Uncertainty and sensitivity analysis of generation expansion planning in hybrid nuclear and renewable energy system

Abstract

In this study, generation expansion planning (GEP) is executed with the lowest cost and pollutant emission, highest efficiency, and lowest loss of load probability (LOLP). A hybrid energy system integrated with nuclear and renewable sources will be a potential candidate for a carbon-free energy network. Integration of nuclear and renewable energy tackles the demerits present when they operate individually. In this new approach, the expansion of modular reactors and renewable energy is discussed in the presence of uncertainties raised by load demand, load duration curve (LCD), energy carrier prices, power plant depreciation rates, weather conditions, and renewables output. Moreover, sensitivity analysis is fulfilled to find the best reactor expansion capacity. Firstly, all uncertainties are modeled by their probability density function (PDF). Then a modified Metropolis Markov chain Monte Carlo (MCMC) coupled with the Wilks method was implemented to predict the stochastic behavior of different uncertainty sources exactly so that reaching to 90% reliability of the final plan. Finally, the approach is integrated into WASP-IV software to provide the optimal number of sources with optimal cost and pollutants. The results indicate that the LOLP index in the last year of the study period is only 1.4%–2.5%, and the number of installed nuclear units are 3 SMART reactor per 2 years just by 10% of tolerance, owing to considering all the uncertainties in the expansion process after sensitivity analysis. Furthermore, the emission of pollutants is about zero.