Design space exploration of centimeter-scale wind turbines using a physics-modified optimization formulation

Abstract

This paper explores the design space of centimeter-scale micro wind turbines to power wireless sensors through an experimentally validated modeling and simulation environment. A stochastic optimizer is used to obtain a functional relationship between the minimum wind velocity required to find a feasible design and multiple constraints relevant to turbine designers, such as the maximum turbine radius, electrical power required, minimum voltage required and available generators. This relationship is created from an optimization formulation that uses knowledge from the underlying physics and previous optimizations. It is shown that the design space of micro wind turbines is significantly different than large wind turbines due to the low Reynolds number regime. Also, a strong coupling exists between the choice of generator and optimal wind turbine geometry to minimize the wind required to meet the requirements. Smaller generators are more appropriate for micro wind turbines only if a constraint is applied on the maximum radius of the turbine and if no minimum voltage is required for a fixed power output.