Gain Adaptive Stepping for Integrating Variable ROC Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing gain adaptive algorithms for control modelers are limited in their ability to effectively manage integrating control variables, leading to inaccuracies, suboptimal model identification, longer commissioning times, and increased likelihood of constraint violations due to the differing response characteristics of integrating control variables compared to self-regulating control variables.
Innovation Solution
Implementing a Rate of Change (ROC) control theory algorithm within the process controller to transform integrating control variables into self-regulating proxies, using gain adaptive stepping based on ROC high and low limits to manage integrating control variables, ensuring they do not exceed constraints and maintaining model accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If existing gain adaptive algorithms are used for control modelers, then general control functionality is maintained, but model accuracy and commissioning efficiency deteriorate when dealing with integrating control variables
Solution Approach 1:
The patent segments the control variable management by distinguishing between self-regulating control variables and integrating control variables. Different gain adaptive algorithms are applied to each segment: standard gain adaptive stepping for self-regulating variables and a specialized ROC-based algorithm for integrating variables. This segmentation allows each type to be handled with the most appropriate method, improving overall model accuracy while maintaining versatility.
Solution Approach 2:
The patent implements dynamic adaptation by automatically selecting different control algorithms based on the type of control variable being managed. The system dynamically adjusts its behavior: using conventional gain adaptive stepping when dealing with self-regulating variables and switching to ROC-based gain adaptive stepping for integrating variables. This dynamic approach enables the system to adapt to different variable characteristics, resolving the contradiction between maintaining general functionality and achieving specialized accuracy.
2Productivity
If conventional control methods are used for integrating control variables, then system simplicity is maintained, but constraint violations increase and commissioning time increases
Solution Approach 1:
The patent introduces Rate of Change (ROC) as an intermediary variable to manage integrating control variables. Instead of directly controlling the integrating variable itself, the system controls its rate of change. This intermediary approach transforms the challenging integrating control problem into a more manageable form that can be handled with gain adaptive stepping, thereby improving constraint compliance while maintaining system efficiency and reducing commissioning time.
3Measurement precision
If gain adaptive stepping is applied to integrating control variables without ROC transformation, then algorithm simplicity is maintained, but control performance and model identification quality deteriorate
Solution Approach 1:
The patent changes the control parameter from the integrating control variable itself to its rate of change (ROC). This parameter transformation converts the integrating variable into a self-regulating proxy, allowing the use of gain adaptive stepping algorithms that were originally designed for self-regulating variables. The parameter change enables improved model identification accuracy while keeping the algorithm structure relatively simple and reusable.
Data Source
AI summary
A system includes processor(s) configured to: receive a rate of change process variable, a rate of change high limit, and a rate of change low limit for an integrating control variable of a process from a process controller used to control the process; determine whether a constraint violation has occurred on integrating control variable of process based on rate of change process variable, rate of change high limit, and rate of change low limit; perform gain adaptive stepping on integrating control variable using rate of change process variable, rate of change high limit, and rate of change low limit; determine at least a first set point for at least a first manipulated variable corresponding to the integrating control variable based on the gain adaptive stepping previously performed; and output the at least the first set point for the at least the first manipulated variable to adjust the process.