Design analysis and optimisation of the self-rectifying impulse turbine for ocean wave energy conversion using computational fluid dynamics and experimental analysis

The worldwide strategic nature of energy, global warming, as well as political instability in major oil producing regions, requires the identification and commercial development of alternative energies, as a matter of great importance. Wave energy is one alternative that offers a very large potential resource. One of the more explored wave energy device concepts is the Oscillating Water Column (OWC). OWC based wave energy plants convert wave energy into low pressure pneumatic power in the form of bi-directional air flow. Self-rectifying turbines, such as the Wells and impulse turbines, are then used to extract mechanical shaft power from this oscillating flow. Due to the poor starting characteristics and narrow operating range of the Wells turbine, the impulse turbine, which is self-starting and has a wider operating range, has the potential to become the standard self-rectifying turbine for OWC based power plants. However, the peak efficiency of the impulse turbine is currently lower than the Wells turbine peak efficiency.

In order to increase the peak efficiency of the impulse turbine, Computational Fluid Dynamics (CFD) techniques have been employed in this research to investigate the effect on turbine performance of specific parameters of the rotor and guide vane arrangement. Although the airflow over the turbine will be bidirectional and unsteady in nature, for this CFD analysis only steady, unidirectional flow was considered. The computational modelling techniques were validated by conducting a comparison of predicted results and actual experimental data collected at the University of Limerick using the original impulse turbine.

2D CFD models were initially used to investigate the effect of blade thickness and inlet/outlet angle at three radial sections of the rotor blade. The 2D CFD analysis of each section revealed that optimisation of individual blade sections could lead to an improvement in the performance of the turbine. It was also found that a taper in blade thickness and a variation in inlet angle from hub to tip resulted in an improvement over the current, constant section, turbine blade.

Upon completion of the 2D CFD work the optimised 2D sections were combined to form a 3D model of the blade. This allowed subsequent 3D CFD analysis to be carried out, to quantitatively evaluate the performance of the proposed geometry compared to the original blade geometry as well as investigating additional parameters such as blade sweep angle and guide vane setting angle for incorporation of the optimised impulse turbine into a variable-pitch controlled guide vane system. Complex 3D CFD models were created that included tip-clearance. These models allowed an analysis of the effects of tip-leakage flows present in the original and optimised impulse turbine. The addition of an oversized tip-cap onto the blade tip was also studied. The 3D CFD models confirmed the improvement in efficiency predicted by the optimisation conducted using 2D models and demonstrated the effectiveness of the tip-caps in reducing tip-leakage flows and subsequent losses in the optimised impulse turbine. The computational models predicted an improvement in the region of 5.6 % in the peak efficiency for the optimised impulse turbine with FGVs.

A comprehensive experimental analysis was conducted using a 0.6 m experimental model of the optimised impulse turbine rotor with blades constructed from ABS plastic and a steel sheet tip-cap attached using a two part structural adhesive. Experimentation was conducted using two guide vane configurations, fixed guide vanes (FGVs) and variable-pitch controlled guide vanes (VPCGVs). Initially steady flow conditions were utilised to establish optimum parameters. In the case of the impulse turbine with FGVs, a series of steady flow experiments with a range of velocities were used to establish a va=7 m/s to be the steady velocity that produced the maximum peak efficiency of 51 %. In the case of VPCGVs steady flow experiments were used to investigate the effect of nozzle and diffuser setting angle on the performance of the optimised impulse turbine. A nozzle angle of 15° and diffuser setting angle of 60° were determined to produce the optimum efficiency, of 67.7 % for this impulse rotor with VPCGVs.

The performance characteristics of the optimised impulse turbine were also compared to that of the original impulse turbine for a range of flow conditions. Experimental data for the original impulse turbine was taken from all available sources. It was determined, that under all conditions considered and with both guide vanes configurations, that the optimised impulse turbine exhibited a higher peak and average efficiency than the original impulse turbine. The improvement achieved was between 2.5 % and 6.2 % depending on flow conditions and guide vanes configuration.