Temporal and spatial simultaneity assessment of wind-solar energy resources in India by statistical analysis and machine learning clustering approach

Abstract

The performance of hybrid power projects significantly relies on simultaneity of energy sources considering the generation fluctuations and grid penetration requirements. The aim of present study is to develop a novel three-module methodology for temporal and spatial simultaneity assessment of wind-solar energy resources useful for technical and economic hybrid (TH and EH) projects. The first module comprises data procurement and system performance evaluation. The second module applies temporal and spatial statistical correlations (Pearson, Spearman, and Kendall) to determine comparative simultaneity (complementarity-synergy). The third module encompasses dimension reduction (principal component analysis) and machine learning classification (Elbow algorithm aided k-means clustering) to classify study region into optimum number of clusters. The proposed methodology is applied over Indian onshore region employing ERA5 reanalysis dataset. The results indicate that the islands of Andaman and Nicobar and south-western parts of India are preferable sites for TH projects. Similarly, the spatial simultaneity clusters signify that the western and south-western part of the country is comparatively preferable for all four types of EH projects. The findings of this study will facilitate project developers, system manufacturers, and policymakers for better understanding of the typical peculiarities of various resources across geographical locations well in advance to deploy hybrid projects.