A new coherency identification technique for use in transient stability studies has been developed. The post-fault swing equations of the system are linearized for purposes of coherency identification and are put in state space form. Using the properties of the exponential matrix and the Cayley-Hamilton theorem, a set of conditions which must be satisfied by coherent machines is obtained. With simplifying assumptions, these conditions expose the essential factors underlying coherency behavior and offer valuable guidelines for further research.;Both the coherency identification algorithm and the new stability equivalent have been tested on sample small-scale systems. The test results indicate that the new techniques provide accurate results, are easy to use, need no previously derived swing curves, involve no numerical integration and require minimal computational effort to implement.;Based on the coherency conditions, a coherency measure is defined and a coherency identification algorithm is developed. A new stability equivalent is also described which uses the coherent groups obtained from the coherency identification algorithm. The methodology is rigorously based and leads to a very much reduced dynamic equivalent for use in large-scale transient stability studies. |