A stator slot temperature fault analysis method based on industrial Internet platform

Abstract

A stator slot temperature fault analysis method based on an industrial internet platform can automatically initiate stator slot temperature fault analysis at any time, replacing manual initiation. The platform consists of four levels: the driver level, the analysis level, the output level, and the data level. The driver level determines the analysis drive method for abnormal generator stator slot temperature, including event-driven and cycle-driven analysis, and outputs fault mode characteristic indicators. The analysis level analyzes the causes and impacts of faults, including fault case analysis, rule analysis, FTA analysis, and big data analysis. The analysis level compares the case matching results, rule analysis results, FTA analysis results, and data analysis results, outputting the most confident result. The output level displays the fault analysis logic and results in the form of a mind map. The data level analyzes each fault analysis result, organizes key fault information, updates fault knowledge, and structures the process and results before entering them into a database.

Description

Technical Field

[0001] The present invention belongs to the field of fault analysis in the power industry, and in particular relates to a stator slot temperature fault analysis method based on an industrial Internet platform. Background Art

[0002] At present, the stator slot temperature of hydro-turbine generators is mainly analyzed manually on a weekly, monthly and annual basis for trend analysis and fault analysis. For any abnormalities that occur, manual collection of relevant measurement point data from each monitoring system is required for statistical analysis, which requires a lot of manpower and time costs and has low analysis efficiency.

[0003] The shortcomings of the existing technology are: the existing hydro-generator stator slot temperature fault analysis method is not highly intelligent, the analysis accuracy is insufficient, and there is a lack of analysis methods combined with industrial Internet platform technology. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a stator slot temperature fault analysis method based on the industrial Internet platform. The present invention can replace manual automatic startup of stator slot temperature fault analysis at any time, and can improve the efficiency of fault analysis and improve the accuracy of analysis.

[0005] In order to solve the above technical problems, the technical solution adopted by the present invention is:

[0006] A stator slot temperature fault analysis method based on the industrial Internet platform is adopted. The industrial Internet platform includes four levels: drive level, analysis level, output level, and data level. The steps of the stator slot temperature fault analysis method are as follows:

[0007] Step 1: Obtain real-time data on stator slot temperature through distributed sensors on the equipment side;

[0008] Step 2: Analyze the real-time data through the trend warning module at the driver level and issue a message indicating that the stator slot temperature of the generator has increased abnormally;

[0009] Step 3: The information analysis module at the driver level processes the above information into trend analysis results and inputs them into the strategy driving module at the driver level to determine the fault mode characteristic indicators;

[0010] Step 4: Based on the data-level database, search for cases with the same or similar failure modes, analyze based on matching rules, determine whether there is a synchronous increasing trend, and output the closest failure mode, failure result, and failure measures based on the matching similarity;

[0011] Step 5: Analyze the cause of the fault based on the FTA database, compare the case matching results, rule analysis results, FTA analysis results, and data analysis results, and determine the cause of the abnormal tank temperature and the impact of the fault. Then, output the fault risk level and the necessary control measures.

[0012] Preferably, in step three, the fault mode characteristic indicators include abnormal slot temperature value-adjacent slot core temperature value, abnormal slot temperature value-adjacent slot tooth pressure plate temperature value, and abnormal slot temperature value-adjacent slot temperature value.

[0013] Preferably, the data level includes a database unit, and the knowledge including rules, algorithms and indicators in the database unit is loaded into the corresponding module to process data and information.

[0014] Preferably, in step five, based on the analysis output, the key fault information is sorted out to determine whether a fault has occurred. If it has occurred, the original fault content is replaced; if it has not occurred, new fault content is generated and entered into the database.

[0015] Preferably, in step five, the FTA database is based on information provided by historical faults of the hydropower system and fault correlation experience, and is organized and established through the FTA engine to establish a fault analysis tree, so as to help find the causes and impacts of similar faults.

