Voltage instability

Abstract

The state of an electric power system may be classified as either stable or unstable. The borderline of stability is at any condition for which a slight change in an unfavourable direction of any pertinent quantity will cause instability. The nature of the unstable response will depend upon the characteristics of the system and the operating condition. Although in the general case every system variable is involved, in particular cases the instability may be manifested primarily by the loss of synchronism of one or more generating units (so-called ‘angle instability’) or by the uncontrollable decay of system voltage over a significant portion of the network (so-called ‘voltage instability’). Since the principal network characteristics of importance and the principal techniques and tools of analysis are in some degree different for the two cases, and since the possibility of ‘voltage instability’ has become of increasing concern with the development of widespread networks, ‘voltage instability’ has been singled out as a separate phenomenon, distinguished from, but closely related to, ‘angle instability’. This close relationship is well illustrated by the common practice of improving stability by voltage support, even though the instability in prospect is ‘angle instability’. Moreover, voltage instability has become fashionable both as a subject for study and as a characterization of many recent cases of system collapse, in some cases even though the basis for such characterization has not been immediately evident. This paper has been written: to define voltage stability, voltage collapse, and voltage security more narrowly so they are not all-embracing phenomena. And also to show the system conditions leading to voltage instability and the system behaviour at its occurrence in the simplest and most direct way. To illustrate the system behaviour a very simple system consisting of a single voltage power source and transfer reactance to a load having reactive-power compensation for voltage support is studied. This system can be shown to exhibit all of the phenomena related to voltage stability and voltage collapse that have been observed in real systems. The system behaviour from the standpoint of voltage stability has been studied and presented in terms of the sensitivity of voltage and load power to changes in physical load (load admittance), system parameters, reactive power compensation, etc. ( dV/ dG L, dP L/ dG L, dV/ dB c, dP L/ d B c , etc.) rather than in terms of the commonly-used P-V, Q-V curves, since these sensitivities are more closely related to what the system operator will see. During voltage collapse, the effects of generator voltage changes, line loss, transformer taps, load loss, etc. can be exhibited by appropriate changes in system voltage, system reactance, assumed voltage ratio (or assumed P-V, Q-V relationships), load admittance, etc. The effects of load P-V sensitivity and types of Q compensation (controlled or fixed) on the estimation of voltage security are also shown. It is demonstrated that from the standpoint of voltage stability reactive power compensation at or very close to the load is very important and that to get the maximum power-transfer controlled compensation (e.g. SVC or Syn. C) is essential. In addition, the quantitative effects of load power factor, system reactance, load P-V, and Q-V characteristics, and required reactive Q are made evident.