Hybrid domain feature optimization-based power quality disturbance identification method

The invention discloses an electric energy quality disturbance identification method based on mixed domain feature optimization, and relates to the field of electric energy quality disturbance detection. Electric energy quality disturbance signals are generated through simulation, and statistical feature subsets of the disturbance signals are extracted; performing S transformation on the power quality disturbance signal to obtain a modular matrix; extracting a time domain curve and a frequency domain curve from the modular matrix, and respectively extracting values from the time domain curve and the frequency domain curve to form a feature subset; constructing a high-dimensional feature complete set by the three feature subsets, and obtaining a low-dimensional optimal feature vector by adopting PSO (Particle Swarm Optimization); and inputting actual sampling data to extract optimal features, and carrying out classification identification on the power quality disturbance by adopting a KNN classifier. Power qua