Fault early warning system and determine method for wind turbine generator system

Abstract

The invention relates to a fault early warning system of a wind turbine generator system. The fault early warning system comprises a real-time monitoring module, an early warning model module, an early warning result feedback module and a knowledge database module. The real-time monitoring module is used for reading parameters of a blower fan in real time and transmitting the parameters to the early warning model module for analysis, if an analysis result reaches early warning standards, an early warning mechanism is triggered, the information is transmitted to the early warning result feedback module, the feedback result is recorded by a worker into the knowledge database module; the results of the early warning result feedback module are continuously accumulated by the knowledge database module, when accuracy of the early warning results is lower than 60%, the result of the knowledge database module is fed back to the early warning model module to carry out model re-training. The fault early warning system is advantaged in that early warning in advance before fault generation can be realized, and improvement measures and maintenance schemes can be provided after early warning according to accurate cases of the knowledge database module; according to knowledge database accuracy calculation, if a historical data training model is not practical, new data is employed by the early warning model as training data for re-training, and accuracy of the early warning model is improved.

Description

technical field [0001] The invention relates to a fault early warning system and a judging method of a wind turbine, belonging to the field of wind power generation. Background technique [0002] With the rapid development of the wind power industry, the development of large-scale wind turbines and the increase in equipment complexity, the frequency and location of faults also increase accordingly. Once an accident occurs, it will inevitably cause casualties and equipment damage. Therefore, it is more and more important to improve the equipment reliability of wind turbines in wind farms and ensure the normal operation of the units. [0003] Through real-time data collection and intelligent fault early warning technology, hidden dangers of accidents can be effectively discovered and system maintenance can be quickly prepared to ensure the safe operation of the unit, so as to prevent problems before they happen, greatly reduce the failure rate of wind turbines, effectively red...

AI Technical Summary

Problems solved by technology

[0004] The invention provides a wind turbine failure early warning system and judgment method, which solves the problem that the current detection system in the wind power industry cannot carry out failure warning before the failure occurs, resulting in maintenance after the failure occurs

Images