Performance and optimization of savonius, darrieus, and combined savonius-darrieus hydrokinetic turbines in low-speed fluid flow

Most wetland rice cultivation schemes in Malaysia and Thailand were designed and built with goodiirrigation and drainage canal networks for the benefit farming activities. These canals have ampleivolume of running water alliyear round and could be harnessed forihydropower. Energy generation from small hydropower plants has been increased. Hydropower could be generated by converting the kinetic energy of flowing water intoimechanical power through the use of turbines. Vertical hydrokinetic turbines, either Savonius or Darrieus, have their own specific operational ranges and their pros and cons. Thus, the combination of Savonius-Darrieus turbines could overcome the weakness and complementithe strength ofithem. This study was conducted to optimize separately the design of Savoniusihydrokinetic turbine (SHKT) and the Darrieus hydrokinetic turbine (DHKT) with the ultimate objective of getting an optimum combined Savonius-Darrieus hydrokinetic turbine (SDHKT) design. ANSYS Fluent was used to optimum design configurations for the individual SHKT- and DHKT-models based on the typical flow conditions in canals. Validation tests on the developed computational approach against the published work of Saad et al., 2020 showed only 1.85% differences of cp,max. The cp of a two-stage model with full-shaft and a constant-stage aspect ratio was higher than a single-stage-model and a two-stage-model with a different stage-aspect ratio. To evaluate and optimize the adopted new designs of the delta-shaped blade DHKT with NACA0012 hydrofoils. Simulations were done for 2-blade models with four designs (MD1-MD4) of varying blade configurations (in-plane- and twisted-blades) and cross-sectional areas (reduced and constant areas). Validation tests on this developed computationalimethod against the experimental work of Doan et al., 2020 on the power coefficient curve showed good agreementiwith deviations of less than 1.25%. The optimum designs were models with twisted blades and reduced and constant cross-sectional areas (MD3 and MD4). The 3-bladed models with similar blade characteristics were tested and compared with the 2-bladed models. The 2-bladed models performed better during the higher range of tip speed ratio (λ), whereas the 3-bladed models were outstanding at the lower range. Overall, the MD4-model was the most appropriate design to operate under the specified conditions. To optimize the design configurations of the combined SDHKT-model, the optimum designs for SHKT and DHKT-models were taken as the fundamental basis. A 2-stage-SHKT model was placed at the center and inside the MD3 and MD4. The attachment angles (φ) and ratios of swept area (Rsa) were investigated using similar computational approaches employed in the earlier DHKT simulations. Validation tests on the developed computational approach against the laboratory work of Sahim et al., 2013 showed good agreement with a maximum deviation value of about 1.5%. The cp,max of φ = 0° was the highest when λ equaled 1 and higher by 0.02% and 0.88% when cp,max of φ = 20° and 30°. Besides, the combined SDHKT-models with φ = 0° and three ratios of Rsa1 = 0.4, Rsa2 = 0.3, and Rsa3 = 0.2 were simulated. The Rsa1 performed better when λ was lower than 0.9. Beyond the value of 0.9, the performances of the Rsa3 were higher than the rests.

Text FK 2024 29 - Declaration Form.pdf Restricted to Repository staff only Available under License Creative Commons Attribution Non-commercial No Derivatives. Download (681kB) Text FK 2024 29 - Full Text.pdf Available under License Creative Commons Attribution Non-commercial No Derivatives. Download (15MB) Text FK 2024 29.pdf Restricted to Repository staff only Available under License Creative Commons Attribution Non-commercial No Derivatives. Download (15MB) Official URL or Download Paper: http://ethesis.upm.edu.my/id/eprint/18994

Actions (login required)

View Item