With the increasing penetration of distributed renewable generation and flexible loads, topology adjustment has been used as a proactive risk-mitigation measure. However, loop-closing operations can readily lead to steady-state overloads and bus-voltage limit violations. To address these issues, an active distribution network reconfiguration method is proposed that accounts for security requirements and source–load uncertainty. First, a probabilistic model considering the correlations among wind power, photovoltaic generation, and loads is established. By combining an improved point estimation method with the Cornish–Fisher expansion, the probabilistic characteristics of nodal voltages and loop-closing currents are efficiently calculated. On this basis, chance constraints are introduced to reformulate the traditional deterministic loop-closing feasibility criterion as probabilistic security constraints. Then, a bi-level solution framework is established. The upper level generates candidate network topologies using the proposed firefly-perturbation chaotic SA-PSO algorithm, while the lower level performs loop-closing current security checks and eliminates infeasible loops by adding cutting constraints. Finally, case studies on a modified IEEE 123-bus distribution system show that, compared to pre-reconfiguration, the proposed method reduces active power loss from 1817 kW to 512 kW, voltage deviation from 0.39 to 0.18, and the load-balancing index from 64.41 to 27.84. In addition, the nodal voltage range is improved from 0.9213 to 1.0797 p.u. before reconfiguration to 0.9743–1.0215 p.u. after reconfiguration. These results verify that the proposed method can effectively enhance operating economy, voltage quality, and operational security under source–load uncertainty.
Hui Li (Wed,) studied this question.