Optimisation and simulation of building energy systems

The Mixed Integer Linear Programming (MILP) technique is a useful tool for the optimisation of energy systems. However, the introduction of integers i n linear models results in a severe drawback because the ranging process is no longer available. Therefore, it is not possible to study what happens t o the solution if input data are changed. In this paper, we compare a MILP model of a building with a simulation model of an identical case. Both mode ls describe a building with a number of possible retrofits. Using the MILP technique, the optimal retrofit strategy is calculated, after which certain input data are changed. The optimisation results in the lowest possible Li fe-Cycle Cost (LCC) of the building, and the paper describes how much the L CC will change if the property owner chooses other solutions. An increase i n a particular data value may cause the LCC to increase or decrease. It may also be unchanged. Only a few data reduce the LCC when their values are in creased. (C) 2000 Elsevier Science Ltd. All rights reserved.