Model predictive control strategy improved by extended Kalman particle filter and fuzzy algorithm for dynamic positioning control of semi-submersible platform

As offshore oil and gas exploration extends into deeper waters, fixed platforms are becoming less suitable, while semi-submersible platforms are gaining wider application owing to their strong wave resistance and deepwater adaptability. This paper presents a dynamic positioning control strategy for semi-submersible offshore platforms, combining an Extended Kalman Particle Filter (EKPF) and Fuzzy Model Predictive Control (FMPC), to cope with input saturation and non-Gaussian measurement noise under complex environmental disturbances. Firstly, the EKPF is utilized to observe the state of the semi-submersible platform, thereby mitigating the adverse effects of measurement noise on the control performance. Meanwhile, a fuzzy algorithm is integrated into the model predictive control (MPC) framework to adaptively adjust the weight matrices in the cost function based on the state error and its rate of change, thereby enhancing adaptability to varying operational environments. Simulation results demonstrate that the use of the EKPF significantly improves control accuracy in both set-point regulation and trajectory tracking of the semi-submersible platform. Furthermore, they indicate that the FMPC strategy provides superior trajectory tracking performance compared to conventional Model predictive control.