Degree Type


Date of Award


Degree Name

Doctor of Philosophy


Electrical and Computer Engineering


Conventional techniques used for the assessment of transient and dynamic stability in multimachine power systems require the solution of a large number of ordinary differential equations describing the dynamics of the system generating units. Detailed representation of generating units results in a system of equations that are numerically stiff and of large dimensionality. As a result, the digital simulation of the entire system becomes excessively complex and uneconomical in terms of computer time;It has been recognized that coherency identification is a fundamental step in obtaining reduced order dynamic equivalent models for power systems. Two methods of coherency identification are presented in this dissertation, one based on a slow coherency approach and the other based on modal analysis. Improvements in the slow coherency approach are accomplished by a sensitivity based identification of the set of group-reference generators in the system, leading eventually to the sets of coherent generators. The second method uses the undamped system equations having a system matrix A. The eigenstructure of A is used to obtain a set of coherency indices, based on the root-mean-square values of variations in angular differences between pairs of generators. These coherency indices are compared with a coherency threshold to determine coherent groups of generators. Numerical examples are presented to illustrate and validate both techniques;With regard to the dynamic stability of multimachine power systems, a method based on singular perturbation theory is proposed for the simulation of models of high dimensionality and widely varying time constants. In this method, the state vector is partitioned into subsets of slow and fast variables, a perturbation parameter, (epsilon), is selected and the equations are expressed in their singular perturbation form. The time responses of the system are obtained by the use of the asymptotic expansions of the outer solution and the boundary layer correction of the perturbed equations;A numerical example using a single machine-infinite bus system is used to illustrate and validate the method. While promising results were obtained, the need for additional refinements was made evident.



Digital Repository @ Iowa State University,

Copyright Owner

Carlos Grande-Moran



Proquest ID


File Format


File Size

186 pages