Research on Sensor Fault Diagnosis of Gas Turbine Control System Based on Data-driven

The sensor is responsible for monitoring the running state of gas turbine in gas turbine control system, so it is of great significance to carry out fault diagnosis of the sensor in gas turbine control system. This paper presents a sensor fault diagnosis method for gas turbine control system based on data-driven. The method first simulates several common faults of the sensor, and uses wavelet packet transform (WPT) to extract fault features, and then use moth-flame optimization (MFO) algorithm to optimize the initial weights and biases of BP neural network (BPNN), establishes the MFO-BPNN fault diagnosis model, and finally inputs the fault characteristics into the diagnosis model to realize the fault diagnosis of the gas turbine control system.