Abstract

ggregated demand response for smart grid services is a growing field of interest especially for market participation. A growing trend in research is to utilize aggregated demand response for multiple smart grid services. This could be multi-market trading (day ahead, intraday and imbalance) or a combination of market and ancillary services such offering reserve power or congestion management. However, there is a potential conflict of interest when offering the same resource for simultaneous services. This work investigates the impact, both from a monetary and network stability perspective, of applying a predictive control trading strategy which actively offers aggregated flexibility of electric vehicles on both the German EPEX day ahead and intraday markets. An artificial neural network was used to forecast the available ramp up and down capacity of a Virtual Power Plant (VPP) of 1000 electric vehicles. Using this information, the available flexibility is traded to ramp up in one quarter and down in the next depending on the price delta seen in the intraday market. A number of simulation runs are done, each with different levels of flexibility traded. In every run, one week of realistic VPP behaviour is simulated. The total earnings on the intraday market are calculated as well as imbalance cost and imbalance power generated over this period. It was seen that with an increased offer of available flexibility, there was an increase in both total revenues up to ∼4200 euros for one week of trade as well as imbalance generated, ∼1.6 MWh. Therefore, there is a clear need for effective regulation that limits imbalance without losing the future grid-stabilising effects of the flexibility aggregator.