Wind turbine generator gearbox fault early warning method and system

Through the analysis of historical fault data and real-time data of the gearbox of the wind turbine, combined with the SCADA system and signal processing methods, potential fault characteristics are identified and thresholds are set, early warning is triggered, and online fine filters are installed at the bottom of the gearbox, the existing technology is solved The problem of difficulty in fault positioning and low maintenance efficiency has achieved timely early warning and efficient maintenance, extending equipment life and improving operational stability.

CN118913682APending Publication Date: 2024-11-08HEBEI HUADIAN GUYUAN WIND POWER CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410969829.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-19
Publication Date
2024-11-08

AI Technical Summary

Technical Problem

The existing wind turbine gearbox fault warning methods are difficult to accurately locate the cause of the fault, which increases the inspection time of maintenance personnel and reduces maintenance and maintenance efficiency.

Method used

Through statistical analysis of historical fault data of wind turbine gearbox, combined with gearbox operating parameters and environmental data, SCADA system, signal processing methods and big data analysis, potential fault characteristics are identified and thresholds are set, early warning is triggered, and at the bottom of the gearbox Install an online fine filter for secondary filtration to improve the quality of oil and reduce impurities.

Benefits of technology

It realizes timely early warning, accurately locates the cause of failure, reduces inspection time, improves maintenance efficiency, extends the service life of the equipment, and improves the quality of lubricant through online fine filters, improving operating stability and overall safety level.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118913682A_ABST
    Figure CN118913682A_ABST
Patent Text Reader

Abstract

The invention discloses a wind turbine generator gearbox fault early warning method and system. The method comprises the following steps: S1, carrying out statistical analysis on historical fault data of a wind turbine generator gearbox, and identifying fault types, occurrence times and proportions thereof; s2, operating parameters of the gear box and environment condition data around the gear box are collected; s3, analyzing the fault reason of the gear box, and analyzing the data of the surrounding environment of the gear box, the data of lubricating oil in the gear box and the data of gear box parts; and S4, collecting and analyzing data by using an SCADA system, and identifying an abnormal trend. The wind turbine gearbox fault early warning method has the advantage of timely early warning, and the problems that in the using process of an existing wind turbine gearbox fault early warning method, actual fault reasons cannot be accurately positioned conveniently, the troubleshooting time of maintenance personnel is prolonged, and the maintenance and overhaul efficiency of a wind turbine gearbox is reduced are solved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of wind power generation, and in particular to a wind turbine gearbox fault early warning method and system. Background Art

[0002] As the operating life of wind turbines increases, the number of wind turbine gearbox failures increases year by year, directly affecting various operating indicators of wind turbines. In order to ensure the stable operation of wind turbines and improve the economic benefits of wind farms, it is necessary to reduce the failure rate of wind turbine gearboxes. In order to reduce the number of wind turbine failures, ensure the smooth and safe operation of wind turbines, and effectively increase power generation, it is necessary to provide advance warning of wind turbine gearbox failures to facilitate maintenance personnel to carry out early inspections and prevent the generator from shutting down.

[0003] The existing wind turbine gearbox fault warning method is not convenient for accurately locating the actual fault cause during use, which increases the troubleshooting time of maintenance personnel and reduces the efficiency of wind turbine gearbox maintenance and overhaul. Summary of the invention

[0004] The purpose of the present invention is to provide a wind turbine gearbox fault warning method and system, which has the advantage of timely warning and solves the problem that the existing wind turbine gearbox fault warning method is not convenient for accurately locating the actual cause of the fault during use, which increases the troubleshooting time of maintenance personnel and reduces the efficiency of wind turbine gearbox maintenance and overhaul.

