Value-At-Risk for Stochastic Optimization of Integrated Energy Systems in HERON

Abstract

Optimizing the economic potential of integrated energy systems (IES) is a challenging process, requiring careful consideration of projected energy demand evolution, local policies and market drivers, and intrinsic uncertainty of both load and variable renewable energy sources (VREs). Previous stochastic portfolio optimization efforts use the expected value of the net present value (NPV) metric to optimize IES-included portfolios. In this paper, we report on a study that contrasts minimizing value-at-risk (VAR) over stochastic scenarios with maximizing expected NPV as economic metrics in optimization. This strategy is demonstrated by optimizing the introduction of advanced Nuclear Power Plant (NPP) generation directly coupled with thermal energy storage (TES) as an IES in NYISO. We consider the statistical benefit of introducing these technologies under two different future policy scenarios. One of these scenarios is taken as “reference” and includes current energy policy in the New York Independent System Operator region (NYISO), while the other scenario assumes a national clean energy standard (CES) in the United States. We show the results of flexible advanced NPP operation with TES in each scenario and discuss the implications of the optimized portfolios. We further contrast the optimized portfolios when VAR is used as an optimization metric instead of expected NPV.