Self-Tuning Controller for Generator Excitation Control

Abstract

It has long been recognized that the generator excitation control is an effective and low cost means for improving the power system stability. Recent developments in control theory and digital technology suggest an extension of the results obtainable through analog control to derive potential improvements through the use of digital control. There has been reluctance to apply real time digital control in electric power engineering, perhaps this is due to lack of experience and confidence in this technology. In addition to advances in technology, motivation for investigating the control of a generator at the end of a long transmission line comes from the common occurrence of generation being remotely located. Such is the case, for example, at Liu Jia Xia in the People's Republic of China where a 330 KV, 332 mile single circuit line connects a 400 MW station with the main grid. It is reasonable to attempt to use excitation control to permit line loading to the transmission limit. Much work has been done with many interesting results by applying linear optimal feedback control theory to the excitation control. Most of the results are based on the assumption that an explicit deterministic mathematical model of the system to be controlled is available. The system model is then linearized around some fixed operating point with the system parameters assumed constant. By numerous off-line simulation studies, suitable weighting factors involved in an index of quality can thus be found by trial and error.