[0016] Preferably, the driver level includes a fault analysis driver unit, which includes a trend warning module, an information analysis module, a strategy driving module, and a defective equipment analysis module. The information analysis module is used to obtain a warning result based on the warning information output by the intelligent warning module in the industrial Internet platform and obtain a trend analysis result based on the information of abnormal increase in stator slot temperature output by the trend warning module of the driver level; based on the warning result and the trend analysis result, the defective equipment analysis module identifies the equipment with defects;

[0017] The driving level is used to determine the analysis driving mode when the generator stator slot temperature is abnormal, including event driving and cycle driving, and output the fault mode characteristic value to the analysis level.

[0018] Preferably, the analysis level includes a fault analysis unit, which includes a fault case analysis module, a rule analysis module, an FTA analysis module, a big data analysis module, and a confidence comparison module; the fault case analysis module is based on the case library of the knowledge precipitation level, and searches whether there are identical or similar cases, and is based on the rules of the analysis level. The rules are to retrieve the fault signal measurement point data, fault phenomenon, fault symptoms, fault mode and treatment measures information, and perform semantic matching with the corresponding fields of the case library, and output the closest fault mode, fault result and fault measure according to the matching similarity to obtain the case matching result; the rule analysis module is based on expert experience and equipment operation mechanism, and the core temperature and tooth pressure plate temperature near the adjacent corresponding slot numbers are If there is a correlation between the temperature difference between adjacent slots, it means that the temperature at that location is indeed abnormal. Based on the relationship between the temperatures, the cause of the fault is determined and the rule analysis results are obtained. The FTA analysis module matches the fault mode characteristic indicators at the driver level to the fault mode in the data-level FTA library. Based on the logical fault tree corresponding to the FTA in the data-level FTA library, the fault cause and fault impact are analyzed to obtain the FTA analysis results. Big data analysis calculates the correlation of the historical data of all measuring points of the unit and obtains the data analysis results from a pure data perspective. The confidence comparison module compares the case matching results, rule analysis results, FTA analysis results and data analysis results, and outputs the result with the highest confidence.

[0019] The analysis level is used to analyze the causes and impacts of faults, including fault case analysis, rule analysis, FTA analysis, and big data analysis. The analysis level compares the case matching results, rule analysis results, FTA analysis results, and data analysis results, and outputs the result with the highest confidence.

[0020] Preferably, the output stage is used to display the logic and results of the fault analysis in the form of a mind map.

[0021] Preferably, the data level includes a database unit, and the knowledge including rules, algorithms and indicators in the database unit is loaded into the corresponding module to process the data and information;

[0022] The data level is used to analyze the results of each fault analysis, sort out the key fault information, update the fault knowledge, and structure the process and results and enter them into the database.

[0023] A stator slot temperature fault analysis system adopts the stator slot temperature fault analysis method based on the industrial Internet platform.

[0024] The present invention can achieve the following beneficial effects:

[0025] The present invention can replace manual automatic startup of stator slot temperature fault analysis at any time, and can improve the efficiency of fault analysis and the accuracy of analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The present invention will be further described below with reference to the accompanying drawings and examples:

[0027] Figure 1 This is a flow chart of the fault analysis method of the present invention;

[0028] Figure 2 This is the structural diagram of the industrial Internet platform of the present invention. DETAILED DESCRIPTION

[0029] The preferred solution is Figures 1 to 2 As shown in the figure, a stator slot temperature fault analysis method based on the industrial Internet platform adopts the industrial Internet platform, which includes four levels, namely the driving level, analysis level, output level and data level; the driving level is used to determine the analysis driving mode when the stator slot temperature of the generator is abnormal, including event driving and cycle driving, and output the fault mode characteristic value to the analysis level; the analysis level is used to analyze the cause and impact of the fault, including fault case analysis, rule analysis, FTA analysis and big data analysis. The analysis level compares the case matching results, rule analysis results, FTA analysis results and data analysis results, and outputs the result with the highest confidence; the output level is used to display the logic and results of the fault analysis in the form of a mind map; the data level is used to analyze each fault analysis result, sort out the key fault information, update the fault knowledge and structure the process and results and enter them into the database.

