Research On Dynamics Analysis And Thrust Allocation Algorithm Of Remotely Operated Vehicle

With the continuous reduction of land resources,people gradually focus the exploration resources of the ocean.Because of the low cost,good control effect and portability of Remotely Operated Vehicle(ROV),it has become one of the important means in the exploration of marine resources.It has also been applied to many other fields and has a very wide application prospect.In order to ensure the excellent control performance of ROV,the motion control algorithm of ROV must have the performance of high efficiency and strong robustness.In order to improve the control performance of ROV,the hydrodynamic coefficients of ROV were obtained by Computational Fluid Dynamics(CFD).The control strategy of Subspace Trust Region Genetic Algorithm(STR-GA)was proposed and designed.The thrust loss curve was fitted based on improved BP Neural Network,and the STR-GA was designed to complete the thrust allocation.Firstly,the dynamic model of the ROV was established,and the hydrodynamic coefficients of the ROV were obtained through simulation experiments.The fixed coordinate system and motion coordinate system were established,and the hydrodynamic model was established through the force analysis of the ROV in the working state.Based on the double constraints of control accuracy and computational efficiency,the ROV model was simplified,the mesh was divided by ICEM,and Fluent was introduced to carry out fluid simulation.The motion of the ROV was divided into horizontal motion and vertical motion,and the hydrodynamic coefficients of the ROV was obtained.Secondly,the motion control algorithm of ROV based on STR-GA was established.According to the motion control requirements of ROV,the motion equation was simplified.The Subspace Trust Region algorithm was used to solve the elite population in the initial population.Designed adaptive mutation and crossover operators.Improved mutation and crossover strategies of elite population and ordinary population.By comparing the results of STR-GA,traditional Genetic Algorithm and Particle Swarm Optimization algorithm to solve the combined thrust of thruster system,the effectiveness and applicability of the proposed STR-GA were verified.Finally,the improved BP Neural Network was used to fit the thrust loss curve of the thruster.On this basis,the combined thrust and combined thrust torque of the thruster system were allocated based on the STR-GA.The influence law between the thrust of each thruster and the deviation distance or angle was studied by MATLAB.By setting different initial deviation distances and angles to carry out simulation experiments,the variation laws of displacement and angular displacement of ROV with time were obtained.The results show that the proposed thrust allocation method had better applicability for ROV.