Modeling and controlling of ship general section attitude adjustment process based on RBF neural network coupled with sliding mode algorithm

Due to the advantages such as high efficiency, high precision, and the ability to reduce welding distortion, the block assembly method in shipbuilding possesses currently holds a dominant position in shipbuilding engineering. However, some key issues including low adjustment precision and slow control response speed urgently need to be resolved for the block assembly adjustment technology. This paper committed to solving the problems of inaccurate tracking of target displacement and slow control response speed in the vertical motion axis of the ship block joining equipment. A docking equipment control method based on the RBF neural network coupled with adaptive sliding mode algorithm was proposed. Firstly, an overview of the overall mechanics and control architecture of the ship block joining equipment was provided. Subsequently, a mathematical model for the transmission at the lifting mechanism was established. A sliding mode controller based on position control for the ship block joining equipment was designed for the transmission system. Then, the RBF neural network was employed to adjust the switching gain of the sliding mode controller and develop a self-adaptive sliding mode controller. Finally, simulations and verifications were conducted for multiple sets of input trajectories with different types. The results demonstrated that the combination of the neural network algorithm and the sliding mode control algorithm model presented in this paper reduces the system response time by 28.125% and improves the average motion tracking accuracy by 30.76%.