Multi-energy systems have proven to be effective tools in addressing the decarbonization needs of the energy system, the increasing penetration of non-programmable energy sources, and the challenges of decentralization and independence in the energy market. Due to their complex structure, which integrates multiple technologies and different energy vectors, the planning and management of such systems greatly benefit from the support of modeling tools.

In general, traditional paradigms of multi-energy planning, and their associated modelling frameworks, focus on the minimization of a single monetary objective. Even when multiple objectives are considered, they are associated with monetization rates to return to a single-objective problem.

The work presented in this report aims to overcome these limitations and develop modeling frameworks that ensure the exploration of various configurations of multi-energy systems according to non-comparable objectives, and extract trade-off solutions through optimization algorithms.

Three different methodologies are presented and tested, integrating the single-objective configuration model CALLIOPE with multi-objective algorithms for exploring the decision space:

1) the exhaustive method,

2) multi-objective evolutionary optimization on weights, and

3) multi-objective evolutionary optimization of system configuration.

These methodologies are tested on a synthetic case study and evaluated for their ability to thoroughly explore the solution space and the input data and computational demands.

The results obtained indicates that all three methods allow for the extraction of multiple optimal configurations of the multi-energy system, adopting considerably different technology combinations depending on the relative importance of the objectives. Multi-objective evolutionary optimization on weights proves to be the most efficient methodology with the most comprehensive results, although it requires the user to define a priori weight ranges for each objective.

The methodologies highlighted in this paper represent a significant step forward in the search for modeling solutions for optimal planning and management of multi-energy systems, which manage to capture the intrinsic complexity of the problems considered, to support the search for truly integrated, efficient, participatory, and sustainable solutions.

The Report is available on the Italian site