Hydrodynamic and control optimization for a heaving point absorber wave energy converter

The thesis aims at studying the non-linear performance of a designed 1/50 scale point absorber wave energy converter (PAWEC) in heave motion. In particular, designs of the PAWEC geometry and control strategy are considered to optimize the power capture. Experimental and computational fluid dynamics (CFD) data are provided to evaluate the studies. Specifically, this thesis can be summarized into four parts.Firstly, a numerical wave tank (NWT) is constructed in a commercial CFD package ANSYS/LS-DYNA. The main objective associated with the NWT is to closely reproduce the physical wave-PAWEC interactions. To achieve this, physical experimental data from two specified WECs are provided to verify the capability of the NWT. One of the devices is the PAWEC designed at University of Hull. Free decay, excitation force and water splashing tests, etc., are conducted. As a result, the developed NWT is validated to be capable of representing the non-linear behaviors of the PAWEC compared with the costly physical experiments. The second part focuses on investigating the extent to which the non-linear hydrodynamic characteristics of the PAWEC need to be considered. By comparing with the CFD data from a series of tests, a non-linear mathematical modeling involving a quadratic viscous term is verified. The results show that the non-linear PAWEC behavior for the conditions of large oscillations (e.g., near resonance or at a large wave heights) can only be predicted realistically by considering a correct viscous effect. This study highlights that the linear counterpart derived from potential flow code ANSYS/AQWA fails to describe the PAWEC behavior and would mislead the control strategy and power take-off (PTO) designs. Additionally, the results show that the viscous damping is significantly larger than the inviscid radiation damping for the flat-bottom cylindrical heaving PAWEC.It is found that the viscous effect can induce clear energy losses during device oscillations which is unwanted for a PAWEC system. Therefore, in the third part, besides the originalflat-bottom PAWEC, two streamline-bottom counterparts are proposed to improve the capability of power capture. Free motion tests are conducted in the NWT regarding the three different geometric devices. The results indicate that for the streamlined devices, the added mass and hydrodynamic damping decrease by up to 60% compared with the flat-bottom device. More importantly by simulating PTO system in the NWT, it is found that there exists a clear mutual interaction among the designs of the device geometry and PTO damping. Applying a proper PTO damping to the streamlined PAWEC can prominently improve the optimal power absorption efficiency by up to 70% underboth regular and irregular waves, compared with the flat-bottom PAWEC. Finally, a fuzzy logic control strategy with particle swarm optimization algorithm (PSO-FLC) is implemented on the developed non-linear modeling to adaptively tune the PTO damping for power absorption maximization. The fuzzy rule base is initialized according to the power capture characteristics achieved through the NWT tests. PSO algorithm is then used to search for more efficient rules. It is found that applying a well-designed fuzzy inference system can adaptively tune the PTO damping for power capture optimization in contrast to the passive control with constant PTO damping.