Multi-Horizon Planning of Multi-Energy Systems

TL;DR

This paper introduces two multi-horizon planning methods for multi-energy systems, incorporating technology learning curves and demonstrating their application on a German test system to optimize the transition towards renewable energy.

Contribution

It presents a novel approach to include technology-dependent learning cost curves in multi-horizon MES planning and improves solution efficiency with Benders decomposition.

Findings

The new method effectively models technology learning in MES planning.

Benders decomposition accelerates solving the mixed-integer linear programming problem.

Application on a German system illustrates practical MES expansion pathways.

Abstract

In order to reach EU's goal of zero emissions in 2050, the energy system will go through a significant transition over the next decades. To substitute fossil energy carriers, renewable energy sources will be mainly integrated in the power system. Thereby, sector coupling will play a major role by making flexibility from other sectors such as heat or transport accessible to the power system. Planning the cost optimal transition requires a whole-system view over multiple horizons and across all sectors. This imposes the need for multi-energy system (MES) models coupled with multi-horizon investment models. This paper presents two multi-horizon planning approaches to determine the cost optimal pathway of a MES. As a major contribution, we propose a new method to incorporate technology-dependent learning cost curves in the planning problem and show that the resulting mixed-integer linear programming problem can be solved faster with a Benders decomposition technique as compared to a closed optimization. As a further contribution, we demonstrate the usefulness of our approach by showing the MES expansion pathway for a small German test system.