Research on fault diagnosis method and application of gas turbine based on acoustic imaging technology

Gas turbine health management is of great significant for enhancing flight safety and reducing maintenance costs. Based on the architecture of gas turbine fault identification system, a multi-arm spiral microphone array with high positioning accuracy, high beam resolution, and strong anti-interference capability is designed in this paper with simulation and experiment. On this basis, by utilizing multi-channel acoustic digital acquisition and analysis equipment, high-precision, wide dynamic range, and high-real-time acquisition of audio data from the combustion chamber of gas turbines are achieved. An audio data analysis strategy based on feature extraction and an autoencoder network construction strategy are proposed, fulfilling the imaging of audio data from the combustion chamber of gas turbines, which can effectively identify gas turbine faults. Experiments conducted on the platform of the multi-arm spiral microphone array indecates that the proposed fault detection strategy has an accuracy rate of over 98%. This is of great theoretical significance and practical value for enhancing the level of gas turbine fault detection, and improving the performance, reliability, and lifespan of gas turbine systems.