Spectral Vibration Monitoring for Adaptive Fault Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current equipment monitoring methods for rotating machinery are inefficient due to the need for tedious and time-consuming threshold setting for detecting equipment deterioration or damage, as they rely on simple vibration indices and lack systematic methods for selecting detection thresholds that vary with load, speed, and other operating parameters.
Innovation Solution
An apparatus and method that processes temporal sensor data to determine spectral sensor data, using statistical algorithms like Hotelling's statistic and Squared Prediction Error to identify frequency ranges of interest, enabling unsupervised monitoring and detection of abnormal vibrations in rotating equipment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If simple vibration indices and predefined thresholds are used for equipment monitoring, then the monitoring system is simple to implement, but the detection accuracy and reliability are insufficient due to lack of systematic threshold selection
Solution Approach 1:
The patent transforms the monitoring approach by changing from simple time-domain vibration indices to frequency-domain spectral analysis. By converting vibration signals to spectral data and analyzing frequency components, the system achieves more reliable detection of equipment deterioration while maintaining practical implementability through automated statistical thresholding.
Solution Approach 2:
The patent replaces manual threshold setting with automated statistical algorithms. Instead of relying on operators to define thresholds based on experience, the system uses statistical process control algorithms to automatically determine thresholds from historical data, improving both reliability and reducing operational complexity.
2Measurement precision
If advanced signal processing methods with specific fault indices are implemented, then the detection precision is improved, but the implementation becomes tedious and time-consuming due to the need to define thresholds for each equipment type
Solution Approach 1:
The patent implements self-service through automated statistical thresholding. The system automatically calculates thresholds based on historical spectral data and statistical process control principles, eliminating the need for manual threshold definition by operators for each equipment type. This maintains high detection precision while dramatically reducing implementation time.
Solution Approach 2:
The patent creates a universal monitoring framework that can be applied to various rotating equipment types (pumps, compressors, motors, etc.) using the same statistical process control algorithms. The system processes spectral data from different equipment types through a unified approach, making the precise detection method universally applicable without equipment-specific threshold calibration.
3Ease of operation
If conservative predefined thresholds are used for vibration monitoring, then the system is easy to operate, but the productivity is reduced due to false alarms and inability to detect early deterioration
Solution Approach 1:
The patent implements feedback through statistical process control algorithms that continuously learn from historical spectral data. The system automatically updates thresholds based on accumulated data and equipment behavior patterns, providing feedback that improves detection accuracy over time while maintaining ease of operation through automated adjustment without manual intervention.
Solution Approach 2:
The patent introduces dynamics by making thresholds adaptive rather than static. Instead of using fixed conservative thresholds, the system dynamically adjusts thresholds based on statistical analysis of historical data and current operating conditions, enabling early detection of deterioration while reducing false alarms and improving monitoring efficiency.
Data Source
Figure 1~2
Figure 3
AI summary
The present invention relates to an apparatus for equipment monitoring. The apparatus comprises an input unit, a processing unit, and an output unit. The input unit is configured to provide the processing unit with a plurality of batches of temporal sensor data for an item of equipment. Each batch of temporal sensor data comprises a plurality of temporal sensor values as a function of time. The processing unit is configured to process the plurality of batches of temporal sensor data to determine a plurality of batches of spectral sensor data. Each batch of spectral sensor data comprises a plurality of spectral sensor values as a function of frequency. The processing unit is configured to implement at least one statistical process algorithm to process the plurality of spectral sensor values for the plurality of batches of spectral sensor data to determine a plurality of index values. For each batch of spectral sensor data there is an index value determined by each of the statistical process algorithms. Each statistical process algorithm has an associated threshold value, and the processing unit is configured to utilise the at least one threshold value and the plurality of index values to determine a batch of spectral sensor data of interest that has an index value greater than the threshold value for the associated statistical process algorithm. The processing unit is configured to determine a frequency range of interest on the basis of the plurality of spectral sensor values for the batch of spectral sensor data of interest.