A Method for Extracting Fault Features of Total Pressure in Aero Engine Based on Improved Grasshopper Optimization Algorithm

A feature selection method for improving the grasshopper optimization algorithm is proposed to address the high dimensionality and redundancy of total pressure data in aircraft engine high-altitude simulation tests. The descent parameter c is dynamically adjusted, and a sine adaptive descent parameter is used to balance the algorithm's global search and local development capabilities, improve the convergence speed and optimization accuracy of the algorithm, and discretize the grasshopper position using a transformation function, Transforming continuous state problems into discrete state problems, reducing data dimensions, improving feature selection efficiency and accuracy, provides a new approach for dimensionality reduction of engine test bench fault data. The algorithm was validated using test functions and standard datasets, and simulation experiments were conducted using real M-section total pressure measurement data. The experimental results show that the improved grasshopper optimization algorithm significantly improves the detection rate and accuracy of feature selection compared to traditional algorithms. After dimensionality reduction, the feature selection accuracy of the M-section total pressure measurement data reached 99.6%, verifying the superiority of the improved grasshopper algorithm in feature selection problems.