| The Autonomous Underwater Vehicle(AUV)plays a very important role in marine development because of its strong autonomy.Most of the current rudder and paddle-control AUV tail rudder layouts are cross rudders.This layout has a simple structure and is easy to control,but it has defects such as low reliability.X rudder is a special AUV rudder layout different from cross rudder.Compared with the cross rudder,it can make the AUV have better maneuverability,and it can achieve roll control.In addition,the four rudders are redundant,AUV can get better reliability.This article starts from the "WL-2" AUV motion model,and modifies the cross rudder to X rudder.Based on this,it conducts research on optimization of control allocation algorithms,fault diagnosis based on data mining technology,and fault-tolerant control of control redistribution.The research contents are as follows:(1)X-rudder AUV motion simulation modeling and motion controller construction.The related parameters of X rudder AUV are defined,and a six-degree-of-freedom dynamic model of X rudder AUV is established.Aiming at the complex working environment and driving characteristics of the X rudder AUV,a variable structure PID controller is designed for each AUV.Build AUV motion simulation model based on Matlab2016/Simulink,and verify AUV motion control simulation of variable structure PID control method.(2)Research on Control Method of X Rudder AUV.Establishing a mathematical model for montrol allocation of X rudder AUV,Aiming at the existing allocation error and allocation accuracy problems of the existing underwater robot control allocation technology,an improved secondary planning control allocation algorithm is proposed.The Lagrangian multiplier method is used to replace the original algorithm for solving the sequential quadratic programming subproblem,which improves the calculation accuracy of control allocation while achieving the allocation accuracy.While the control system error input changes greatly,the sudden change of the rudder angle will not be conducive to the life of the steering gear and cause the actuator to respond in a timely manner,by adding dynamic terms and dynamically changing the actuator’s constraints to optimize control allocation results.In the X-rudder AUV motion control simulation,the improved control allocation algorithm results in good control allocation.(3)Research on X rudder AUV fault diagnosis and fault-tolerant control.When using the wavelet method and the threshold method to diagnose AUV faults,the threshold is not easy to determine and subjective,and the boundary between AUV weak fault and normal state is often fuzzy.Aiming at these shortcomings,an AUV fault diagnosis method based on SVM is proposed.By referring to the ideal state of the AUV output by the sliding mode state observer,the actual state of the AUV is output by the dynamics model,take the residual data output from the two AUV motion states under different operating conditions,one part is used to train the SVM to obtain the fault diagnosis model,and the other part is used to test and verify the model.Subsequently,a fault location method based on short sea trials is proposed,the fault rudder is located by defining the moment direction matrix and matching with its columns.Finally,a fault-tolerant control strategy based on control redistribution is proposed,and torque redistribution is performed by updating the control allocation efficiency matrix.The effectiveness of the fault-tolerant control strategy is verified in the simulation of the fault-tolerant control of the X rudder AUV under different operating conditions. |
