Much work has been done in applying computer solutions to the problem of power systems fault and stability studies. However, with a few notable exceptions, previous efforts have largely been confined to attempts to solve the problems in the classical manner by having the computer simply take over the tedious portion of the calculations. A few pioneers have succeeded in finding new methods of solving some of the problems by the use of the machine equations, without resorting to the simplifying assumptions necessary for the classical method of solution and their inevitable inaccuracies. These pioneers, however, have only dealt with the balanced condition type of fault, such as a three-phase-fault or a sudden change in load on a machine, and have ignored the far more common although less severe types of fault wherein the system becomes unbalanced due to a fault involving only one or two phases. In contrast, this thesis deals with the unbalanced type of fault on a system. The necessary equations are derived and simple problems are solved for three major types of faults, i.e., the single-phase-to-ground fault, the phase-to-phase fault, and the two-phase-to-ground fault. In addition, a program is written to simulate a field study in order that a comparison can be made. The programs used are written for the single-machine on an infinite bus, and are written so that initial steady conditions and machine and system constants may be entered easily, thus making the thesis method usable by practicing power systems engineers. The thesis method results in a great variety of useful data, which, since they may be transformed into symmetrical components form allows the use of classical methods of system reduction and re-expansion so that fault currents and voltages may be found at any point in the system. All that is required for a system analysis is the reduction of the system, the choice of program for the type of fault under consideration, and the re-expansion of the system. It is hoped that the results of this thesis may prove useful not only to those who wish to continue research in this area, but also to those who are practicing engineers concerned with system fault studies and who are in need of an improved method for the study of unbalanced faults.

Ph. D.