Optimal reconfiguration of distribution systems with representation of uncertainties through interval analysis

• An interval approach is proposed for radial distribution networks reconfiguration.
• Uncertainties over load demand and wind distributed generation are considered.
• An artificial immune optimization algorithm is used for reconfiguration.
• The algorithm seeks to minimize interval loss while maintaining a radial operation.
• A method for comparing intervals based on the midpoint and radius is applied.

The present work presents an approach for optimal reconfiguration of electrical distribution systems (EDS) to minimize energy losses considering uncertainties in the load demand and in the wind based distributed generation (DG). The optimization algorithm applied to solve the reconfiguration problem is based on the bio-inspired metaheuristic Artificial Immune Systems (AIS). An interval power flow model is used to obtain an interval energy loss from the representation of the uncertainties. The interval loss is used to guide the AIS algorithm through the search space. Network and operational constraints as the radiality and connectivity of the network as well as different load levels are considered. Well-known test systems are used to assess the impact of the uncertainties representation in the reconfiguration problem.