Experimental study of redesigned draft tube of an Agnew microhydro turbine

Abstract

In this study, a surrogate-based optimization has been carried out for the components of an Agnew microhydro turbine. A neural network was constructed as the surrogate, while the Non-dominated Sorting Genetic Algorithm (NSGA-II) was used as the optimizer. The optimal design was found by numerical simulations, and the final design was manufactured and installed at the turbine outlet. The performance of the turbine components was then measured according to ASME performance test code. Comparison was carried out between the original draft tube and its modified under different operating conditions. The test results have confirmed that the pressure recovery factor of the new component increases by 20.3% and the loss coefficient diminishes by 4.0%, with regard to the original design under the best operating conditions.