Research On Identification Method Of AUV System Based On Numerical Simulation

With the continuous advancement of science and technology,the Autonomous Underwater Vehicle(AUV)plays an increasingly important role in marine exploration and resource development.Control technology is the basic and key technology of AUV,and AUV system model is one of the important contents of AUV control technology research.There are many hydrodynamic parameters in the AUV system model,and these parameters have strong coupling and nonlinear characteristics.The identification method of the system model largely determines the accuracy of the system model.Therefore,studying the AUV system model identification method has important research significance for improving AUV motion control performance.In this paper,based on the AUV system identification method,the "UVIC-I" AUV independently researched and developed by the underwater vehicle intelligent control technology research team of Harbin Engineering University is used as the research object,and its system model is established.The unknown parameters in the established model were identified.The specific research work is as follows:(1)Study the system model rigid body dynamics matrix identification.In order to analyze the dynamic performance of "UVIC-I" AUV and support the subsequent simulation research,this paper studies the "UVIC-I" AUV system model identification problem.The system model is divided into two parts:the rigid body dynamic matrix and the hydrodynamic matrix.Aiming at the identification problem of rigid body dynamic matrix,the rigid body dynamic matrix is divided into two parts:rigid body mass inertia matrix and rigid body centripetal force and Coriolis force matrix.The unknown parameters are identified by numerical simulation.(2)Study system model hydrodynamic matrix identification.According to the identification problem of hydrodynamic matrix,the hydrodynamic matrix is divided into three parts:additional mass matrix,water resistance matrix and restoring force matrix.For the identification problem of the additional mass matrix,the method of identifying the unknown parameters by the empirical formula derivation method is used.For the identification problem of the water resistance matrix,the translational freedom and the rotational freedom drag coefficient are respectively identified by the fluid dynamics analysis.For the identification of the translational resistance resistance parameter,the relative flow rate method is used for identification.For the identification of the rotational freedom resistance torque parameter,the method of dynamic reference model is used for identification.For the identification problem of the restoring force matrix,the unknown parameters are identified by the analysis of the combined force of gravity and buoyancy.(3)Single degree of freedom verification of the system model.In order to provide accurate thrust data for the system model single degree of freedom verification,the problem of the problem is caused by the poor accuracy of the initial model of the primary function thruster.It is determined that the initial model of the thruster is set to the quadratic function and The cubic function is used to identify the propeller dynamics model and compare the results of the propeller dynamics model identification of three different initial models.In order to verify the accuracy of the system model identified in this paper in single degree of freedom,a method for verifying the relationship between the resistance and velocity of AUV according to uniform motion is proposed.The accuracy of the system model is verified.In order to improve the accuracy of the system model and study the cause of the analysis error,the method of averaging is determined to correct the identified system model.(4)Multi-degree of freedom verification of the system model.In order to verify the accuracy of the identified AUV system model in multi-degree-of-freedom motion,a method for comparing and analyzing the motion trajectory by inputting the same control voltage in the AUV system model and the actual experiment is determined,and the system model is verified.The accuracy.In order to further verify the identification method of this paper,the accuracy of the identification method is verified by the method of comparing the system model with the system model obtained by least squares identification.