[0005] To achieve the above object, the present invention provides the following technical solution: a wind turbine gearbox fault early warning method, comprising the following steps:

[0006] S1. Statistical analysis of historical fault data of wind turbine gearboxes to identify fault types, occurrence frequency and proportion;

[0007] S2, collecting gearbox operating parameters and environmental condition data around the gearbox;

[0008] S3. Analyze the cause of the gearbox failure based on the gearbox surrounding environment data, gearbox internal lubricating oil data and gearbox parts data;

[0009] S4. Collect and analyze data using the SCADA system to identify abnormal trends;

[0010] S5. Apply signal processing methods and big data analysis methods to analyze data patterns through algorithms, identify potential fault characteristics, and set thresholds to trigger warnings when monitoring data approaches or exceeds the thresholds;

[0011] S6. Set up maintenance plan according to early warning signals;

[0012] S7, gearbox modification, install an online fine filter at the oil drain valve at the bottom of the gearbox to form a bypass with the crude oil circulation system, and perform secondary filtration through the fine filter to gradually improve the oil quality and reduce impurities;

[0013] S8. After the fault is handled, the effect is evaluated and fed back to the early warning system, the fault database is updated, and the early warning model is optimized;

[0014] S9. Regularly review and adjust early warning strategies to ensure the accuracy and effectiveness of the early warning system.

[0015] As a preferred method for early warning of a wind turbine gearbox fault of the present invention, the historical faults in S1 include gearbox oil pressure being too low, gearbox heating time being overdue, gearbox oil temperature being too high, high and low speed modes of the gearbox oil cooling pump being not turned on, gearbox oil temperature sensor circuit failure, gearbox bearing temperature being too high, large gearbox vibration and abnormal sound, gear oil blockage, abnormal gearbox oil viscosity change, and abnormalities detected in gearbox components.

[0016] As a preferred method for early warning of a wind turbine gearbox fault of the present invention, the operating parameters in S2 include oil pressure, oil temperature, vibration level and noise, and the environmental condition data around the gearbox include ambient humidity and ambient temperature.

[0017] As a preferred method for early warning of a wind turbine gearbox fault of the present invention, when collecting the gearbox operation data in S2, an oil pressure sensor, an oil temperature sensor, a vibration sensor and a noise sensor are installed to monitor the gearbox oil pressure, oil temperature, vibration frequency and noise, and transmit them to the monitoring system in real time. When collecting the environmental condition data around the gearbox, an air humidity sensor and an air temperature sensor are installed to monitor the humidity and temperature of the environment around the gearbox.

[0018] As a preferred method for early warning of a wind turbine gearbox fault of the present invention, the abnormal trends in S4 include increased oil temperature, increased vibration and increased noise.

[0019] As a preferred method for early warning of a wind turbine gearbox fault of the present invention, when the SCADA system in S4 collects and analyzes data, it first processes the collected data to remove invalid data, fill in missing values, smooth noise, ensure data quality, and store the processed data in a database.

[0020] As a preferred method for wind turbine gearbox fault warning of the present invention, the signal processing method in S5 includes a low-pass filtering method, a high-pass filtering method, a band-pass filtering method, a singular value decomposition method and an empirical mode decomposition method, and the big data analysis method includes a time series analysis method, a machine learning algorithm, a deep learning method and anomaly detection algorithm.

[0021] As a preferred method for early warning of a wind turbine gearbox fault of the present invention, the maintenance plan in S6 includes replacing a filter element, cleaning an oil circuit, and inspecting and replacing worn parts of the gearbox.

[0022] As a preferred method for early warning of a wind turbine gearbox fault of the present invention, the fine filter in S7 uses a fine filter system with a flow rate of 120L / H, a model of HDU27 / 27, a filtration accuracy of 3 microns, and a dirt holding capacity of 4Kg.

[0023] A wind turbine gearbox fault early warning system, comprising:

[0024] Data acquisition module, the data acquisition module includes an oil pressure sensor, an oil temperature sensor, a vibration sensor, a noise sensor, an air humidity sensor and an air temperature sensor;

[0025] SCADA systems;

[0026] Signal processing and big data analysis module;

[0027] Early warning and maintenance management module;

[0028] Remote monitoring and communication module;

[0029] Knowledge base and fault database module;

[0030] User interface and report generation module.

