State Monitoring and Early Warning Technology of Gas Turbine Based on Mixed Gaussian distribution under Complex Working Conditions

摘要

Aiming at the problem that existing monitoring and early warning methods can not effectively realize gas turbine state monitoring and early warning under complex operating conditions, a monitoring and early warning technology based on nonlinear fitting and mixed Gaussian distribution is proposed, and a detailed analysis is carried out by taking the average gas temperature as an example. Firstly, based on the least squares nonlinear fitting method, the nonlinear relationship between the average gas temperature and power is obtained as the baseline of the normal state. Then through the statistical analysis of the probability distribution of the residual after fitting, it is found that it meets the two Weighted Gaussian mixture distribution model. Finally, the EM algorithm is used to estimate the parameters of the two-weighted mixed Gaussian distribution of the residuals, and the upper and lower thresholds are set according to the three principles of the two-weighted Gaussian distribution. The verification results of the normal data set and the abnormal data set showed that the set threshold value of the sub-operating conditions is reasonable and credible, and can effectively identify the abnormal state of the average gas temperature under variable operating conditions. The comparison and analysis of abnormal data sets with the existing monitoring and warning methods highlights the superiority of the monitoring and warning methods mentioned in this article, and also has a certain reference value for the monitoring and warning of other equipment under complex working conditions.