Multivariate modelling and analysis for location suitability evaluation of onshore wind power station

The efficient utilization of wind energy resources and wind power generation is a key pathway to achieving the United Nations’ 2030 Sustainable Development Agenda and China’s “dual carbon goals.” Proper planning and siting are critical factors in realizing these objectives. This study focuses on Guangdong Province, a region rich in wind energy resources in China, and designs an onshore wind power station (OWPS) siting optimization method that combines suitability evaluation with deep reinforcement learning. First, the spatial distribution patterns of existing onshore wind farms in Guangdong Province were analyzed using kernel density analysis, and a suitability evaluation system for wind farm siting was constructed based on environmental and socioeconomic indicators. The results show that: (1) OWPS siting suitability is influenced by various factors from the natural environment and socioeconomic conditions, among which GDP, elevation, and urban development level have the most significant impact on OWPS siting suitability, with entropy values of 0.5917, 0.1597, and 0.1489, respectively. (2) The evaluation of OWPS siting suitability levels reveals significant spatial heterogeneity at the current stage in Guangdong Province, with the western Pearl River region, including cities such as Zhuhai, Jiangmen, and Yangjiang, showing high OWPS siting suitability and substantial wind energy potential. This study provides theoretical support for optimizing wind power siting in Guangdong Province, promoting the efficient use of clean energy, and fostering sustainable development.