An integrated, systematic data-driven supply-demand side management method for smart integrated energy systems. (15th November 2021)

• Record Type: Journal Article
• Authors: Su, Huai; Chi, Lixun; Zio, Enrico; Li, Zhenlin; Fan, Lin; Yang, Zhe; Liu, Zhe; Zhang, Jinjun
• Is Part Of: Energy. Volume 235(2021)
• Publication Date: 2021-11-15
• Subjects: Supply-demand side management -- Integrated energy system -- Machine learning -- Intelligent decision algorithm -- Multi-objective optimization -- Forecasting; Power resources -- Periodicals; Power (Mechanics) -- Periodicals; Energy consumption -- Periodicals; 333.7905
• DOI: 10.1016/j.energy.2021.121416
• Languages: English
• ISSNs: 0360-5442

Abstract

Abstract: Different energy systems become highly connected to provide better flexibility. However, this change poses new challenges for system management considering the diversity of demands, complexities of the energy networks, uncertainties, etc. This work develops a smart Supply-Demand Side Management method to overcome these challenges. The main objectives of this Supply-Demand Side Management framework are improving system efficiency and smoothing energy load, through flexible supply planning and dynamic pricing. Firstly, the customer response analysis method is proposed by combining the Deep Learning model and the economic model. Then, the energy network simulation model is used to coordinate the Supply-Demand Side Management strategies and the overall energy system capacity. A method is proposed to introduce the compressibility of natural gas in the management framework to offset the uncertain disturbances. Finally, a multi-objective decision method is developed to find the optimal strategy. The results of the application on a typical integrated energy system show that the proposed method can reduce the energy load fluctuation by 4%–8% under different planning horizons, and improve the system efficiency by reducing energy loss and increasing the profitability. The results also present a possibility of the development toward resilient Integrated Energy Systems by managing the buffer capacity of natural gas pipeline networks. Highlights: A management framework is proposed for the multi-concerns in energy systems. Deep-learning and economic model are combined to analyze customer behaviors. Natural gas compressibility management is employed for improving system resilience. Physical property of transportation network is introduced in the management method. An intelligent multi-objective decision method is developed.