Research On Time-frequency Entropy Enhancement And Fault Identification Methods Of Thruster Fault Signal Of Underwater Vehicle

“Made in China 2025”clearly proposes to build powerful marine engineering equipment.As the only marine equipment capable of working in the deep sea,underwater vehicle is becoming more and more important.As the core component of the underwater vehicle system,its thruster has the longest working time,the largest load and the highest potential safety failure.The research on the fault diagnosis technology of the thruster is an important prerequisite to ensure the safe and reliable operation of the underwater vehicle,which has important research significance and practical value.Aiming at the problem of thrusters fault diagnosis of underwater vehicle,this paper studies it from three aspects: thrusters fault feature extraction,thruster fault feature enhancement and thruster fault degree identification.Meanwhile a set of underwater vehicle prototype system is made to verify the effectiveness of the proposed research method.The problem of fault feature signals extraction of thruster is studied.Aiming at the problem that the mapping relationship between fault eigenvalues and fault degree is not unique,when extracting fault features from a single signal source in underwater vehicle speed signals or thruster control signals.Time-frequency entropy and time-domain energy are extracted as fault features from the fusion signal of underwater vehicle speed signal and thruster control voltage change rate signal evidence theory.The fault feature samples with unique mapping relationship with different fault degrees of the thruster are obtained.The effectiveness of this method is verified by the pool experimental data of underwater vehicle.The problem of fault feature enhancement of thruster is studied.In order to solve the problem that the fault feature value is not sensitive to the fault degree of the thruster when use the traditional SPWVD-SE method to extract fault features from the dynamic fault signal,a TFEE method is proposed.The time-frequency distribution energy of dynamic signal is enhanced from three aspects: frequency domain,time domain and time-frequency domain.The sensitivity of fault eigenvalues to thruster faults and the distance between fault eigenvalues corresponding to different fault degrees are increased.The effectiveness of this method is verified by the pool experimental data of underwater vehicle.The problem of fault degree identification of thruster is studied.In order to solve the problem of low precision and large error of fault feature extracted by single source dynamic signal in the process of thruster fault degree identification,a SVDD fault identification algorithm based on fusion signal is proposed to improve the accuracy of fault identification.In view of the fact that when the GRA method is used to identify the fault degree of the thruster,the fault identification result can only be based on the preset discrete level standard fault degree.A fault identification method with BC-RGRG is proposed,which can identify not only the unknown fault degree value which falls on the preset standard fault degree,but also the unknown fault degree specific value between the preset standard fault degree.The effectiveness of these methods are verified by the pool experimental data of underwater vehicle.