Adaptive PID control of overhead cranes based on ISOA–NN

Positioning and anti-swing are the key issues of overhead crane control systems. An adaptive PID control strategy based on improved seagull optimization algorithm and neural network is proposed for overhead cranes under variable working conditions and disturbances. An adaptive PID controller based on a neural network is firstly designed to adjust parameters of the controller online. Then, an improved seagull optimization algorithm with an adaptive inertia coefficient is introduced to optimize the initial weight of the neural network. The optimization process of controller parameters and initial weights of neural network are given in the form of two algorithms. The results show that, compared with the adaptive PID control methods based on fuzzy rules and neural network technique with random initial weights, the proposed adaptive PID control strategy can make the crane system more adaptable to payload changes and more robust against disturbances. Additionally, compared with the particle swarm optimization algorithm, the improved seagull optimization algorithm makes the crane system more stable and has better positioning and anti-swing performance.