Optimal location and size of distributed generation to reduce power losses and improve voltage profiles using differential evolution optimization method

There are numerous advantages that can be obtained when Distributed Generation (DG) is integrated into the distribution systems. These advantages include improving the voltage profiles and reducing the power losses of the distribution system. Such advantages can be accomplished and confirmed if the DG units are optimally located and sized in the distribution systems. In fact, there are several algorithms used for optimizing the size and finding the best location to install DG units in the power system. Some existing algorithms need to be improved while others, need to add a new parameter for improving the performance of optimization methods and making it more effective and efficient. This research aimed to reduce both total real and reactive power losses and improve voltage profiles of the distribution system by proposing a differential evolution (DE) algorithm to optimize DG size and location by taking into consideration different types of DG units. The multi-objective function, which represents the summation of product five indices by corresponding weights, was utilized to identify the candidate buses to reduce the search space of the algorithm. The suggested algorithm of DE was tested using IEEE 30 bus test system and IEEE 57 bus test system taking into consideration three types of DG units. The results obtained by using the DE method were compared with those obtained by genetic algorithm (GA) method. It was observed that the DE method gives a better result in terms of improving the voltage profile and reducing real and reactive power losses compared to GA method. In the IEEE 30 bus test system, the percentages of reduction in the real power losses obtained by the DE method were 20.58 %, 22.08 % and 20.40 % compared with 15.03 %, 16.81 % and 19.66 % obtained by the GA method for type 1, type 2 and type 3 respectively, while the percentages of reduction in the reactive power losses were 23.28 %, 19.20 % and 23.14 % by using the DE method compared with 17.01 %, 14.41 % and 22.49 % by using the GA method for type 1, type 2 and type 3 respectively. The voltage profile of the system also improved after optimizing DG sizes and locations using the DE method. However, in the IEEE 57 bus test system, the percentages of reduction in the real power losses were 17.08 %, 21.53 % and 20.20 % by DE method compared with 16.72 %, 15.91 % and 16.86 % by the GA method for type 1, type 2 and type 3, while the percentages of reduction in the reactive power loss were 9.64 %, 11.67 % and 9.42 % by DE method compared with 9.32 %, 10.81 % and 9.08 % by the GA method for three types of DG units respectively