Hybrid stochastic robust optimization and robust optimization for energy planning – A social impact-constrained case study

Abstract

The uncertainties inherent in future projections affect energy planning schemes in different ways. In this study, both stochastic robust optimization and robust optimization were incorporated simultaneously into a proposed model to deal with multiple uncertainties. Risks that need to be immunized against all possible outcomes were dealt with using robust optimization, while other uncertainties were treated using scenario-based stochastic robust optimization. The ranges of optimal solutions determined from the proposed model were practical enough to generate the various alternatives, but robust enough to accommodate any risk-free requirements. Energy planning typically focuses on three main objectives: the security of the energy supply, the environmental protection and economic competitiveness. In this study, social acceptance which is one of the crucial influences, is also considered. To demonstrate the potential of the proposed model, a case study involving energy decisions in Thailand is featured. Furthermore, the model is applied to the energy planning of Vietnam as an alternative case study. Here, given the prominent role of social impact, it is especially critical to limit the variation in social damage that may result from planning uncertainties. The empirical analysis conducted in these cases includes both fossil fuel-based and renewable energy in the grid. The results show that strengthening system reliability, with a 92.6% reduction in capacity deviation, produces only a 5.08% increase in total cost. Numerical results from the model could help policy makers effectively address the trade-off between system stability and economy correlated with budgetary limits and determine effective weight coefficients for the preferred control levels.