Heidary, Roohollah, Groth, Katrina, Modarres, Mohammad Fusing more frequent and accurate structural damage information from one location to assess damage at another location with less information (Inproceedings) Proceedings of the 14th Probabilistic Safety Assessment and Management Conference (PSAM 14), Los Angeles, CA, 2018.

BibTeX

@inproceedings{Heidary2018PSAM,
title = {Fusing more frequent and accurate structural damage information from one location to assess damage at another location with less information},
author = {Roohollah Heidary and Katrina M Groth and Mohammad Modarres},
url = {http://psam14.org/proceedings/paper/paper_301_1.pdf},
year = {2018},
date = {2018-09-16},
booktitle = {Proceedings of the 14th Probabilistic Safety Assessment and Management Conference (PSAM 14)},
address = {Los Angeles, CA},
keywords = {augmented particle filtering, data fusion, gamma process, hierarchical Bayesian model, pitting corrosion},
pubstate = {published},
tppubtype = {inproceedings}
}

Abstract

This paper proposes a two-phase data fusion framework to be used within a prognostic and health management-based degradation model to estimate remaining useful life (RUL) of a segment of corroded oil and gas pipelines due to internal pitting. The existing degradation models for internal pitting corrosion are based on the assumption that operational conditions remain the same during the life of the pipeline. In contrast, this framework addresses the actual case where operational conditions change over time. The change in operational conditions is reflected in on-line inspection data of a specific active pit (pit M) and this framework is proposed to link this change to the growing behavior of other pits. In this way, dummy measurements of pit i are simulated based on on-line inspection data of pit M as well as the similarity between pit i and pit M. This similarity is defined as the average of the ratio of the estimated depth of pit i and pit M which is modified by a location factor. A hierarchical Bayesiangamma process model and augmented particle filtering method are used respectively in the first and
second phase of this framework to estimate the RUL of the pipeline.

Recent Posts

Welcome to the SyRRA Lab

Posted on by
1

The Systems Risk and Reliability Analysis (SyRRA) Lab, directed by Dr. Katrina M. Groth, conducts research to address emerging safety, reliability and security issues for engineered systems, with primary applications in energy and transportation. Our multi-disciplinary research involves integrating information, data, statistics, and models from multiple domains to understand the causes of risk and failure of systems.

We leverage state-of-the-art computational tools and Bayesian methods to develop scientific, evidence-based models, and we add human-focused thinking to understand the needs of decision makers and use those models and data to support decision making for problems with many types of uncertainty.