[0030] Fault analysis method flow Figure 1 As shown,

[0031] Step 1: Obtain real-time data on stator slot temperature through distributed sensors on the equipment side;

[0032] Step 2: Analyze the real-time data through the trend warning module of the drive level and issue a message indicating that the stator slot temperature of the generator has abnormally increased;

[0033] Step 3: The information analysis module at the driver level processes the above information into trend analysis results and inputs them into the strategy driving module at the driver level to determine the fault mode characteristic indicators;

[0034] Step 4: Based on the data-level database, search for cases with the same or similar failure modes, analyze based on matching rules, determine whether there is a synchronous increasing trend, and output the closest failure mode, failure result, and failure measures based on the matching similarity;

[0035] Step 5: Analyze the cause of the fault based on the FTA database, compare the case matching results, rule analysis results, FTA analysis results, and data analysis results, and determine the cause of the abnormal tank temperature and the impact of the fault. Then, output the fault risk level and the necessary control measures.

[0036] Fault mode characteristic indicators include abnormal slot temperature value-nearby slot core temperature value, abnormal slot temperature value-nearby slot tooth pressure plate temperature value, abnormal slot temperature value-nearby slot temperature value

[0037] The following is a detailed introduction to each unit.

[0038] 1. The driving level includes a fault analysis driving unit, which includes a trend warning module, an information analysis module, a strategy driving module and a defective equipment analysis module. The information analysis module can obtain warning results based on the warning information output by the intelligent warning module in the industrial Internet platform and obtain trend analysis results based on the information of abnormal increase in stator slot temperature output by the trend warning module of the driving level; based on the warning results and trend analysis results, the defective equipment analysis module clarifies the equipment with defects.

[0039] In order to ensure the normal operation of the fault analysis driving unit, the knowledge including rules, algorithms and indicators in the database needs to be loaded into the policy driving module to process the data and information.

[0040] The fault analysis drive unit can be driven in two modes: event-driven and periodic-driven. Event-driven: fault analysis is initiated after an event such as a warning or alarm occurs. Periodic-driven: fault analysis is initiated at a fixed period such as daily, weekly, or monthly.

[0041] 2. Fault analysis unit, including fault case analysis module, rule analysis module, FTA analysis module, big data analysis module, and confidence comparison module.

[0042] Fault case analysis module: Fault case analysis is based on the database to search for identical or similar cases. It is also based on rules. The rules are to retrieve information such as fault signal measurement point data, fault phenomena, fault symptoms, fault modes and treatment measures, and perform semantic matching with the corresponding fields in the case library. The closest fault mode, fault result and fault measures are output based on the matching similarity to obtain the case matching result.

[0043] Rule analysis module: Based on expert experience and equipment operating mechanisms, if there is a correlation between the core temperature and the tooth pressure plate temperature near adjacent corresponding slot numbers, or the temperature difference between adjacent slot temperatures, it means that the temperature at that location is indeed abnormal. Based on the relationship between the temperatures, the cause of the fault is determined and the rule analysis results are obtained.

[0044] FTA analysis module: matches the fault mode characteristic indicators of the driver level to the FTA fault mode in the database, analyzes the fault cause and fault impact based on the logical fault tree corresponding to the FTA in the database, and obtains the FTA analysis results.

[0045] Big data analysis module: Calculate the correlation of historical data of all measuring points of the unit and obtain data analysis results from a pure data perspective.

[0046] Confidence comparison module: compares case matching results, rule analysis results, FTA analysis results, and data analysis results, and outputs the result with the highest confidence.

[0047] In order to ensure the normal operation of the fault analysis unit, the knowledge including rules, algorithms and indicators in the database is loaded into the corresponding module of the fault analysis unit to process the data and information.

[0048] The key fault information is sorted out to determine whether it is an existing fault. If it is an existing fault, the original fault knowledge is overwritten; if not, new fault knowledge is generated and entered into the database.

[0049] The database may include FMEA database, case database, rule database, FTA database, algorithm database, sample database, and indicator database.

