A structure for predicting wind speed using fuzzy granulation and optimization techniques

With the increasing scarcity of global energy, the rapid development of science and technology, and the growing demand for environmental protection, wind energy is receiving increasing attention as the cleanest source of energy. Due to its pollution-free nature and widespread availability, it has become a preferred source of electricity generation in many countries. However, wind speed prediction plays a vital role in wind power generation. Traditional prediction models, due to randomness and uncertainty, often produce unstable and inaccurate results, leading to power and economic losses. Therefore, this study proposes a hybrid prediction system based on an information processing strategy and a multi-objective optimization algorithm. By preprocessing the data and optimizing the combination of five individual models, the singularity of a single model is overcome, a Pareto-optimal solution is obtained, and accurate and stable prediction results are provided. To verify the effectiveness of the proposed combined model in predicting wind speed, various experiments on a wind speed series were conducted based on a wind power station located in Penglai, China. The results show that the combined model proposed in this study has better prediction performance than conventional models.