Probabilistic Stress Wave Analysis for Varying Machine Conditions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing stress wave analysis techniques require similar operating conditions for data comparison, making it impractical to monitor machine health due to difficulties in imposing uniformity, especially in varying operational environments.
Innovation Solution
A method that generates baseline data by analyzing stress waves under normal conditions and compares it to current data, using probabilistic analysis and delta function calculations to filter out normal operating variations, allowing for accurate assessment of machine status without the need for uniform operating modes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If stress wave data is collected under varying operating conditions, then the system can monitor machine health in practical applications, but the data comparison becomes unreliable due to normal operating variations
Solution Approach 1:
The system dynamically adapts to varying operating conditions by continuously updating baseline data and adjusting monitoring parameters. Instead of requiring static, uniform conditions, the system evolves its reference standards to match actual operating environments, enabling reliable monitoring across diverse conditions while maintaining measurement precision through adaptive normalization.
Solution Approach 2:
The system changes key parameters including collecting stress wave data over extended periods (e.g., 24 hours) to capture full operational cycles, and transforms raw stress wave measurements into normalized metrics that account for operating condition variations. This parameter transformation enables meaningful comparisons despite changes in load, speed, or environmental conditions.
2Reliability
If baseline data is collected over extended periods to capture normal variations, then the system can distinguish actual faults from normal operation, but the monitoring time and resource requirements increase
Solution Approach 1:
The system performs preliminary baseline collection during normal operation before fault conditions develop. By establishing reference data during known healthy operation, the system prepares comparison standards in advance, allowing rapid fault detection without requiring extended monitoring periods when problems occur. The baseline is built proactively during routine operation.
Solution Approach 2:
The system continuously collects and updates baseline data during normal operation, transforming the baseline collection from a discrete, time-consuming task into an ongoing process that occurs naturally during machine operation. This continuous accumulation of reference data improves reliability over time without adding dedicated monitoring time, as the baseline development is integrated into normal operational monitoring.
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 approach enables accurate and practical monitoring of machine health by smoothing out operating changes and noise, providing a reliable indication of current status and remaining useful life, even in non-uniform operating conditions.
Implementation Method 1
a sensor mounted on or in the apparatus for monitoring stress waves generated by the apparatus
Data Source
AI summary
A statistical process, and system for implementing the process, is described for the analysis of stress waves generated in operating machinery or equipment. This technique is called Probabilistic Stress Wave Analysis. The process is applied to a population of individual “feature” values extracted from a digitized time waveform (such as a 2 second Stress Wave Pulse Train, or a 2 month history of Stress Wave Energy). Certain numeric descriptors of the statistical distributions of computed features are then employed as inputs to decision making routines (such as neural networks or simple threshold testing) to accurately classify the condition represented by the original time waveform data, and thereby determine a status of the operating machine/equipment.