Reliability centered planning for distributed generation considering wind power volatility

• Published in: Electric power systems research Vol. 81; no. 8; pp. 1654 - 1661
• Main Authors: Novoa, Clara, Jin, Tongdan
• Format: Journal Article
• Language: English
• Published: Amsterdam Elsevier B.V 01.08.2011; Elsevier
• Subjects: Applied sciences; Direct energy conversion and energy accumulation; Distributed generation; Electric power plants; Electrical engineering. Electrical power engineering; Electrical power engineering; Energy; Exact sciences and technology; Loss-of-load probability; Miscellaneous; Moment methods; Natural energy; Stochastic programming; Weibull distribution; Wind energy; Wind power; Loss-of-load probability; Distributed generation; Wind power; Weibull distribution; Moment methods; Stochastic programming; Performance evaluation; Cost minimization; Stochastic model; Positioning; Wind generator; Distributed power generation; Wind energy; Probabilistic approach; Statistical moment; Moment method; Distributed system; Dimensioning; Case study; Genetic algorithm; Heuristic method; Economic aspect; Planning; Electric power production; Reliability
• ISSN: 0378-7796; 1873-2046
• DOI: 10.1016/j.epsr.2011.04.004

[Display omitted] ► We propose a probabilistic model to plan distributed generation systems with variable wind power. ► Moment methods are shown to be effective to characterize power volatility and load uncertainty. ► Genetic algorithm and heuristics methods solve the non-linear integer planning issues efficiently. ► Research shows the loss-of-load probability can be controlled even with 30% wind power penetration. This paper investigates a stochastic planning model to minimize the lifecycle cost of distributed generation (DG) systems under the energy reliability criterion, namely the loss-of-load probability. In particular, our study focuses on the DG system penetrated by renewable wind technology. The optimization is formulated to determine the wind turbine capacity and their placement in the DG system with the intent to minimize the capital, operational and environmental costs. Statistical moments including mean and variance are utilized to characterize the wind power volatility and the load uncertainty. Genetic algorithm combined with heuristic search is used to find the best sitting and sizing of the distributed energy recourses. Our study is among the first attempts in the literature to model and optimize DG system based on continuous probabilistic theory. The moment methods are shown to be effective in characterizing the stochastic behavior of wind power and load dynamics. Case studies are provided to demonstrate the application and performance of the planning method.