Predicting Windup in Industrial Process Control Systems

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional process control systems face challenges in predicting and managing windup states in industrial processes due to user-specified limits becoming stale, leading to inconsistent and unpredictable behavior, particularly when actual physical limits are exceeded, causing oscillations and difficulties in maintaining setpoints.

Innovation Solution

The method involves identifying regions where the output value of a downstream controller falls and calculating achievable manipulated variable limits based on these regions, allowing for more accurate prediction and control by tracking actual physical limits rather than user-specified limits, thereby preventing windup and improving optimization solutions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If user-specified limits are used for manipulated variable control, then the control system operates within defined boundaries, but the limits become stale and obsolete over time leading to windup states

Engineering Contradiction:
Improvecontrol stabilityVSAvoidlimit accuracy
Core Design Contradiction:
ReliabilityVSLoss of information

Solution Approach 1:

The system continuously monitors the actual physical limits of manipulated variables and feeds this information back to update the achievable limits. This feedback mechanism ensures that the control system uses current, accurate limit information rather than stale user-specified limits, preventing windup states while maintaining operational reliability.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The system proactively determines achievable manipulated variable limits based on actual physical constraints before windup conditions occur. By calculating and applying these achievable limits in advance, the system prevents the downstream controller from entering windup states, thereby maintaining control stability without waiting for limit obsolescence to manifest.

Inventive Principle:
Principle #10Preliminary action

2Productivity

If the MPC controller pushes manipulated variables towards user-specified limits, then optimization solutions are achieved, but the downstream controller hits actual physical limits and enters windup state

Engineering Contradiction:
Improveoptimization efficiencyVSAvoidcontroller stability
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The system dynamically adjusts the achievable manipulated variable limits based on the actual physical state of the process. Rather than using fixed user-specified limits, the system continuously updates the achievable limits to reflect current physical constraints, allowing the MPC controller to optimize while respecting real-time physical boundaries, thus preventing windup and maintaining stability.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system changes the limit parameters from static user-specified values to dynamic achievable limits that are continuously updated based on actual physical measurements. This parameter transformation allows the optimization algorithm to work with realistic constraints, improving both productivity by avoiding suboptimal conservative limits and reliability by preventing windup conditions.

Inventive Principle:
Principle #35Parameter changes

3Manufacturing precision

If downstream controller operates close to windup state, then manipulated variable reaches physical limits, but process variable drifts away from setpoint causing prediction difficulties

Engineering Contradiction:
Improvesetpoint accuracyVSAvoidprediction accuracy
Core Design Contradiction:
Manufacturing precisionVSDifficulty of detecting and measuring

Solution Approach 1:

The system takes preliminary action to prevent the downstream controller from approaching the windup state by enforcing achievable manipulated variable limits based on actual physical constraints. By acting before windup occurs, the system maintains the process variable close to the setpoint and preserves the accuracy of predictions regarding the effects of manipulated variables on controlled variables.

Inventive Principle:
Principle #9Preliminary anti-action

Data Source

PatentEP2751628B1Apparatus and method for predicting windup and improving process control in an industrial process control system
Publication Date: 2019.11.20 HONEYWELL INTERNATIONAL INC
  • EP2751628B1 patent drawingFigure 1
  • EP2751628B1 patent drawingFigure 2~3
  • EP2751628B1 patent drawingFigure 4A~4B

AI summary

A method includes identifying one of multiple regions in a range where an output (OP) value used to implement a manipulated variable is located. The manipulated variable is associated with an industrial process, and the OP value represents an output of a downstream controller. The method also includes calculating an achievable manipulated variable (MV) limit for the manipulated variable based on the region in which the OP value is located. For example, when the OP value is located in one region, the achievable MV limit could match a user-specified limit or be based on a gain between the OP value and a value of a process variable. When the OP value is located in another region, the achievable MV limit could track the value of the process variable with a gap.