Research On Remaining Useful Life Prediction Of Gearbox Of Wind Turbine Based On Genetic Algorithm Optimized Support Vector Machine

In recent years,the demand for energy is increasing,and the problem of energy shortage is becoming more and more serious.Wind energy has a bright future as a new type of clean energy.Wind turbine as the main implementation of wind energy into the implementation of the main body,its health is directly related to the production of security and economic benefits.Wind Turbine gearbox as a key component in the operation of the fan,usually installed in the top of the fan,the failure of maintenance is very difficult.Most of the failure of the gearbox is performance degradation of the bearing failure,so it is necessary to predict the remaining life of the wind turbine gear box bearing in a scientific and effective way to take maintenance measures at the right time to control and correct the performance.Practical significance.In this paper,the wind turbine bearing is used as the research object,and the data-driven technology is used as the entry point to study the residual life prediction method.Firstly,this paper reviews the methods of equipment residual forecasting,classifies the remaining life prediction methods,and then analyzes and compares these methods separately.Then,the concept of failure prediction of mechanical equipment,the content of fault prediction and the research status at home and abroad are introduced,and the definition of failure evolution process and the definition of remaining life are given.Which laid the foundation for the establishment of the remaining life prediction model of the wind turbine gear box bearing.Secondly,this paper establishes a support vector machine model for the prediction of the remaining life of wind turbine gearbox.First,the vibration characteristics of the vibration signal extracted from the bearing experiment platform are extracted.In order to reduce the redundancy in the feature and eliminate the noise in the feature,a data dimension reduction method based on principal component analysis is proposed.The reduced data is input to the support vector machine to predict the remaining life.Aiming at the problem that the parameters in the supportvector machine have great influence on the generalization ability and the parameters are difficult to be adjusted,this paper proposes a support vector machine parameter optimization method based on genetic algorithm.Finally,in order to verify the validity of the proposed algorithm,this paper validates the algorithm.The experiment of this paper is carried out first.In order to verify the improvement of the prediction accuracy of PCA,the contrast experiment of PCA reduction is carried out.At the same time,in order to verify the validity of SVM parameters in genetic algorithm,the SVM parameters and SVM parameters are optimized by grid method.Finally,the results of this paper are compared and analyzed.The results show that the support vector machine(SVM)is based on the genetic algorithm of PCA data.The generalization performance of SVM is strong and the residual life prediction accuracy is higher.In this paper,a series of work has been done on the residual life prediction of the gearbox bearing of the wind turbine,and some achievements have been made,but some work needs to be further improved.Based on the data-driven remaining useful life prediction method,the mechanism of bearing degradation is analyzed.The next step can deeply excavate the mechanism of bearing degradation to provide theoretical support for data-driven method.In addition,this paper mainly optimizes the two key parameters of the support vector machine,and the other parameters can also be optimized according to the idea of this paper.The research work of this paper can provide some reference and reference for the prediction of the remaining life of gearbox bearing in wind turbine.