AI large model-based wind driven generator fault auxiliary monitoring system

Abstract

The invention relates to the technical field of wind driven generator monitoring, and particularly discloses a wind driven generator fault auxiliary monitoring system based on an AI large model, and the system carries out the real-time monitoring of the working parameters of the core parts of a wind driven generator set, obtains the vibration, temperature and noise parameters of the three core parts during working in real time, and carries out the real-time monitoring of the working parameters. The three core components are numbered once, monitored parameters are conveniently stored and called, the parameters monitored in real time are analyzed and processed based on an AI large model processing module, whether the wind generating set breaks down or not is monitored in an auxiliary mode according to the analysis result, the fault type is found out in an auxiliary mode, and if no fault exists currently, the fault type is found out in an auxiliary mode. The AI large model processing module assists in analyzing the current state of the wind generating set and predicting whether potential safety hazards exist or not, and finally, the early warning module gives out different early warnings according to the auxiliary analysis result of the AI large model processing module.

Description

Technical Field

[0001] The present invention relates to the technical field of wind turbine monitoring, and specifically discloses a wind turbine fault auxiliary monitoring system based on an AI large model. Background Art

[0002] With the rapid development of wind power technology, the scale and number of wind turbines are constantly expanding. However, wind turbines are prone to various faults during operation, which affect power generation efficiency and equipment safety.

[0003] Traditional fault monitoring methods mainly rely on manual inspections and regular maintenance, but this method is inefficient and has a certain lag. When a fault is discovered, it has already affected the normal operation of the wind turbine. In addition, in order to determine the specific fault type of the wind turbine, the current traditional fault monitoring method sometimes requires the wind turbine to be disassembled for inspection. The disassembly process may also cause new faults, and only the current faults of the generator can be monitored, and it is impossible to predict whether the generator will have hidden faults in the future. Therefore, the present invention provides a wind turbine fault auxiliary monitoring system based on an AI large model. Summary of the invention

[0004] The purpose of the present invention is to provide a wind turbine fault auxiliary monitoring system based on an AI large model to solve the problems faced in the above-mentioned background technology.

[0005] The purpose of the present invention can be achieved through the following technical solutions:

[0006] An AI large model-based wind turbine fault auxiliary monitoring system, the system comprising a working parameter monitoring module, a data storage module, an AI large model processing module and an early warning module;

[0007] The operating parameter monitoring module is used to monitor the real-time data of the operating parameters during the operation of the wind turbine;

[0008] The data storage module is used to store the data acquired by the working parameter monitoring module in the system database, so as to facilitate the AI ​​large model processing module to retrieve and analyze;

[0009] The AI ​​large model processing module is used to process and analyze the working parameters of the system database;

[0010] The early warning module is used to issue corresponding early warnings according to the results of the fault assessment module.

[0011] As a further description of the solution of the present invention, the working process of the working parameter monitoring module includes:

[0012] Sensors are installed at reserved installation locations at key parts of the wind turbine to monitor the vibration, temperature and noise parameters of the blades, main shaft and generator in real time during the operation of the wind turbine;

[0013] The blades, main shaft and generator are numbered i respectively, and the numbers i are 1, 2 and 3 respectively;

[0014] The vibration parameters obtained by sensors at the blade, main shaft and generator are amplitude, and vibration frequency ;

[0015] The temperature parameters obtained by sensors at the blades, main shaft and generator are ;

[0016] The noise parameters obtained by sensors at the blade, main shaft and generator are ;

[0017] The monitored data is stored in the system database.

[0018] As a further description of the solution of the present invention, the specific working process of the AI ​​large model processing module includes:

[0019] According to the real-time working parameters monitored by the working parameter monitoring module, the real-time parameters are input into the AI ​​big model to generate real-time vibration curve, real-time temperature curve and real-time noise curve respectively;

[0020] The state coefficients of blades, main shaft and generator are calculated through curve changes.

