Research On Modeling And Fault Diagnosis Of Inlet Guide Vane System For Gas Turbines

Currently,China is vigorously promoting the research and development of gas turbine technology,and its localization has significant significance for the industrial and national defense fields.Among them,the inlet guide vane(IGV)of the compressor,which controls the angle and flow of the airflow entering the compressor by changing the angle of the guide vane,is one of the key components of gas turbines and plays an important role in the safe and reliable operation of gas turbine units.This article takes the double connecting rod IGV as the research object,and focuses on the modeling and fault diagnosis methods of the IGV,with the following main work and achievements:(1)Research on the mechanism of IGV was conducted,and dynamic models of hydraulic drive sub-module and mechanical drive sub-module were established.A modular modeling method was used to build a Simulink simulation model of IGV,and parameter identification and testing were conducted using the Xinhua engine simulation machine.The results indicate the rationality and accuracy of the constructed IGV mechanism model.(2)By analyzing the fault mechanisms of three typical failures,namely blade fracture,IGV connecting rod loosening,and internal leakage of the hydraulic cylinder,an IGV fault model was built.Typical fault simulations were carried out based on the Simulink simulation platform and the Xinhua engine simulation machine.The results show that the IGV fault model constructed in this paper has a high degree of accuracy.(3)A fault diagnosis method based on multi-feature information fusion and improved least squares support vector machine is proposed,and the IGV performance simulation and fault diagnosis system is established.This method uses complementary ensemble empirical mode decomposition to decompose the IGV fault signal,uses kurtosis and mutual information entropy to screen signal components,and uses the energy entropy,permutation entropy,power spectral entropy,and approximate entropy of the retained components as fault feature vectors.The whale algorithm is used to improve the LSSVM parameters,and a fault classification model is constructed for fault diagnosis.The IGV performance simulation and fault diagnosis system are tested on the Xinhua gas turbine simulation machine,and the effectiveness of the proposed method is verified.