Abstract
For the load swing and trolley tracking and positioning issues of overhead crane system during lifting operations with uncertain inherent parameters and partial unknown parameters, a dynamic sliding mode robust control algorithm based on RBF neural network is suggested. The Lyapunov stability of the closed loop error is Theoretically testified by the control algorithm .The control algorithm takes the sliding mode control idea as its structural framework. The design of the sliding mode surface takes into account the impact of the alteration in the control output, makes the sliding mode switching term adaptive by setting fuzzy rules ,and further derives the self adaption laws of the RBF neural networks to adapt the intricated as well as unknown dynamic parts in the system of overhead crane, rendering the operation of the system of overhead crane with no need for any system parameters as the signal input. For the sake of testifying the superior control accuracy under the influence of this tactics, a comparison was made with the layered sliding mode control method, with results demonstrated that the output power of the controller based on above algorithm designed in this paper proves more stable, and there is no long-term high-frequency jitter, while its anti-swing control effect on the load is also standing a better robust performance of any other controllers. It is also exceedingly significant to point out that when the system parameters change to some extent. While the trolley based on this control system can also achieve stable tracking and positioning under zero overshoot.