Fault diagnosis and repair method and system in industrial pilot plant test platform tool

Abstract

The invention relates to the technical field of fault diagnosis, and provides a fault diagnosis and repair method and system in an industrial pilot test platform tool, and the method comprises the steps: calculating a feature dissimilation value between working condition features, and recognizing working condition fault data in tool working condition data; querying a tool adaptation system corresponding to the pilot plant test platform tool, scheduling a system record log corresponding to the tool adaptation system, mining system warning information from the system record log, calculating a warning priority corresponding to the system warning information, and analyzing key warning factors in the system warning information; analyzing a system fault corresponding to the tool adaptation system, and analyzing a tool fault corresponding to the pilot plant test platform tool; and calculating a fault effect value corresponding to the steady-state fault, formulating a fault repair scheme corresponding to the steady-state fault, and executing fault diagnosis processing corresponding to the pilot-scale test platform tool to obtain a diagnosis result. The invention aims to improve the accuracy of fault diagnosis and repair in the industrial pilot plant test platform tool.

Description

Technical Field

[0001] The present invention relates to the technical field of fault diagnosis, and in particular to a fault diagnosis and repair method and system in an industrial pilot platform tool. Background Art

[0002] Industrial pilot platform tools play a vital role in the development of modern industries. They can accelerate the research and development and promotion of new products and new technologies. With the continuous development of the industry, the demand for pilot platform tools is also growing. However, the current maintenance and fault diagnosis and repair of industrial pilot platform tools are relatively lagging behind. The current fault detection of pilot platform tools mainly relies on simple tool status prompts or limited system log information for analysis. This detection method is relatively limited and one-sided. Many potential faults cannot be discovered in a timely and accurate manner. It often happens that the fault seems to be solved but the tool still cannot operate normally. Repeated fault analysis and repair attempts are required, which greatly affects the efficiency and reliability of the pilot platform, resulting in low accuracy and efficiency of fault diagnosis and repair of pilot platform tools. Therefore, there is an urgent need for a method that can effectively improve the accuracy of fault diagnosis and repair of industrial pilot platform tools. Summary of the invention

[0003] The present invention provides a method and system for fault diagnosis and repair in an industrial pilot platform tool, the main purpose of which is to improve the accuracy of fault diagnosis and repair in the industrial pilot platform tool.

[0004] To achieve the above-mentioned purpose, the present invention provides a fault diagnosis and repair method in an industrial pilot platform tool, comprising:

[0005] Obtain tool operating data corresponding to the pilot platform tool of the industry, extract the operating condition features corresponding to the tool operating condition data, calculate the feature alienation value between the operating condition features, and identify the operating condition fault data in the tool operating condition data based on the feature alienation value, wherein the calculation of the feature alienation value between the operating condition features includes: vectorizing the operating condition features to obtain an operating condition feature vector, calculating the average feature vector corresponding to the operating condition feature vector, combining the average feature vector and the operating condition feature vector to calculate the standard feature vector corresponding to the operating condition feature vector, and combining the standard feature vector to calculate the feature alienation value between the operating condition features by the following formula:

[0006]

[0007] Among them, A represents the characteristic alienation value between working condition characteristics, B′ a represents the standard feature vector corresponding to the a-th operating condition feature, B′ a+1represents the standard feature vector corresponding to the a+1th operating condition feature, a represents the serial number corresponding to the operating condition feature, and r represents the number of features corresponding to the operating condition feature;

[0008] Query the tool adaptation system corresponding to the pilot platform tool, schedule the system record log corresponding to the tool adaptation system, mine the system warning information from the system record log, calculate the warning priority corresponding to the system warning information, and analyze the key warning factors in the system warning information in combination with the warning priority;

[0009] In combination with the key warning factors and the system warning information, the system fault corresponding to the tool adaptation system is analyzed, and according to the working condition fault data, the tool fault corresponding to the pilot platform tool is analyzed;

[0010] Analyze the steady-state faults in the tool fault and the system fault, calculate the fault effect value corresponding to the steady-state fault, formulate a fault repair plan corresponding to the steady-state fault based on the fault effect value and the steady-state fault, and according to the fault repair plan, perform the fault diagnosis processing corresponding to the pilot platform tool to obtain a diagnosis result.

[0011] Optionally, extracting the operating condition characteristics corresponding to the tool operating condition data includes:

[0012] Performing data cleaning on the tool working condition data to obtain cleaning working condition data;

[0013] Identify the data labels corresponding to the cleaning condition data, and calculate the label weights corresponding to the data labels;

[0014] According to the label weight, the cleaning condition data is filtered to obtain target condition data;

[0015] Determine the operating condition data in the target condition data, and extract the condition characteristics corresponding to the operating condition data.

[0016] Optionally, combining the average feature vector and the operating condition feature vector to calculate a standard feature vector corresponding to the operating condition feature vector includes:

[0017] Measure the vector number corresponding to the characteristic vector of the working condition;

[0018] Combining the average eigenvector, the operating condition eigenvector and the vector number, the standard eigenvector corresponding to the operating condition eigenvector is calculated by the following formula:

[0019]

[0020] Where D represents the standard eigenvector corresponding to the working condition eigenvector, Ei represents the eigenvector of the i-th operating condition, represents the average eigenvector, i represents the vector sequence number corresponding to the operating condition eigenvector, and n represents the vector number of the operating condition eigenvector.

