Valve Friction Monitoring via Segmented Hammerstein Model
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for monitoring control loops in industrial plants are inefficient due to high computing complexity and unreliable friction estimation in valves, which affects control quality and requires invasive and resource-intensive diagnostics.
Innovation Solution
A diagnostic device and method that divides the estimation of the Hammerstein model into simpler subproblems, using a setpoint jump for linear submodel estimation and a constant setpoint for nonlinear submodel estimation, with reduced computing complexity and non-invasive data-based monitoring, allowing for real-time analysis without disrupting plant operations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional Hammerstein model estimation methods are used for friction monitoring, then comprehensive friction parameters can be obtained, but computing complexity becomes too high for real-time application
Solution Approach 1:
The patent segments the complex Hammerstein model estimation into two separate submodel estimations: a linear dynamic submodel estimated during transient phases and a static nonlinear submodel estimated during steady states. This segmentation reduces computing complexity by avoiding simultaneous estimation of all parameters, while still achieving comprehensive friction monitoring through sequential analysis of both submodels.
2Reliability
If invasive diagnostic methods are used to monitor valve friction, then reliable friction data can be obtained, but plant operations are disrupted and resource consumption increases
Solution Approach 1:
The diagnostic system uses the control loop's own operational data (manipulated variables and process variables already being measured for control purposes) to estimate friction parameters. No additional sensors, test equipment, or invasive measurements are required - the system serves itself using existing data streams, thereby maintaining plant productivity while achieving reliable friction monitoring.
Solution Approach 2:
The patent enables continuous friction monitoring by processing control data in real-time during normal plant operations. The estimation occurs continuously as new control data becomes available, without requiring shutdowns or interruption of plant processes, thus maintaining both reliability of monitoring and productivity of plant operations.
3Measurement precision
If comprehensive model estimation is performed continuously, then accurate friction parameters are available, but computational resources are wasted during phases when friction behavior is not informative
Solution Approach 1:
The system performs model estimation periodically based on the operational phase rather than continuously. The linear submodel is estimated during transient phases when the process is changing, and the static nonlinear submodel is estimated during steady states. This periodic, phase-based estimation approach ensures accurate parameter estimation only when the data is informative, avoiding waste of computational energy during inappropriate phases.
Data Source
AI summary
A diagnostic device and diagnostic method for monitoring the operation of a control loop with a controlled system having a valve as an actuator, wherein in the case of a substantially stepped profile of a setpoint, a linear submodel is identified, which is subsequently used, with a substantially constant setpoint, in order to identify a nonlinear submodel, which is arranged upstream of the linear submodel in a Hammerstein model for the controlled system so as to facilitate a quantitative assessment of the friction behavior of a valve with a comparatively low level of computing complexity.