Stochastic optimal operation model for a distributed integrated energy system based on multiple-scenario simulations

Highlights: • Latin hypercube sampling and K-means clustering are combined to build scenarios. • Uncertainties of load prediction and power output are considered for modeling. • Energy efficiency index is transfer into fuzzy economic cost for optimizing. • Total energy efficiency and new energy consumption capacity are considered. Problems related to the uncertainties of the sources and loads in integrated energy systems (IESs) are becoming more prominent with the interconnection of large-scale renewable energy sources and multi-energy loads. Moreover, such scenarios pose great challenges for the optimal operation of IESs. A distributed IES in an industrial park is regarded as the research object, and a stochastic optimal operation model based on multiple-scenario simulations is proposed to consider the prediction uncertainties arising in the case of distributed power generation and multi-energy loads. Specifically, scenario analysis for stochastic optimization is applied to address these prediction uncertainties in a two-part approach: operation scenario generation based on Latin hypercube sampling (LHS) and the reduction of multiple scenarios into a smaller number of more general scenarios based on K-means. Afterwards, a day-ahead stochastic optimal operation model for a distributed IES with the total operating economy as the decision-making objective is proposed based on typical operation scenarios. Moreover, the overall energy efficiency and new energy consumption capacity are all considered. In this way, the safe and economical operation of the IES can be guaranteed even under the negative influence of uncertainties. The validity and rationality of the proposed model are verified by analysis of examples.