[0021] Optionally, the calculating the warning priority corresponding to the system warning information includes:

[0022] Performing information screening on the system warning information to obtain filtered warning information;

[0023] Calculate the information gain value corresponding to the filtered warning information, and analyze the warning subject corresponding to the filtered warning information;

[0024] Calculate the subject criticality corresponding to the warning subject;

[0025] The warning priority corresponding to the system warning information is calculated by combining the information gain value and the subject criticality.

[0026] Optionally, the calculating the subject criticality corresponding to the warning subject includes:

[0027] Query the subject function attributes corresponding to the warning subject, and extract key function attributes from the subject function attributes;

[0028] Quantifying the key functional attributes to obtain functional quantification values, and calculating the attribute correlation between the key functional attributes;

[0029] The subject criticality corresponding to the warning subject is calculated by combining the function quantification value and the attribute association degree.

[0030] Optionally, analyzing key warning factors in the system warning information in combination with the warning priority includes:

[0031] Identify warning information characters in the system warning information;

[0032] Extracting key characters from the warning information characters to obtain warning key characters;

[0033] Analyzing the meaning of characters corresponding to the warning key characters, and determining the warning factors in the system warning information based on the meaning of the characters;

[0034] In combination with the warning priorities, key warning factors are extracted from the warning factors.

[0035] Optionally, the combining the key warning factor and the system warning information to analyze the system failure corresponding to the tool adaptation system includes:

[0036] Extracting the information factor corresponding to the system warning information and its corresponding information value;

[0037] Combining the information factor and the information value, constructing a system state diagram corresponding to the tool adaptation system;

[0038] Analyze the synergy between the key warning factors;

[0039] In combination with the system state diagram and the factor synergy, the system failure corresponding to the tool adaptation system is analyzed.

[0040] Optionally, analyzing the synergy between the key warning factors includes:

[0041] Analyze the factor characteristics corresponding to the key warning factors, and calculate the characteristic interaction coefficients between the factor characteristics;

[0042] Based on the factor characteristics and the attribute interaction coefficient, the factor synergy coefficient between the key warning factors is calculated by the following formula:

[0043]

[0044] Among them, F represents the synergy coefficient between key warning factors, G b represents the feature interaction coefficient of the b-th factor feature, represents the vector corresponding to the b-th factor feature, represents the vector corresponding to the b+1th factor feature, b represents the serial number of the factor feature, and q represents the number of factor features;

[0045] The factor synergy between the key warning factors is determined according to the factor synergy coefficient.

[0046] Optionally, calculating the fault effect value corresponding to the steady-state fault includes:

[0047] Monitoring the comprehensive performance corresponding to the steady-state fault, and reading the performance value corresponding to the comprehensive performance;

[0048] Calculate the performance deviation value corresponding to the comprehensive performance by combining the preset performance standard value and the performance value;

[0049] Determine a performance indicator corresponding to the comprehensive performance, and construct a performance evaluation matrix corresponding to the comprehensive performance by combining the performance indicator and the performance value;

[0050] Based on the performance evaluation matrix, calculating the performance importance corresponding to the comprehensive performance;

[0051] The fault effect value corresponding to the steady-state fault is calculated by combining the performance deviation value and the performance importance.

[0052] A fault diagnosis and repair system in an industrial pilot platform tool, characterized in that the system comprises:

[0053] A fault data identification module is used to obtain tool condition data corresponding to the pilot platform tool of the industry, extract the condition characteristics corresponding to the tool condition data, calculate the characteristic alienation value between the condition characteristics, and identify the condition fault data in the tool condition data based on the characteristic alienation value, wherein the calculation of the characteristic alienation value between the condition characteristics includes: vectorizing the condition characteristics to obtain a condition characteristic vector, calculating the average characteristic vector corresponding to the condition characteristic vector, combining the average characteristic vector and the condition characteristic vector to calculate the standard characteristic vector corresponding to the condition characteristic vector, and combining the standard characteristic vector to calculate the characteristic alienation value between the condition characteristics by the following formula:

[0054]

[0055] Among them, A represents the characteristic alienation value between working condition characteristics, B′ a represents the standard feature vector corresponding to the a-th operating condition feature, B′ a+1 represents the standard feature vector corresponding to the a+1th operating condition feature, a represents the serial number corresponding to the operating condition feature, and r represents the number of features corresponding to the operating condition feature;

[0056] The warning factor analysis module is used to query the tool adaptation system corresponding to the pilot platform tool, schedule the system record log corresponding to the tool adaptation system, mine the system warning information from the system record log, calculate the warning priority corresponding to the system warning information, and analyze the key warning factors in the system warning information in combination with the warning priority;

[0057] A fault analysis module, used to analyze the system fault corresponding to the tool adaptation system in combination with the key warning factors and the system warning information, and to analyze the tool fault corresponding to the pilot platform tool according to the working condition fault data;

[0058] A fault diagnosis module is used to analyze the steady-state faults in the tool faults and the system faults, calculate the fault effect value corresponding to the steady-state faults, formulate a fault repair plan corresponding to the steady-state faults based on the fault effect value and the steady-state faults, and execute the fault diagnosis processing corresponding to the pilot platform tool according to the fault repair plan to obtain a diagnosis result.

