Due to increasing energy prices and the greenhouse effect, a more efficient energy supply is desirable, preferably based on renewable sources. To cope with the decrease of flexibility due to the introduction of renewables in production side of the supply chain, a more flexible consumer side is required. In literature, a lot of management methodologies are proposed to exploit the flexibility of consuming devices. In this paper an auction based realtime control is compared with an ILP based realtime control, both based on cost functions expressing the preferences of devices. Furthermore, the influence of different planning approaches on these strategies is studied. The approaches are rated based on their ability to work towards an objective. The results are related to characteristics from the control engineering perspective. Analyses and simulations show that planning improves the results, but a more direct realtime control approach (auction) is better capable of coping with prediction errors. However, the auction should slightly be altered to decrease communication and to be able to cope with (local) grid constraints.
|Publisher||IEEE Power & Energy Society|
|Conference||IEEE PowerTech 2015|
|Period||29/06/15 → 2/07/15|