Collaborative energy management of interconnected regional integrated energy systems considering spatio-temporal characteristics

Multi-regional integrated energy systems (MRIES), which encompass renewable distributed generation (RDG) and electric vehicles (EV) dispersed across multiple regions, are pivotal for promoting low-carbon energy supplies, offering both economic and environmental benefits. However, the complexity introduced by diverse energy conversions and spatiotemporal coupling in MRIES poses significant challenges for energy management. This paper introduces a novel low-carbon energy management framework (EMF) for multi-regional IES, based on spatiotemporal correlations and considering three-dimensional (3D) integrated demand-side response (IDR). The proposed framework employs a novel Cross-Gated Spatio-Temporal Graph Convolutional Network (CGSTG) to achieve high-precision joint power forecasting separately for the RDG and multi-type loads. In the day-ahead stage, 3D IDR involving flexible loads, e.g., EVs, is considered with multi-regional IES sharing electricity and heat. By engaging in IDR across the temporal (horizontal), spatial (depth), and energy conversion (vertical) dimensions, each IES can enhance its coordinated scheduling capability and operational economy. During the intra-day stage, rolling optimization is performed based on the model predictive control (MPC) framework, incorporating a ladder-type carbon trading mechanism (CTM). Comparative experiments demonstrate that the proposed method can effectively reduce operational costs and carbon emissions with significant improvement of coordinated scheduling capability, leading to economic and environmental benefits of MRIES.