Engine fault diagnosis method and system based on multi-source information fusion

• Main Authors: ARAMUSI, YANG YANHE, HU NING, BI XIAOYANG
• Format: Patent
• Language: Chinese; English
• Published: 07.11.2023
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; ELECTRIC DIGITAL DATA PROCESSING; PHYSICS

The invention provides an engine fault diagnosis method and system based on multi-source information fusion. The method comprises the following steps: acquiring a vibration signal and an acoustic emission signal respectively acquired by a vibration sensor and an acoustic emission sensor; based on a first fault diagnosis model, performing fault diagnosis on the vibration signal; based on a second fault diagnosis model, performing fault diagnosis on the acoustic emission signal; generating a probability matrix based on the probability that the vibration signal and the acoustic emission signal obtained by the two fault diagnosis models belong to each fault type, and taking the probability matrix as an original basic probability distribution function; correcting the original basic probability distribution function based on the weighting coefficient matrix; the weighting coefficient matrix is generated based on the accuracy of different fault diagnosis models when different sample signals of various fault types ar