Estimation of Confidence Intervals for Distribution Network Reliability Indices SAIFI and SAIDI

The reliability indices of distribution networks, SAIDI and SAIFI, depend on the number of service interruptions per year, the number of customers affected by each interruption, and the restoration time of power supply. All three factors are stochastic and can vary widely even when the network structure and the technical condition of the electrical equipment remain unchanged. This leads to random year-to-year fluctuations in SAIDI and SAIFI values. Determining the range of these stochastic variations is necessary for comparing the indices and assessing their compliance with regulatory standards. Currently, formal methodologies for estimating confidence intervals for reliability indices are lacking both in Russia and internationally. This paper proposes an algorithm for calculating confidence intervals for SAIDI and SAIFI based on daily segmentation and numerical simulation. The application of the algorithm is demonstrated using the distribution networks of Moscow and the Moscow Region.