Machine Learning for Enhanced CBM+

Abstract

The proposed program focuses first on developing autoencoders using normaloperating data for anomaly detection, while also recognizing the need for signal fault detection. Special attention will be given to exploring the effectiveness of the approach when applied to individual assets and to fleets. Next, transfer learning will be employed to extend the trained autoencoder models and their components to diagnostic capabilities. The diagnostics developed using autoencoders will be compared to the results of other traditional methods, such as physics based diagnostic classifiers. Furthermore, detection and isolation metrics will be utilized to determine the effectiveness of the autoencoder based diagnostic approach.

Document Details

Document Type
DoD Grant Award
Publication Date
Aug 20, 2019
Source ID
N000141912600

Entities

People

  • Michael Thurston

Organizations

  • Office of Naval Research
  • Rochester Institute of Technology
  • United States Navy

Tags

Readers

  • Fault Tolerant Diagnosis of Black and White Balloon Isolation Tests Using ¥.
  • Neural Network Machine Learning.

Technology Areas

  • AI & ML
  • AI & ML - Neural Networks