A heuristic methodology to economic dispatch problem incorporating renewable power forecasting error and system reliability. (March 2016)

• Record Type: Journal Article
• Authors: Lujano-Rojas, J.M.; Osório, G.J.; Matias, J.C.O.; Catalão, J.P.S.
• Is Part Of: Renewable energy. Volume 87:Part 1(2016)
• Page Start: 731
• Page End: 743
• Publication Date: 2016-03
• Subjects: Insular power systems -- Power system reliability -- Probabilistic economic dispatch -- Wind power forecasting error; Renewable energy sources -- Periodicals; Power resources -- Periodicals; Énergies renouvelables -- Périodiques; Ressources énergétiques -- Périodiques; 333.794
• DOI: 10.1016/j.renene.2015.11.011
• Languages: English
• ISSNs: 0960-1481

Abstract

Abstract: With the constant increment of wind power generation driven by economic and environmental factors, the optimal utilization of generation resources has become a critical problem discussed by many authors. Within this topic, determination of optimal spinning reserve (SR) requirements is a key and complex issue due to the variable and unpredictable nature of renewable generation besides of generation unit reliability. Cost/benefit relationship has been suggested as a way to determine the optimal amount of power generation to be committed by taking into account renewable power forecasting error and system reliability. In this paper, a technique that combines an analytical convolution process with Monte Carlo Simulation (MCS) approach is proposed to efficiently build cost/benefit relationship. The proposed method uses discrete probability theory and identifies those cases at which convolution analysis can be used by recognizing those situations at which SR does not have any effect; while in the other cases MCS is applied. This approach allows improving significantly the computational efficiency. The proposed technique is illustrated by means of two case studies of 10 and 140 units, demonstrating the capabilities and flexibility of the proposed methodology. Highlights: An analytical convolution process is combined with Monte Carlo simulation. The proposed method uses discrete probability theory. This approach allows improving the computational efficiency. The proposed technique is illustrated by means of two case studies. The capabilities and flexibility of the proposed methodology are demonstrated.