[0059] The present invention can obtain the state description features related to the working condition in the tool working condition data by extracting the working condition features corresponding to the tool working condition data, and provide a basis for the identification of subsequent working condition fault data. The present invention can obtain specific information suggesting that there are problems or potential risks in the system by mining the system warning information from the system record log. By calculating the warning priority corresponding to the system warning information, the importance of the warning corresponding to the system warning information can be understood, and then the analysis of the key warning factors can be provided. The present invention combines the key warning factors and the system warning information to analyze the system fault corresponding to the tool adaptation system, and can accurately obtain the fault of the tool adaptation system, which provides a basis for subsequent fault repair processing. The present invention can determine the permanent fault in the tool fault and the system fault by analyzing the tool fault and the steady-state fault in the system fault, so as to improve the subsequent fault repair efficiency, and calculate the fault effect value corresponding to the steady-state fault, so as to understand the fault severity corresponding to the steady-state fault, and then provide a basis for the formulation of the subsequent fault repair plan. Therefore, the fault diagnosis and repair method and system in the industrial pilot platform tool provided by the embodiment of the present invention can improve the accuracy of fault diagnosis and repair in the industrial pilot platform tool. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 A flowchart of a fault diagnosis and repair method in an industrial pilot platform tool provided by an embodiment of the present invention;

[0061] Figure 2 A functional module diagram of a fault diagnosis and repair system in an industrial pilot platform tool provided by an embodiment of the present invention.

[0062] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0063] 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.

[0064] The embodiment of the present application provides a fault diagnosis and repair method in an industrial pilot platform tool. In the embodiment of the present application, the execution subject of the fault diagnosis and repair method in the industrial pilot platform tool includes but is not limited to at least one of the electronic devices such as the server, the terminal, etc. that can be configured to execute the method provided in the embodiment of the present application. In other words, the fault diagnosis and repair method in the industrial pilot platform tool can be executed by software or hardware installed on the terminal device or the server device, and the software can be a blockchain platform. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc. The server can be an independent server, or it can be a cloud server that provides cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (Content Delivery Network, CDN), and basic cloud computing services such as big data and artificial intelligence platforms.

[0065] Reference Figure 1 FIG. 1 is a flow chart of a method for diagnosing and repairing a fault in an industrial pilot platform tool provided by an embodiment of the present invention. In this embodiment, the method for diagnosing and repairing a fault in an industrial pilot platform tool includes steps S1-S4.

[0066] S1. Obtain tool operating condition data corresponding to the pilot platform tool of the industry, extract the operating condition features corresponding to the tool operating condition data, calculate the feature alienation values ​​between the operating condition features, and identify the operating condition fault data in the tool operating condition data based on the feature alienation values.

[0067] The present invention can obtain the state description features related to the working conditions in the tool working condition data by extracting the working condition features corresponding to the tool working condition data, thereby providing a basis for the identification of subsequent working condition fault data. The pilot platform tool is a variety of tools used in an intermediate test platform with a certain scale and function built in the process of industrial research and development to verify the feasibility and stability of new products and new technologies, usually including various detection equipment, data analysis software, simulation systems, etc. The tool working condition data is the operation record data of the tool corresponding to the pilot platform tool during use, and the working condition features are the key feature performances corresponding to the tool working condition data, which can reflect the operating status and performance characteristics of the pilot platform tool under specific working conditions.

[0068] As an embodiment of the present invention, the extracting of the operating condition characteristics corresponding to the tool operating condition data includes: performing data cleaning on the tool operating condition data to obtain cleaning operating condition data, identifying data labels corresponding to the cleaning operating condition data, calculating label weights corresponding to the data labels, filtering the cleaning operating condition data according to the label weights to obtain target operating condition data, determining the operating condition data in the target operating condition data, and extracting the operating condition characteristics corresponding to the operating condition data.

[0069] Among them, the cleaning condition data is the data obtained after removing abnormal data in the tool condition data, the data label is the attribute description corresponding to the cleaning condition data, the label weight represents the importance corresponding to the data label, and the operating condition data is the relevant data about the tool operation in the target condition data.

[0070] Optionally, data cleaning of the tool operating condition data can be achieved through a box plot method; identification of data labels corresponding to the cleaning condition data can be achieved through an OCR recognition algorithm; label weights corresponding to the data labels can be implemented by a weight calculator, which is compiled by a script language; when the label weight is lower than a preset weight value, the cleaning condition data is filtered to obtain target condition data, and the preset weight value can be set to 0.8, or it can be set according to the actual application scenario; determination of the operating condition data in the target condition data can be achieved through a manual labeling method; extraction of the operating condition features corresponding to the operating condition data can be achieved through a principal component analysis method.

[0071] The present invention can understand the degree of difference between the working condition characteristics by calculating the characteristic alienation value between the working condition characteristics, thereby improving the recognition accuracy of the working condition fault data, wherein the characteristic alienation value represents the degree of difference between the working condition characteristics, and the working condition fault data represents the fault data in the tool working condition data. Furthermore, the average alienation value corresponding to the characteristic alienation value is calculated. When the characteristic alienation value is greater than the average alienation value, the corresponding tool working condition data is the working condition fault data.

