A Stochastic Framework for Multi-stage Generation Expansion Planning under Environmental and Techno-economic Constraints

• Published in: Electric power components and systems Vol. 44; no. 17; pp. 1917 - 1934
• Main Authors: Shivaie, Mojtaba, Ameli, Mohammad T.
• Format: Journal Article
• Language: English
• Published: Philadelphia Taylor & Francis 20.10.2016; Taylor & Francis Ltd
• Subjects: Algorithms; augmented Powell heuristic method; Decision theory; distribution companies; Electric utilities; Electrical engineering; Electricity; Expansion; generation companies; generation expansion planning; information gap decision theory; Markets; Mathematical models; melody search algorithm; Nonlinearity; Optimization; Planning; pool-based electricity market; Risk; risk-averse decision maker; risk-taker decision maker; stochastic framework; Stochasticity
• ISSN: 1532-5008; 1532-5016
• DOI: 10.1080/15325008.2016.1199067

In this article, a new stochastic framework is proposed for multi-stage generation expansion planning subject to environmental/techno-economic constraints. In this framework, with a new point of view, impacts of expansion plans on strategic behaviors of both generation companies and distribution companies are evaluated. To do so, the proposed framework is developed as a three-level model. The lower level indicates a pool-based electricity market, the intermediate level provides strategic interactions among all market participants, and the upper level represents the generation expansion planning by considering the outcomes of the previous levels. To minimize risks of planning arising from severe uncertainties, a well-founded information gap decision theory is used to model both risk-averse and risk-taker decision makers. The newly developed model is a non-convex non-linear mixed-integer optimization problem. A new melody search algorithm and an innovative augmented Powell heuristic method are employed to obtain the final optimal solution. The feasibility and capabilities of the proposed methodology are illustrated on the 46-bus south Brazilian electric power grid. The obtained results from case studies validate the profitability of the newly developed framework.