[0021] As a further description of the solution of the present invention, the specific process of calculating the state coefficients of the blades, the main shaft and the generator through the curve changes includes:

[0022] First, the vibration state coefficients of the blade, main shaft and generator are calculated by the following formula: :

[0023] ;

[0024] Then, the temperature state coefficients of the blades, main shaft and generator are calculated by the following formula: :

[0025] ;

[0026] Then, the noise state coefficients of the blades, main shaft and generator are calculated by the following formula: :

[0027] ;

[0028] In the formula, is the starting time of the wind turbine generator set. The end time of the wind turbine generator set. and is the weight coefficient, for The maximum amplitude of the real-time vibration curve during the period, for The maximum amplitude of the standard vibration curve corresponding to the time period, for Average frequency of real-time vibration curve during the period, for The average frequency of the standard vibration curve corresponding to the time period, is the real-time temperature curve. is the corresponding standard temperature curve, is the real-time noise curve, is the corresponding standard noise curve, t is the time, is the temperature characteristic reference value, is the noise characteristic reference value;

[0029] Finally, the fault types of wind turbines are analyzed based on the state coefficients.

[0030] As a further description of the solution of the present invention, the specific process of analyzing the fault type of the wind turbine generator set according to the state coefficient includes:

[0031] The vibration state coefficients of the blades, main shaft and generator are Preset vibration state coefficient thresholds for blades, main shaft and generator Compare:

[0032] If exists Greater than or equal to When , it means that the blades of the wind turbine generator set are vibrating abnormally. , it means that the main shaft vibration of the wind turbine is abnormal. , it means that the generator vibration of the wind turbine generator set is abnormal;

[0033] The temperature state coefficients of the blades, main shaft and generator are Compare with preset blade, shaft and generator temperature condition coefficient thresholds Compare;

[0034] If exists Greater than or equal to When , it means that the blade temperature of the wind turbine is abnormal. , it means that the main shaft temperature of the wind turbine is abnormal. , it means that the generator temperature of the wind turbine generator set is abnormal;

[0035] The noise state coefficients of the blades, main shaft and generator are Compare with preset noise condition coefficient thresholds for blades, main shaft and generator Compare;

[0036] If exists Greater than or equal to When , it means that the blade noise of the wind turbine is abnormal. , it means that the main shaft noise of the wind turbine is abnormal. , it means that the generator noise of the wind turbine is abnormal.

[0037] As a further description of the solution of the present invention, the specific working process of the AI ​​large model processing module also includes:

[0038] like Less than , Less than and Less than When the wind turbine generator system is in operation, the overall status of the wind turbine generator system is analyzed:

[0039] The overall state analysis coefficient A of the wind turbine generator set is calculated by the following formula:

[0040] ;

[0041] In the formula, and is the weight coefficient, is the vibration weight coefficient of the corresponding i-th component, is the temperature weight coefficient of the corresponding i-th component, is the noise weight coefficient of the corresponding i-th component;

[0042] The overall state analysis coefficient A and the preset overall state analysis coefficient threshold Compare, if A is greater than or equal to , it means that the overall state of the wind turbine is abnormal and there is a potential fault.

[0043] As a further description of the solution of the present invention, the early warning module sends out different early warning signals according to different fault types when working.

[0044] As a further description of the solution of the present invention, the data storage module stores different types of data in categories and names them according to numbers.

[0045] Beneficial effects of the present invention: The present invention monitors the working parameters of the core components of the wind turbine in real time, obtains the vibration, temperature and noise parameters of the three core components when they are working in real time, and numbers the three core components once, which is convenient for storing and retrieving the monitored parameters, and analyzes and processes the real-time monitored parameters based on the AI ​​large model processing module, and assists in monitoring whether the wind turbine has faults and assists in finding out the type of fault according to the analysis results. If there is no fault at present, the AI ​​large model processing module assists in analyzing the current state of the wind turbine and predicts whether there are still safety hazards. Finally, the early warning module issues different early warnings according to the analysis results assisted by the AI ​​large model processing module.

[0046] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for describing the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0048] Figure 1 It is a schematic diagram of the overall structure of the multi-source heterogeneous data synchronization and reconstruction system of the new energy station of the present invention. DETAILED DESCRIPTION

[0049] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0050] See also Figure 1 As shown, a wind turbine fault auxiliary monitoring system based on an AI large model, the system includes a working parameter monitoring module, a data storage module, an AI large model processing module and an early warning module;

[0051] The operating parameter monitoring module is used to monitor the real-time data of the operating parameters during the operation of the wind turbine;

[0052] The data storage module is used to store the data acquired by the working parameter monitoring module in the system database, so as to facilitate the AI ​​large model processing module to retrieve and analyze;

[0053] The AI ​​large model processing module is used to process and analyze the working parameters of the system database;

[0054] The early warning module is used to issue corresponding early warnings according to the results of the fault assessment module.

