights through real-time monitoring by the sensor group; Performing feature extraction and dimensionality reduction screening on the multi-dimensional working signal streams of the N indicator lights to obtain a set of key working features of the N indicator lights; Determine an indicator light fault evaluation index set according to the instrument panel display application standard, calculate the fault degree of the N indicator light working key feature sets based on the indicator light fault evaluation index set, and obtain the N indicator light working fault degrees; Compare and judge the working fault degrees of the N indicator lights according to a preset fault degree judgment threshold value to generate M indicator light fault parameters, where M≤N; Perform fault data mining based on the N indicator lights to obtain a dashboard fault database, perform correlation analysis on the dashboard fault database, and construct a dashboard fault logic diagram; Using the instrument panel fault logic diagram to perform fault diagnosis analysis on the M indicator light fault parameters to determine vehicle instrument panel fault diagnosis information; The obtaining of the working fault degrees of N indicator lights comprises: Arrange the N indicator light working key feature sets according to the timing information to obtain N indicator light working timing feature sets; According to the indicator light fault evaluation index set, the indicator light fault evaluation index set includes a numerical response conformity, a color brightness conformity, and a flashing frequency conformity; According to the indicator light fault evaluation index set, the N indicator light working timing feature sets are evaluated and calculated respectively to obtain N indicator light evaluation index compliance sets; Based on the N indicator light evaluation index compliance sets, the N indicator light working key feature sets are evaluated for failure degree to obtain the N indicator light working failure degrees; According to the N indicator light evaluation index compliance sets, N indicator light working characteristic matrices are constructed: ； Among them, Rn is the working characteristic matrix of the nth indicator light, T is the number of timing characteristics, R T is the numerical response conformity under the Tth time series, L T is the color brightness conformity at the Tth time sequence, F T is the flicker frequency compliance at the Tth timing; Performing forward and standardization processing on the N indicator light working characteristic matrices respectively to obtain N standard indicator light working characteristic matrices; Constructing a fault degree evaluation function, and performing a fault degree evaluation on the N standard indicator light working characteristic matrices based on the fault degree evaluation function to obtain the working fault degrees of the N indicator lights; The fault degree evaluation function is specifically: ,in, ; ； ； Among them, B n is the fault degree score of the nth indicator light, is the fault degree score of the i-th time series feature, is the decision weight of the i-th time series feature, , and They are the weights of numerical response compliance, color brightness compliance, and flashing frequency compliance, which can be determined according to the degree of influence on the indicator light failure. , and are the maximum values ​​of the elements in each matrix column, , and are the minimum values ​​of the elements of each matrix column, , and It is the numerical response conformity, color brightness conformity, and flicker frequency conformity under the i-th timing.

2. The method for online identification of display faults of a vehicle instrument panel according to claim 1, characterized in that: The obtained N indicator light working key feature sets include: Performing filtering, cleaning and amplification and enhancement preprocessing on the N indicator light multidimensional working signal streams respectively to obtain N available indicator light multidimensional working signal streams; Extracting correlation features from the multi-dimensional working signal streams of the N available indicator lights according to the indicator light application scenario to obtain a set of correlation working features of the N indicator lights; Decentralizing the N indicator light-associated work feature sets respectively to obtain N covariance matrices, and performing principal component solving based on the N covariance matrices to obtain N characteristic principal component information; Based on the N characteristic principal component information, projection dimension reduction screening is performed on the N indicator light associated working feature sets to obtain the N indicator light working key feature sets.

3. The method for online identification of display faults of a vehicle instrument panel according to claim 1, characterized in that: The construction of the dashboard fault logic diagram includes: Assigning identifiers to the N indicator lights to obtain N indicator light identifiers; Associating and mapping the instrument panel fault database according to the N indicator light identifiers to obtain N indicator light fault association data sets; Use an association rule mining algorithm to perform rule mining and fault identification on the N indicator light fault association data sets to obtain an indicator light fault rule pattern set; The N indicator light identifiers are used as nodes, and a cascade analysis is performed on the indicator light fault rule pattern set to construct the instrument panel fault logic diagram.

4. A method for online identification of display faults of a vehicle instrument panel as claimed in claim 3, characterized in that: The step of obtaining a set of indicator light fault rule patterns includes: Use an association rule mining algorithm to perform fault association rule mining on the N indicator light fault association data sets to obtain multiple indicator light fault association rules; Set the rule confidence threshold according to the rule mining requirements and indicator light fault data characteristics; Performing rule filtering on the plurality of indicator light fault association rules based on the rule confidence threshold to obtain an available indicator light fault association rule set; Fault identification is performed on the N indicator light fault association data sets based on the available indicator light fault association rule set to obtain an indicator light fault rule pattern set.

5. The method for online identification of display faults of a vehicle instrument panel as claimed in claim 4, characterized in that: The step of obtaining a set of indicator light fault rule patterns includes: Matching the N indicator light fault association data sets based on the available indicator light fault association rule set, and recording the association rule indicator light fault data set; The indicator light fault rule pattern set is obtained by performing fault identification classification and association relationship integration on the association rule indicator light fault data set.

Images