[0072] As an embodiment of the present invention, the calculating of the characteristic alienation value between the operating condition characteristics includes: performing vectorization processing on the operating condition characteristics to obtain an operating condition characteristic vector, calculating an average characteristic vector corresponding to the operating condition characteristic vector, combining the average characteristic vector and the operating condition characteristic vector to calculate a standard characteristic vector corresponding to the operating condition characteristic vector, and combining the standard characteristic vector to calculate the characteristic alienation value between the operating condition characteristics by the following formula:

[0073]

[0074] Among them, A represents the characteristic alienation value between working condition characteristics, B′ a represents the standard feature vector corresponding to the a-th operating condition feature, B′ a+1 represents the standard feature vector corresponding to the a+1th operating condition feature, a represents the serial number corresponding to the operating condition feature, and r represents the number of features corresponding to the operating condition feature.

[0075] Among them, the operating condition feature vector is the expression vector corresponding to the operating condition feature, the standard feature vector is the vector after normalization of the operating condition feature vector, and the average feature vector is the average value corresponding to the operating condition feature vector. Furthermore, the vectorization processing of the operating condition feature can be achieved through the word2vec algorithm; the calculation of the average feature vector corresponding to the operating condition feature vector can be achieved through the average function.

[0076] Further, as an optional embodiment of the present invention, the combining the average characteristic vector and the operating condition characteristic vector to calculate the standard characteristic vector corresponding to the operating condition characteristic vector includes: measuring the vector number corresponding to the operating condition characteristic vector, combining the average characteristic vector, the operating condition characteristic vector and the vector number, and calculating the standard characteristic vector corresponding to the operating condition characteristic vector by the following formula:

[0077]

[0078] Where D represents the standard eigenvector corresponding to the working condition eigenvector, E i represents the eigenvector of the i-th operating condition, represents the average eigenvector, i represents the vector sequence number corresponding to the operating condition eigenvector, and n represents the vector number of the operating condition eigenvector.

[0079] S2. Query the tool adaptation system corresponding to the pilot platform tool, schedule the system record log corresponding to the tool adaptation system, mine the system warning information from the system record log, calculate the warning priority corresponding to the system warning information, and analyze the key warning factors in the system warning information in combination with the warning priority.

[0080] The present invention can obtain specific information suggesting that there are problems or potential risks in the system by mining system warning information from the system record log. By calculating the warning priority corresponding to the system warning information, the importance of the warning corresponding to the system warning information can be understood, thereby providing a basis for the subsequent analysis of key warning factors. The tool adaptation system is a software system used in conjunction with the pilot platform tool, the system record log is the relevant record information of the operation behavior and operation of the tool adaptation system, the system warning information is the information of possible risks or problems recorded in the system record log, and the warning priority indicates the importance of the warning corresponding to the system warning information. Optionally, the tool adaptation system can be obtained by querying the official website corresponding to the pilot platform tool; the scheduling of the system record log corresponding to the tool adaptation system can be implemented by a scheduling tool, such as the ELK Stack tool; mining the system warning information from the system record log can be implemented by the Apriori algorithm.

[0081] As an embodiment of the present invention, the calculation of the warning priority corresponding to the system warning information includes: performing information filtering processing on the system warning information to obtain filtered warning information, calculating the information gain value corresponding to the filtered warning information, analyzing the warning subject corresponding to the filtered warning information, calculating the subject criticality corresponding to the warning subject, and combining the information gain value and the subject criticality to calculate the warning priority corresponding to the system warning information.

[0082] Among them, the filtered warning information is representative information in the system warning information, the information gain value indicates the contribution degree corresponding to the key warning information, the warning subject is the warning object corresponding to the key warning information, such as a specific system component, and the subject criticality indicates the criticality corresponding to the warning subject. Optionally, the information screening process of the system warning information can be implemented by a naive Bayes classification algorithm; the information entropy and conditional entropy corresponding to the key warning information are calculated, and the difference between the information entropy and the conditional entropy of each information is calculated to obtain the information gain value, and the information entropy and conditional entropy can be calculated by the Shannon entropy algorithm and the conditional entropy formula respectively; the warning subject corresponding to the key warning information can be implemented by a semantic analysis method; the information gain value and the subject criticality are standardized to obtain a standard gain value and a standard criticality, the standard gain value and the standard criticality are summed, and the warning priority corresponding to the system warning information is determined according to the summation result.

[0083] Further, as an optional embodiment of the present invention, the calculation of the subject criticality corresponding to the warning subject includes: querying the subject functional attributes corresponding to the warning subject, extracting key functional attributes from the subject functional attributes, quantifying the key functional attributes to obtain functional quantification values, calculating the attribute correlation between the key functional attributes, and calculating the subject criticality corresponding to the warning subject by combining the functional quantification values ​​and the attribute correlations.

[0084] Among them, the subject functional attribute is a description of the functional role corresponding to the warning subject, the key functional attribute is the core functional attribute in the subject functional attributes, the functional quantification value is the expression value corresponding to the key functional attribute, and the attribute correlation degree indicates the correlation degree between the characteristic functional attributes. Optionally, the subject functional attribute can be obtained by querying the subject introduction information corresponding to the warning subject; the extraction of the key functional attributes in the subject functional attributes can be achieved by factor analysis; the role played by the key functional attribute in handling a certain event can be determined, the role can be quantified, and the functional quantification value can be obtained; the attribute correlation between the key functional attributes can be calculated by the Pearson correlation coefficient method; the functional quantification value and the corresponding attribute correlation degree are multiplied and summed to obtain the subject criticality corresponding to the warning subject.

[0085] By combining the warning priority and analyzing the key warning factors in the system warning information, the present invention can understand the factors in the system warning information that may cause major failures and safety hazards, thereby providing a basis for subsequent analysis of system failures, wherein the key warning factors are the factors in the system warning information that may cause major failures and safety hazards.

