As complex systems with nonlinear thermodynamic coupling of various gas-path components

heavy duty gas turbines are the key thermodynamic engines for clean energy utilization and efficient conversion in natural gas power plants. Gas-path components are the components with the highest failure rate in gas turbines

and their faults are highly insidious and destructive compared to mechanical and auxiliary system faults. Aiming at the problems in diagnosing gas-path faults

a machine learning based diagnostic method for all gas-path components of the heavy duty gas turbine with the aid of thermodynamic model is proposed for the first time. On the basis of established high-accuracy gas turbine thermodynamic model

a comprehensive rule base is established for the relationship between the internal fault modes of gas-path components and the external fault symptoms of gas path measurable parameters. A mathematical model for all gas-path component fault diagnosis suitable for machine learning framework is established based on typical nonlinear gas-path fault diagnostic principles. The proposed method can be used to comprehensively diagnose the different types and severity of faults in all gas-path components (i.e.

intake and exhaust system

compressor

combustion chamber

and turbine) under various operating conditions after grid connection from inlet guide vane (IGV) minimum opening load to IGV fully open base load. Case analysis shows that the proposed method can achieve a success rate of 100% for diagnosing different types of faults and can achieve an overall success rate of over 97% for diagnosing the types and severity of faults when the performance of multiple components deteriorates. This approach can be used in real-time online fault diagnosis application scenarios.