godwillokwuchukwu/LNG-Regasification-Plant-Optimization

This master's thesis presents a comprehensive machine learning framework for optimizing LNG regasification process plant in Nigeria. # Master's Thesis Project Documentation: Machine Learning for LNG Regasification Plant Optimization ## Table of Contents 1. Executive Summary 2. Introduction 3. Literature Review 4. Methodology - 4.1 Data Preparation - 4.2 Data Cleaning - 4.3 Machine Learning Implementation 5. Results and Analysis - 5.1 Data Overview - 5.2 Flow Rate Analysis - 5.3 Model Performance - 5.4 BOG Generation and Impact 6. Discussion 7. Conclusions and Recommendations 8. References 9. Appendices --- ## Executive Summary This master's thesis presents a comprehensive machine learning framework for optimizing LNG regasification plant in Nigeria. The study addresses four key objectives: (1) accurate prediction of LNG flow rates, (2) energy consumption optimization, (3) quantification of Boil-Off Gas (BOG) generation, and (4) assessment of BOG impact on system efficiency. Utilizing operational data from September 2021 to August 2025 (982,121 records), we developed and compared three machine learning models: ARIMA, LSTM, and XGBoost. The LSTM model achieved the highest performance with an R² of 0.4205 for flow rate prediction. Key findings include a BOG-flow correlation of 0.074 and BOG-induced flow variations of 69.09 m³/h, representing significant efficiency losses. This research provides actionable insights for LNG plant operators, demonstrating how predictive analytics can enhance operational efficiency and reduce energy waste. Recommendations include real-time BOG monitoring systems and hybrid ML model deployment for operational optimization. --- ## Introduction ### Background Liquefied Natural Gas (LNG) regasification is a critical process in the natural gas supply chain, converting LNG from liquid to gaseous form for distribution. In Nigeria, facilities such as Greenville LNG and Ama Nigeria Breweries play vital roles in energy supply. However, challenges including Boil-Off Gas (BOG) generation, variable flow rates, and energy inefficiency persist. Boil-Off Gas, generated durin …