Practical meter placement algorithms for improving state estimation performance.

摘要

In the early 1990s, Congress and the Federal Energy Regulatory Commission (FERC), which oversees the transmission and wholesale of electricity, decided to deregulate the electric power industry to bring more competition to the market. The traditional vertically integrated structure of the utilities was broken into separate entities of generation, transmission, and distribution. These recent changes resulted in a competitive market where a reliable, secure, and cost efficient operation has become more important. In an effort to improve the performance of their energy management functions, the utilities are upgrading their measurement sets by installing new meters, such as phasor measurement units.;In this thesis, methods to improve the functionality of state estimation by observability analysis and bad data detection are discussed. Identification of unobservable branches and detection of observable islands are studied as part of the observability problem. Critical measurement elimination by placing additional meters is investigated to provide the system with bad data detection capability.;Two measurement placement algorithms are developed to upgrade the measurement configuration and to eliminate critical measurements. For each algorithm, two types of measurements, either real power injections or phasor measurement units (PMUs), can be placed as candidate measurements. The simulations of IEEE 14 and 30 bus systems illustrate that the algorithms are effective in finding an optimal set of additional measurements.;The methods developed in this thesis are not only used to install new meters, but they are also used to detect the weak spots in a power system. These algorithms work for both designing a measurement configuration for a new system or upgrading the measurement design of an existing system.