Survey of Fault Detection and Classification in Power Conversion Electronics

Abstract

This work summarizes the research trends in the area of anomaly detection and classification within electrical power conversion systems with special emphasis on motor drive inverters. To begin, an example of Army interest in and brief overview of the concepts and terminology of this field are given. These are followed with a general description of fault detection and classification methods and a summary of selected published works to provide the reader with a sense of the state of these methods as applied to power electronics. The analytical framework for anomaly detection and classification has been under development since the early years of the 20th century and all of the algorithms discussed in this review may be readily found in software libraries. It is, therefore, a reasonable expectation that machine learning (ML)-enabled products should be widely available. However, a significant oversight or underappreciation is the ability to produce an optimized feature vector that is statistically significant to the available training data. In order to successfully apply ML to a given problem, an optimally-selected feature vector and statistically significant training data are required. Based on this review, a study of optimal feature selection for fault detection in power conversion systems is warranted.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Nov 13, 2020
Accession Number
AD1115443

Entities

People

  • Charles Iv W Tipton

Organizations

  • United States Army Research Laboratory

Tags

Communities of Interest

  • Advanced Electronics
  • Energy and Power Technologies
  • Materials and Manufacturing Processes

DTIC Thesaurus Topics

  • Algorithms
  • Anomaly Detection
  • Artificial Intelligence
  • Bayesian Networks
  • Change Detection
  • Computational Science
  • Detection
  • Dimensionality Reduction
  • Information Science
  • Inverters
  • Machine Learning
  • Neural Networks
  • Power Electronics
  • Signal Processing
  • Supervised Machine Learning
  • Three Dimensional
  • Two Dimensional

Readers

  • Computational Modeling and Simulation
  • Integrated Circuit Design and Technology.
  • Neural Network Machine Learning.

Technology Areas

  • AI & ML
  • AI & ML - DoD AI Strategy
  • Microelectronics