• Reserve Margin
• security assessment
• Stochastic Optimization

In the last years the increasing penetration of renewable energy in electric power sectors has introduced new problems for the power system reliability that have to be faced using market-based mechanisms. Reliability standards in power systems are traditionally established as a series of technical requirements to be achieved during planning and operation. In general, reliability requirements are satisfied providing a group of services, called ancillary services, which are necessary to safe the reliability of the system and to guarantee the production and delivery of electric power. Ancillary services include coordinated system operation, i.e. frequency regulation, energy balance and generation reserves.
This report focuses on the short-term tuning of the operating reserve, based on a stochastic optimization; the considered market model was inspired to the Italian one.

Electricity is produced and delivered on real time and for now there is still no convenient method to readily store a significant amount of it. This makes necessary to maintain a continuous and almost instantaneous balance between production and consumption. A way to ensure energy balance is to allocate adequate generating reserve margins beyond the forecasted load curve; so that the system can deal with unexpected differences between supply and demand. Generating margins are ensured by providing stand-by or spinning plant capacity and they represent reserves of generation that can be rapidly used in case of necessity.
The Transmission System Operator estimates a reasonable amount of capacity to be reserved and kept available, so that usual contingencies will not cause severe blackouts. Usually analytical methods are used, but the final decision regarding reserve levels depends on the operator’s judgment and experience of what is the acceptable risk of system failure.
This work proposes and discusses a stochastic method to size the generating reserve, by evaluating the real cost of the reserve procurement and also the expected value of the “day of delivery” costs related to the real time balancing and possible energy not supplied. The expected value of this cost is obtained by modelling the future behavior of the system with a finite number of possible scenarios, associated to their probabilities.