Dynamic gradient method for PEBS detection in power system transient stability assessment

In methods for assessing the critical clearing time based on the transient energy function, the dominant procedures in use for detecting the exit point across the potential energy boundary surface (PEBS) are the ray and the gradient methods. Because both methods rely on the geometrical characteristics of the post-fault potential energy surface, they may yield erroneous results. In this paper, a more reliable method for PEBS detection is proposed. It is called the dynamic gradient method to indicate that from a given system state, a small portion of the trajectory of the gradient system is approximated and tested for convergence toward the post-fault stable equilibrium point. It is shown that a trade-off between computing time and reliability can be found as the number of machines in the system becomes greater. The method is illustrated on 3-machine and 10-machine systems. (C) 2000 Elsevier Science Ltd. All rights reserved.