Limits and potentials of Mixed Integer Linear Programming methods for optimization of polygeneration energy systems

Abstract

Abstract The simultaneous production of different energy vectors from hybrid polygeneration plants is a promising way to increase energy efficiency and facilitate the development of distributed energy systems. The inherent complexity of polygeneration energy systems makes their economic, environmental and energy performance highly dependent on system synthesis, equipment selection and capacity, and operational strategy. Mixed Integer Linear Programming (MILP) is the state of the art approach to tackle the optimization problem of polygeneration systems. The guarantee of finding global optimality in linear problems and the effectiveness of available commercial solvers make MILP very attractive and widely used in optimization problems of polygeneration systems. Nevertheless, several drawbacks affect the MILP formulation, such as: the impossibility of taking into account nonlinear effects; the necessity of considering all the time periods at once; the risk of high-dimensionality of the problem. To tackle these limitations, several techniques have been developed, such as: piecewise linearization methods; rolling horizon approaches; dimensionality reduction by means of energy demands clustering algorithms. In this paper, limits and potentials of MILP methods for the optimization problem of polygeneration energy systems are reviewed and discussed.