[0055] As a further description of the solution of the present invention, the working process of the working parameter monitoring module includes:

[0056] Sensors are installed at reserved installation locations at key parts of the wind turbine to monitor the vibration, temperature and noise parameters of the blades, main shaft and generator in real time during the operation of the wind turbine;

[0057] The blades, main shaft and generator are numbered i respectively, and the numbers i are 1, 2 and 3 respectively;

[0058] The vibration parameters obtained by sensors at the blade, main shaft and generator are amplitude, and vibration frequency ;

[0059] The temperature parameters obtained by sensors at the blades, main shaft and generator are ;

[0060] The noise parameters obtained by sensors at the blade, main shaft and generator are ;

[0061] The monitored data is stored in the system database.

[0062] Through the above technical solution, the working parameter monitoring module of the present invention monitors the key part parameters of the wind turbine generator set through sensors, wherein the key parts include:

[0063] Blades, which capture wind energy and convert it into mechanical energy, are the main mechanical components of wind turbines;

[0064] The main shaft transmits the rotational motion of the wind rotor to the generator and is the main transmission component of the wind turbine;

[0065] Generator: The generator rotates and generates electricity driven by the wind rotor shaft, converting mechanical energy into electrical energy. It is the core component of the wind turbine generator set.

[0066] The present invention monitors the working parameters of the three core components in real time, obtains the vibration, temperature and noise parameters of the three core components when they are working in real time, and numbers the three core components once, which is convenient for storing and retrieving the monitored parameters, and analyzes and processes the real-time monitored parameters based on the AI ​​large model processing module, and assists in monitoring whether the wind turbine generator set has faults according to the analysis results, and assists in finding out the type of fault. If there is no fault at present, the AI ​​large model processing module assists in analyzing the current state of the wind turbine generator set and determines whether there are still safety hazards. Finally, the early warning module issues different early warnings according to the analysis results assisted by the AI ​​large model processing module.

[0067] As a further description of the solution of the present invention, the specific working process of the AI ​​large model processing module includes:

[0068] According to the real-time working parameters monitored by the working parameter monitoring module, the real-time parameters are input into the AI ​​big model to generate real-time vibration curve, real-time temperature curve and real-time noise curve respectively;

[0069] The state coefficients of blades, main shaft and generator are calculated through curve changes.

[0070] As a further description of the solution of the present invention, the specific process of calculating the state coefficients of the blades, the main shaft and the generator through the curve changes includes:

[0071] First, the vibration state coefficients of the blade, main shaft and generator are calculated by the following formula: :

[0072] ;

[0073] Then, the temperature state coefficients of the blades, main shaft and generator are calculated by the following formula: :

[0074] ;

[0075] Then, the noise state coefficients of the blades, main shaft and generator are calculated by the following formula: :

[0076] ;

[0077] In the formula, is the starting time of the wind turbine generator set. The end time of the wind turbine generator set. and is the weight coefficient, for The maximum amplitude of the real-time vibration curve during the period, for The maximum amplitude of the standard vibration curve corresponding to the time period, for Average frequency of real-time vibration curve during the period, for The average frequency of the standard vibration curve corresponding to the time period, is the real-time temperature curve. is the corresponding standard temperature curve, is the real-time noise curve, is the corresponding standard noise curve, t is the time, is the temperature characteristic reference value, is the noise characteristic reference value;

[0078] Finally, the fault types of wind turbines are analyzed based on the state coefficients.

[0079] Through the above technical solution, this embodiment inputs the real-time parameters monitored by the working parameter monitoring module into the AI ​​big model to generate real-time vibration curve, real-time temperature curve and real-time noise curve respectively; then, based on the real-time vibration curve, the formula Calculate the vibration state coefficients of the blades, main shaft and generator respectively , based on the real-time temperature curve through the formula Calculate the temperature state coefficients of the blades, main shaft and generator separately , based on the real-time noise curve through the formula Calculate the noise state coefficients of the blades, main shaft and generator separately .

[0080] It should be noted that and is the weight coefficient set according to the empirical data fitting, is the temperature characteristic reference value, The noise characteristic reference value is determined according to the sensor measurement error changing with the environment.

