Reactive Absorption Soft Sensors for Parameter Estimation Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Reactive absorption units face challenges in efficiently estimating and controlling operating parameters, particularly in the removal of acidic gases like CO2 and H2S, due to incomplete measurement data and energy inefficiencies.
Innovation Solution
A method utilizing machine learning techniques, specifically 1D convolutional neural networks (1D-CNN) and kinetic models, to estimate missing parameters and optimize operations by predicting temperature profiles and adjusting process variables, enabling closed-loop or open-loop control.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional measurement and control methods are used in reactive absorption units, then the system structure is simple, but the estimation of operating parameters is incomplete and control precision is poor
Solution Approach 1:
The patent replaces traditional mechanical measurement systems with a virtual measurement system based on machine learning models. The system uses trained neural network models to estimate unmeasured operating parameters (such as amine loading, temperature profiles, gas composition) from available process data, eliminating the need for complex physical measurement devices while achieving high estimation accuracy.
Solution Approach 2:
The patent introduces machine learning models as intermediary components between available process measurements and required operating parameters. These models act as virtual sensors that translate readily available process data into accurate estimates of difficult-to-measure parameters, bridging the gap between simple measurements and complex process understanding.
2Reliability
If more complete measurement data is collected to improve control, then measurement precision improves, but device complexity and cost increase
Solution Approach 1:
The patent substitutes physical measurement devices with virtual measurement models. Instead of installing additional sensors and measurement equipment to obtain complete process data, the system uses machine learning models to virtually measure unmeasured parameters from existing process data, maintaining control reliability while avoiding increased hardware complexity.
Solution Approach 2:
The patent creates virtual copies of physical measurement capabilities through machine learning models. These models learn from historical process data and replicate the function of physical sensors, providing estimated values for parameters that would be difficult or expensive to measure directly, thereby achieving complete process understanding without complete physical measurement systems.
3Productivity
If energy consumption is increased to improve absorption efficiency, then productivity improves, but energy costs increase
Solution Approach 1:
The patent implements model-based feedback control that uses machine learning predictions to optimize energy consumption. The system continuously estimates operating parameters and uses this information to adjust process conditions (such as amine circulation rate, stripper temperature, gas flow rate) to maintain optimal absorption performance while minimizing energy usage, achieving a balance between productivity and energy efficiency.
Solution Approach 2:
The patent uses machine learning to identify optimal operating parameter settings that maximize absorption efficiency while minimizing energy consumption. By analyzing relationships between process parameters and performance outcomes, the system dynamically adjusts operating conditions to achieve high productivity at lower energy costs, avoiding the simple approach of increasing energy input to improve throughput.
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
Enhances the efficiency of reactive absorption units by maximizing throughput and minimizing energy consumption while preventing gas breakthrough, thus improving operational robustness and reducing energy costs.
Implementation Method 1
An aqueous (typically amine) solution is circulated within the column and acts as an absorbent. The amine solution absorbs the acidic gases from the gas stream, effectively removing them.
Implementation Method 2
Reactive absorption units, particularly amine absorption, are a common technology used in chemical engineering and the oil and gas industry for the removal of acidic gases
Implementation Method 3
In this unit, heat is applied to release the captured gases, regenerating the amine solution for reuse. This heat driven process is called stripping or desorption.
Data Source
AI summary
A system and method for estimating a parameter for a reactive absorbance unit are provided. An exemplary method includes creating a kinetic model of an absorbance process, setting a range for each of a plurality of input parameters, based, at least in part, on operational data measured from the reactive absorbance unit. A sampling technique is used to generate a plurality of input vectors in the range of each of the plurality of input parameters. A plurality of output vectors is generated from the plurality of input vectors. A predictive model is trained with the plurality of output vectors and the plurality of input vectors. The parameter is estimated from the predictive model. The parameter is used in a control model for the reactive absorbance unit.