Design and Application of Data Aquistion Interface Circuit

TL;DR

This paper presents a design of a data acquisition interface circuit for condition monitoring, demonstrating its application in fault detection of induction motors using neural networks, with two test scenarios illustrating frequency response differences.

Contribution

It introduces a new interface circuit design for machine condition monitoring and applies neural network-based fault detection to improve diagnosis accuracy.

Findings

Frequency amplitude differences observed in two fault scenarios

Neural network approach enhances fault detection accuracy

Proposed circuit simplifies condition monitoring tasks

Abstract

A commitment to condition monitoring involves the operators of plant in the conduct of a range of activities. These activities may be compli-cated in nature and indeed may often be performed automatically under computer control. They can, however, always be down into a rela-tively small number of easily identifiable functional tasks. This makes it much easier to identify the common elements of machine condition monitoring schemes. A proposed interface circuit design and application will be further explain in this paper, the implemented monitoring unit circuit also illustrated, see appendix A. Two scenarios presented in this paper, first ten turns assume to be shorted, and in the second thirty turns shorted to show the difference in the amplitude of frequencies at each case. This paper present. An improvement in three-phase squirrel-cage induction motor stator inter-turn fault detection and diagnosis based on a neural network approach is presented.