Research On Bridge Crane Control Based On Fuzzy And Neural Network Method

The fast,accurate positioning and anti-swaying of the bridge crane is of great significance to people’s work efficiency and personal safety.Bridge crane is a typical underactuated system,especially in the case of variable rope length,variable load and variable target position,resulting in greatly increased difficulty in system control.At present,there is still no more appropriate method for automatic control.This article first models the crane system,establishes a nonlinear mathematical model of the system to change the rope length,and points out that the traditional linearization method is no longer the optimal choice,and applies the intelligent control method to set the length of the rope,change the length of the rope,and change the load.The experiment and analysis were carried out to change the target position,and the actual control process of the crane was simulated as much as possible.Secondly,a bridge crane experiment system was designed to simulate the length of the rope and the rope length control process of the crane,and the real-time control experiments were performed on the MATLAB platform.Then using the fuzzy control,variable universe fuzzy control and adaptive neuro-fuzzy inference system to control the crane system,it shows the superiority of the intelligent method in dealing with nonlinear mathematical model.At the same time,the neural network algorithm in the intelligent method is combined with the LQR control algorithm in the traditional method.A controller based on LQR and BP neural network(abbreviated as LQR-BP)was designed to decompose the complex nonlinear system process into two linear processes: vertical movement and horizontal movement,and the bridge crane system was controlled accordinglyFinally,based on the control experiments of the fixed and variable rope lengths,the control effects of each control algorithm are comprehensively analyzed,and the advantages and disadvantages of the control effects of the various control algorithms in the variable target position and variable load conditions are analyzed.Under a certain linearization process,it can be considered as a linear process,and the process of changing the rope length is a nonlinear process.