Var planning problem considering conditional value-at-risk assessment

Abstract

This paper presents the reactive power planning solution under risk assessment through the CVaR (Conditional-Value-at-Risk) using stochastic programming. Load uncertainty is modeled by distribution function. Uncertainty in the reactive power availability of existing and new reactive power sources is modeled through probabilistic constraints with a ρ-quartile measure. The taps settings of the under-load tap-changing transformers are modeled as discrete settings. The problem solution in this paper includes a reasonable number of possible future scenarios that calculate a set of solutions which allow to find the best flexible planning and adapting to future scenarios of power system operation such that the planning has found local optimum solution quality. The tradeoff between risk mitigation and cost minimization is analyzed. The efficacy of the proposed model is tested and justified by the simulation results using the CIGRE-32 electric power system.