Analog Sensor Fault Detection via Slope and Statistical Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for detecting failures in analog sensors, particularly in aerospace, are inadequate as they fail to detect certain failure modes that do not result in sensor signals exceeding thresholds, leading to potential equipment damage and safety risks.
Innovation Solution
A method that acquires and analyzes sensor data to detect discontinuities, exceedances of predefined thresholds, calculates slopes, and performs statistical evaluations to identify anomalies, generating a signal representative of sensor failure without requiring a reference sensor or signal, using standard deviation, form factor, and signal-to-noise ratio calculations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If threshold-based monitoring is used to detect sensor failures, then the detection method is simple and economical, but certain failure modes cannot be detected when the sensor signal does not exceed the threshold
Solution Approach 1:
The monitoring method is segmented into multiple independent detection techniques: threshold violation detection, discontinuity detection, slope calculation, and statistical evaluation. Each technique targets specific failure modes, and their combination provides comprehensive coverage without requiring a single complex system
Solution Approach 2:
The monitoring system performs multiple functions using the same sensor data: it detects threshold violations, identifies discontinuities, calculates slopes, and conducts statistical analyses. This multi-functional approach enables detection of various failure modes without adding separate sensors or reference signals
2Reliability
If statistical evaluation methods are used to detect sensor failures, then comprehensive monitoring is achieved, but the computational complexity and processing requirements increase
Solution Approach 1:
The system applies statistical evaluations selectively rather than continuously. Discontinuities are detected by comparing consecutive measurements, and slopes are calculated only when necessary. This partial application reduces computational burden while maintaining detection accuracy
Solution Approach 2:
The monitoring system changes parameters dynamically: it adjusts monitoring intensity based on operational conditions, uses different statistical measures (standard deviation, form factor, signal-to-noise ratio) depending on the situation, and modifies threshold levels adaptively to optimize detection while minimizing processing requirements
Data Source
Figure 1~4
Figure 2
Figure 3
AI summary
The method involves detecting discontinuity in measurements by an analog sensor, and detecting exceedance of predefined operating threshold from measured values. Slopes are calculated in evolution of the values and exceedance of predefined slope threshold is detected from the slopes. Statistical evaluations are performed on the measurements for detecting operation anomaly from the evaluations. Verification is made if one of situations defined in one of detection, calculation and evaluation steps is proved and a signal representative of failure of sensor is generated if the situation is proved. Independent claims are also included for the following: (1) a device for detecting a failure of an analog sensor, comprising an indicator unit (2) an aircraft comprising a display device for displaying state of sensor.