摘要
Abstract
Given the complexity of the design and the poor robustness of the current crane anti-swing system,a robust control strategy based on the radial basis function(RBF)is proposed.This strategy aims to simplify the design of the anti-swing system and enhance the robustness of system control.Leveraging the self-iteration advantage of neural networks and designing an adaptive update rate,the strategy addresses information that is difficult for sensors to accurately collect in the crane anti-swing system,such as real-time estimation of the movement state of crane's lifting load.Consequently,a crane anti-swing control strategy based solely on the position feedback of the crane trolley is proposed.To further enhance control robustness,the asymptotic stability of the closed-loop error of the control strategy is proven.This is achieved by designing a neural network-based sliding surface with an adaptive update rate and constructing a Lyapunov function.This approach strengthens the robustness of the anti-swing system.To validate the practical effectiveness of the designed strategy,it is compared with a hierarchical sliding mode control scheme using the control variable method.The results demonstrate the strategy's advantages in multiple performance metrics,ease of implementation(i.e.,the strategy does not require knowledge of load speed or other information),and robust control.关键词
桥式起重机/鲁棒性控制策略/Lyapunov定理/神经网络滑模/反馈Key words
bridge crane/robust control strategy/Lyapunov theorem/neural network sliding mode/feedback分类
信息技术与安全科学
