Spinning reserve quantification considering confidence levels of forecast in systems with high wind and solar power penetration

The increasing penetration of renewable energy sources, featured by intermittent outputs, emphasizes the investigation of the potential effects on frequency deviation in the normal operation and solving the associated problems. This paper proposes a method for assessing the impacts of forecast errors in wind and solar generation outputs on a unit commitment problem. An iterative process is used across different spinning reserve requirements until the hourly average and intrahour variation forecast errors in demand and renewable energy generation are fully compensated at the lowest cost. The process, containing time series simulations, is applied to various confidence levels of the forecasted renewable energy generation so that violations of the frequency limit will not occur in the normal operation. The effectiveness and applicability of the proposed method are well illustrated through a case study on the 2030 power system of the island of Hokkaido in Japan.