Process resilience analysis based data-driven maintenance optimization: Application to cooling tower operations

Prerna Jain

doi:10.1016/j.compchemeng.2018.10.019

Return

Process resilience analysis based data-driven maintenance optimization: Application to cooling tower operations

2019-02-01

PRE

Prerna Jain *

Efstratios N. Pistikopoulos

M. Sam Mannan

DOI：10.1016/j.compchemeng.2018.10.019

Original

Original request for help

Save

Abstract

En 中文

In a process plant system, safe and reliable operations are highly sensitive to utilities such as power, steam, cooling water, nitrogen, and instrument air. These play an important role or act as safety barriers. This is because a disturbance in their supply is likely to affect process operations downstream, may reduce the production efficiency, and may lead to a sudden shutdown or contribute to an unsafe condition. The focus of this paper is to introduce and evaluate a model for survival of a process system under upset conditions using the Process Resilience Analysis Framework (PRAF). Resilience metrics within a data-driven and model-based optimization approach using Bayesian regression are employed integrating both technical (process parameter variations) and social (human and organizational) factors. Based on an optimization objective function accounting for the overall system performance in terms of energy consumption, maintenance costs, safety impact, environmental impact, asset damage, and production loss, the proposed methodology aims to determine the optimal maintenance policy for optimal and safer plant operations. The implementation of the survival model within the PRAF is demonstrated on a cooling tower operation example problem, where optimal operation and maintenance (preventive, corrective, and predictive) strategies are determined based on trade-offanalysis of process revenue, safety impact, and maintenance costs. (C) 2018 Elsevier Ltd. All rights reserved.

Keywords:

Maintenance

Optimization

Cooling tower

Resilience

Bayesian regression

AI Summary

Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.