[0031] Compared with the prior art, the present invention has the following beneficial effects:

[0032] 1. The present invention can identify potential fault characteristics early and trigger early warning signals through statistical analysis of historical data and monitoring of real-time data, which enables maintenance work to be carried out in advance, avoids downtime losses caused by sudden faults, and extends the service life of wind turbines.

[0033] 2. The present invention can more accurately locate the cause of the fault, reduce troubleshooting time and improve maintenance efficiency by comprehensively using multi-dimensional analysis such as gearbox operating parameters, environmental data, lubricating oil status and parts status.

[0034] 3. The present invention makes maintenance planning more reasonable by utilizing the SCADA system and data analysis technology, and improves maintenance quality based on actual data.

[0035] 4. Through the effect evaluation and feedback mechanism after fault handling, the system of the present invention can continuously learn and improve, update the fault database and optimize the early warning model, thereby improving the accuracy and adaptability of the early warning system.

[0036] 5. The present invention improves the quality of lubricating oil by installing an online fine filter, reduces failures caused by oil quality problems, ensures that the gearbox operates more stably and efficiently, and improves the overall safety level of the wind farm.

[0037] 6. The present invention reduces the need for manual intervention through an integrated monitoring system and data analysis process, realizes automation and intelligence of operation and maintenance management, reduces labor costs, and improves management efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 It is a schematic diagram of the flow of the fault early warning method of the present invention;

[0039] Figure 2 Schematic diagram of the fault warning system of the present invention. DETAILED DESCRIPTION

[0040] Example 1

[0041] See also Figure 1 , a wind turbine gearbox fault early warning method, comprising the following steps:

[0042] S1. Statistical analysis of historical fault data of wind turbine gearboxes to identify fault types, occurrence frequency and proportion;

[0043] Historical faults include gearbox oil pressure too low, gearbox heating time timeout, gearbox oil temperature too high, gearbox oil cooling pump high and low speed modes not turned on, gearbox oil temperature sensor circuit failure, gearbox bearing temperature too high, gearbox vibration and abnormal sound, gear oil blockage, abnormal gearbox oil viscosity change and abnormalities detected in gearbox components.

[0044] S2, collecting gearbox operating parameters and environmental condition data around the gearbox;

[0045] Operating parameters include oil pressure, oil temperature, vibration level and noise, and environmental conditions around the gearbox include ambient humidity and ambient temperature;

[0046] When collecting gearbox operation data, install oil pressure sensors, oil temperature sensors, vibration sensors and noise sensors to monitor the gearbox oil pressure, oil temperature, vibration frequency and noise, and transmit them to the monitoring system in real time. When collecting environmental condition data around the gearbox, install air humidity sensors and air temperature sensors to monitor the humidity and temperature of the environment around the gearbox.

[0047] S3. Analyze the cause of the gearbox failure based on the gearbox surrounding environment data, gearbox internal lubricating oil data and gearbox parts data;

[0048] Abnormal trends include increased oil temperature, increased vibration, and increased noise.

[0049] S4. Collect and analyze data using the SCADA system to identify abnormal trends;

[0050] When the SCADA system collects and analyzes data, it first processes the collected data to remove invalid data, fill in missing values, smooth noise, ensure data quality, and store the processed data in a database.

[0051] S5. Apply signal processing methods and big data analysis methods to analyze data patterns through algorithms, identify potential fault characteristics, and set thresholds to trigger warnings when monitoring data approaches or exceeds the thresholds;

[0052] Signal processing methods include low-pass filtering method, high-pass filtering method, band-pass filtering method, singular value decomposition method and empirical mode decomposition method. Big data analysis methods include time series analysis method, machine learning algorithm, deep learning method and anomaly detection algorithm.

[0053] S6. Set up maintenance plan according to early warning signals;

[0054] The maintenance schedule includes replacing filters, cleaning oil lines, and inspecting and replacing gearbox wear parts.

