Multiple DG Placements in Distribution System for Power Loss Reduction Using Hybrid PSO-CSA-GSS Algorithm

Dr. Radhakrishna K. R.

Abstract: The increasing penetration of Distributed Generation (DG) in modern radial distribution systems offers significant benefits in terms of real power loss reduction, voltage profile enhancement, and improved operational efficiency. Nevertheless, suboptimal DG placement and sizing can adversely affect system stability and network performance, particularly in large-scale distribution systems. To overcome these challenges, this paper presents a novel hybrid metaheuristic optimization framework that integrates Particle Swarm Optimization (PSO), Cuckoo Search Algorithm (CSA), and Golden Section Search (GSS) for optimal DG allocation. In the proposed approach, PSO is employed for global exploration to identify promising DG locations across the distribution network, while CSA enhances population diversity and mitigates premature convergence through L?vy flight-based exploration. Subsequently, GSS is applied as a deterministic local search technique to accurately fine-tune DG sizes within identified candidate buses, thereby improving convergence precision and solution robustness. The optimization objective is formulated to minimize total real power losses, subject to practical operational constraints including bus voltage limits, power balance equations, and DG capacity restrictions. The proposed PSO-CSA-GSS hybrid methodology is evaluated on the IEEE 118-bus radial distribution system, representing a large-scale and computationally challenging benchmark. Comprehensive simulation results demonstrate that the proposed method achieves superior performance compared with conventional algorithms and standalone metaheuristic techniques, yielding substantial reductions in real power losses, improved voltage stability margins, and faster convergence characteristics. The results confirm the scalability, robustness, and effectiveness of the proposed hybrid optimization strategy, making it a viable solution for planning and operation of future active distribution networks with high DG penetration.

Keywords: Distributed Generation Allocation, Radial Distribution Networks, Hybrid Metaheuristic Optimization, Power Loss Minimization, Voltage Stability Enhancement, Particle Swarm Optimization, Cuckoo Search Algorithm, Golden Section Search, Large-Scale Distribution Systems, IEEE 118-Bus System