Method for monitoring state of underwater vehicle based on fuzzy support vector domain description

Abstract

The invention provides a method for monitoring the state of an underwater vehicle based on fuzzy support vector domain description, and belongs to the technical field of fault diagnosis for the underwater vehicle. The method comprises the steps of respectively extracting feature information from AUV (Autonomous Underwater Vehicle) longitudinal velocity signal wavelet approximate components and propeller control signals when the operating state of a propeller is to be monitored based on a modified Bayesian classification algorithm, constructing a sample to be monitored based on the extracted feature information, calculating a fuzzy membership coefficient, substituting the fuzzy membership coefficient into a fuzzy support vector domain description monitoring model so as to acquire a monitoring coefficient, and judging the operating state of an AUV propeller based on the monitoring coefficient. The method provided by the invention not only can judge whether the AUV propeller has a fault or not, but also can judge the fault severity of the AUV propeller, and is especially suitable for being applied to state monitoring for the autonomous underwater vehicle propeller.

Description

technical field [0001] The invention relates to an underwater robot state monitoring method based on fuzzy support vector domain description, and belongs to the technical field of underwater robot fault diagnosis. Background technique [0002] Autonomous underwater vehicles (AUV) work unmanned and untethered in complex marine environments, and safety is one of the important research contents in the process of AUV research and practical application. The thruster is one of the main fault sources of AUV, and the research on the condition monitoring technology of AUV thruster has important research significance and practical value for improving the safety of AUV. However, in the actual state monitoring process of the AUV thruster, there are often many data samples in the normal operation state of the thruster, but few data samples in the fault state of the thruster. Aiming at this problem, support vector domain description algorithm (SVDD) based on single value classification t...

AI Technical Summary

Problems solved by technology

However, SVDD does not consider the importance of AUV data samples in the measurement process, and treats the data samples when the AUV thruster is in normal operation and the data samples when the AUV thruster failure is different in severity, so that it can only judge whether the AUV thruster is faulty or not. , but cannot judge the severity of AUV thruster failure

Images