Optimal placement and sizing of distributed generation units in distribution networks using an enhanced particle swarm optimization framework

Introduction. Optimal planning of distributed generation (DG) units is a critical research topic due to the growing integration of renewable energy and the need to enhance distribution network performance. Classical optimization methods often struggle with the nonlinear, nonconvex, and highly coupled nature of DG allocation problems. Problem. The IEEE 33-bus distribution network experiences significant voltage drops and high active and reactive power losses under normal operating conditions. Determining the optimal placement and sizing of DG units is a complex problem involving multiple interacting variables and operational constraints. Goal. This study aims to improve technical performance by minimizing total active power losses and voltage deviation while ensuring voltage stability and network reliability. Methodology. The particle swarm optimization (PSO) algorithm is enhanced using the Dehghani method (DM) – a population-based modification framework allowing all individuals, including the worst member, to contribute in improving the best solution. The improved PSO-DM algorithm is applied to the IEEE 33 bus system under four cases: the base case without DG and scenarios with 2, 3 and 4 DG units. The objective function includes active power loss minimization and total voltage deviation. Results. The 4-DG configuration significantly improves system performance: active power losses decrease from 210.67 kW to 53.9 kW (74.4 % reduction), reactive losses drop from 142.84 kVAr to 38.42 kVAr (73.1 % reduction), the minimum bus voltage rises from 0.9037 to 0.9741 p.u. and total voltage deviation decreases from 1.8037 p.u. to 0.5129 p.u. (71.6 % improvement). These results demonstrate that PSO-DM effectively balances exploration and exploitation, yielding superior DG allocation solutions. Scientific novelty. Integrating DM into PSO introduces a cooperative solution-refinement mechanism that enhances convergence speed and search accuracy. Practical value. The PSO-DM framework provides a reliable and computationally efficient tool for DG planning in modern smart distribution networks. References 22, tables 1, figures 3.