Wind turbine generator gearbox fault monitoring method and system based on deep neural network

Abstract

The invention discloses a wind turbine generator gearbox fault monitoring method and system based on a deep neural network. The method comprises the following steps: A. acquiring gearbox state-related data in wind turbine generator SCADA real-time data, including gearbox lubricating oil pressure, gearbox lubricating oil temperature, gearbox bearing temperature and unit output power; B, inputting the gearbox lubricating oil temperature, the gearbox bearing temperature and the unit output power into a prediction model based on the deep neural network, and acquiring a prediction value of the gearbox lubricating oil pressure outputted by the prediction model; C, calculating a fitting error between an actual value of the gearbox lubricating oil pressure and a predicted value of the gearbox lubricating oil pressure; and D, monitoring the fitting error so as to determine a gearbox state of a wind turbine generator, and triggering the alarm when the gearbox state of the wind turbine generator is determined as a fault state. By adopting the gearbox fault monitoring method and system, a gearbox fault can be effectively monitored in real time, a monitoring result is less influenced by an external environment, and the accuracy is high.

Description

technical field [0001] The invention relates to the technical field of wind turbine fault early warning technology, in particular to a fault monitoring method and system for a wind turbine gearbox based on a deep neural network. Background technique [0002] As the service life of wind farms increases, the cost control of wind farm operation and maintenance becomes more important. Several main subsystems of wind turbines, such as gearboxes, generators, bearings, etc., are the main application objects of condition monitoring and fault warning. With early warning of faults, wind farm operators can effectively adjust the operation mode and perform equipment maintenance and replacement in advance, which can significantly reduce operating costs. For gearboxes, which account for most of the cost share of wind turbines, failures will cause excessive downtime, so the development of accurate and efficient gearbox fault detection models is essential. [0003] The traditional method ...

AI Technical Summary

Problems solved by technology

The oil temperature of the gearbox lubricating oil is the main monitoring index of the gearbox during monitoring, but the oil temperature of the gearbox is easily affected by the surrounding environment and noise, thus affecting the monitoring results

Images