Research On Intelligent Control Method Of Marine Electric Propulsion System Based On Sea State Identification

As the core power equipment of a ship system,the control performance of an electric propulsion system is greatly affected by the changes of sea states.It is necessary to identify sea states and adopt appropriate control methods according to the changes of sea states in order to realize the effective control of the electric propulsion system.In this paper,intelligent sea states identification and intelligent control for propulsion systems are studied based on maximum likelihood evidential reasoning(MAKER)rule and evidence reasoning(ER)rule to solve the problems of strong interdependence between input features and weak adaptive ability of control model in these methods.The main work is as follows:(1)Intelligent sea states identification based on MAKER rule.In order to solve the problem of strong interdependence among input features of the electric propulsion system,a method to identify the sea states by fusing the interdependence evidence based on MAKER rule is proposed.Firstly,it is necessary to obtain samples corresponding to the state variables of the propulsion system.And the reference evidence matrix related to the ventilation mode(type)is obtained from the sample value.Then,the reliability of evidence is obtained based on the basic belief assignment,and MAKER rule is used for evidence fusion and reasoning to identify the type of ventilation,and identify the sea state based on the correspondence between the type of ventilation and sea state.Finally,the validity of the model is verified based on the data set reflecting sea states.(2)Research on the speed control method of the electric propulsion system in normal sea state.In order to adaptively update the parameters of the PID controller for propulsion system under normal sea state which is identified by(1),a PID control method for the speed of the propulsion system based on the ER rule(ER-PID speed control)is proposed.Firstly,a PID controller is developed for propulsion system to estimate the PID parameters based on ER rule.Then,the sequential linear programming(SLP)algorithm is used to optimize the reference evidence matrices and importance weights in real time.Finally,the control accuracy and robustness of the proposed method are verified by comparing those of the PID control method based on back propagation(BP)neural network.(3)Research on the speed control method of the electric propulsion system in extreme sea state.In order to reduce the influence of propeller ventilation on the control performance of the propulsion system under extreme sea state identified by(1),the control method in(2)is improved and an anti-spin PID control method for the speed of the propulsion system based on the ER rule(Anti-spin ER-PID speed control)is further proposed.Anti-spin ER-PID limits the motor input torque by introducing the anti-spin coefficient,which reduces the influence of the mismatch between the given torque and the propeller load torque caused by ventilation,and the excessive speed of propeller on control performance.Finally,the scenario when the ship encounters extreme sea state is simulated.The effectiveness of the proposed method is verified in the anti-spin speed control experiment.