Session: 08-09-01: Multi-Energy Systems

Paper Number: 145230

145230 - A Two-Level Optimization Approach for the Synthesis, Design and Operation of Multi-Energy Systems Integrated With Energy Networks 

In recent years, there has been a growing consensus that decarbonization of the global energy system should not be addressed by individual sectors (electricity, heating, transportation, etc.), but requires a holistic approach aimed at integrating multiple carriers. Accordingly, researchers are increasingly focusing on multi-energy systems (MESs). An MES is an energy system of any spatial extent that involves multiple energy carriers (electricity, heat, fossil fuels, biomass, etc.) and delivers energy in various forms to end users. It typically covers the entire supply chain from producers to consumers, including energy conversion, storage, transportation, and end uses. The proper design and operation of an MES is one of the main challenges in implementing such systems and has fostered the development of dedicated mathematical optimization tools.

The supply chain of an MES is rarely considered in its entirety in the optimization problem, thus leading to solutions that may not be global optima. In particular, the number, type, size, and operation of energy conversion and storage plants are usually optimized by neglecting the energy networks that connect them to end users or, at most, by imposing specific network constraints a priori. On the other hand, the optimal design of energy networks is mainly addressed by deciding in advance on the design of the energy conversion plants feeding the system.

To fill this gap, the objective of this paper is to propose a novel approach for the synthesis, design, and operation optimization of an MES in its entirety. Synthesis refers to the layout and topology of the system, i.e., the placement of energy conversion and storage plants and the choice of energy network branches that connect the plants to each other and to end users. Design, on the other hand, refers to the sizing of the components (networks and plants). Finally, operation refers to the scheduling of plants and networks over time and the management of associated power flows. To handle this complex optimization problem, a two-level evolutionary algorithm has been developed, as the decomposition of the complete problem into two parts allows the computational load to be drastically reduced. The upper level deals with the synthesis problem, defined entirely by binary decision variables. The lower level deals with the design problem, defined by continuous decision variables, and the operation problem, defined by both continuous and binary decision variables. The lower level is set up as a Mixed Integer Linear Programming (MILP) problem and contains a reduced number of operation variables due to time series aggregation of energy sources and demand curves. The objective function consists in minimizing the life-cycle cost of the systems under consideration, while an increasingly stringent cap on carbon emissions can be imposed as a secondary objective, thus obtaining a Pareto front of cost-optimal solutions for different levels of decarbonization.

District energy systems of increasing spatial extent are considered to demonstrate the potential of the proposed method compared to a benchmark optimization approach, in which the complete problem is set up as a MILP without any decomposition. It turns out that the two-level evolutionary algorithm developed here can drastically reduce the computational time compared with the benchmark. In addition, it can handle problems of larger size, which are practically not solvable with the traditional MILP approach.

Presenting Author: Gianluca Carraro University of Padova

Presenting Author Biography: He received the Master degree in energy engineering at the University of Padova, Italy, in 2017, and the Ph.D. degree in Industrial Engineering (curriculum Energy Engineering) at the University of Padova in March 2021 with a Thesis entitled “Design, operation and control of power systems for low-to-medium temperature heat recovery: from organic Rankine cycle to supercritical CO2 systems”.
In 2021 he was a post-Doctoral researcher with a Research Grant on the project “Smart optimization of multi-energy systems, storages and interactions with energy networks". He is currently working at the University of Padova as an Assistant Professor (RTDa).
Research interests include two main topics: i) experimental evaluation, study of theoretical aspects and of the dynamic behaviour of energy production and recovery systems powered by renewable energy and waste heat; 2) design and off-design modeling, and optimization of the design and operation of multi-energy systems.

Authors:

Enrico Dal Cin University of Padova
Gianluca Carraro University of Padova
George Tsatsaronis Technische Universität Berlin