[0050] The FMEA database uses the FMEA analysis method to identify the failure modes and risks of hydropower station equipment, and establish the influence relationship between equipment failure modes, thereby forming a fault association knowledge base.

[0051] The case database uses IT technology to record the various characteristics of industrial failures, faithfully documenting them, summarizing failure experience, and organizing them into a typical failure case library. When a failure occurs again, a case matching algorithm is used to quickly match historical failure cases and retrieve case information. This allows troubleshooting to be performed using historical experience, including failure causes and treatment measures, thereby improving the efficiency of fault analysis and resolution.

[0052] The rule database manages business rules by business theme and stores them as a business rule repository. A rule consists of a set of conditions and one or more actions based on those conditions. Conditions are computable logical expressions (consisting of constants or variables). Conditions can be calculated using mathematical functions, string functions, built-in APIs, scripts, or other custom methods. Actions are output to the console, function execution, external API calls, or other custom methods.

[0053] The FTA database is based on information provided by historical faults and fault correlation experience of the hydropower system. It is organized and established through the FTA engine to establish a fault analysis tree, helping to find the causes and impacts of similar faults, thereby improving fault analysis efficiency and taking effective preventive measures to prevent similar faults from recurring.

[0054] The algorithm database is a collection of general operator algorithms and hydropower operator algorithms.

[0055] The sample database is a data set with certain business value. It needs to meet the requirements of data screening and data processing processes. At the same time, it needs to be closely integrated with the application model to define sample sets of various types and ranges to support the needs of advanced application models. Sample definition is the process of reflecting the value of the sample library and is the core capability of the data sample library. According to different business meanings, it can be divided into several categories: healthy samples, fault samples, and simulation samples. The definition of a healthy sample: from the entire set of time series data, abnormal data is found according to the algorithm, and after manual confirmation, the abnormal data is eliminated, which is a healthy sample; the definition of a fault sample: the operating data when the fault occurs is extracted and saved to provide a reference for fault warning and analysis; the definition of a simulation sample: the data of actual equipment operation does not meet the data input requirements of fault analysis. Usually, simulation samples are used to generate simulation data for fault analysis simulation.

[0056] The indicator database categorizes and manages all indicators that can provide external data services according to different application scenarios. It also provides query and retrieval capabilities for indicator definition information, as well as comprehensive query and retrieval capabilities for indicator results. All indicators in the indicator database can be retrieved and used through online query or real-time interface access.

[0057] The above embodiments are merely preferred technical solutions of the present invention and should not be construed as limiting the present invention. The scope of protection of the present invention shall be the technical solutions set forth in the claims, including equivalent alternatives to the technical features of the technical solutions set forth in the claims. In other words, equivalent alternatives and improvements within this scope are also within the scope of protection of the present invention.