[0086] As an embodiment of the present invention, analyzing the key warning factors in the system warning information in combination with the warning priority includes: identifying the warning information characters in the system warning information, performing key character extraction on the warning information characters to obtain warning key characters, analyzing the character meanings corresponding to the warning key characters, determining the warning factors in the system warning information based on the character meanings, and extracting the key warning factors from the warning factors in combination with the warning priority.

[0087] Among them, the warning information characters are text information in the system warning information, the warning key characters are core characters in the warning information characters, the character meaning is the character content of the warning key characters, and the warning factors are risk signals and problem directions in the system information.

[0088] Optionally, the recognition of warning information characters in the system warning information can be achieved through the above-mentioned OCR recognition technology; the key character extraction of the warning information characters can be achieved through the TF-IDF algorithm; the analysis of the character meanings corresponding to the warning key characters can be achieved through the GloVe model; the warning words are identified from the character meanings, and the warning factors in the system warning information are determined according to the warning words; in combination with the level of the warning priority, the key warning factors are extracted from the warning factors, and the warning factors with low warning priority are filtered out.

[0089] S3. Analyze the system fault corresponding to the tool adaptation system in combination with the key warning factors and the system warning information, and analyze the tool fault corresponding to the pilot platform tool according to the operating condition fault data.

[0090] The present invention combines the key warning factors and the system warning information to analyze the system fault corresponding to the tool adaptation system, and can accurately obtain the fault of the tool adaptation system, providing a basis for subsequent fault repair processing, wherein the system fault is the fault corresponding to the tool adaptation system.

[0091] As an embodiment of the present invention, the combination of the key warning factors and the system warning information to analyze the system fault corresponding to the tool adaptation system includes: extracting the information factor corresponding to the system warning information and its corresponding information value, combining the information factor and the information value to construct a system status diagram corresponding to the tool adaptation system, analyzing the factor synergy between the key warning factors, and combining the system status diagram and the factor synergy to analyze the system fault corresponding to the tool adaptation system.

[0092] Among them, the information factor is the information variable corresponding to the system warning information, the information value is the value corresponding to the information factor, the system state diagram is the state visualization diagram corresponding to the tool adaptation system, and the factor synergy represents the interaction relationship between the key warning factors. Furthermore, the extraction of the information factor corresponding to the system warning information and its corresponding information value can be realized by an extraction function, and the extraction function is compiled by a programming language; the construction of the system state diagram corresponding to the tool adaptation system can be realized by the Visio drawing tool; combining the system state diagram and the factor synergy, the steps of analyzing the system fault corresponding to the tool adaptation system are as follows: the system state diagram can display each The system status diagram can be used to analyze the connection status, data transmission status, resource usage, etc. of each software tool, and then identify the corresponding abnormal status. For example, if the connection status of a tool is shown in red (indicating connection failure) in the status diagram, this is an obvious abnormal state. If the key warning factors are high network latency, slow response of specific tools, and increased data transmission error rate, and if the system status diagram shows that a tool frequently has error prompts, the synergy of the factors can be used to find that the database connection on which the tool relies is unstable and the memory resources are tight, and then it can be determined that these factors jointly cause the failure of the tool, thereby affecting the entire tool adaptation system. By combining the system status diagram and the synergy of the factors, system failures can be analyzed more comprehensively.

[0093] Further, as an optional embodiment of the present invention, the analyzing the factor synergy between the key warning factors includes: analyzing the factor characteristics corresponding to the key warning factors, calculating the feature interaction coefficient between the factor characteristics, and calculating the factor synergy coefficient between the key warning factors based on the factor characteristics and the attribute interaction coefficient by the following formula:

[0094]

[0095] Among them, F represents the synergy coefficient between key warning factors, G b represents the feature interaction coefficient of the b-th factor feature, represents the vector corresponding to the b-th factor feature, represents the vector corresponding to the b+1th factor feature, b represents the serial number of the factor feature, and q represents the number of factor features;

[0096] The factor synergy between the key warning factors is determined according to the factor synergy coefficient.

[0097] Among them, the factor feature is the unique attribute corresponding to the key warning factor, the feature interaction coefficient represents the degree of mutual influence between the factor features, and the factor synergy coefficient represents the quantitative value of the interaction between the key warning factors. Furthermore, the analysis of the factor features corresponding to the key warning factors can be achieved through a decision tree algorithm; the calculation of the feature interaction coefficient between the factor features can be achieved through a mutual information algorithm; according to the level of the factor synergy coefficient, the factor synergy between the key warning factors is determined. If the factor synergy coefficient is high (for example, greater than or equal to 0.7), it indicates that the key warning factors have strong synergy.

[0098] The present invention analyzes the tool fault corresponding to the pilot platform tool based on the operating condition fault data, and then obtains the fault type of the pilot platform tool, which facilitates the subsequent analysis and processing of steady-state faults, wherein the tool fault is the fault type of the pilot platform tool. Further, the data trend corresponding to the operating condition fault data is analyzed, and the tool operation trend corresponding to the pilot platform tool is analyzed based on the data trend, and it is determined whether the operation trend is normal, thereby analyzing the tool fault of the pilot platform tool.