[0055] S7, gearbox modification, install an online fine filter at the oil drain valve at the bottom of the gearbox to form a bypass with the crude oil circulation system, and perform secondary filtration through the fine filter to gradually improve the oil quality and reduce impurities;

[0056] The fine filter uses a fine filtration system with a flow rate of 120L / H, model HDU27 / 27, a filtration accuracy of 3 microns, and a dirt holding capacity of 4Kg.

[0057] S8. After the fault is handled, the effect is evaluated and fed back to the early warning system, the fault database is updated, and the early warning model is optimized.

[0058] S9. Regularly review and adjust early warning strategies to ensure the accuracy and effectiveness of the early warning system.

[0059] Through statistical analysis of historical data and monitoring of real-time data, potential fault characteristics can be identified early and early warning signals can be triggered in time, which allows maintenance work to be carried out in advance, avoids downtime losses caused by sudden failures, and extends the service life of wind turbines.

[0060] The comprehensive use of multi-dimensional analysis such as gearbox operating parameters, environmental data, lubricating oil status and parts condition can more accurately locate the cause of the fault, reduce troubleshooting time and improve maintenance efficiency.

[0061] Utilize SCADA system and data analysis technology to make maintenance plan formulation more reasonable, based on actual data, and improve maintenance quality.

[0062] Through the effect evaluation and feedback mechanism after fault handling, the system can continuously learn and improve, update the fault database and optimize the early warning model, thereby improving the accuracy and adaptability of the early warning system.

[0063] By installing an online fine filter, the quality of the lubricating oil is improved, the failures caused by oil problems are reduced, the gearbox operation is ensured to be more stable and efficient, and the overall safety level of the wind farm is improved.

[0064] The integrated monitoring system and data analysis process reduce the need for manual intervention, realize the automation and intelligence of operation and maintenance management, reduce labor costs and improve management efficiency.

[0065] Example 2

[0066] See also Figure 2 , a wind turbine gearbox fault warning system, comprising:

[0067] Data acquisition module, which includes oil pressure sensor, oil temperature sensor, vibration sensor, noise sensor, air humidity sensor and air temperature sensor; it collects various operating parameters and environmental data of the wind turbine gearbox in real time, and is used to accurately obtain information on oil pressure, oil temperature, vibration level, noise, ambient humidity and ambient temperature;

[0068] SCADA system, the core of data collection and analysis, is responsible for receiving real-time data from the data acquisition module, performing data preprocessing, and storing the processed data in the database. In addition, it also has data analysis capabilities to identify abnormal trends in the data, such as increased oil temperature, increased vibration, and increased noise;

[0069] Signal processing and big data analysis module: The signal processing and big data analysis module combines a variety of signal processing technologies and big data analysis methods to conduct in-depth analysis of pre-processed data, identify potential fault characteristics, and issue warnings based on the thresholds set by the algorithm;

[0070] Early warning and maintenance management module: The early warning and maintenance management module triggers early warning signals based on data analysis results and formulates corresponding maintenance plans according to the early warning level, such as replacing filter elements, cleaning oil circuits, and inspecting and replacing worn parts. The module also has an execution tracking and feedback system for maintenance plans to evaluate maintenance effects and continuously optimize early warning models;

[0071] Remote monitoring and communication module: The remote monitoring and communication module ensures that all collected data can be transmitted to the central control room or cloud platform in real time, supports remote monitoring and fault diagnosis, and also enables system instructions to be quickly issued to on-site equipment;

[0072] The knowledge base and fault database module stores historical fault data, fault types and processing scheme information, provides basic data support for fault identification and early warning model optimization, and improves the prediction accuracy and processing efficiency of the system through continuous feedback and updates;

[0073] User interface and report generation module: The user interface and report generation module provides operators with an intuitive operation interface, displays real-time monitoring data, early warning information and maintenance plans, and can also generate fault analysis reports and system performance evaluation reports.

