Wind Farm Layout Optimization Based on Dynamic Opposite Learning-Enhanced Sparrow Search Algorithm

• Published in: International journal of energy research Vol. 2024; no. 1
• Main Authors: Zhu, Yun, Guo, Yahui, Hu, Tianyu, Wu, Chengke, Zhang, Lidong
• Format: Journal Article
• Language: English
• Published: Bognor Regis Hindawi 2024; John Wiley & Sons, Inc
• Subjects: Accuracy; Algorithms; Alternative energy sources; Clean technology; Electric power generation; Electricity; Layouts; Local optimization; Machine learning; Optimization algorithms; Renewable resources; Search algorithms; Turbines; Wind direction; Wind farms; Wind power; Wind power generation; Wind speed
• ISSN: 0363-907X; 1099-114X
• DOI: 10.1155/2024/4322211

In recent years, the proportion of wind power in new energy generation has gradually increased. The natural wind in wind farms is subject to velocity attenuation by the wake effect, so improving the efficiency of wind farm power generation has become a problem that must be solved for wind power generation. Considering the uncertainty of wind farms, we regard wind farm layout optimization (WFLO) as a strongly nonlinear problem. In this paper, we improve the sparrow search algorithm (SSA) using dynamic opposite learning (DOL) strategy. Twenty-eight benchmark test results prove that compared with other algorithms, the improved algorithm DOLSSA has excellent robustness and the ability of searching for a better solution when solving a strongly nonlinear optimization problem, and the DOL strategy effectively improves the shortcomings of the original algorithm which is prone to local optimization and space limitation. In this paper, the authors establish the dynamic rotational coordinates of wind farms and set six different physical scenarios by considering the wind direction and wind speed variables, and the results prove that the performance of DOLSSA is optimal.