Reactive Absorption Control Using ML for Throughput and Energy Balance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Reactive absorption units face inefficiencies in controlling acid gas removal processes due to unmeasured parameters and imbalances, leading to suboptimal throughput and energy consumption, with potential gas breakthroughs.
Innovation Solution
Implementing a method that uses machine learning techniques, specifically 1D convolutional neural networks (1D-CNN) and kinetic models, to estimate missing parameters, reconcile data imbalances, and optimize control parameters for reactive absorption units, allowing for closed-loop or open-loop system control.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional control methods are used for reactive absorption units, then the system operation is simple, but the throughput is suboptimal and energy consumption is high
Solution Approach 1:
The patent implements a closed-loop control system that continuously monitors operating parameters (temperatures, pressures, flow rates) and uses this feedback to adjust control variables in real-time, optimizing throughput while minimizing energy consumption. The system compares actual measurements with model predictions and automatically adjusts operations to maintain optimal performance.
Solution Approach 2:
The optimization system dynamically adjusts operating parameters such as amine circulation rate, reboiler duty, and column pressures based on real-time conditions and economic objectives. By continuously changing these parameters to optimal values, the system maximizes throughput while minimizing energy consumption, resolving the contradiction between productivity and energy use.
2Measurement precision
If more measurements and monitoring are implemented to improve control accuracy, then the measurement precision increases, but the device complexity increases
Solution Approach 1:
The patent introduces a process model as an intermediary that reconciles measurements from multiple sources and estimates unmeasured parameters. This model acts as a mediator between raw sensor data and control decisions, improving measurement precision by filling in missing information without requiring direct physical measurements for every parameter, thus avoiding excessive device complexity.
Solution Approach 2:
The system replaces physical measurement devices for unmeasured parameters with virtual measurements generated by the process model. Instead of installing additional sensors and measurement equipment (which would increase device complexity), the system uses mathematical models and data reconciliation to estimate parameters like acid gas loading and tray temperatures, achieving high measurement precision through computational methods.
3Productivity
If real-time optimization calculations are performed to maximize throughput, then the productivity increases, but the computational requirements and system complexity increase
Solution Approach 1:
The system performs preliminary actions by pre-calculating optimal operating trajectories and control strategies based on economic objectives and process constraints. The model predictive controller uses pre-developed process models and optimization algorithms to determine optimal control moves in advance, enabling real-time throughput maximization without requiring complex computational resources during actual operation.
Solution Approach 2:
The optimization system is designed to be dynamic and adaptive, adjusting its computational effort and optimization frequency based on process conditions and changes. The system dynamically recalculates optimal operations only when necessary, balancing throughput maximization with computational efficiency, thereby managing system complexity while maintaining high productivity.
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 operational efficiency by maximizing throughput and minimizing energy consumption while preventing gas breakthroughs, thereby improving the robustness and performance of reactive absorption units.
Implementation Method 1
a reboiler to provide heat energy to the stripper
Implementation Method 2
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
Implementation Method 3
the reboiler includes a steam control valve to adjust a duty cycle of the reboiler
Data Source
AI summary
A system and method for controlling a reactive absorbance unit are provided. An exemplary method includes obtaining operating data for the reactive absorbance unit, reconciling data imbalances, and estimating unmeasured parameters. An optimization calculation is performed, and control parameters are adjusted based on the optimization calculation.