Vibration Sensor Fault Detection via Spectral Coherence Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for monitoring vibration sensors in mechanical systems require manual intervention, system shutdown, or additional surveillance equipment, leading to inefficiencies and increased costs, especially when dealing with numerous sensors.
Innovation Solution
A method involving grouping vibration sensors receiving characteristic frequency signals via equivalent propagation paths, synchronously acquiring and analyzing vibration signals to calculate spectral coherence coefficients, and constructing a coherence matrix to identify potentially defective sensors without manual intervention or additional equipment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual testing of vibration sensors is performed by dismantling and applying known stress, then sensor functionality can be verified, but the system must be shut down and technician intervention is required
Solution Approach 1:
The system performs self-diagnosis by using its own operational vibration signals to monitor sensor health, eliminating the need for external technician intervention and system shutdown. The vibration signals naturally present during system operation are sufficient to detect sensor faults without requiring external test equipment or manual dismantling.
Solution Approach 2:
The method continuously monitors sensor health during normal operation before actual failure occurs, enabling early detection of degradation trends. This preliminary monitoring allows for proactive maintenance scheduling while the system is still operational, rather than waiting for complete failure.
2Reliability
If electronic testing is implemented with integrated monitoring means, then continuous monitoring is possible, but system complexity and cost increase significantly
Solution Approach 1:
The existing vibration signal acquisition system serves dual purposes: both for normal system monitoring and for sensor health diagnosis. The same sensors that measure system vibrations are used to detect their own faults, eliminating the need for separate test signal generation equipment and response analysis systems.
Solution Approach 2:
The vibration signal processing chain is made multi-functional by extracting diagnostic information about sensor health from the same signals used for system monitoring. This universal use of existing infrastructure avoids adding dedicated test equipment while enabling continuous sensor health assessment.
3Measurement precision
If traditional monitoring methods are used without sophisticated analysis, then implementation is simple, but measurement precision and fault detection capability are insufficient
Solution Approach 1:
The method replaces complex mechanical test setups with signal processing analysis. Instead of applying physical test stresses and measuring mechanical responses, the system uses spectral coherence analysis of existing vibration signals to detect sensor faults, substituting mechanical testing with computational methods.
Solution Approach 2:
The approach transforms the vibration signals from time-domain waveforms to frequency-domain spectra through Fourier transformation. This parameter transformation enables the use of spectral coherence metrics to detect sensor faults with high precision, leveraging frequency domain characteristics that are not apparent in raw time-domain signals.
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
Enables reliable and cost-effective monitoring of vibration sensors by using existing vibration signals, allowing for the detection of faulty sensors without system shutdown or additional surveillance means, improving operational efficiency and reducing maintenance costs.
Implementation Method 1
Vibration sensors, such as accelerometers or microphones, are generally installed on these systems in order to monitor their operation, by measuring these vibration signals
Implementation Method 2
one calculates for each pair of vibration sensors of this group G i a spectral coherence coefficient of the spectra of the vibration signal received by this pair of vibration sensors
Data Source
Figure 1~4
AI summary
The method involves providing a set of vibratory sensors (1a-1f) such as accelerometers, that receive vibratory signals (S1, S2). The vibratory signals are provided with respective characteristic frequencies of a set of groups (G1, G2) of sensors. A step of synchronous acquisition of the vibratory signals received by each vibratory sensor of the groups is carried out. A number of sensors that is probably defective for each group is calculated. An analysis of each group is compiled so as to confirm the vibratory sensors considered as defective for a mechanical system (5) e.g. aircraft.