Distillation Column Control Using ML for Variable Feed Separation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing refinery operations face challenges in optimizing fluid production and separation due to varying feedstock properties, equipment changes, and the need for expert personnel to maintain first-principles models, leading to inefficiencies and non-uniform optimization across refineries.
Innovation Solution
Implementing systems and methods that utilize machine learning models to enhance fluid production and separation by collecting data from sensors and analyzers, applying it to trained models to predict settings for refining equipment, and adjusting operations in real-time to achieve targeted products.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If first-principles models are used to optimize refinery operations, then manufacturing precision is improved, but device complexity and ease of operation worsen due to requiring expert personnel and extended execution time
Solution Approach 1:
The patent creates a digital twin (virtual model) that replicates the behavior of the physical refinery process. This digital twin can be updated and optimized without affecting the actual operation, allowing complex first-principles modeling to be performed on copies rather than the live system, thus reducing operational complexity while maintaining precision
Solution Approach 2:
The system performs preliminary optimization calculations using the digital twin before implementing changes to the actual refinery operations. This allows complex computations to be completed in advance on the virtual model, reducing the execution time and complexity burden on the live system while maintaining high production accuracy
2Manufacturing precision
If first-principles models are used to optimize refinery operations, then manufacturing precision is improved, but loss of time increases due to extended execution time
Solution Approach 1:
The system performs preliminary optimization calculations using the digital twin before implementing changes to the actual refinery operations. This allows complex computations to be completed in advance on the virtual model, reducing the execution time and complexity burden on the live system while maintaining high production accuracy
Solution Approach 2:
The patent creates a digital twin (virtual model) that replicates the behavior of the physical refinery process. This digital twin can be updated and optimized without affecting the actual operation, allowing complex first-principles modeling to be performed on copies rather than the live system, thus reducing operational complexity while maintaining precision
3Ease of operation
If traditional control systems are used across multiple refineries, then ease of operation is maintained, but adaptability worsens due to non-uniform optimization across different refineries
Solution Approach 1:
The patent implements a universal digital twin framework that can be deployed across multiple refineries. This single platform adapts to each specific refinery's characteristics through data-driven modeling, providing consistent optimization capabilities across different facilities while maintaining ease of operation through a standardized interface
Solution Approach 2:
The system dynamically adjusts model parameters based on each refinery's specific conditions, feedstock variations, and operational characteristics. This allows the same control platform to adapt to different refineries by changing parameters rather than requiring custom solutions for each facility, achieving both uniformity and adaptability
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The system enables accurate and efficient production of targeted products by optimizing refinery operations in real-time, reducing energy consumption, and improving the efficiency of fluid separation processes.
Implementation Method 1
one or more distillation columns to receive a feed and separate the feed into a plurality of products
Data Source
AI summary
Embodiments of systems and methods for enhancing control of a distillation operation are disclosed. The method includes obtaining data for a plurality of ongoing and continuous distillation operations from one or more of (a) a plurality of sensors or (b) a plurality of analyzers configured to analyze fluid output via the distillation operations. The method may include determining one or more parameters for each one or more of one or more distillation columns or distillation control devices based on application of the data to a machine learning model. The method may include in response to determination of the one or more parameters, operating each of the one or more distillation columns or distillation control devices based on the one or more parameters, thereby to enhance operation of the one or more distillation column.