Abstract
This paper proposes a sensor-data-driven prognosis approach for the predictive maintenance of a liquefied natural gas (LNG) satellite plant. By using data analytics of sensors installed in the satellite plants, it is possible to predict the remaining time to refill the tank of the remote plants. In the proposed approach, the first task of data validation and correction is presented in order to transform raw data into reliable validated data. Then, the second task presents two methods for the prognosis of gas consumption in real time and the forecast of remaining time to refill the tank of the plant. The obtained results with real satellite plants showed good performance for direct implementation in a predictive maintenance plan.
Subject
Artificial Intelligence,Applied Mathematics,Industrial and Manufacturing Engineering,Human-Computer Interaction,Information Systems,Control and Systems Engineering
Reference41 articles.
1. Economic and Technological Analysis of Commercial LNG Production in the EU
2. A Novel Multicriteria Evaluation of Small-Scale LNG Supply Alternatives: The Case of Greece
3. General Overview of Spanish LNG Sectorhttps://ec.europa.eu/energy/sites/ener/files/documents/prieto_-_lng_experience_spain.pdf
4. LNG Marked in Spainhttps://www.kaasuyhdistys.fi/wp-content/uploads/2018/12/LNG-market-in-SpainReganosa-Rodrigo-Diaz-Ibarra.pdf
5. Optimization of a small-scale LNG supply chain
Cited by
2 articles.
订阅此论文施引文献
订阅此论文施引文献，注册后可以免费订阅5篇论文的施引文献，订阅后可以查看论文全部施引文献