Affiliation:
1. PETRONAS
2. PETRONAS Digital Sdn Bhd
3. Malaysia LNG Sdn Bhd
Abstract
Abstract
To capture the invaluable tacit knowledge from skilled operators and leverage on insights generated from over two decades of plant operation data, an AI-driven live advisory was developed to provide operators with real-time parameters control advisory during start-up, in response to actual process and equipment conditions. The solution had been deployed to ten start-ups in Malaysia LNG, successfully achieving 44% reduction in duration and 17% reduction in carbon emission from reduced gas usage. The shorter start-up duration translated into significant value creation from production opportunity, yielding an impressive return on investment of 35, in just 15 months. Furthermore, six of the start-ups also emerged as the top executions when benchmarked against all historical occurrences in the past two decades, demonstrating its ability to enable consistent and optimised start-ups.
Reference7 articles.
1. Profitability, reliability and condition based monitoring of LNG floating platforms: A review,;Gowid;Journal of Natural Gas Science and Engineering,2015
2. Stacked regressions;Breiman;Machine Learning,1996
3. The Cascade Model in Action;Abeysena;English Language Teacher Research Partnerships. A collection of research papers from the Sri Lankan context,2016
4. Simulation and analysis of Double Pipe Heat Exchanger,;Joshi;International Journal for Research in Applied Science and Engineering Technology,2021
5. 9.6 SHAP (SHapley Additive exPlanations) | Interpretable Machine Learning;Molnar;[online] Christophm.github.io,2022