The Monitoring And Path Fault Diagnosis Of Gas Turbine System

The gas turbine is always working in harsh environment, not only the routine maintenance is difficult but also because of environmental issues are more prone to a variety of failures. Once the failure will bring great harm and damage to the hull, even more can damage the safety of staff. So the effective monitoring and diagnostic platform gas turbine not only enables online monitoring of the data, but also to help operators detect abnonnal turbine,detecting the malfunction in time and proven reasons, and it will help to do troubleshooting and maintenance work to improve the safe operation of the gas turbine mechanism to ensure the safe operation of the ship. In this paper, three-axis turbine for the study, we have established gas turbine gas path fault model, mechanism analysis, research and design of fault diagnosis algorithm fault diagnosis system. The main work is as follows:(1)Establishing gas path fault coefficient matrix. Through analysising the air line components thermodynamic, Obtain thermodynamic formula, and the use of the small deviations method of linear equations; The performance parameters analyzed when a fault occurs, isolated independent variation for establishing fault coefficient matrix .By measuring the coefficient matrix can be linked to parameters and performance parameters. According to laboratory data, the coefficient of the coefficient matrix solving matrix can be obtained for a gas turbine fault simulation.(2)Simulation of gas turbine gas path fault. According to the literature and information to determine nine kinds of typical gas turbine gas path fault: Compressor blade fouling?Compressor top gap increases?Compressor blades wear?Compressor blades Mechanical damage?Hot corrosion of turbine blades?Turbine blade fouling?Turbine blade wear?Mechanical damage turbine blades?Combustion fault?Based Gas Turbine Fault model matrix,to simulate nine gas path fault.(3) Applicating clustering anfis system to the gas turbine fault diagnosis. First, according to the fault information transfer theory, put forward the basic idea of fault diagnosis. Then discuss the architecture of Fuzzy Inference System, then establishing the adaptive fuzzy network. training and testing with the data. The diagnostic results show that the Adaptive Fuzzy Neural system have rapid diagnosis and has good diagnostic accuracy.(4)Design gas turbine monitoring and fault diagnosis software.In the Labview platform ,building gas turbine fault diagnosis system; The entire system is divided into three modules: Network communication module, Data management module and Fault diagnosis module. In order to process high portability and legibility, all modules to be designed in the form of sub-VI.