Speaker :- Saroj Kumar
Abstract :- The dynamic security assessment (DSA) is the process of determination of the stability of a power system when subjected to large disturbances (contingencies). A large disturbance such as a fault, followed by a loss of a system component, may lead to cascading outages. Hence, it is important to predict, online, the stability of the system when subjected to large disturbances. The prediction of stability can be done by system simulation and one can take preventive action or decide remedial action if instability is predicted. There is a time constraint for online analysis since it has to be performed for a large number of contingencies in every few tens of minutes. In order to reduce the amount of computation, screening of contingencies is performed. Screening involves analyzing the system described by a simpler model so that the computation requirement is reduced. Screening identifies those contingencies which are sure to not cause instability and hence can be eliminated from further scrutiny.
The computation required is in the numerical solution of nonlinear differential-algebraic equations. The computation time and the error in the solution depend on the numerical method and the values of the step size and the tolerance used in the solution of nonlinear algebraic equations. The decision, on the numerical method and the values of step size and tolerance, should be made with a compromise between speed and accuracy. There is a need for quantifying the error. In this work, practical measures of error in the numerical solution are proposed. The proposed measures of error are used to compare a few numerical methods with different step sizes and tolerances. Based on the proposed measures of error, the most accurate numerical method satisfying the time constraint can be chosen for online DSA. An alternative to system simulation is the use of energy function method which involves determination of a quantity called critical energy, by a short term system simulation. A comparison of the performance, in terms of accuracy and speed, of different numerical methods in the determination of critical energy for online DSA, is also presented. Case studies on 10 generator New England systems and 17 generator IEEE test system are reported.