Electric Power Systems Research, 16 (1989) 209 - 216 209 Security Constrained Optimal Reactive Power Dispatch VENKATARAMANA AJJARAPU Department of Electrical Engineering, Iowa State University, Ames, IA 50011 (U.S.A.) J. CARR Acres Consulting Services Ltd., 5259 Dorchester Road, Niagara Falls, Ont,, L2E 6W1 (Canada) R. S. RAMSHAW Department of Electrical Engineering, University of Waterloo, Waterloo, Ont., N2L 3G1 (Canada) (Received November 14, 1988) ABSTRACT This paper presents a systematic methodol- ogy to allocate reactive power devices in a power system. This is achieved through the application of an active set analysis based linear programming technique. The proce- dure takes into consideration outages which cause voltage problems and also existing reac- tive power controllers are fully utilized before any new reactive power devices are added. Linearized sensitivity relationships of the power system are used to obtain an objective function for minimizing the cost of installa- tion. The constraints include the limits on dependent variables (reactive power of the generators, load bus voltages) and control variables (generator voltages, tap positions, switchable reactive power sources). The methodology is demonstrated by two power system examples. INTRODUCTION A power system routinely experiences con- tingencies such as the outage of transmission lines, transformers and generators, etc. Some contingencies may result in a new system state which has overloaded transmission lines and/ or voltages outside acceptable operating limits. A system operator has to take appropriate action to rectify these situations. Frequency excursions too far from the nominal 60 Hz operating point can be alle- viated by rescheduling the real power genera- tion in the system, and in critical situations by resorting to load shedding [1]. Overvoltages and undervoltages are alleviated by proper adjustment of reactive power control variables, such as generator voltages, transformer taps and switchable reactive power sources. Proper planning of reactive power devices for contin- gencies is an important aspect of the secure operation of power systems. A common definition of a secure system is one that will continue to operate after the loss of one or more components (transmission lines, transformers, generators} because of a contingency [2]. A precise definition of security, as pointed out by Carpentier [3], is that a system is n secure if it continues to operate satisfactorily when all its n elements are intact. The system is n-k secure if the sys- tem continues to operate after k elements have been lost. Analysis is therefore needed to determine whether a particular contingency operating condition is secure. A complete contingency analysis requires the evaluation of a large number of possible contingencies on a power system. If all the theoretically possible con- tingencies were considered, it would be im- possible to find a secure system. Hence, we have to find a way to select contingencies based on their probability of occurrence and severity. A performance index has to be de- fined to measure the severity of a particular contingency. Another aspect of this analysis is that of testing all contingencies every time an evaluation of the system performance is needed. Almost all contingency analysis programs rely on a list of selected contingencies to be studied. Even then the testing of all the con- tingencies every time an evaluation of the 0378-7796/89/$3.50 © Elsevier Sequoia/Printed in The Netherlands