Heuristics for the Power Edge Set Problem

Abstract

A power grid is an interconnected network to distribute electricity generated from power stations to consumers. If the supply and the demand of the power grid is not matched, there will be power outages. To avoid this situation, the power grid should be monitored. A device called the Phasor Measurement Unit (PMU) is used to monitor the power grid. The PMUs are expensive devices. Hence, optimizing the number of PMUs required to completely monitor the power grid is a well-known problem. A PMU installed on a power station can monitor the PMU installed station and one or more adjacent power stations. A multi-channel PMU can monitor the PMU installed station and more than one adjacent power stations. Whereas, a single-channel PMU can monitor the PMU installed station and exactly one adjacent power station. The multi-channel PMU placement problem formulated as a graph problem known as the Power Dominating Set (pds) problem. The single-channel PMU placement problem is known as the Power Edge Set (pes) problem. Both the PDS problem and the PES problem are NP-hard. The PDS problem is well studied, many heuristics were known. The PES problem being a relatively new problem, there was attention to study the problem from the theoretical point of view. So far, no attempt has been made to solve the problem on real-world data. In this paper, we propose heuristics for the PES problem. The proposed heuristics are implemented and tested on some realworld data sets.