Aero-engine fault mode modeling and detecting method based on BP optimized by IWOA

The invention provides an IWOA-optimized BP-based aero-engine fault mode modeling and detection method, and the method comprises the steps: obtaining a normal data set, carrying out the correlation analysis of the normal data set, and removing parameters with large correlation to form a simplified data set, carrying out fault processing on parameters in the simplified data set to form a fault simulation data set representing different faults of the engine, carrying out division, establishing an engine fault mode model based on a BP neural network, training the model to obtain a mean square error, comparing the mean square error with a set ideal mean square error, obtaining an optimal hidden layer node number, and obtaining an optimal hidden layer node number; constructing a BP neural network according to the optimal hidden layer node number and determining an initial weight and an initial threshold, optimizing the BP neural network based on IWOA to obtain an optimal weight and an optimal threshold for the BP