Valve Stiction Detection Using Bispectral Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for detecting valve stiction in process control loops are invasive, costly, and ineffective, particularly for non-invasive detection and quantification, especially in loops with integrators or compressible fluids, and often produce false positives due to sinusoidal disturbances.
Innovation Solution
A non-invasive, data-based method using normalized bispectrum and bicoherence analysis to detect non-linearity in control loops, filtering signals to identify limit cycles, and employing clustering techniques to quantify stiction without requiring valve positioner data, suitable for both online and offline automated implementation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If invasive valve travel test is used to detect stiction, then detection accuracy is improved, but operational disruption and cost increase
Solution Approach 1:
The patent replaces mechanical/invasive testing methods with signal processing and data analysis techniques. By analyzing the relationship between process variable (PV) and controller output (OP) signals using cross-correlation and spectral analysis, the system detects stiction without physically manipulating the valve, thereby eliminating operational disruption while maintaining detection accuracy
Solution Approach 2:
The patent introduces signal processing algorithms as an intermediary between the control system and stiction detection. Instead of directly testing the valve mechanically, the system uses PV and OP signals as intermediaries to infer stiction conditions through mathematical analysis, avoiding direct intervention in valve operation
2Ease of operation
If Horch's cross-correlation method is used for non-invasive stiction detection, then operational disruption is reduced, but applicability is limited to flow control loops and produces false positives
Solution Approach 1:
The patent changes the analysis parameters from simple cross-correlation to spectral analysis and frequency domain methods. By examining the spectral characteristics and frequency relationships between PV and OP signals, the system achieves broader applicability across different control loop types (including level, pressure, and temperature controls) while reducing false positives caused by sinusoidal disturbances
Solution Approach 2:
The patent creates a universal detection method that works across multiple control loop types and operating conditions. The spectral analysis approach can identify stiction in flow, level, pressure, and temperature control loops, making the system adaptable to various process conditions without requiring method modification
3Ease of operation
If traditional non-invasive methods are used, then operational disruption is minimized, but quantification capability is lost
Solution Approach 1:
The patent transitions from one-dimensional time-domain analysis to two-dimensional frequency-domain analysis. By examining both the magnitude and phase relationships across different frequencies, the system can not only detect stiction but also quantify its severity based on the spectral characteristics and energy distribution in the frequency domain
Data Source
AI summary
An effective non-intrusive data-based monitoring method may reduce the cost of control loop performance maintenance by screening and short-listing those control loops or valves that need maintenance. The invention comprises a data-based, model-free, non-invasive method that can automatically detect and quantify stiction present in control valves. The method does not require the implementation of any additional valve travel test or, as commonly known, bump test of the control loop. The method may detect and quantify stiction using controlled variable (pv), controller output (op) and set point (sp) data. It does not require valve positioner (mv) data.