The integration of distributed photovoltaic (PV) generation in distribution networks plays a key role in the energy transition, yet its impact on technical losses and voltage profiles depends strongly on optimal siting and sizing. This study proposes a rigorous methodology that combines node-specific spatiotemporal solar data, hourly power-flow simulations and evolutionary optimization to determine the optimal PV placement in radial distribution systems. PV generation is modeled using PVGIS, enabling distinct hourly profiles for each network node. The IEEE 33-bus system is simulated in OpenDSS over a representative 24-hour period, and daily energy loss minimization is formulated as the objective function. Three metaheuristic algorithms - Genetic Algorithm (GA), Particle Swarm Optimization (PSO) and Differential Evolution (DE) - are applied and benchmarked against an exhaustive brute-force search to ensure global optimality and robustness. Results demonstrate that node-specific PV modeling significantly improves accuracy and that the optimal placement yields substantial reductions in daily technical losses.
Cziker et al. (Thu,) studied this question.