Generation of solar radiation data in unmeasurable areas for photovoltaic power station planning

In order to obtain the solar radiation data (SRD), which are crucial for the photovoltaic power station planning, for the unmeasurable areas in China, both an inverse distance weighting (IDW) method and an artificial neural network (ANN) method are adopted to generate monthly global solar radiation (MGSR) data for the studied cities in this paper. For both methods, typical MGSR data for all the cities concerned are acquired by applying the typical meteorological year (TMY) method. The results show that the IDW method is only suitable for the case where the sampled cities and the studied city have relatively concentrated distribution and similar altitudes, while the ANN method performs well not only for the forementioned case but also for the case where the cities have comparatively dispersed distribution and different altitudes.