The Use of Histograms for Detection of Electrical Insulation Breakdown

Abstract

Seeded fault experiments were performed on a 50 Hp, 3 phase, 1750 RPM induction motor operated with a variable speed (pulse width modulation, or PWM, type) controller. A hole was cut in the stator housing, allowing access with an industrial heat gun. A single coil of one phase, at the point where it enters the stator laminations, was heated to a series of raised local temperatures. Current and voltage data were simultaneously measured on the target phase and recorded at a sample rate of 2.5 MHz each. Various feature extraction methods were applied to the voltage, current and instantaneous power (voltage times current) time series data. Single point statistics (rms, skewness, and kurtosis), histograms, and Fourier spectra are presented here.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Apr 05, 2001
Accession Number
ADP013504

Entities

People

  • Jun Wang
  • Sally Mcinerny

Organizations

  • University of Alabama

Tags

Communities of Interest

  • Biomedical
  • Energy and Power Technologies
  • Space

DTIC Thesaurus Topics

  • Coefficients
  • Data Science
  • Data Sets
  • Electrical Insulation
  • Energy Consumption
  • Frequency
  • Frequency Bands
  • Histograms
  • Induction Motors
  • Information Science
  • Insulation
  • Motors
  • Spectra
  • Standards
  • Statistical Analysis
  • Statistical Distributions
  • Test Facilities

Fields of Study

  • Physics

Readers

  • Approximation Theory.
  • Computer Vision.
  • Electrical Engineering

Technology Areas

  • AI & ML
  • AI & ML - Bayesian Inference
  • Microelectronics
  • Microelectronics - Microelectromechanical Systems