ADP based long-term renewable generation planning while considering the effects of hourly SCUC

Abstract

This paper proposes an Approximate Dynamic Programming (ADP) based approach for solving the long-term renewable generation expansion planning problem by considering the effects of hourly security-constrained unit commitment (SCUC). Compared with traditional approaches, the proposed approach can assist Independent System Operators (ISO) to determine the dynamic expansion for multiple years on the basis of current year information, hence it can reduce the computation burden from future states forecasting. The objective of the proposed long-term power system planning is to minimize the total costs of investment and operation for the base case with forecast values while considering variable cost from uncertainties. Numerical case studies on a 6-bus system illustrate the effectiveness of the proposed ADP based long-term planning model for the integration of renewable energy under various uncertainties.