Abstract
The advancement of deep-sea exploration equipment, with extended operational range and duration, necessitates the development of hydrokinetic turbines capable of efficiently harnessing low-speed current energy in deepwater environments. In response, we propose a novel Banki hydrokinetic turbine with superior self-starting capabilities. Nevertheless, knowledge gaps persist regarding the hydrodynamic mechanisms governing Banki hydrokinetic turbines. Furthermore, prototypes exhibit considerable potential for performance enhancement. Employing a parametric investigation, supported by a validated numerical model, this study elucidates the influence of critical geometric parameters on the turbine's performance. Building upon these findings, a synergistic approach involving the optimal Latin hypercube method, radial basis function neural network, and non-dominated sorting genetic algorithm is utilized for multi-objective optimization. Water flume experiments confirm that the optimized Banki hydrokinetic turbine can achieve a static torque coefficient surpassing 0.7, a magnitude exceeding tenfold that of lift-type horizontal axis hydrokinetic turbines. Remarkably, the optimized Banki hydrokinetic turbines exhibit the ability to self-initiate rotation at a current velocity as low as 0.2 m/s when configured with a turbine radius of 200 mm. This study reveals the precise geometric configuration of the optimized Banki hydrokinetic turbine, serving as a reference for the practical deployment of in-situ power generation systems capitalizing on this turbine technology.