In-process motor testing results using model based fault detection approach

Abstract

Rapid progress in process automation and tightening quality standards result in a growing demand being placed on fault detection and diagnostics (FDD) methods to provide both speed and reliability of motor quality testing. This paper presents the findings of a decade-long research and development efforts in the field of experimental modeling technique and its practical applications for the fault detection purposes, first in the fields of aerospace and defense, and now in the context of high-volume electric motor manufacturing. Underlying this patented technology is a set of proprietary algorithms that enable precise tracking of the parameters pertaining to the physical structure of the motor. The derivation of condition information from changes in the physical structure, rather than from symptoms of faults such as noise and vibration, allows detecting a wide variety of faults and drastically simplifies the assessment of fault types.