A REFINED METHOD OF NEGLECTING STATOR TRANSIENTS IN INDUCTION MOTOR LOADS (REDUCED ORDER MODELLING, MACHINE)

FRANCIS DENNIS RODRIGUEZ, Purdue University

Abstract

Various methods of deriving reduced order models of induction machines have previously been developed and investigated. The resulting models are used widely in power system stability studies where, due to the complexity of the overall system, it becomes difficult to use the detailed form of the induction machine equations. A common method of deriving reduced order models of induction machines involves neglecting the rate of change of stator flux link-ages (p(psi) terms) in the stator voltage equations, i.e. the so called method of neglecting stator transients. It has previously been shown that the resulting reduced order model is reasonably accurate when used to predict the dominant mode (low frequency) behavior of induction machines following various system disturbances. This can be attributed to the fact that in many cases, the changes in stator flux linkages are much faster than the changes in rotor flux linkages. Consequently, for many disturbances, the stator transients subside before the rotor variables begin to change. This suggests that the steady-state version of the stator voltage equations may be used when solving for the slower rotor variables. It is clear, however, that in an actual machine, the rotor flux linkages will change before the stator transients completely subside. Therefore, the reduced order equations provide only an approximation to the actual system dynamics and are accurate when the violation of this assumption is minor. In this research, a refinement of the standard procedure of neglecting stator transients is set forth and investigated. In this approach, order reduction is accomplished by neglecting the rate of change of only the fast component of the stator flux linkages when solving for the slower rotor variables. It is shown that the resulting model provides a more accurate indication of the dominant mode behavior of the induction machine than the standard reduced model. Furthermore, this improvement is accomplished with a straightforward modification of the established procedure of neglecting stator transients and without increasing the dynamic order of the reduced model. The research work is extended further to include systems of interconnected induction motors and/or generators. Of particular interest is the modelling of the station loads in nuclear power plants where a major concern involves the integrity of the large induction motor drives following electrical disturbances. It is shown that the response of the resulting multi-machine reduced order model more closely follows the response of the full order model (exclusive of the fast modes) than the standard reduced order model.

Degree

Ph.D.

Subject Area

Electrical engineering

COinS