Request for AI-Powered Machine Learning Solution for Refiner
زمان تحویل: 45 روز
112 روز پیش
4 پیشنهاد
From: Oil Design and Construction Company (ODCC)
Date: 2/7/2025
Subject: Proposal for AI-Based Machine Learning System for Refinery Process Optimization
Introduction to ODCC
Oil Design and Construction Company (ODCC) is a leading EPC contractor in Iran, specializing in refinery construction and petrochemical projects. Established in 1991, ODCC has successfully executed large-scale oil and gas projects, including Crude Distillation Units (CDU), Vacuum Distillation Units (VDU), Naphtha Hydrotreaters (NHT), Diesel Hydrotreaters (DHT), Sulfur Recovery Units (SRU), Fluid Catalytic Cracking (FCC), and Isomerization Units.
To improve refining efficiency, product yield, and process optimization, we seek a machine-learning-based solution capable of analyzing complex refinery process data and predicting outcomes based on varying feedstock compositions.
Project Scope & Objectives
The AI solution should:
1. Process Large-Scale Refinery Data – Analyze historical and real-time data from CDU, VDU, and other process units to identify efficiency improvements.
2. Predict Product Yields – Based on different crude oil types, forecast product output composition (gasoline, diesel, fuel oil, LPG, etc.).
3. Optimize Process Parameters – Adjust operational settings for maximum yield, energy efficiency, and cost reduction.
4. Feedstock Adaptability – Automatically generate refining process predictions when new crude oil blends are introduced.
5. Anomaly Detection & Process Safety – Identify deviations, inefficiencies, or potential failures in refinery operations.
6. Real-Time Simulation & Decision Support – Provide engineers with AI-driven recommendations to enhance operational decisions.
Key Challenges & Considerations
• Complex Feedstock Variability: ODCC processes various crude oil types with different chemical compositions.
• Multi-Unit Process Integration: The AI model must learn from CDU, VDU, NHT, DHT, FCC, SRU, and other refining units.
• Real-Time Processing Needs: The system must work with live operational data while also leveraging historical datasets.
• Customization & Industrial Compatibility: The software must integrate with ODCC’s existing refinery control systems (DCS, SCADA, and historian databases).
Available Data & Resources
ODCC has extensive process information, operational records, and crude oil test reports, including:
• Crude Oil Assay Data – Composition, density, sulfur content, etc.
• Process Operating Conditions – Temperature, pressure, flow rates.
• Historical Production Data – Product yields, quality, and refining efficiencies.
• Energy Consumption & Catalyst Performance – Key factors in refining economics.
We are prepared to provide full access to these datasets to train the AI model effectively.
Expected Deliverables
We expect the AI company to develop:
1. A refinery process optimization software using machine learning models.
2. User-friendly dashboard for refinery engineers to analyze results.
3. Integration with ODCC’s existing control and data systems.
4. Simulation tool for predicting product outputs and refining performance.
5. Training & Technical Support for ODCC process engineers.
Collaboration & Development Approach
We seek a collaborative development approach, where your AI specialists work alongside ODCC’s process engineers, control system experts, and refinery operators. The project will follow an iterative testing model, ensuring continuous refinement and accuracy improvement.
Next Steps
Please provide a detailed proposal, including:
• Your technical approach to developing this AI system.
• Implementation timeline and project milestones.
• Estimated budget and licensing structure.
• Case studies of similar projects (if available).
We believe this AI-driven refinery optimization system will significantly enhance ODCC’s process efficiency, cost control, and product yield optimization. We look forward to discussing this collaboration in further detail.
Best regards,
