Network reconfiguration and control for loss reduction using genetic algorithm

Power distribution networks typically have tie and sectionalizing switches whose states determine the topological configuration of the network. In this work a feeder reconfiguration algorithm is presented for the purpose of power loss reduction in distribution networks. The methodology developed a combined optimization technique. Also capacitor placement employing Genetic Algorithm is presented in achieving the proper arrangement of capacitor in a distribution network. The proposed algorithm has been implemented in technical MATLAB package and tested with two examples namely, 18- and 49-bus systems. All the data for 18-bus system test are taken from previous work, and all the data for 49-bus system test are taken from an existing Iraqi distribution network. The network reconfiguration problem is formulated as single objective optimization problem with equality and inequality constraints. The proposed solution to this problem is based on a general combinatorial optimization algorithm known as Genetic Algorithm, and the load flow equations in distribution network. Two methods of load flow solution with different accuracy are employed, i.e., simplified method, which uses the approximation of P and Q at the start of the implemented program, and fast-decoupled method, which gives an exact solution. Standard size capacitors according to IEEE are used in the capacitor placement tests using Genetic Algorithm for 18-bus system where 7 capacitors are needed to ensure minimal total power losses to be obtained. While in the 49-bus system the existing 7 capacitors are rearrange using Genetic Algorithm in obtaining minimal total power losses. It is found that a better capacitor placement is obtained for the 49-bus system compared to its existing network in Iraq. Note that the 18-bus system is originally without any capacitor. Two selection methods that are used in Genetic Algorithm are the roulette wheel and tournament selections. Tests have shown that the differences are demonstrated only in execution time to arrive the best fitness function of the Genetic Algorithm. Whereas the results obtained from both selection methods are similar to each other. Tests results show that Genetic Algorithm is a suitable algorithm as it is an optimization technique with, high accuracy, and it avoids local minimum by searching in several regions to arrive to the global optimum solution. Thus, the outcome of this study shows that an efficient technique in solving network reconfiguration problem for power loss reduction in the distribution network has been successfully found