A two-stage scheduling method for integrated community energy system based on a hybrid mechanism and data-driven model

The integrated community energy system (ICES) is an effective means to promote the synergies among multiple energy carriers. However, the off-design performance of equipment challenges the accurate and economical scheduling of the ICES. To solve this problem, a two-stage scheduling method for the ICES based on a hybrid mechanism and data-driven model is proposed in this paper. Combing the mechanism energy hub (EH) model with a data-driven efficiency correction model, a hybrid-driven dynamic energy hub (DEH) with variable equipment efficiency is built first. The EH describes the multi-energy coupling relationship; the embedded efficiency correction model adopts data-driven approaches of polynomial regression (PR) and backpropagation neural networks (BPNNs) to accurately extract nonlinear characteristics of equipment efficiency. On this basis, a two-stage scheduling model for the ICES is developed. In the day-ahead stage, the PR method is applied to calculate equipment efficiency which varies with load rate. The day-ahead scheduling model is established with the aim of minimizing the operating cost. In the intraday stage, considering the effects of load rate, temperature, and atmospheric pressure, the BPNNs method is employed to further correct equipment efficiency using the latest data. Furthermore, a rolling optimization (RO) strategy is used to address the uncertainties of equipment efficiency and load demand to improve the accuracy and economy of the scheduling scheme. Case studies demonstrate that the proposed method can improve the solution speed and accuracy of the scheduling model, and enhance the operating economy of the ICES.

Automated summary

This paper proposes a two-stage scheduling method for the integrated community energy system (ICES) based on a hybrid mechanism and data-driven model. The method combines the mechanism energy hub model with a data-driven efficiency correction model to accurately extract nonlinear characteristics of equipment efficiency. Case studies demonstrate that the proposed method can improve the solution speed and accuracy of the scheduling model, and enhance the operating economy of the ICES.