[0081] As a further description of the solution of the present invention, the specific process of analyzing the fault type of the wind turbine generator set according to the state coefficient includes:

[0082] The vibration state coefficients of the blades, main shaft and generator are Preset vibration state coefficient thresholds for blades, main shaft and generator Compare:

[0083] If exists Greater than or equal to When , it means that the blades of the wind turbine generator set are vibrating abnormally. , it means that the main shaft vibration of the wind turbine is abnormal. , it means that the generator vibration of the wind turbine generator set is abnormal;

[0084] The temperature state coefficients of the blades, main shaft and generator are Compare with preset blade, shaft and generator temperature condition coefficient thresholds Compare;

[0085] If exists Greater than or equal to When , it means that the blade temperature of the wind turbine is abnormal. , it means that the main shaft temperature of the wind turbine is abnormal. , it means that the generator temperature of the wind turbine generator set is abnormal;

[0086] The noise state coefficients of the blades, main shaft and generator are Compare with preset noise condition coefficient thresholds for blades, main shaft and generator Compare;

[0087] If exists Greater than or equal to When , it means that the blade noise of the wind turbine is abnormal. , it means that the main shaft noise of the wind turbine is abnormal. , it means that the generator noise of the wind turbine is abnormal.

[0088] Through the above technical solution, this embodiment uses the vibration state coefficients of the blades, the main shaft and the generator. Preset vibration state coefficient thresholds for blades, main shaft and generator Compare and judge whether the current wind turbine generator set has abnormal vibration, and determine the abnormal vibration location based on the judgment result, and use the temperature state coefficient of the blades, main shaft and generator to determine the abnormal vibration location. Compare with preset blade, shaft and generator temperature condition coefficient thresholds Compare and judge whether the current wind turbine generator set has temperature anomaly, and determine the temperature anomaly location based on the judgment result, and use the noise state coefficient of the blades, main shaft and generator to Compare with preset noise condition coefficient thresholds for blades, main shaft and generator By comparison, it is determined whether the current wind turbine generator set has abnormal noise, and the abnormal noise location is determined according to the determination result.

[0089] As a further description of the solution of the present invention, the specific working process of the AI ​​large model processing module also includes:

[0090] like Less than , Less than and Less than When the wind turbine generator system is in operation, the overall status of the wind turbine generator system is analyzed:

[0091] The overall state analysis coefficient A of the wind turbine generator set is calculated by the following formula:

[0092] ;

[0093] In the formula, and is the weight coefficient, is the vibration weight coefficient of the corresponding i-th component, is the temperature weight coefficient of the corresponding i-th component, is the noise weight coefficient of the corresponding i-th component;

[0094] The overall state analysis coefficient A and the preset overall state analysis coefficient threshold Compare, if A is greater than or equal to , it means that the overall state of the wind turbine is abnormal and there is a potential fault.

[0095] Through the above technical solution, in the case that all key components are normal, this embodiment also predicts whether the wind turbine has safety hazards based on the overall state analysis coefficient of the wind turbine. Calculate the overall state analysis coefficient A of the wind turbine generator set and compare the overall state analysis coefficient A with the preset overall state analysis coefficient threshold Compare, if A is greater than or equal to , it means that the overall state of the wind turbine is abnormal and there is a potential fault.

[0096] It should be noted that and is the weight coefficient, which is set according to the empirical data fitting.

[0097] As a further description of the solution of the present invention, the early warning module sends out different early warning signals according to different fault types when working.

[0098] As a further description of the solution of the present invention, the data storage module stores different types of data in categories and names them according to numbers.

[0099] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A wind turbine fault auxiliary monitoring system based on AI large model, characterized in that: The system includes a working parameter monitoring module, a data storage module, an AI large model processing module and an early warning module; The operating parameter monitoring module is used to monitor the real-time data of the operating parameters during the operation of the wind turbine; The data storage module is used to store the data acquired by the working parameter monitoring module in the system database, so as to facilitate the AI ​​large model processing module to retrieve and analyze; The AI ​​large model processing module is used to process and analyze the working parameters of the system database; The early warning module is used to issue corresponding early warnings according to the results of the fault assessment module.

