Wind turbine generator gearbox bearing temperature state monitoring method based on deep learning model

Abstract

The invention discloses a wind turbine generator gearbox bearing temperature state monitoring method based on a deep learning model. The method comprises the following steps of introducing deep learning into wind turbine generator state monitoring; adopting partial least squares (PLS) method to select variables; establishing a deep convolutional neural network to establish a relation model betweenthe gearbox bearing temperature and the influence variable of the gearbox bearing temperature, predicting the gearbox bearing temperature by using the model in a monitoring stage, and when a residualerror between a gearbox bearing temperature prediction value calculated by the model and an actual value is greater than a set threshold value, sending out a gearbox bearing temperature abnormity alarm. The method is used for analyzing the temperature data of the gearbox bearing, and the purposes of artificial intelligence monitoring and fault early warning of the temperature of the gearbox bearing of the wind turbine generator are efficiently and accurately achieved. Example analysis verifies the practicability and universality of the method.

Description

technical field [0001] The invention belongs to the field of state monitoring of a gearbox of a wind turbine, and in particular relates to a method for monitoring the temperature state of a gearbox of a wind turbine. Background technique [0002] In recent years, the air environment in some areas of my country has been deteriorating, and severe smog has occurred frequently. The traditional energy structure dominated by fossil fuels such as coal and oil needs to be adjusted urgently. The scientific and efficient development of renewable energy is imminent. As an important part of renewable energy, wind power is developing rapidly in my country, and its cumulative installed capacity and new installed capacity both rank first in the world. [0003] The operating conditions of wind turbines are harsh, such as large changes in external temperature difference and random changes in wind speed. These uncertain external factors lead to a high failure rate of wind turbines, resulting...

AI Technical Summary

Problems solved by technology

The current vibration analysis technology is aimed at the time-varying and complex working conditions of variable speed and variable load of gearbox bearings. The accuracy of fault diagnosis is low, and the rate of false alarms and missing alarms is high.

The gearbox oil analysis technology diagnoses the state of the gearbox bearings by collecting gearbox oil samples during the shutdown of the wind turbine, and analyzing the water content in the lubricating oil, the number and diameter of metal particles in the laboratory, but the oil analysis It can only be diagnosed offline, and cannot realize online real-time monitoring and diagnosis of gearbox bearings

There is also a multi-layer forward neural network to model and monitor the bearing temperature of the wind turbine gearbox. However, due to the simple structure of the forward neural network and low modeling accuracy, it is difficult to timely and accurately monitor the abnormal changes in the gearbox bearing temperature. Early warning

Images