Operational optimization of energy systems, 25 years – established and promising use cases. (September 2018)

• Record Type: Journal Article
• Authors: Stange, Peter; Matthees, Anja; Sander, Thomas
• Is Part Of: Energy procedia. Volume 149(2018)
• Page Start: 15
• Page End: 24
• Publication Date: 2018-09
• Subjects: optimization of operation -- mixed-integer linear programming -- energy supply -- district heating -- demand forecast; Power resources -- Congresses; Power resources -- Periodicals; Power resources; Conference proceedings; Periodicals; 333.7905
• DOI: 10.1016/j.egypro.2018.08.165
• Languages: English
• ISSNs: 1876-6102

Abstract

Abstract: The operation of complex energy systems for the supply of heat and electricity leads to several questions regarding their optimal control, e.g. when to use which generator, when to load or unload energy storages or when to buy or sell energy. Usually it is a complex task to answer these questions with the aim of optimizing a specific objective and respecting all arising physical, technical and economic constraints. Since 25 years we are solving this problem for an energy provider of a medium-sized city with the aim of minimizing the operational costs. For this purpose, an own modelled mixed-integer linear optimization problem (MILP) has to be solved in association to the continuous operation of the energy system. The model includes but is not limited to several combined heat and power generators, heat accumulators, steam generators and auxiliary coolers. In this presentation we will give an outline about the wide range of given conditions that are successfully implemented for this application. Further we show our approach to generate realistic heat demand and power consumption forecasts which are both essential preconditions for obtaining reliable optimization results. In addition to the well – established MILP model in this specific use case we will outline some further promising applications of mathematical optimization in the context of energy systems. This includes the more precise modelling of energy storages, the computation of the optimal design of energy systems and the consideration of different or multiple targets in optimization. Moreover, we outline the problem of uncertain boundary conditions due to the growing amount of temporally hard to predict energy production and demand.