PETRONAS Pipeline Predictive Analytics for Corrosion (PrAnCorr) is a cutting-edge cloud-based innovation that uses big data from near real-time operating data to predict internal and external corrosions for onshore and offshore pipelines. The technology which made up of multilayer system architecture allows seamless data integration which transform the traditional method of managing pipeline integrity from one where pipelines are typically inspected every 5 years to one in which pipeline corrosion behavior can predicted on near real time, thus reducing dependency of having in line inspection. The motivation of the development which began in 2019 was derived from significant increasing trends of OPEX spending to control and monitor corrosion development and production losses due to corrosion failures. The formation of multidisciplinary team within PETRONAS had demonstrated proven results for a comprehensive life cycle machine learning system through collaborative efforts. The system is powered by 2 main processing engines namely iCON (process simulation tool) as a real-time process simulation engine and Machine Learning engine. The prediction utilizes more than 13 operating variables simultaneously generated through integration from various data sources within i-PIMS (PETRONAS Integrated Pipeline Integrity Assurance Solutions) system, capable to perform continuous data ingestion & cleaning which eventually revealed corrosion behavior along the pipelines. PrAnCorr was successfully deployed in March 2021 as a module in i-PIMS application and has already been used to predict corrosion for various pipeline system i.e., dry gas, multiphase, wet gas, condensate etc. The establishment of PrAnCorr is a significant step forward in the achievement of pipeline integrity performance, especially given that the potential savings derived from optimization of I&M. The positive value creation is estimated for long term application by 20% OPEX reduction over the life cycle. PrAnCorr is prospective for PETRONAS as a future of pipeline integrity solution that spans the company's operations with the capability from obtaining a quality pipeline real-time condition to subsequently performing integrity assessments.
Managing pipeline integrity revolves around abundance of data and information from monitoring of safe operating limits, inspections and maintenance. Those data and information may come from real-time/on-line systems, manually input ad-hoc, manually input from inspections and maintenance carried out on a particular pipeline. The oil and gas pipeline industry have sort of mature with respect to having a software/tool in aiding and assisting personnel in performing risk, fitness-for-service, repair assessments, and executing management of change for a pipeline system; and many more assessments/analyses. Nevertheless with the invent of analytics, there is strong need to explore the new ways of working with those abundance data and information OR maximising the utilisation of the data and information for the benefit of assessing or evaluating pipeline's risk and integrity to predict 'accurately' risk and integrity so that specific and cost-effective actions and mitigations can be deployed within a stipulated period of time. In those regard, PETRONAS is actively working with industry to establish predictive analytics for critical offshore and onshore pipelines' threats/anomalies i.e. internal corrosion and free span for offshore; and external corrosion and geotechnical hazard for onshore. This paper will dwell on the principles, concepts and methodology of predictive analytics tools for the development. It is envisaged that eventually those threats/anomalies will be analytics-managed to eliminate unwanted incidents to PETRONAS's offshore and onshore pipelines. In addition, analytics-pipeline integrity management will also likely to provide 'accurate' prediction of 'future' pipeline integrity.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.