Title:
A Hybrid Feature Selection Framework for Catalytically Cracked Gasoline Yield Prediction via Opinion Discriminative Analysis and Relief Algorithm
Author:
Supervisor:
Department:
Mechanical Engineering
Year:
2026
URL:
Abstract:
The rapid digitization of the oil and gas industry has led to the widespread adoption of both conventional and artificial intelligence models for process optimization and predictive turnaround mainten...
View more
Multimedia Content:
01
Methodology
1. Response Surface Methodology (RSM)
Response Surface Methodology was developed as a statistical baseline model to investigate the relationships between FCCU process variables and gasoline yield. ...
View more
02
Citation Analysis
1. Fuel
Scope: Refining technologies, catalytic cracking, fuel production, process optimization, and energy conversion.
Relevance: Directly related to FCCU operation, gasoline yield optimization, ...
View more
03
Impact
1. Technical Impact
This research introduces a novel hybrid refinery optimization framework that integrates Response Surface Methodology (RSM), ensemble machine learning models, deep learning archi...
View more
04
Benefits to Society
1. Environmental Benefits
The proposed framework helps refineries optimize operating conditions and improve process efficiency, which can reduce unnecessary energy consumption and minimize process ...
View more
05
Applications
1. FCCU Real-Time Optimization
Benefit: Improves gasoline yield prediction and operational efficiency.
Implementation: The hybrid predictive framework can be integrated into refinery control syste...
View more
06
Infographic
Summary of this Thesis
07
Patents / Products
Product Name:
ACO+RF+DNN+XGB Hybrid Refinery Prediction Framework
Significance:
The framework represents the principal innovation of the thesis by integrating feature selection, expert validation...
View more
08
Experts
1. Professor Dr. Faiz Ahmad
Institution: Universiti Teknologi PETRONAS (UTP), Malaysia
Contribution: Main supervisor and lead researcher overseeing the development of the hybrid refinery optimizat...
View more