A Thermodynamic Entropy Based Approach for Prognosis and Health Management

Abstract

Data-driven stochastic and probabilistic methods that underlie reliability prediction and structural integrity assessment remain unchanged for decades. This paper provides a method to explain the Prognostics and Health Management (PHM) in terms of fundamental concepts of science within the irreversible thermodynamic framework. The common definition of damage, which is widely used to measure the reduction of reliability over time, is based on observable markers of damage at different geometric scales. Observable markers are typically based on evidences of any change in the physical or spatial properties or the materials, and exclude unobservable and highly localized damages. Thermodynamically, all forms of damage share a common characteristic: energy dissipation. Energy dissipation is a fundamental measure of irreversibility that within the context of non-equilibrium thermodynamics is quantified by entropy generation. The definition of damage in the context of thermodynamics allows for incorporation of all underlying dissipative processes including unobservable markers of damage. Using a theorem relating entropy generation to energy dissipation associated with damage producing failure mechanisms, this paper presents an approach that formally describes and measures the resulting damage.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Oct 02, 2014
Accession Number
AD1002224

Entities

People

  • Anahita Imanian
  • Mohammad Modarres

Organizations

  • University of Maryland

Tags

Communities of Interest

  • Advanced Electronics
  • Energy and Power Technologies

DTIC Thesaurus Topics

  • Artificial Intelligence
  • Bayesian Networks
  • Chemical Reactions
  • Computational Science
  • Engineering
  • Failure Mode And Effect Analysis
  • Information Processing
  • Information Science
  • Materials
  • Mechanical Engineering
  • Mechanics
  • Neural Networks
  • Reliability
  • Reliability Engineering
  • Self Organizing Systems
  • Statistical Analysis
  • Supervised Machine Learning

Readers

  • Adaptive Control and Estimation with Uncertainty in Dynamic Systems.
  • Fault Tolerant Diagnosis of Black and White Balloon Isolation Tests Using ¥.
  • Theoretical Analysis.