Investigation of in-pipe drag-based turbine for distributed hydropower harvesting: Modeling and optimization

Abstract

Hydropower systems can provide a considerable proportion of sustainable and clean energy. In the present study, the performance of a drag-based vertical axis in-pipe turbine is optimized to harvest the existing excessive potential energy from small diameter (100 mm) pipelines more effectively. The enhancement of the efficiency and output power of the turbine in this research makes it as a more promising and sustainable device for distributed power generation. To achieve a comprehensive solution all the factors affecting in-turbine performance has been taken into account simultaneously. Due to the complicated relation between turbine design parameters and its performance, artificial neural networks (ANN), which is a popular tool for modeling devices, has been deployed to generate a predictor model for turbine performance. The predictor model is formed on a dataset provided by 3D transient numerical simulations, which are validated by previous experiments. The optimization analysis is based on the in-pipe turbine non-dimensional variables, which are introduced to provide more simplification. For the particular water head loss less than 5 m and the flow rate bounded to the typical range, the proposed in-pipe turbine produces a power of 200 W with an efficiency over 33%, which shows a considerable improvement compared to the previously developed drag-based in-pipe turbines.