This study presents a data-driven assisted real-time optimization model which is an innovative approach to address the challenges posed by integrating Submerged Arc Furnace (SAF) systems with renewable energy sources, specifically photovoltaic (PV) and wind power, with modern intelligent energy terminals. Specifically, the proposed method is divided into two stages. The first stage is related to data-driven prediction for addressing local time-varying renewable energy and electricity market prices with predicted information, and the second stage uses an optimization model for real-time SAF dispatch. Connections between intelligent energy terminals, demand-side devices, and load management systems are established to enhance local renewable resource utilization. Additionally, mathematical formulations of the operating resistance in SAF are explored, and deep neuron networks are employed and modified for dynamic uncertainty prediction. The proposed approach is validated through a case study involving an intelligent energy terminal with a 12.5 MVA SAF system and 12 MW capacity renewable generators in an electricity market with fluctuating prices. The findings of this research underscore the efficacy of the proposed optimization model in reducing operational costs and enhancing the utilization of localized renewable energy generation. By integrating four distinct dissatisfaction coefficients into the optimization framework, we demonstrate the model's adaptability and efficiency. The application of the optimization strategy delineated herein results in the SAF system's profitability oscillating between $111 and $416 across various time intervals, contingent upon the coefficient settings. Remarkably, an aggregate daily loss recovery amounting to $1,906.84 can be realized during the optimization period. Such outcomes not only signify considerable economic advantages but also contribute to grid stability and the diminution of renewable energy curtailment, thereby underscoring the dual benefits of economic efficiency and sustainability in energy management practices.
© 2024. The Author(s).