Abstract:

In the large-scale multi-moment distribution network reconfiguration (DNR) problem, a large number of switches to be optimized seriously reduces the solution efficiency of distribution network reconfiguration. To address this problem, a distributional robust distribution network reconfiguration model based on a compressed switch candidate set was proposed, which was divided into two stages. The first stage took the minimization of active network loss of the system as the objective function and used the optimal matching loop flow method to compress the switch candidate set; the second stage took the minimization of the sum of the power purchase cost and the switch action cost as the objective function, constructed the chance constraints on the capacity limits of the power point, and adopted a Wasserstein ball-based distributional robust method to deal with the uncertainty of distributed generation. It transformed the model into a mixed-integer second-order conic planning problem by deterministically transforming the worst-case expectation and chance constraints in the objective function by using a dual transformation method. Finally, numerical experiments were conducted on the 33 node and Liaoning Panjin 45 node systems, which proved that the model proposed in this paper could effectively improve the computational efficiency, and the decision maker could adjust the economy and conservatism of the model by changing the number of samples and the confidence level, compared with the robust model and the stochastic planning model.

Key words: distribution network reconfiguration, optimal matching loop flow, distributional robust optimization, Wasserstein ball, chance constraint