A hybrid approach for optimization of electric power distributed networks with photovoltaic sources

This paper presents a hybrid approach, which associates a mathematical programming technique with a meta-heuristic, to optimize the electricity distribution network with penetration of photovoltaic (PV) generation. This optimization considers the distribution network reconfiguration (DNR) and the adjustment of the power factor of photovoltaic (PV) inverters about the reactive power, controlling the voltage profile and minimizing network losses. The methodology uses particle swarm optimization (PSO) for DNR and an optimal power flow (OPF) to adjust the value of the PV inverters and maintenance of network operating limits, this being a great difference from other methodologies. The objective of the hybrid approach is to obtain the best result for the problem, taking advantage of the best characteristics of both techniques. The hybrid approach was compared with the PSO, highlighting its superiority in terms of the solutions found. The tests were performed using the AMPL Modeling Language in the 33-node system with the addition of PV generation. The approach considers the variations in the powers generated by the PV systems and the variations in the loads over 24 h. The results indicate that the combination of PV inverter control and DNR can reduce the active power losses by more than 37% in 24 h.