Prediction of residual service life of gas turbine bearings based on support vector regression

Abstract

A method for predicting residual service life of gas turbine bearing based on support vector regression By collecting the health status data of gas turbine bearing and extracting the time-domain and frequency-domain features after preprocessing, Then the time-domain and frequency-domain features are fused by the data fusion tool through the principal component analysis method, and the low-dimensional feature indices which characterize the bearing degradation are obtained and used to train the residual service life prediction model based on support vector regression. Finally, the residual service life prediction model is used to predict the bearing life in real time. The prediction error of the invention is lower than the prediction result of the ordinary support vector regression model andthe neural network model, and the prediction result precision is high.

Description

technical field [0001] The invention relates to a technology in the field of thermal power generation, in particular to a method for predicting the remaining service life of a gas turbine bearing based on support vector regression. Background technique [0002] The bearing parts of the gas turbine are an important part of the operation of the gas turbine. Predicting the remaining life of bearings has become a hot research issue in the academic field of prediction. The methods used to predict the remaining life of gas turbine bearings are mainly divided into two categories. On the one hand, the mechanical model of the working state of the bearing is established, and the working state of the bearing is analyzed through various factors such as material, speed, frequency, temperature, and humidity. , to predict the remaining life of the bearing through the physical model. On the other hand, by setting various sensors to detect the state monitoring data of the bearing during op...

AI Technical Summary

Problems solved by technology

The disadvantage of the former is that the establishment of the model is very complicated, and it is easy to ignore some influencing factors, resulting in inaccurate prediction results.

Compared with the physical model, the latter is less difficult to establish a model, but it also has the problem of strong model correlation and weak causality, and the interpretability of the model is poor, and it is easy to fall into misunderstandings

Images