Research On Key Technologies Of Wind Power Generation Based On Deep Learning

In recent years,the role of renewable energy sources in the power system has become more and more important.Renewable energy can play a role in protecting the ecological environment and alleviating electrical pressure.They have received more and more attention from all over the world,and have been widely developed and used.Wind power is a mature and promising renewable energy power generation method.The key technologies of wind power generation mainly include two aspects: wind power prediction and wind turbine fault diagnosis.Applying deep learning to the field of wind power generation enables wind power generation to be better connected to the grid and enables rapid development of key technologies.Research on wind power forecasting.Due to the instability of the wind itself,wind power generation has volatility and discontinuity,which will cause serious difficulties for the power grid to dispatch wind power generation.Accurate ultra-short-term prediction of wind power is of great help to alleviating the pressure of peak and frequency regulation in the power system,and is of great significance for wind power grid integration.In order to further improve the accuracy of wind power ultra-short term prediction,a wind power prediction based on AM-LSTM model was proposed.This model combined long-term and short-term memory network(LSTM)with attention model(AM).In combination,the LSTM network could handle the nonlinear relationship between time series variables such as wind speed,wind direction and wind power,and the attention model could optimize the weight of the LSTM network to make the prediction result more accurate.The real wind farm historical data was used in experiments.The experiments results show that the proposed AM-LSTM prediction model can effectively utilize multivariate time series data to predict the ultra-short-term power generation of wind farms,which is more accurate on forecast effect than the traditional BP neural networks and the LSTM networks.This prediction model provides a scientific reference for the power dispatching of wind farms.The fault diagnosis of the wind gearbox bearing is studied.Gearbox bearings play an important role in wind power generation system.Their regular and stable operation will increase wind turbine power generation and improve the economic efficiency of wind farms.They often fail because they work under complex wind conditions.Therefore,it is necessary to find the fault early.The vibration signal of the gearbox bearing has the characteristics of volatility and continuity.Traditional bearing fault diagnosis methods are often based on signal analysis and feature selection,and the process is relatively complex.Deep learning methods can extract and select features automatically,thereby reducing the workload.A fault diagnosis method based on deep learning is proposed in this study.This method combines a one-dimensional convolutional neural network(1DCNN),support vector machine(SVM)classifier.First,1DCNN adaptively extracts features.Then,the extracted features are input into the SVM classifier.Finally,the particle swarm optimization(PSO)is used to optimize the SVM classifier.The results show that the proposed fault diagnosis method is effective for fault diagnosis of wind turbine gearbox bearings.This method improves the precision and accuracy of diagnosis when compared to other methods.The study of wind power prediction and gearbox bearing fault diagnosis will help to further solve the key technical problems of wind power generation.In this project,through the research on wind power prediction and gearbox bearing fault diagnosis,deep learning is used for wind power prediction and gearbox bearing fault diagnosis,so that the clean energy of wind power can better serve mankind.