Due to the high cost of electric power, achieving the optimal reduction of active (real) losses during power transmission is very important. The reduction of these losses is crucial in Peru because the values of leakage losses in coastal lines are very high and produce large transverse active losses. In this work it has been investigated the reduction of active power losses through the optimal dispatch of reactive power and the optimal control of voltages using the Particle Swarm Optimization (PSO) technique. This work has been implemented using Matlab, and MATPOWER has been used as an engine to solve the load flow.The efficiency of the process is confirmed using the 14-bar IEEE test power system achieving a loss reduction of 8.38%. The loss of active power converges after 100 iterations from 13,393 to 12.27 MW.During the investigation it was confirmed that the PSO can solve with great efficiency problems where the objective function is very complex, so it can be applied to the problem of reducing losses in the national network and it will also be able to solve major problems such as reactive power planning and planning of the expansion of the transmission network.