Back Propagation Neural Network Based Positional Positive Solution for Wave Compensation Device

In this paper, a method for solving the kinematic positive solution of a wave compensation device based on a rope traction parallel mechanism using a BP neural network is proposed. By using the kinematic inverse solution of the rope traction parallel mechanism derived from the Monte Carlo method as the training sample of the BP neural network, the kinematic positive solution of the position of the load platform based on the rope traction parallel wave compensating device is solved, which overcomes the disadvantage that the accuracy of the solution when using the numerical solution method has a direct relationship with the selection of the initial value, as compared with the traditional numerical method. It not only eliminates the tedious mathematical derivation process, but also obtains the positive solution of the position in accordance with the accuracy requirements. The position solved by neural network can be directly applied in the control system of active wave compensation device.