Claims

1. A stator slot temperature fault analysis method based on the industrial Internet platform, characterized by: An industrial Internet platform is used, which includes four levels: drive level, analysis level, output level, and data level. The steps of the stator slot temperature fault analysis method are as follows: Step 1: Obtain real-time data on stator slot temperature through distributed sensors on the equipment side; Step 2: Analyze the real-time data through the trend warning module at the driver level and issue a message indicating that the stator slot temperature of the generator has increased abnormally; Step 3: The information analysis module at the driver level processes the above information into trend analysis results, and then inputs the trend analysis results into the strategy driving module at the driver level to determine the fault mode characteristic indicators; Step 4: Based on the data-level database, search for cases with the same or similar failure modes, analyze based on matching rules, determine whether there is a synchronous increasing trend, and output the closest failure mode, failure result, and failure measures based on the matching similarity; Step 5: Analyze the cause of the fault based on the FTA database. Compare the case matching results, rule analysis results, FTA analysis results, and data analysis results to determine the cause of the abnormal tank temperature and the impact of the fault. Output the fault risk level and the necessary remediation measures. The analysis level includes a fault analysis unit, which includes a fault case analysis module, a rule analysis module, an FTA analysis module, a big data analysis module, and a confidence comparison module. The fault case analysis module searches for identical or similar cases based on the case library at the knowledge precipitation level and uses analysis level rules to retrieve fault signal measurement point data, fault phenomena, fault symptoms, fault modes, and treatment measures, performs semantic matching with corresponding fields in the case library, and outputs the closest fault mode, fault result, and fault measures based on matching similarity to obtain a case matching result. The rule analysis module analyzes based on expert experience and equipment operating mechanisms. If there is a correlation between the core temperature, the tooth pressure plate temperature, and the temperature difference between adjacent slots, it indicates that the temperatures at the adjacent slots are abnormal. Based on the relationship between the temperatures, the cause of the fault is determined and the rule analysis results are obtained. The FTA analysis module matches the fault mode characteristic indicators of the driver level to the fault mode in the data-level FTA library. Based on the logical fault tree corresponding to the FTA in the data-level FTA library, it analyzes the fault cause and fault impact to obtain the FTA analysis results. Big data analysis is to calculate the correlation of historical data of all measuring points of the unit and obtain data analysis results from a pure data perspective; The confidence comparison module compares the case matching results, rule analysis results, FTA analysis results, and data analysis results, and outputs the result with the highest confidence; The analysis level is used to analyze the causes and impacts of faults, including fault case analysis, rule analysis, FTA analysis, and big data analysis. The analysis level compares the case matching results, rule analysis results, FTA analysis results, and data analysis results, and outputs the result with the highest confidence.

2. The stator slot temperature fault analysis method based on the industrial Internet platform according to claim 1 is characterized in that: In step three, the fault mode characteristic indicators include abnormal slot temperature value-adjacent slot core temperature value, abnormal slot temperature value-adjacent slot tooth pressure plate temperature value, and abnormal slot temperature value-adjacent slot temperature value.

3. The stator slot temperature fault analysis method based on the industrial Internet platform according to claim 1 is characterized in that: The data level includes a database unit, and the knowledge including rules, algorithms and indicators in the database unit is loaded into the corresponding module to process data and information.

4. The stator slot temperature fault analysis method based on the industrial Internet platform according to claim 1 is characterized in that: In step 5, based on the analysis output, the key fault information is sorted out to determine whether the fault has occurred. If it has occurred, the original fault content is replaced; If it is not a fault that has occurred before, a new fault content is generated and entered into the database.

5. The stator slot temperature fault analysis method based on the industrial Internet platform according to claim 1 is characterized in that: In step 5, the FTA database is based on information provided by historical faults of the hydropower system and fault correlation experience, and is organized and established through the FTA engine to help find the causes and impacts of similar faults.

6. The stator slot temperature fault analysis method based on the industrial Internet platform according to claim 1 is characterized in that: The driver level includes a fault analysis drive unit, which includes a trend warning module, an information analysis module, a strategy drive module, and a defective equipment analysis module. The information analysis module is used to obtain a warning result based on the warning information output by the intelligent warning module in the industrial Internet platform and obtain a trend analysis result based on the information of abnormal increase in stator slot temperature output by the trend warning module of the driver level; Based on the early warning results and trend analysis results, the defective equipment analysis module identifies the equipment with defects; The driving level is used to determine the analysis driving mode when the generator stator slot temperature is abnormal, including event driving and cycle driving, and output the fault mode characteristic value to the analysis level.

7. The stator slot temperature fault analysis method based on the industrial Internet platform according to claim 1 is characterized in that: The output level is used to display the logic and results of fault analysis in the form of a mind map.

8. The stator slot temperature fault analysis method based on the industrial Internet platform according to claim 1 is characterized in that: The data level includes database units, which load the knowledge including rules, algorithms and indicators in the database units into the corresponding modules to process data and information; The data level is used to analyze the results of each fault analysis, sort out the key fault information, update the fault knowledge, and structure the process and results and enter them into the database.

9. A stator slot temperature fault analysis system, characterized by: A stator slot temperature fault analysis method based on an industrial Internet platform according to any one of claims 1 to 8 is adopted.

Images