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 for model maintenance
Solution Approach 1:
The patent introduces an intermediary system comprising sensors, analyzers, and a controller that mediates between the complex first-principles models and the operators. This intermediary layer automatically collects process data, applies the models, and generates optimization recommendations, eliminating the need for operators to directly manage the complex models while still achieving their precision benefits
Solution Approach 2:
The system enables self-service optimization by automatically collecting process data through sensors and analyzers, applying first-principles models through a controller, and generating optimization recommendations without requiring expert personnel intervention. The system serves itself by maintaining and updating the models based on incoming process data
2Manufacturing precision
If first-principles models are used to optimize refinery operations, then manufacturing precision is improved, but device complexity worsens due to requiring expert personnel and complex model maintenance
Solution Approach 1:
The patent segments the optimization system into distinct functional modules: sensors for data collection, analyzers for sample analysis, a controller for model application, and a user interface for recommendation display. This segmentation isolates the complex first-principles models within the controller module, allowing them to be maintained independently without increasing overall system complexity for operators
Solution Approach 2:
The controller acts as an intermediary that encapsulates the complexity of first-principles models. It automatically receives process data from sensors, applies the models, and outputs optimization recommendations, shielding the rest of the system from model complexity while maintaining high production accuracy
3Ease of operation
If traditional controllers and monitoring devices are used, then ease of operation is maintained, but productivity worsens due to extended execution time and expert personnel requirements
Solution Approach 1:
The system performs self-service optimization by automatically collecting data from sensors, analyzing samples, applying first-principles models through the controller, and generating recommendations without requiring expert personnel intervention. This automation dramatically reduces execution time while maintaining operational simplicity
Solution Approach 2:
The system enables continuous optimization by constantly collecting process data through sensors and analyzers, continuously applying first-principles models through the controller, and continuously generating updated recommendations. This continuous operation eliminates the extended execution times associated with traditional periodic optimization approaches
4Manufacturing precision
If uniform optimization is attempted across multiple refineries, then manufacturing precision is improved, but adaptability worsens due to different equipment service cycles and maintenance patterns
Solution Approach 1:
The patent applies local quality by allowing each refinery or processing unit to have its own first-principles models tailored to its specific equipment characteristics, service cycles, and maintenance patterns. The system collects local process data through sensors and analyzers, applies locally-adapted models through controllers, and generates location-specific optimization recommendations, achieving both uniform precision standards and local adaptability
5Productivity
If real-time data collection and model application are implemented, then productivity is improved, but device complexity worsens due to additional sensors, analyzers, and control systems
Solution Approach 1:
The patent applies universality by designing multi-functional integrated units that combine sensors, analyzers, and controllers into unified systems. These universal devices perform multiple functions: process monitoring, sample collection, data analysis, and optimization recommendation generation, thereby improving productivity without proportionally increasing system complexity
Solution Approach 2:
The system merges previously separate functions (data collection, sample analysis, model application, and recommendation generation) into an integrated real-time optimization system. By combining these functions into a coordinated workflow managed by the controller, the system achieves high productivity while managing complexity through functional integration rather than proliferation of separate components
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
Enables accurate and efficient production of targeted products by dynamically adjusting refining operations based on real-time data, reducing the need for expert intervention and improving equipment efficiency.
Implementation Method 1
one or more distillation columns to receive a feed and separate the feed into a plurality of products
Implementation Method 2
separate the feed into a plurality of products including one or more distillates and/or a bottom or residue
Implementation Method 3
separate the feed into a plurality of products including one or more distillates and/or a bottom or residue
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.