[0099] S4. Analyze the steady-state faults in the tool fault and the system fault, calculate the fault effect value corresponding to the steady-state fault, formulate a fault repair plan corresponding to the steady-state fault based on the fault effect value and the steady-state fault, and according to the fault repair plan, execute the fault diagnosis processing corresponding to the pilot platform tool to obtain a diagnosis result.

[0100] The present invention can determine the permanent faults in the tool failure and the system failure by analyzing the steady-state faults in the tool failure and the system failure, so as to improve the subsequent fault repair efficiency, and calculate the fault effect value corresponding to the steady-state fault, so as to understand the fault severity corresponding to the steady-state fault, thereby providing a basis for the formulation of subsequent fault repair plans, wherein the steady-state fault is a permanent fault in the tool failure and the system failure, and the fault effect value represents the severity corresponding to the steady-state fault. Furthermore, the analysis of the steady-state faults in the tool failure and the system failure can be achieved through a fault tree analysis method.

[0101] As an embodiment of the present invention, the calculation of the fault effect value corresponding to the steady-state fault includes: monitoring the comprehensive performance corresponding to the steady-state fault, and reading the performance value corresponding to the comprehensive performance, combining a preset performance standard value and the performance value, calculating the performance deviation value corresponding to the comprehensive performance, determining the performance indicator corresponding to the comprehensive performance, combining the performance indicator and the performance value, constructing a performance judgment matrix corresponding to the comprehensive performance, calculating the performance importance corresponding to the comprehensive performance based on the performance judgment matrix, and calculating the fault effect value corresponding to the steady-state fault in combination with the performance deviation value and the performance importance.

[0102] Among them, the comprehensive performance is the overall performance of the pilot platform tool and the tool adaptation system under the steady-state fault, the performance value is the specific quantitative value corresponding to the comprehensive performance, the preset performance standard value is the performance quantitative value under normal state corresponding to each performance in the comprehensive performance, the performance indicator is the performance type corresponding to the comprehensive performance, the performance evaluation matrix is ​​the performance multi-dimensional evaluation matrix corresponding to the comprehensive performance, and the performance importance indicates the importance of the comprehensive performance. Furthermore, the monitoring of the comprehensive performance corresponding to the steady-state fault can be achieved through monitoring tools, such as performance monitoring software, system log analysis tools, network traffic monitors, resource utilization monitoring tools, etc.; calculate the preset performance standard value and The difference between the performance values ​​is used to obtain the performance deviation value corresponding to the comprehensive performance; the determination of the performance index corresponding to the comprehensive performance can be achieved through the target analysis method, by clarifying the performance target to be achieved, and decomposing the target into specific measurable performance indicators; the performance judgment matrix corresponding to the comprehensive performance can be implemented by a matrix function, such as a zero matrix function; based on the performance judgment matrix, the performance importance corresponding to the comprehensive performance is calculated, and the performance judgment matrices are compared pairwise. The comparison results can be represented by numbers 1-9 and their reciprocals, where 1 means that two indicators are equally important, and 9 means that one indicator is extremely important than the other indicator. The calculation formula of the fault effect value is: fault effect value = Σ performance deviation value × performance importance.

[0103] The present invention formulates a fault repair plan corresponding to the steady-state fault based on the fault effect value and the steady-state fault, thereby facilitating improving the accuracy of fault diagnosis and repair corresponding to the pilot platform tool, wherein the fault repair plan is a fault solution corresponding to the steady-state fault, and further, the fault cause corresponding to the steady-state fault is analyzed, and a repair plan corresponding to the steady-state fault is formulated according to the fault cause, and based on the fault effect value, the execution order of the repair plan is set to obtain the fault repair plan corresponding to the steady-state fault.

[0104] The present invention can obtain the state description characteristics related to the working condition in the tool working condition data by extracting the working condition characteristics corresponding to the tool working condition data, and provide a basis for the identification of subsequent working condition fault data. The present invention can obtain specific information suggesting that there are problems or potential risks in the system by mining the system warning information from the system record log. By calculating the warning priority corresponding to the system warning information, the importance of the warning corresponding to the system warning information can be understood, and then the analysis of the key warning factors is provided. The present invention combines the key warning factors and the system warning information to analyze the system fault corresponding to the tool adaptation system, and can accurately obtain the fault of the tool adaptation system, which provides a basis for subsequent fault repair processing. The present invention can determine the tool fault and the permanent fault in the system fault by analyzing the tool fault and the steady-state fault in the system fault, so as to improve the subsequent fault repair efficiency, and calculate the fault effect value corresponding to the steady-state fault, so as to understand the fault severity corresponding to the steady-state fault, and then provide a basis for the formulation of the subsequent fault repair plan. Therefore, the fault diagnosis and repair method in an industrial pilot platform tool provided by an embodiment of the present invention can improve the accuracy of fault diagnosis and repair in an industrial pilot platform tool.

[0105] like Figure 2 1 is a functional module diagram of a fault diagnosis and repair system in an industrial pilot platform tool provided by an embodiment of the present invention.

[0106] The fault diagnosis and repair system 100 in the industrial pilot platform tool described in the present invention can be installed in an electronic device. According to the functions implemented, the fault diagnosis and repair system 100 in the industrial pilot platform tool can include a fault data identification module 101, a warning factor analysis module 102, a fault analysis module 103 and a fault diagnosis module 104. The module described in the present invention can also be called a unit, which refers to a series of computer program segments that can be executed by an electronic device processor and can complete fixed functions, which are stored in the memory of the electronic device.

