Optimal placement of distributed generation sources in power distribution systems using the combination of differential evolution algorithm and search mechanism

In this study, a combination of differential evolution algorithm and a search mechanism with infinite mean and variance which can effectively explore the search space was employed, instead of using random walks to simultaneously specify the size and location of distributed generation units and capacitors installed in a radial distribution network. Thus, quasi-opposition-based learning was used to generate the initial population of the combined algorithm. Objective functions have been formulated such that power loss and loss supply costs are minimized, also the voltage profile improves, and finally, a stable increase in the voltage is obtained. The results of stimulation with IEEE ۳۳-Bus standard distribution system compare the efficacy of the suggested method to the other algorithms. The convergence rate and efficiency of this method will be analyzed in comparison with previous methods. Loss reduction and maintenance of the buses’ voltage within the proper ranges are of significant importance.