Inverse dynamic compensation via simulation of feedback control system by artificial neural networks for crane system

Abstract

This paper propose a vibration control system for overhead cranes considering nonlinear behavior. Overhead cranes are widely used in the industrial field to carry heavy loads. It is a non-linear system, and vibration of the load occurs as it moves. As one of remarkable nonlinear control methods, there is NNDIC (Neural Network Direct Inverse Control). This control method is important as an intelligent approach to counter the problem of mathematical models. However, it show a high sensitivity to noise and disturbance. In order to ensure the stability and improve performance, this study applies IDCS (Inverse Dynamic Compensation via ‘simulation of feedback control Systems’) to NNDIC. The IDCS allows a suitable control without noise and disturbance by inputting control variable generated in simulation area into a real crane. Simulations and experiments verified that the proposed method suppresses the vibration of lifted load and solves the problem of stability against noise and disturbance.