Common rail oil injector fault diagnosis method based on adaptive waveform continuation optimization EEMD (ensemble empirical mode decomposition)

The invention belongs to the technical field of mechanical health management, and discloses a common rail oil injector fault diagnosis method based on adaptive waveform continuation optimization EEMD, which comprises the following steps: acquiring pressure fluctuation signals of a high-pressure oil pipe of a common rail oil injector in different states, and dividing the acquired pressure signals into training samples and test samples, the states of the common-rail fuel injector include normal fuel injector state, clamping stagnation of a fuel injector needle valve and blockage of a fuel injector orifice; adaptive decomposition is carried out on the pressure signals according to an adaptive waveform continuation optimization EEMD algorithm, and redundant signal interference is reduced; calculating a hierarchical weighted permutation entropy (HWPE) of the filtered pressure signal, and taking an information entropy value as a fault feature of the pressure signal; and completing SVM multi-classifier training by u