Analysis For Submarine Maneuvering Motion Simulation By Intelligent Control

Maneuverability is one of the most important performances of modern submarine. Good maneuverability is very important for ensuring the safety of submarine navigation, for full Playing the submarine work effectiveness and combat effectiveness. It is composed of maneuvering, balancing, submerging and surfacing, and hovering subsystems. Here focuses on both the hovering and maneuvering subsystems of submarine.A set of motion model fitted for the hovering control system design is derived by reasonable analysis on standard submarine 6-DOF motion model. Based on the simplified motion model, the fuzzy hovering control system is designed. And a mathematic expression of disturbing force is established, providing validation basis for the submarine hovering based on fuzzy control.According to the motion character of hovering, fuzzy controller about fix depth was designed based on the ruler of fuzzy controller. The result of simulation showed that the fuzzy controller was able to control the system very well, and had the advantage of small overshoot, fast adjustment, robust and so on.Because of domain limit, regular fuzzy control could not carry out variable depth hovering. A hovering control system based on fuzzy logic controller with variable region was adopted. The paper provided double inputs and single output fuzzy logic control algorithm with indirectly-inherited variable region, completed variable depth hovering. The following simulations showed that new control system can achieve control aims in variable depth, and improve the control accuracy which had a better robustness and anti-interference capacity.According to the problem that the high frequency disturbance of wind and wave induced operating rudders, the paper adopted direct model with reference to the control means of self-adapting neural network was adopted. The reference model was instructed by identifying off line of submarine's course. The weight value of RBF neural network was set by the output of reference model and the actual output of submarine model. In this way, the self-adapting control of submarine at surface course was achieved. And Chebyshev 2nd filter was designed to against wave disturb. The simulation demonstrated that the integration of Chebyshev 2nd filter algorithm and self-adapting control algorithm of direct model with reference to neural network can solve the problem of void steering of course control under wind and wave disturb.Aimed at submarine moving couple on vertical plane, the paper integrated the neural network identifier, neural network filter, neural network controller. The identified neural network model can provide gradient information to the controller, and made the model output of neural network instead of the submarine output, to achieve the purpose of anti-interference. BP algorithm was improved with the Levnberg-Merqnardt optimizing algorithm led in. The learning convergence rate of BP network was accelerated and learning accuracy was improved. RBF network was improved that RBF network with local feedback and real time recursive algorithm were adopted to filter the vertical plane. The simulation indicated that the control project can solve the problem of void steering of course control of vertical plane in submarine under wind and wave disturb.