New technologies and emerging challenges are drastically changing how the energy needs of our society have to be met. By consequence, energy models have to adapt by taking into account such new aspects while aiding in decision making processes of the design of energy systems. In this work the problem of the design and operation of a multi-energy system is tackled by means of a mixed integer linear programming (MILP) formulation. Given the large size of the problem to be solved, a matheuristic approach based on constraint relaxations and variable fixing is proposed in order to not restrict the applicability to small cases. Two variable fixing policies are presented and performance analysis comparison on them has been done. Tests have been performed on small and realistic instances and results show the correctness of the approach and the quality of the heuristic proposed in term of solution quality and computational time.
A Matheuristic for the Design and Management of Multi-energy Systems / Bartolini, Andrea; Comodi, Gabriele; Marinelli, Fabrizio; Pizzuti, Andrea; Rosetti, Roberto. - STAMPA. - 1162:(2020), pp. 171-188. (Intervento presentato al convegno 8th International Conference on Operations Research and Enterprise Systems, ICORES 2019 tenutosi a cze nel 2019) [10.1007/978-3-030-37584-3_9].
A Matheuristic for the Design and Management of Multi-energy Systems
Bartolini Andrea;Comodi Gabriele;Marinelli Fabrizio;Pizzuti Andrea;Rosetti Roberto
2020-01-01
Abstract
New technologies and emerging challenges are drastically changing how the energy needs of our society have to be met. By consequence, energy models have to adapt by taking into account such new aspects while aiding in decision making processes of the design of energy systems. In this work the problem of the design and operation of a multi-energy system is tackled by means of a mixed integer linear programming (MILP) formulation. Given the large size of the problem to be solved, a matheuristic approach based on constraint relaxations and variable fixing is proposed in order to not restrict the applicability to small cases. Two variable fixing policies are presented and performance analysis comparison on them has been done. Tests have been performed on small and realistic instances and results show the correctness of the approach and the quality of the heuristic proposed in term of solution quality and computational time.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
