An Approximate Solution Algorithm for Load Balancing of Distribution Systems

Abstract

This paper proposes a new algorithm to obtain an approximate optimal solution for the load balancing of large-scale radial distribution system. Since the problem is formulated as a combinatorial optimization problem because of the discreteness of load section connection, it is difficult to solve a large-scale combinatorial optimization problem accurately within the reasonable computational time. So for, in order to find an approximate optimal solution quickly, the authors have published a solution method based on the network flow incremental algorithm. Although the incremental algorithm can find the optimal solution to the problem, the result is not always radial (feasible). So as to overcome this deficiency, in this paper, the authors propose a new algorithm that can find the load balanced radial distribution feeder configuration. The proposed algorithm picks up load by a unit of a section load. Therefore, the problem is elementarily a combinatorial optimization problem. However, to avoid the permutation, the algorithm employs a heuristic algorithm based on the mathematical concept of the optimality of the incremental algorithm. In picking up section load, the algorithm can take into account not only the current load balance but also the expected load which feeders should pick up afterward. To realize this algorithm, a combination of labels is introduced to each node. Through numerical examples, the validity of the proposed method is examined.