Rapid Calculation of Active Distribution Network Reliability Based on Dynamic Equipment Failure Rate

Abstract

The active distribution network has been fully developed because it can achieve efficient energy utilization through the effective control of distributed generation, electrical energy storage, and active loads. However, the research on the reliability of active distribution network operation still stays in the medium and long-term calculation, without considering the influence of dynamic equipment characteristics and active distribution network operation characteristics on reliability. In order to improve the comprehensiveness and calculation speed of the reliability calculation of an active distribution network, this paper proposes a fast calculation method for the reliability of the active distribution network based on dynamic equipment failure rates. Firstly, based on fuzzy-theory and data-driven methods, the cloud model and the Proportional Hazard Model are used to model the dynamic failure rates of exposed and enclosed types of equipment, respectively. This considers the impacts of the dynamic failure characteristics of the equipment on the reliability calculation and improves the comprehensiveness of the calculation. Then, the improved K-means algorithm is used to achieve a faster and more suitable scenario reduction and increase the computation rate. Finally, considering the island operation and contact line transfer under the active distribution network, simplifying the network by improving the minimum path method, the reliability calculation method of the active distribution network combining source–load uncertainty, switch fault information, and load transfer probability is further proposed. The cases show that the dynamic fault rate model is closer to an engineering reality and has generality. The improved K-means algorithm is faster and more accurate than the traditional algorithm. The final proposed fast reliability calculation method reduces the time by 72.76 times compared to the traditional method. It fully reflects the operational characteristics of the active distribution network and provides a reference for the optimal dispatch of the active distribution network.