[0107] In this embodiment, the functions of each module / unit are as follows:

[0108] The fault data identification module 101 is used to obtain tool condition data corresponding to the pilot platform tool of the industry, extract the condition features corresponding to the tool condition data, calculate the feature alienation value between the condition features, and identify the condition fault data in the tool condition data based on the feature alienation value, wherein the calculation of the feature alienation value between the condition features includes: vectorizing the condition features to obtain a condition feature vector, calculating the average feature vector corresponding to the condition feature vector, combining the average feature vector and the condition feature vector to calculate the standard feature vector corresponding to the condition feature vector, and combining the standard feature vector to calculate the feature alienation value between the condition features by the following formula:

[0109]

[0110] Among them, A represents the characteristic alienation value between working condition characteristics, B′ a represents the standard feature vector corresponding to the a-th operating condition feature, B′ a+1 represents the standard feature vector corresponding to the a+1th operating condition feature, a represents the serial number corresponding to the operating condition feature, and r represents the number of features corresponding to the operating condition feature;

[0111] The warning factor analysis module 102 is used to query the tool adaptation system corresponding to the pilot platform tool, schedule the system record log corresponding to the tool adaptation system, mine the system warning information from the system record log, calculate the warning priority corresponding to the system warning information, and analyze the key warning factors in the system warning information in combination with the warning priority;

[0112] The fault analysis module 103 is used to analyze the system fault corresponding to the tool adaptation system in combination with the key warning factors and the system warning information, and analyze the tool fault corresponding to the pilot platform tool according to the working condition fault data;

[0113] The fault diagnosis module 104 is used to analyze the steady-state faults in the tool faults and the system faults, calculate the fault effect value corresponding to the steady-state faults, formulate a fault repair plan corresponding to the steady-state faults based on the fault effect value and the steady-state faults, and execute the fault diagnosis processing corresponding to the pilot platform tool according to the fault repair plan to obtain a diagnosis result.

[0114] In detail, each module described in the fault diagnosis and repair system 100 in the industrial pilot platform tool described in the embodiment of the present application is used in the same manner as described above. Figure 1 The technical means are the same as the fault diagnosis and repair method in the industrial pilot platform tool described in, and can produce the same technical effects, so they will not be repeated here.

[0115] In the several embodiments provided by the present invention, it should be understood that the provided methods and systems can be implemented in other ways. For example, the method embodiments described above are only illustrative, for example, the division of the modules is only a logical function division, and there may be other division methods in actual implementation.

[0116] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. A fault diagnosis and repair method in an industrial pilot platform tool, characterized in that: The method comprises: Obtain tool operating data corresponding to the pilot platform tool of the industry, extract the operating condition features corresponding to the tool operating condition data, calculate the feature alienation value between the operating condition features, and identify the operating condition fault data in the tool operating condition data based on the feature alienation value, wherein the calculation of the feature alienation value between the operating condition features includes: vectorizing the operating condition features to obtain an operating condition feature vector, calculating the average feature vector corresponding to the operating condition feature vector, combining the average feature vector and the operating condition feature vector to calculate the standard feature vector corresponding to the operating condition feature vector, and combining the standard feature vector to calculate the feature alienation value between the operating condition features by the following formula: Among them, A represents the characteristic alienation value between working condition characteristics, B a represents the standard feature vector corresponding to the a-th operating condition feature, B a+1 represents the standard feature vector corresponding to the a+1th operating condition feature, a represents the serial number corresponding to the operating condition feature, and r represents the number of features corresponding to the operating condition feature; Query the tool adaptation system corresponding to the pilot platform tool, schedule the system record log corresponding to the tool adaptation system, mine the system warning information from the system record log, calculate the warning priority corresponding to the system warning information, and analyze the key warning factors in the system warning information in combination with the warning priority; In combination with the key warning factors and the system warning information, the system fault corresponding to the tool adaptation system is analyzed, and according to the working condition fault data, the tool fault corresponding to the pilot platform tool is analyzed; Analyze the steady-state faults in the tool fault and the system fault, calculate the fault effect value corresponding to the steady-state fault, formulate a fault repair plan corresponding to the steady-state fault based on the fault effect value and the steady-state fault, and according to the fault repair plan, perform the fault diagnosis processing corresponding to the pilot platform tool to obtain a diagnosis result.

2. The fault diagnosis and repair method in an industrial pilot platform tool according to claim 1, characterized in that: The extracting the working condition characteristics corresponding to the tool working condition data includes: Performing data cleaning on the tool working condition data to obtain cleaning working condition data; Identify the data labels corresponding to the cleaning condition data, and calculate the label weights corresponding to the data labels; According to the label weight, the cleaning condition data is filtered to obtain target condition data; Determine the operating condition data in the target condition data, and extract the condition characteristics corresponding to the operating condition data.

3. The fault diagnosis and repair method in an industrial pilot platform tool according to claim 1, characterized in that: The combining the average feature vector and the operating condition feature vector to calculate a standard feature vector corresponding to the operating condition feature vector includes: Measure the vector number corresponding to the characteristic vector of the working condition; Combining the average eigenvector, the operating condition eigenvector and the vector number, the standard eigenvector corresponding to the operating condition eigenvector is calculated by the following formula: Where D represents the standard eigenvector corresponding to the working condition eigenvector, E i represents the eigenvector of the i-th operating condition, represents the average eigenvector, i represents the vector sequence number corresponding to the operating condition eigenvector, and n represents the vector number of the operating condition eigenvector.

