Spatial Analysis and Evaluation of the Development Potential of Roadside Distributed Photovoltaics : — A Case Study of Gansu Province

This study investigates the development potential of roadside distributed photovoltaics in Gansu Province by constructing a multi-dimensional evaluation index system that integrates natural conditions, infrastructure, and policy orientation. A combination of methods, including Analytic Hierarchy Process (AHP), spatial autocorrelation (Moran's I), and hotspot analysis (Getis-Ord $G^*$), is employed to systematically reveal the spatial differentiation patterns. The results indicate that: (1) There is a significant regional variation in the photovoltaic development potential along roadsides in Gansu Province. The northern Hexi Corridor (with the central area of Jiuquan City) is identified as an area of extremely high potential due to its high solar radiation ($\mathrm{GHI}\gt 1800 \mathrm{kWh} / \mathrm{m}^2$), favorable photovoltaic output (PVOUT $\gt 1900 \mathrm{kWh} / \mathrm{kWp}$), flat terrain (slope $\leq 5^{\circ}$), and low vegetation coverage ($\leq 30 \%$). In contrast, Longnan and Tianshui Cities face limitations due to low GHI ($\leq 1200 \mathrm{kWh} / \mathrm{m}^2$), steep terrain (slope $\gt 15^{\circ}$), and high vegetation coverage ($\gt 60 \%$), resulting in lower development potential. (2) Spatial correlation analysis reveals that the northern Hexi Corridor forms a “high-high” agglomeration area, with a high overlap between the road network and power grid, facilitating high energy consumption efficiency. In contrast, the Qilian Mountains and Longnan areas are hindered by complex terrain and sparse road networks, showing “low-low” or “low-high” agglomeration patterns. (3) Hotspot analysis confirms that the hotspot areas in the Hexi Corridor are contiguous, with a confidence level of $\geq 99 \%$, and exhibit significant long-term energy output stability. This study provides a technical pathway for regional distributed photovoltaic planning and proposes practical strategies, such as the integration of photovoltaics and transportation in multiple scenarios. Additionally, it emphasizes the need for further integration of dynamic meteorological and socio-economic factors to optimize the model.