Study On Main Engine Cylinder Water Temperature Control Based On Neural Network

Main engine cylinder water system is an important part of ship cooling system,as well as an important auxiliary system to ensure the safe operation of main engine.The control performance of main engine cylinder water temperature directly influences the work performance of main engine and will have effects on the power,economy and reliability of ship.The parameters of conventional PID controller are fixed values which will not change with the process characteristics of the controlled system and unable to ensure the control quality under complex and variable conditions.Therefore,this paper take the main engine cylinder water temperature control as the research object.An intelligent PID controller that can overcome the limitations of conventional PID controller is proposed in order to achieve the purpose of precise control over the main engine cylinder water temperature.First of all,mathematical models are established for the simplified main engine cylinder water system on the basis of the relevant theories of thermodynamics and heat transfer.Simulation models are established in MATLAB/Simulink on the basis of mathematical models.Simulation experiments are carried out under different simulation conditions.Simulation results are compared with the real ship design data to verify the accuracy of the simulation models of the main engine cylinder water system.Secondly,the limitations of conventional PID controller in practical engineering applications are deeply analyzed.The PID controller based on BP neural network is proposed on the basis of neural network relevant theories.BP neural network is designed according to the characteristics of the main engine cylinder water system.The BP neural network PID controller optimized by twice determination method is used for the simulation experiments to verify the effectiveness of the controller.Finally,the conventional PID controller and the optimized BP neural network PID controller are used to carry out simulation experiments on the main engine cylinder water system,respectively.The simulation results show that the optimized BP neural network PID controller is capable of overcoming the limitations of the conventional PID controller,can meet the requirements of control quality under complex and variable conditions and make the system have good steady and dynamic performance.