Deep Learning-based Self-scheduling of Virtual Energy Hub Considering Phase Change Material-based Thermal Energy Storage

Decentralized operation of power systems provides a great opportunity for DERs of multiple energy carriers to participate in the energy markets. A virtual energy hub (VEH) can operate the DERs locally and aggregate their output profile to participate in the retail and wholesale markets. In this paper, deep learning-based self-scheduling of a VEH for participating in the electrical and thermal energy markets is presented. A bidirectional long short-term memory network is used to forecast the uncertainties lied in the power production of the wind turbine, electrical demand, and electrical energy price. Novel modelling for Phase Change Material-based Thermal Energy Storage system (PCMTESS) is presented in this paper. The proposed model is simulated in the GAMS software to evaluate its performance and behaviour in response to energy price variation.