[0074] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A wind turbine gearbox fault early warning method, characterized in that: The following steps are involved: S1. Statistical analysis of historical fault data of wind turbine gearboxes to identify fault types, occurrence frequency and proportion; S2, collecting gearbox operating parameters and environmental condition data around the gearbox; S3. Analyze the cause of the gearbox failure based on the gearbox surrounding environment data, gearbox internal lubricating oil data and gearbox parts data; S4. Collect and analyze data using the SCADA system to identify abnormal trends; S5. Apply signal processing methods and big data analysis methods to analyze data patterns through algorithms, identify potential fault characteristics, and set thresholds to trigger warnings when monitoring data approaches or exceeds the thresholds; S6. Set up maintenance plan according to early warning signals; S7, gearbox modification, install an online fine filter at the oil drain valve at the bottom of the gearbox to form a bypass with the crude oil circulation system, and perform secondary filtration through the fine filter to gradually improve the oil quality and reduce impurities; S8. After the fault is handled, the effect is evaluated and fed back to the early warning system, the fault database is updated, and the early warning model is optimized; S9. Regularly review and adjust early warning strategies to ensure the accuracy and effectiveness of the early warning system.

2. A wind turbine gearbox fault early warning method according to claim 1, characterized in that: The historical faults in S1 include low gearbox oil pressure, gearbox heating time timeout, gearbox oil temperature too high, gearbox oil cooling pump high and low speed modes not turned on, gearbox oil temperature sensor circuit failure, gearbox bearing temperature too high, large gearbox vibration and abnormal sound, gear oil blockage, abnormal gearbox oil viscosity change and abnormal detection of gearbox components.

3. A wind turbine gearbox fault early warning method according to claim 1, characterized in that: The operating parameters in S2 include oil pressure, oil temperature, vibration level and noise, and the environmental condition data around the gearbox include ambient humidity and ambient temperature.

4. A wind turbine gearbox fault early warning method according to claim 1, characterized in that: When collecting the gearbox operation data in S2, an oil pressure sensor, an oil temperature sensor, a vibration sensor and a noise sensor are installed to monitor the gearbox oil pressure, oil temperature, vibration frequency and noise, and transmit them to the monitoring system in real time. When collecting the environmental condition data around the gearbox, an air humidity sensor and an air temperature sensor are installed to monitor the humidity and temperature of the environment around the gearbox.

5. A wind turbine gearbox fault early warning method according to claim 1, characterized in that: The abnormal trends in S4 include increased oil temperature, increased vibration and increased noise.

6. A wind turbine gearbox fault early warning method according to claim 1, characterized in that: When the SCADA system in S4 collects and analyzes data, it first processes the collected data to remove invalid data, fill in missing values, smooth noise, ensure data quality, and store the processed data in a database.

7. A wind turbine gearbox fault early warning method according to claim 1, characterized in that: The signal processing methods in S5 include low-pass filtering methods, high-pass filtering methods, band-pass filtering methods, singular value decomposition methods and empirical mode decomposition methods, and the big data analysis methods include time series analysis methods, machine learning algorithms, deep learning methods and anomaly detection algorithms.

8. A wind turbine gearbox fault early warning method according to claim 1, characterized in that: The maintenance plan in the S6 includes replacing the filter element, cleaning the oil circuit, and inspecting and replacing the gearbox wear parts.

9. A wind turbine gearbox fault early warning method according to claim 1, characterized in that: The fine filter in S7 uses a fine filter system with a flow rate of 120L / H, model HDU27 / 27, a filtration accuracy of 3 microns, and a dirt holding capacity of 4Kg.

10. A wind turbine gearbox fault warning system, applicable to a wind turbine gearbox fault warning method according to any one of claims 1 to 9, characterized in that: include: Data acquisition module, the data acquisition module includes an oil pressure sensor, an oil temperature sensor, a vibration sensor, a noise sensor, an air humidity sensor and an air temperature sensor; SCADA systems; Signal processing and big data analysis module; Early warning and maintenance management module; Remote monitoring and communication module; Knowledge base and fault database module; User interface and report generation module.