A hybrid approach of parallel tabu search and ordinal optimization to meter placement for improving topological observability

This paper presents an efficient method for improving topological observability in power system static state estimation. The proposed method makes use of a hybrid approach of parallel tabu search (PTS) and ordinal optimization (OO) to speed up computational time. Topological observability examines the relationship between a given set of measurements and network configurations. As the framework of security control, topological observability is important in checking whether a given set of measurements allows power system operator to carry out static state estimation. If a network is not topologically observable, they do not run the program of static state estimation. Mathematically, topological observability results in the existence of a full-ranked spanning tree in graph theory. In this paper, a reliable meter placement method is proposed to maximize topological observability in meter placement. The formulation may be expressed as a combinatorial optimization problem. To solve it, this paper presents a hybrid approach of PTS and OO to reduce the number of solution candidates in the subneighborhoods of PTS. The proposed method OO-PTS is applied to the IEEE 118-node system and compared with SA, GA and TS.