Redundant Signal Path Filter Testing for Hidden Fault Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Detecting failures in redundant signal paths within industrial control systems is challenging, especially when the input signal remains constant, as existing methods rely on signal changes to identify faulty components, potentially allowing failed components to go undetected for extended periods.
Innovation Solution
A method and system that generate a test signal at a predefined frequency and inject it into both signal paths to measure and compare the frequency responses of filters in each path, identifying a failure when the difference exceeds a predefined threshold, allowing for precise identification of faulty components without relying on signal changes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional fault detection methods relying on signal changes are used, then the system can detect faults when signals vary, but failed components remain undetected during extended periods of constant signal values
Solution Approach 1:
The system performs preliminary fault detection by comparing frequency responses of redundant signal paths using test signals before actual operation. This allows detection of component failures that would otherwise remain hidden during constant signal conditions, as the frequency response comparison reveals path differences independent of input signal variations.
Solution Approach 2:
A test signal acts as an intermediary to excite the signal paths and reveal faults. By injecting a known test signal and measuring the frequency response, the system can detect component failures without relying on natural signal changes from the process being monitored. The test signal mediates between the constant process signal and the fault detection requirement.
2Reliability
If redundant signal paths are implemented for high SIL ratings, then the system can detect differences between paths, but the specific location of faults cannot be precisely identified
Solution Approach 1:
The fault detection process is segmented into distinct measurement steps: injecting test signals at specific frequencies, measuring frequency responses at different points in the signal path, and comparing responses to identify the location of faults. This segmentation allows precise identification of which component (filter, amplifier, etc.) is faulty by isolating frequency response characteristics to specific segments of the signal path.
Solution Approach 2:
The system uses frequency response characteristics as a signature or fingerprint for each signal path component. By analyzing how different frequency components are affected, the system can identify the specific location and nature of faults, similar to how color changes can indicate different conditions. Each component has a characteristic frequency response that reveals its health status.
Data Source
AI summary
A system and method for detecting a failure in a redundant signal path during operation of the redundant path is disclosed. A test signal is sequentially injected into each signal path while an input signal is conducted by the other signal path not receiving the test signal. The test signal is selected at a frequency to verify operation of a filter connected in series along each path. A processor generates the test signal, injects the test signal at the input of the filter, and receives the output of the filter. The processor then generates a frequency response of the filter in each signal path as a function of the output from the filter and of the original test signal. The frequency response obtained along each of the redundant signal paths is compared to each other to detect a failure of one of the filters present along the respective signal paths.