Adaptive Disturbance Rejection Control of Overhead Cranes Based on Neural Network

A neural network-based online optimization algorithm for double linear active disturbance rejection control is proposed to address the problems of large load swing and long convergence time of overhead cranes in the lifting process. Firstly, a double linear active disturbance rejection controller (DLADRC) is designed by adding the control of load swing angle to the conventional linear active disturbance rejection controller (LADRC). Then, a neural network double linear active disturbance rejection controller (NN-DLADRC) is designed by combining the DLADRC with a neural network, which realizes the online tuning of the controller bandwidth and the observer bandwidth. In addition, considering the actual working environment of the crane, simulations were performed under a variety of working conditions and disturbances. The proposed control strategy enhances the adaptability of the crane under different working conditions and disturbances, and realizes the accurate positioning of the trolley and the swing suppression of the load. The effectiveness and superiority of the proposed control strategy is verified by simulations.