2. According to claim 1, a wind turbine fault auxiliary monitoring system based on AI large model is characterized in that: The working process of the working parameter monitoring module includes: Sensors are installed at reserved installation locations at key parts of the wind turbine to monitor the vibration, temperature and noise parameters of the blades, main shaft and generator in real time during the operation of the wind turbine; The blades, main shaft and generator are numbered i respectively, and the numbers i are 1, 2 and 3 respectively; The vibration parameters obtained by sensors at the blade, main shaft and generator are amplitude, and vibration frequency ; The temperature parameters obtained by sensors at the blades, main shaft and generator are ; The noise parameters obtained by sensors at the blade, main shaft and generator are ; The monitored data is stored in the system database.

3. According to claim 1, a wind turbine fault auxiliary monitoring system based on AI large model is characterized in that: The specific working process of the AI ​​large model processing module includes: According to the real-time working parameters monitored by the working parameter monitoring module, the real-time parameters are input into the AI ​​big model to generate real-time vibration curve, real-time temperature curve and real-time noise curve respectively; The state coefficients of blades, main shaft and generator are calculated through curve changes.

4. According to claim 3, a wind turbine fault auxiliary monitoring system based on AI large model is characterized in that: The specific process of calculating the state coefficients of the blades, the main shaft and the generator through the curve changes includes: First, the vibration state coefficients of the blade, main shaft and generator are calculated by the following formula: : ； Then, the temperature state coefficients of the blades, main shaft and generator are calculated by the following formula: : ； Then, the noise state coefficients of the blades, main shaft and generator are calculated by the following formula: : ； In the formula, is the starting time of the wind turbine generator set. The end time of the wind turbine generator set. and is the weight coefficient, for The maximum amplitude of the real-time vibration curve during the period, for The maximum amplitude of the standard vibration curve corresponding to the time period, for Average frequency of real-time vibration curve during the period, for The average frequency of the standard vibration curve corresponding to the time period, is the real-time temperature curve. is the corresponding standard temperature curve, is the real-time noise curve, is the corresponding standard noise curve, t is the time, is the temperature characteristic reference value, is the noise characteristic reference value; Finally, the fault types of wind turbines are analyzed based on the state coefficients.

5. According to claim 4, a wind turbine fault auxiliary monitoring system based on AI large model is characterized in that: The specific process of analyzing the fault type of the wind turbine generator set according to the state coefficient includes: The vibration state coefficients of the blades, main shaft and generator are Preset vibration state coefficient thresholds for blades, main shaft and generator Compare: If exists Greater than or equal to When , it means that the blades of the wind turbine generator set vibrate abnormally. , it means that the main shaft vibration of the wind turbine is abnormal. , it means that the generator vibration of the wind turbine generator set is abnormal; The temperature state coefficients of the blades, main shaft and generator are Compare with preset blade, shaft and generator temperature condition coefficient thresholds Compare; If exists Greater than or equal to When , it means that the blade temperature of the wind turbine is abnormal. , it means that the main shaft temperature of the wind turbine is abnormal. , it means that the generator temperature of the wind turbine generator set is abnormal; The noise state coefficients of the blades, main shaft and generator are Compare with preset noise condition coefficient thresholds for blades, main shaft and generator Compare; If exists Greater than or equal to When , it means that the blade noise of the wind turbine is abnormal. , it means that the main shaft noise of the wind turbine is abnormal. , it means that the generator noise of the wind turbine is abnormal.

6. The wind turbine fault auxiliary monitoring system based on AI large model according to claim 5 is characterized in that: The specific working process of the AI ​​large model processing module also includes: like Less than , Less than and Less than When the wind turbine generator system is in operation, the overall status of the wind turbine generator system is analyzed: The overall state analysis coefficient A of the wind turbine generator set is calculated by the following formula: ； In the formula, and is the weight coefficient, is the vibration weight coefficient of the corresponding i-th component, is the temperature weight coefficient of the corresponding i-th component, is the noise weight coefficient of the corresponding i-th component; The overall state analysis coefficient A and the preset overall state analysis coefficient threshold Compare, if A is greater than or equal to , it means that the overall state of the wind turbine is abnormal and there is a potential fault.

7. The wind turbine fault auxiliary monitoring system based on AI large model according to claim 1 is characterized in that: The early warning module sends out different early warning signals according to different fault types when working.

8. The wind turbine fault auxiliary monitoring system based on AI large model according to claim 1 is characterized in that: The data storage module stores different types of data in categories and names them according to numbers.

Images