Research on Control Schemes for Dynamic Positioning of Surface Ships
【摘要】：The upsurge of world's inhabitants and fast growth of the economy have led to the bulky consumption of both land and marine (offshore) resources. As an outcome, the problem of energy deficiencies has come to be more essential in today's life. Consequently, developing and utilizing the rich marine resources has remained to be more noteworthy and of practical implication. However, with the ongoing exploration of marine resources (such as hydrocarbon products) in the deep sea, the conventional anchoring and mooring positioning system can no longer meet the increasing requirements of marine vessel positioning operations. Thus, this is a reason the ship (marine vessel) dynamic positioning (DP) systems were introduced. However, a marine vessel at sea subjected to various hydrodynamic forces and moments, and environmental disturbances (such as wind, water waves, and ocean currents) as well. The presence of these terms has caused difficulty in achieving the dynamic positioning operations at deep sea. Hence, a more precise DP system is required to achieve the operation goals. It means that accurate controllers should be designed to meet the requirements of the DP operations. As some literatures show, there have been many studies conducted concerning control systems design for DP systems. Thus far, more and more studies are required so as to enhance the performance and applicability of the DP systems.In control theory, there are many different control methods of which some of them have been proposed for DP system. Alternatively, we have witnessed in some literatures, the combination of two or more control techniques to form an enhanced controller for DP system. Likewise, this thesis presents research on control schemes for dynamic positioning of surface ships. The emphasis of this thesis is to design and propose various control schemes for DP system. Firstly, we propose a high-gain observer control scheme for DP system. The key aim is to design a high-gain observer for estimation of unmeasured states (velocity) and eliminate the constant known disturbances. Equally, we design an output feedback controller based on the high-gain observer for dynamic position system. The main theme is to design feedback control law that can force the ship to its desired position and heading. Vectorial backstepping technique is employed to derive an output feedback control law. The output feedback controller utilizes estimated states from the high-gain observer for feedback. Simultaneously, an adaptive law is derived to estimate the unknown constant disturbances. Finally, the global uniformly asymptotic stability (GUAS) performance is guaranteed.Secondly, a Relaxed LMI Stability based fuzzy controller for DP system of surface ships is proposed. Takagi-Sugeno (T-S) fuzzy model is applied to formulate the ship's mathematical model into the fuzzy model. Then, we employ parallel distributed compensation (PDC) design technique to formulate the fuzzy rules. Relaxed LMI and decay rate stability conditions are used to derive the stability of the resulting fuzzy controller. Linear matrix inequality (LMI) solver is used to compute a positive definite matrix and control gains that stabilize the fuzzy controller.Thirdly, a design of a robust fuzzy controller and the robust observer is presented. It assumes two cases:case one considers when the states are available for measurements and the environmental disturbances are unknown but bound with, the upper limit. Thus, the design of a robust fuzzy controller using H∞ attenuation design method is given. The role of H∞ attenuation method in the fuzzy controller is to abolish the disturbance to a prescribed minimum level. The second case assumes that the state variables are not known for measurements. Hence, robust fuzzy observer is design to estimate the unavailable states. Then, optimal H∞ attenuation method is utilized in the first case. Applying the LMI stability conditions and Lyapunov stability theory, the stability of the controller can be guaranteed. In addition, both cases prove the uniformly upper bound (UUB) and the performance of H∞ control can be guaranteed.Lastly, simulations for each proposed control scheme are carried out, and simulation results are given. Obtained results demonstrate the performance and validity of each control schemes.