Neural network-based modelling of wind/solar farm siting: a case study of East-Azerbaijan
Meysam Asadi, Kazem Pourhossein
Abstract
The location of wind/solar power plants is a critical part of design process. Multi-criteria decision making (MCDM), the well-known procedure of site selection, suffers from the local-scoring property. This paper proposes a combined approach of MCDM and artificial neural networks (ANN) to alleviate this deficiency. Here, the weighting of site selection criteria has been performed using the analytic hierarchy process (AHP), and then a multi-layer perceptron (MLP) is used for implementing the global scoring capability. By using this procedure, adding any new alternative site location cannot affect the scores of the others. In other words, the proposed procedure is global-scale and robust. Scores derived by this procedure for two candidate sites can be interpreted as real differences in these sites.