GRMWBO: design of a general-purpose repowering model for wind farms via bioinspired optimization

Abstract

Wind farm optimization requires efficient design of placement models which can assist in identification of deployment locations for higher conversion efficiency. To perform this task, various methods are proposed by researchers, and most of them are applicable for new deployments. Due to limited availability of wind farm sites and cost constraints, it is impractical to replace old equipments. Thus, repowering of wind farms is necessary, and must be planned under existing deployment constraints. Existing repowering methods apply static modelling techniques, which limits their scalability and usability for real-time deployments. To overcome this limitation, a novel general-purpose repowering model via bioinspired optimization is discussed in this text. The model proposes design of a genetic algorithm that considers inflow wind speed, wind direction angle, height of hub, surface roughness, horizontal-axis wind turbine rotor diameter, thrust coefficient, turbine efficiency, air density, and wake expansion rate to estimate location of wind-based generators. These locations are optimized for maximum output power, and minimum deployment costs, which can be estimated for multiple farm areas. This text also proposes various case studies and hypothesis about these deployments, and showcases performance optimizations w.r.t. optimum placement for a given number of turbines. Based on this evaluation, it was observed that the proposed GRMWBO model is capable of improving power efficiency by 18.5%, reduce cost by 8.3%, and reduce number of wind-based generators by 2.9% when compared with various state-of-the-art models. It was also observed that the proposed model showcased consistent performance under different use cases. Thus, making it applicable for a wide variety of real-time deployments.