4. The fault diagnosis and repair method in an industrial pilot platform tool according to claim 1, characterized in that: The calculating the warning priority corresponding to the system warning information includes: Performing information screening on the system warning information to obtain filtered warning information; Calculate the information gain value corresponding to the filtered warning information, and analyze the warning subject corresponding to the filtered warning information; Calculate the subject criticality corresponding to the warning subject; The warning priority corresponding to the system warning information is calculated by combining the information gain value and the subject criticality.

5. The fault diagnosis and repair method in an industrial pilot platform tool according to claim 4, characterized in that: The calculating the subject criticality corresponding to the warning subject includes: Query the subject function attributes corresponding to the warning subject, and extract key function attributes from the subject function attributes; Quantifying the key functional attributes to obtain functional quantification values, and calculating the attribute correlation between the key functional attributes; The subject criticality corresponding to the warning subject is calculated by combining the function quantification value and the attribute association degree.

6. The fault diagnosis and repair method in an industrial pilot platform tool according to claim 1, characterized in that: The step of analyzing the key warning factors in the system warning information in combination with the warning priority includes: Identify warning information characters in the system warning information; Extracting key characters from the warning information characters to obtain warning key characters; Analyzing the meaning of characters corresponding to the warning key characters, and determining the warning factors in the system warning information based on the meaning of the characters; In combination with the warning priorities, key warning factors are extracted from the warning factors.

7. The fault diagnosis and repair method in an industrial pilot platform tool according to claim 1, characterized in that: The combining the key warning factor and the system warning information to analyze the system failure corresponding to the tool adaptation system includes: Extracting the information factor corresponding to the system warning information and its corresponding information value; Combining the information factor and the information value, constructing a system state diagram corresponding to the tool adaptation system; Analyze the synergy between the key warning factors; In combination with the system state diagram and the factor synergy, the system failure corresponding to the tool adaptation system is analyzed.

8. The fault diagnosis and repair method in an industrial pilot platform tool according to claim 7, characterized in that: The analyzing the synergy between the key warning factors includes: Analyze the factor characteristics corresponding to the key warning factors, and calculate the characteristic interaction coefficients between the factor characteristics; Based on the factor characteristics and the attribute interaction coefficient, the factor synergy coefficient between the key warning factors is calculated by the following formula: Among them, F represents the synergy coefficient between key warning factors, G b represents the feature interaction coefficient of the b-th factor feature, represents the vector corresponding to the b-th factor feature, represents the vector corresponding to the b+1th factor feature, b represents the serial number of the factor feature, and q represents the number of factor features; The factor synergy between the key warning factors is determined according to the factor synergy coefficient.

9. The fault diagnosis and repair method in an industrial pilot platform tool according to claim 1, characterized in that: The calculating the fault effect value corresponding to the steady-state fault includes: Monitoring the comprehensive performance corresponding to the steady-state fault, and reading the performance value corresponding to the comprehensive performance; Calculate the performance deviation value corresponding to the comprehensive performance by combining the preset performance standard value and the performance value; Determine a performance indicator corresponding to the comprehensive performance, and construct a performance evaluation matrix corresponding to the comprehensive performance by combining the performance indicator and the performance value; Based on the performance evaluation matrix, calculating the performance importance corresponding to the comprehensive performance; The fault effect value corresponding to the steady-state fault is calculated by combining the performance deviation value and the performance importance.

10. A fault diagnosis and repair system in an industrial pilot platform tool, characterized in that: The system comprises: A fault data identification module is used to obtain tool condition data corresponding to the pilot platform tool of the industry, extract the condition characteristics corresponding to the tool condition data, calculate the characteristic alienation value between the condition characteristics, and identify the condition fault data in the tool condition data based on the characteristic alienation value, wherein the calculation of the characteristic alienation value between the condition characteristics includes: vectorizing the condition characteristics to obtain a condition characteristic vector, calculating the average characteristic vector corresponding to the condition characteristic vector, combining the average characteristic vector and the condition characteristic vector to calculate the standard characteristic vector corresponding to the condition characteristic vector, and combining the standard characteristic vector to calculate the characteristic alienation value between the condition characteristics by the following formula: Among them, A represents the characteristic alienation value between working condition characteristics, B a represents the standard feature vector corresponding to the a-th operating condition feature, B a+1 represents the standard feature vector corresponding to the a+1th operating condition feature, a represents the serial number corresponding to the operating condition feature, and r represents the number of features corresponding to the operating condition feature; The warning factor analysis module is used to query the tool adaptation system corresponding to the pilot platform tool, schedule the system record log corresponding to the tool adaptation system, mine the system warning information from the system record log, calculate the warning priority corresponding to the system warning information, and analyze the key warning factors in the system warning information in combination with the warning priority; A fault analysis module, used to analyze the system fault corresponding to the tool adaptation system in combination with the key warning factors and the system warning information, and to analyze the tool fault corresponding to the pilot platform tool according to the working condition fault data; A fault diagnosis module is used to analyze the steady-state faults in the tool faults and the system faults, calculate the fault effect value corresponding to the steady-state faults, formulate a fault repair plan corresponding to the steady-state faults based on the fault effect value and the steady-state faults, and execute the fault diagnosis processing corresponding to the pilot platform tool according to the fault repair plan to obtain a diagnosis result.

Images