Abstract:In order to improve the compensation accuracy of ships working on the sea,BP neural network PID control method was adopted,and the output error of ship lift motion was simulated.The sketch of active lift compensation system was established,the working principle of ship lift motion was analyzed,and the driving transfer function of hydraulic cylinder was given.BP neural network algorithm was used and the weighted value of BP neural network was modified by gradient descent method.The output error of the control system was compensated by learning rate,so that the parameters of the PID controller could be adjusted online.Under the influence of different loads,the compensation accuracy of ship lift motion was simulated by using MATLAB software,and compared with the compensation accuracy of PID control.The results show that the output error of ship lift motion is larger and the response speed of control system is slower by using PID controller,while the output error of ship lift motion is smaller and the response speed of control system is faster by using BP neural network PID controller.At the same time,with the increase of load quality,the output error will increase.Using BP neural network PID control system,ships working on the sea has fast response speed,high compensation accuracy and high positioning accuracy.