Comprehensive Reactive Power Optimization Dispatch for Active Distribution Networks Under Uncertain Conditions

With a large number of distributed power sources incorporating the distribution network system, the distribution network reactive power optimization problems become more complex and diverse. In this paper, a reactive power optimization dispatch for the active distribution grid under source-load uncertainty conditions are explored. Monte Carlo-like and K-means clustering methods are applied to generate typical source-load probabilistic scenarios, and a fuzzy chance constraints and an artificial bee colony-improved particle swarm algorithm are used to solve the optimal reactive power problem. Finally, the effectiveness of the proposed method is verified by the IEEE 33-nodes system incorporating distributed generations.