Design and optimisation of the impulse turbine with an internal variable-pitch controlled guide vane system using computational fluid dynamics and experimental analysis

Following a review of previous research on wave power conversion, it is evident that Oscillating Water Columns are one of the few devices to successfully convert wave power to electrical power at a commercial stage. Furthermore, it is evident that the impulse turbine is a key component in this conversion process and the aim of this research is to improve its performance by designing and implementing an internally housed variable-pitch controlled guide vane system. A validated 3-D CFD technique was used to analyse the performance of varying nozzle and diffuser guide vane setting angles under steady flow conditions at 0.5 and 0.6 hub-to-tip ratios. 0.5H/T recorded peak performance at 15⁰ nozzle and 82.5⁰ diffuser angles while 0.6H/T recorded 15⁰ nozzle and 72.5⁰ diffuser angles. It was observed that turbine performance was more sensitive to variation in nozzle than diffuser angle, as both torque and pressure loss were affected by the nozzle. Concurrent research by Walsh (2011) required investigation on the effect of a spherical hub and casing design on turbine performance using 3-D CFD, designs also implemented in this research. Results showed a reduction in peak efficiency of 1.1 percentage points at 0.5H/T while 0.6H/T was reduced by 0.13 percentage points. Flow visualisation confirmed that this reduction in performance was caused by an area of stagnation in the hub which increased pressure loss. A 2nd generation, internally housed variable-pitch controlled guide vane system was designed and tested to validate the potential performance improvements outlined in the CFD studies. A number of concepts were evaluated before an iterative design procedure was used to develop the selected concept. The system increased efficiency from 42.82% with fixed guide vanes to 56.56% under real sea, “Site 2” conditions while mechanical shaft power was increased from 97.8 W to 200.6 W. Steady flow testing confirmed nozzle angles of between 15⁰ and 17.5⁰ produced optimum performance although peak efficiency of 64.2% was recorded at 17.5⁰. Optimum diffuser angle of 72.5⁰ was identified. 0.6H/T experimental analysis by Walsh (2011) identified optimum angles of 15⁰ nozzle and 78⁰ diffuser, validating the CFD steady flow analysis. Scaling of model test results to full scale prototype was also investigated and two methodologies were identified, the Froude and polynomial scaling methods. These were successfully used to scale bi-directional model results to varying fullscale diameters and produce turbine design charts. Little variation between methods was noted with the exception of the inclusion of turbulent scatter using the Froude method. A CFD methodology was defined and validated through experimentation to successfully simulate bi-directional flow in turbomachinery problems, modernising and improving the accuracy of CFD analysis. This method used a sliding mesh with rescaled guide vane zone and produced performance trends similar to experimental for a fixed 30⁰ guide vane system. At 0.5H/T peak efficiency of 44.35% was computed compared to 43.14% experimental, while at 0.6H/T 45.52% was computed and 45.3% was recorded experimentally. The mixing plane method computed only unidirectional results and these were not in good agreement with experimental data.IRC