Prognostic and Health Management using Copula Correlation: a Power System application

Loredana Cristaldi, Michy Alice, Enrico Ragaini

Abstract

A copula is a function that joins multivariate distribution functions to their margins (i.e. marginal distribution functions). Copulas are widely used in finance and economics for time series analysis and approaches based on them have found application in engineering as well, typically in civil and reliability engineering. In this paper, two data-driven prognostic algorithms based on copula application to power systems are proposed. The first is related to the estimation of the Remaining Useful Life (RUL) of a product, and the second aims at evaluating the performance and predict the behavior of an energy-consuming load. Obtained results are encouraging and candidate this approach as a useful method in Prognostics and Health Management (PHM) and energy monitoring applications.

Keywords

Health condition (HC), Remaining Useful Life (RUL), Copula correlation, marginal distributions, load modelling, pseudo-random load curves

Download

IMEKO-TC10-2020-033.pdf

DOI

-

IMEKO TC

TC10 - Measurement for Diagnostics, Optimization and Control

Back to the proceedings