The objective of this research project is to investigate and devise a methodology for identifying areas in power systems that are prone to voltage instability under particular operating conditions and contingencies. These areas, which are prone to instability due to their lack of reactive power reserves, are referred to as Voltage Control Areas (VCAs). Once VCAs are identified, methods of determining their adequate reactive power reserve requirements to ensure secure system operation under all conditions. The Electric Power Research Institute (EPRI) contracted Powertech Labs Inc. (PLI) to carry out this research project. PLI conducted a rather exhaustive literature survey which revealed that the existing methods for the identification of VCA have had only a limited success in application because they have, in general, not produced results for practical systems. This, in general, is due to the fact-that voltage security problem is highly nonlinear and VCAs may also change in shape and size for different system conditions and contingencies. To deal with these issues, a more practical approach was adopted by this project to clearly establish the VCAs for a given system under all system conditions. The approach is based on a PV Curve method combined with Modal Analysis. The general approach is as follows: - define a system operating space based on a wide range of system load conditions, dispatch conditions, and defined transactions (source-to-sink transfers), - define a large set of contingencies that spans the range of credible contingencies, - using PV curve method, push the system through every condition, under all contingencies until the voltage instability point is found for each condition, - at the point of instability for each case (nose of the PV curve) perform modal analysis to determine the critical mode of instability as defined by a set of bus participation factors corresponding to the zero eigenvalue, - store the results of the modal analysis in a database for analysis using data mining techniques to identify the VCAs and track them throughout the range of system changes, - establish the reactive reserve requirements for each identified VCA. The above approach was successfully implemented and tested on the PSE Operator power system. A database with user-friendly interface for storing/retrieving result of modal analysis for each scenario/contingency was designed based on the Microsoft Access. VCA identification was carried using several data mining techniques. Finally the VCAs were identified using a clustering method developed by PLI. For the studied scenarios and contingencies in the PSE Operator system, two VCAs were identified.
