Under complex wind scenarios: Considering large-scale wind turbines in wind farm layout optimization via self-adaptive optimal fractional-order guided differential evolution

Abstract

Wind energy has become a crucial solution for reducing environmental pollution through electricity generation. However, the wake effect between different wind turbines significantly impacts the efficiency of power generation in wind farms. Wind farm layout optimization (WFLO) seeks to mitigate the wake effect and improve overall energy conversion efficiency by optimizing turbine placement. As the number of turbines increases, maintaining high efficiency becomes more challenging, testing the performance of optimization algorithms. But existing algorithms often overlook valuable historical information, which can be crucial for improving optimization efficiency. Therefore, this paper proposes self-adaptive optimal fractional-order guided differential evolution algorithm (SaOFGDE) optimizing large-scale and complex WFLO. In SaOFGDE, classification crossover rate update approach based on historical information weighting is proposed to enhance the mutation efficiency of individuals. Secondly, the optimal fractional-order calculus method is designed to make high use of the information of historical individuals, which can provide more accurate derivative approximations to help individuals make optimal decisions. SaOFGDE is tested on four extremely complex WFLO in four large wind farms, comparing with eight well-established algorithms. The results show that SaOFGDE still improves the conversion efficiency by an average of 2.64 %, 5.10 %, 3.04 % and 3.86 % respectively under the largest wind turbine setting.