A study on the profitability of Virtual Power Plants and their potential for compensation of imbalances

Abstract: In the current climate crisis, there is an increasing need for the integration of renewableenergy in higher penetration indexes, which is boosting innovation not only in the technology but also in new business models.  Virtual Power Plants (VPPs) are an aggregation model for generation that is said to be able to decrease imbalances from renewable generation while improving economic performance.  While the regulation in some countries still does not allow this type of activity there are also many others in which aggregation of demand and even VPPs are growing in numbers. This study aims to prove the previous statement and quantify the amount of imbalances that can be offset and the consequent penalties avoided for a VPP located  in  the  north  of  Portugal  composed  of  a  pumped-storage  hydro  plant  and  onshore wind generation.  In order to do so, two case studies have been compared:  a baseline case in which each unit is operated independently and another in which all the units are aggregated under the VPP model.  With this aim, a simplified bidding strategy has been simulated for both cases and three different error levels, to finally compare the results through four key performance indicators (KPIs):  increase in profit, increase revenues, decrease in imbalance and decrease in penalty cost.  The optimization problem was formulated as a mixed-integer linear programming (MILP) problem and it was carried out in two steps:  one for the day ahead session and a second for the intraday market.  It aims to diversify the generation port- folio of the hydro power plant and divide it among the available products:  energy sold in the day ahead market, capacity reserves for the secondary reserves, and energy sold as tertiary reserves.  It was decided to follow a deterministic approach, considering in the strategy a tree of scenarios and their associated probabilities.  In order to formulate this scenarios historical data was used, due to the high dependence between the market variables.  The results show that an average annual 16% decrease in energy deviations could be achieved which implies a 16.3% decrease in the penalty costs.  Moreover, it was also found that the combined oper- ation of the assets in the intraday market, together with the penalty reduction, would lead to a 1% increase in profit with a 0.1% decrease in revenues, concluding that a more detailed and data intensive model would be required to analyse the full advantage of the VPP modelwhen operating the assets together also during the day ahead session.