Multistage decision support framework for sites selection of solar power plants with probabilistic linguistic information

Energy consumption is constantly improving due to the increasing development of community economy and material living standards. Solar energy is the first green energy, and its unique advantage is of great importance in solving the energy crisis and environmental degradation. Furthermore, selecting appropriate locations is necessary to take full advantages of solar energy. This study intends to set up a multistage decision support framework for sites selection of solar power plants with probabilistic linguistic (PL) information. The proposed framework considers not only the disposal of linguistic information but also the interrelationships among criteria. A probabilistic linguistic term set (PLTS) is used to deal with a large volume of linguistic evaluation information. Then, probabilistic linguistic normal cloud (PLNC) is proposed to transform qualitative concepts into quantitative values so as to handle PLTS effectively and reduce information loss and distortion. A new distance measure that connects a Heronian mean (HM) operator with distance measure is also presented. An improved maximizing deviation method is introduced on the basis of this new model to deal with the interrelationships among criteria. In addition, a ranking method VlseKriterijumska optimizacija I Kompromisno (VIKOR) which considers the maximum utility of a group and the minimum regret value of individuals is proposed. Next, the proposed framework is successfully implemented in a case study. Results show that Gansu and Xinjiang are more suitable locations for building solar power plants. Sensitivity analysis is carried out to explore the change in ranking order due to variations in parameters. The proposed assessment model is then compared with the extant methods to deal with PL information and thereby certify its effectiveness, feasibility and advantages. Ultimately, the policy recommendations and contributions of this article are laid out. (C) 2019 Elsevier Ltd. All rights reserved.