Overview + Relevance
Energy storage and demand response (DR) are options for coping with the rising share of intermittent renewable generation. Most existing DR comes from heavy industry shifting a few large loads. Could small consumers also be motivated to shift their consumption? If many consumers are willing to shift small loads by a few minutes, and this issmartly coordinated, Swarm demand response (sDR) might have a lot of potential as a form of virtual storage.<
With sDR, electrical devices delay or interrupt operation according to user preferences and market price signals. A higher share of load could be shifted by a short time of 5 minutes, but some might be shifted for longer (e.g. 30 minutes) without any costs or inconveniences. After this initial delay, other members of the swarm take over, building up a chain of short load shifts. We ask whether this could be equivalent to conventional longer term storage. If the chain breaks when successors are not willing to shift their loads in turn, this could produce abrupt jumps in demand.
We analyse sDR in a model in which individual energy demands can be shifted (backwards or forwards) by varying amounts of time, building on Geske et al (2017).<strong> </strong>This sDR model is then integrated into a highly resolved (5 Minutes) total cost minimizing electricity system model that minimizes generators’ operation and investment costs. The model is calibrated to match German load data and solved over a looping 24-hour period to give annualized equivalent results. By the time of the conference, we expect to use longer time series and more disaggregated demand data.
We find that the virtual storage provided by sDR storage is almost equivalent to conventional storage in terms of system costs. Prices are able (in principle) to find a chain of demand shifters for as long as they are needed, so that there should not be sudden demand spikes as the chain breaks. This suggests that if sDR were adopted, it could be modelled as if it were additional energy storage, but cheaper.
Demand response should not be modelled without considering the subsequent pick-up in load. We show that this need not negate the value of small-scale demand response, as long as the swarm is organised to allow a chain of small shifts over the period of need. <
J. Geske, R. Green, Q. Chen and Y. Wang, “Smart demand side management: Storing energy or storing consumption — It is not the same!,” 2017 14th International Conference on the European Energy Market (EEM), Dresden, 2017, pp. 1-7.
Categories: Academic PapersGeske-Green-Swarm-demand-response.pptx 3.93 MBGeske-Green-Swarm-demand-resoponse.pdf 609.58 KB