Research On Model Switching Fault Diagnosis And Adaptive Fault Tolerant Control Technology For Underwater Delivery System

Exploration of the ocean is the premise of understanding and development of the ocean,which needs various types of marine exploration equipment.Underwater delivery system uses autonomous underwater vehicle as the carrier,equipped with underwater glider,profile drift buoy and other detection equipment.The underwater delivery system is a new marine exploration system.The autonomous underwater vehicle can make full use of strong mobility and accuracy detection operation.So,the autonomous underwater vehicle can put detection equipment into the designated area.According to the different types and number of detection equipment,the underwater delivery system has the characteristics of typical nonlinear,strong coupling and multi-dynamic models,and the underwater delivery system is more susceptible to random interference such as the ocean currents and the model uncertainty.The model switching fault seriously affects the implementation of the task and even threatens the system safety.Therefore,the paper researches the multi-model switching fault diagnosis and adaptive fault tolerance control for the underwater delivery system.The research will have important theoretical research significance and practical application value for ensuring the safety and efficiency of underwater delivery system.First,the dynamic model of the underwater delivery system and a simulation experiment platform were built.The paper researched the representation and conversion method between the earth-fixed frame and the body-fixed frame for the underwater delivery system.According to the test carrier developed by the project,the paper established its kinematics and dynamics model and gave the parameter values in the model.Based on the simulation experiment platform,the paper verified the effectiveness of the horizontal,vertical,and spatial three-dimensional model for the underwater delivery system.Secondly,the paper researched the motion control of the underwater delivery system and the buffeting of the control quantity.Aiming at the long convergence time of the traditional non-singular terminal sliding mode control,this paper proposed an adaptive reaching law to improve the control accuracy;For the random disturbance of the ocean current and the mode uncertainty made poor motion control accuracy for the underwater delivery system,the paper adopted adaptive fuzzy control method to approximate the uncertainty of the system and updated the control law.Aiming at the chattering problem of the sliding mode control,the paper proposed a function combining fractional power continuous function with hyperbolic tangent function.The function replaced the traditional sign function and weakened the chattering of the control quantity.Then,the paper researched underwater multi-model switching fault diagnosis.The underwater delivery system was a multi-model system.For multi-model switching fault,this paper designed a set of fault diagnosis strategy,and proposed a fault diagnosis method based on the residual analysis for state quantities and control quantities.Compared with the actual model and switching model,the paper analyzed the residuals of the state quantity to diagnosis the multi-model switching fault.The simulation results showed that the paper proposed the method can effectively diagnosis the multi-model switching fault.Finally,the paper research on the adaptive fault-tolerance control method for insufficient thruster output for the underwater delivery system.Aiming at the fault of insufficient thruster output for the underwater delivery system,the paper designed an adaptive fault-tolerant controller based on the adaptive backstepping sliding mode control method that does not depend on the fault diagnosis results.Compared with the adaptive fuzzy PID controller by simulation experiments,the paper designed adaptive controller based on adaptive backstepping sliding mode control method has a better compensation effect on the fault of insufficient thruster output.