•  
  •  
 

Keywords

Distribution network, Network reconfiguration, Distributed generation, Active power losses, Hybrid algorithm

Document Type

Article

Abstract

Distribution networks face increasing pressure from rising loads and renewable energy integration, making the simultaneous optimization of reconfiguration and distributed generation placement a critical engineering challenge, and conventional single-objective methods are insufficient for simultaneously addressing power losses, voltage profile degradation, and the structural constraints of radial networks. This paper proposes a hybrid framework that combines Ant Colony Optimization with the Truncated Newton Constrained method to jointly solve network reconfiguration alongside the optimal placement, sizing, and type selection of renewable distributed generation—both solar PV and wind—within a single unified model. The approach relies on a four-part weighted objective function covering active power loss, voltage quality, economic cost, and reactive power loss, while a spanning-tree constraint keeps the network radial and the exact DistFlow model handles power flow calculations. Real weather data from Baghdad was used to drive the renewable generation models, rather than relying on idealized standard conditions. Tested on the IEEE 33-bus and 69-bus benchmark systems, the hybrid ACO-TNC algorithm delivered results that surpassed what had been reported in prior studies, with the 69-bus case in particular going beyond the best figures previously documented in the literature for reconfiguration-only scenarios. It also consistently outperformed Particle Swarm Optimization across every metric tested, and did so more efficiently on the larger network. These results demonstrate the strong capability of this hybrid approach in solving the reconfiguration problem while simultaneously determining optimal DG locations and sizes for renewable energy sources—handling the whole task as one integrated problem rather than splitting it into separate steps, and doing so without any manual intervention in setting the generation unit parameters.

DOI

10.30684/2412-0758.2411

First Page

118

Last Page

146

Download

Share

COinS