Optimal design of multi-energy grid systems Finance a by: Hikmet Eliyev Optimal design of multi-energy systems with seasonal storage

Introduction

Recently, the energy sector is experiencing a wave of tremendous transformations: the need to reduce environmental impact has led to the introduction of transformation and storage technologies based on renewable energy sources [1]. In this context, multi-energy systems (MES) represent a new paradigm that uses the interaction between different energy carriers (for example, electricity and heat) at the design and operation stage, which allows to improve the technical, economic and environmental characteristics of the system [2]. , Within this framework, seasonal storage systems have recently attracted great attention because of their ability to compensate for the seasonal instability of renewable energy sources [3]. However, compensating for fluctuations in renewable energy on a seasonal scale is a particularly difficult task: on the one hand, several systems, such as a hydrogen storage and a large heat reserve, can compensate for seasonal fluctuations in the production of renewable energy; on the other hand, optimal design and operation are complicated by a large number of decision variables due to the required length and resolution of the time horizon.

Several papers provide comprehensive reviews of typical language and computer tools used to study MES and their integration with renewable energy sources and storage technologies. For example, Alarcon-Rodriguez et al. focused on multipurpose planning of distributed energy resources [4]; Connolly et al. presented a review of computer tools implemented to analyze the integration of renewable energy into various energy systems [5], while Keirstead et al. [6] and Allegrini et al. [7] focused on urban energy system models; Mancarella presented a review of concepts and models for planning and analysis of multi-energy systems [2]. When storage technologies are available, the optimal design of MES is greatly complicated by the need to consider the system at the design stage in order to accurately use storage systems. Although several non-linear approaches have been proposed, for example, Elsido et al. [8], Mixed Integer Linear Programming (MILP) has been especially approved as the basis of optimization for the design and operation of MES, as it captures system features with reasonable computational complexity. The problem of choosing the best technology and adherence to a unit using the MILP composition has been thoroughly investigated in the past. For example, Marnay et al. The case of commercial construction of a microgrid with the accumulation of heat and electricity is presented [9]; Hawks and Leach expanded the study to include hospital and residential buildings [10]. Later, Angrisani et al. investigated the energy, economic, and environmental characteristics of microtrigeneration systems [11]. Fazlollahi et al. presented methods for the multipurpose design of complex energy systems [12], and Ahmadi et al. presented thermodynamic modeling and multi-purpose optimization of the power system for the simultaneous production of electricity, heating, cooling and hot water [13]. Taking into account that these works were mainly focused on small but centralized systems (i.e., one hub for different end users), a number of studies also studied the distribution of energy between different nodes of decentralized energy systems (i.e. several hubs for different end users) ) For example, Genon et al. presented an environmental assessment of small district heating systems [14], while Weber and Shah optimized the structure of the heating main in addition to selecting technology and block distribution [15]. In addition, various case studies have been proposed using such tools: Mehleri ​​et al. investigated the optimal design and operation of various areas in the Greek energy sector [16], [17]; Omu et al. focused on possible measures to reduce the carbon footprint of the UK energy sector [18], and Bracco et al. presented the design and operation of a microgrid built on a campus in Savona, Italy [19]. Finally, interest in multi-energy systems at various levels (from domestic to national) has sparked the development of commercial tools for developing MILP. These include EnergyPlan, an optimization tool developed at Aalborg University that simulates the operation of national energy systems [20], and DER-CAM, developed by Lawrence Berkeley National Laboratory [21]. Recent improvements to the DER-CAM tool include building upgrades [22] and detailed models of intelligent heat storage systems [23].

In these works, the remarkable complexity of the optimization problem required significant simplifications of the model. In particular, in all these past studies, a one-year time interval based on settlement days was considered.