Valve Vibration Diagnosis Using Autoregressive Change Rates
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing valve diagnosis methods face challenges such as difficulty in attaching sensors to valve stems, need for expensive drive unit sensors, indirect vibration estimation, and requirement for structural analysis models, which lead to inaccurate predictions and increased costs.
Innovation Solution
A valve diagnosis method using an autoregressive model to calculate rates of change in vibration data, allowing for non-destructive monitoring and prediction of valve state changes without the need for structural analysis models or expensive sensors, enabling timely maintenance measures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If an acceleration sensor is attached to the valve stem to detect vibration, then valve stem damage can be diagnosed, but it becomes difficult to attach the sensor when the valve stem is exposed to high temperature or located inside the valve
Solution Approach 1:
The patent uses the valve body itself as an intermediary structure to mount the acceleration sensor. Instead of attaching the sensor directly to the valve stem in harsh environments, the sensor is placed on the valve body which serves as a mediator to detect vibrations that propagate through the valve assembly, enabling indirect but effective monitoring of valve stem conditions.
Solution Approach 2:
The patent replaces direct mechanical attachment of sensors to the valve stem with an alternative approach using the valve body structure. The vibration detection system substitutes direct stem monitoring with body-mounted sensing that captures vibration characteristics through the valve assembly, avoiding the need for direct high-temperature or confined-space sensor installation.
2Reliability
If a permanent driving force sensor is attached to the drive unit to predict deterioration, then valve state can be monitored, but the drive unit becomes expensive
Solution Approach 1:
The patent enables the valve system to monitor its own state using vibration sensors that detect operational characteristics. The system uses its own operational vibrations as the monitoring signal, eliminating the need for separate expensive driving force sensors in the drive unit, thereby achieving cost-effective self-diagnosis.
Solution Approach 2:
The patent replaces expensive mechanical driving force sensors with vibration-based detection using acceleration sensors. By substituting direct force measurement with vibration analysis, the system achieves reliable valve state monitoring without the high cost of permanent drive unit instrumentations.
3Measurement precision
If diagnosis is performed using intermediate torque data, then valve can be diagnosed, but the valve cannot be diagnosed without moving the valve
Solution Approach 1:
The patent enables continuous vibration monitoring of the valve during normal operation. The acceleration sensor continuously captures vibration data as the valve operates, allowing real-time diagnosis without requiring valve movement or shutdown, thus maintaining continuous productive operation while monitoring valve health.
Solution Approach 2:
The patent replaces torque-based diagnosis requiring valve movement with vibration-based diagnosis using acceleration sensors. The vibration measurement system substitutes mechanical torque measurement, enabling diagnosis during normal valve operation without requiring actuation or movement, thus eliminating downtime.
4Ease of operation
If fluid pressure is measured to indirectly evaluate valve vibration, then monitoring can be performed, but the vibration is not accurately evaluated and structural analysis models are required
Solution Approach 1:
The patent replaces indirect fluid pressure-based vibration estimation with direct vibration measurement using acceleration sensors. The sensor directly detects mechanical vibrations on the valve body, substituting the complex pressure-to-vibration inference process with direct physical measurement, thereby eliminating the need for structural analysis models while achieving accurate vibration evaluation.
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
This method effectively predicts valve state changes, allowing for proactive maintenance and reducing unnecessary replacements and downtime, while providing accurate and cost-effective monitoring of valve conditions.
Implementation Method 1
a vibration signal from the acceleration sensor (14) is received to output vibration data (6)
Data Source
AI summary
A diagnosis method is provided for appropriately predicting a state change of a valve with a simple configuration. The valve diagnosis method comprises: calculating a value of a first rate of change, with respect to a predetermined damage index during reference data acquisition, of the predetermined damage index during diagnosis; calculating a value of a second rate of change, with respect to a root mean square value calculated from a difference between predicted waveform data of vibration during reference data acquisition and vibration data serving as reference, of a root mean square value calculated from a difference between predicted waveform data of vibration during diagnosis and vibration data during diagnosis; in a plane, arranging a plot of a diagnostic result based on a value of the first rate of change being calculated and a value of the second rate of change being calculated; and determining a state of a valve to be diagnosed based on a value of the first rate of change being calculated, a value of the second rate of change being calculated, and an arrangement position on the plane of a plot of the diagnostic result.