Three Input Voting Algorithm for Fault Isolation and Output Smoothing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing high integrity sensing systems face challenges in accurately identifying and isolating faulty input channels, especially in unstable environments or with significant offset and cyclic noise, using median value selection and limit test approaches, which require complex data processing and additional test logic.
Innovation Solution
A method for input channel voting that sorts three input values by magnitude, calculates differences, and compares them to a failure threshold, using two comparison tests and one failure threshold to determine intermediate values, which are then averaged to produce an output, simplifying fault tolerance and smoothing outputs with mid-value biasing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If median value selection is used, then fault isolation capability is improved, but output stability deteriorates due to step changes from offset differences and cyclic noise
Solution Approach 1:
The patent implements feedback by continuously monitoring the difference between the middle value and the average of all three inputs. When this difference exceeds a threshold, it indicates noise or offset issues, and the system switches to averaging mode to smooth the output. This feedback mechanism allows the system to adapt dynamically between fault detection mode and stability mode.
Solution Approach 2:
The patent makes the voting algorithm dynamic by allowing it to switch between two operational modes: median-based mode for fault isolation and average-based mode for stability. The mode selection is determined by the current signal conditions (difference between middle value and average), enabling the system to adapt its behavior based on real-time data characteristics.
2Measurement precision
If limit test and average approach is used, then measurement precision is improved, but device complexity increases due to additional test logic and higher failure thresholds
Solution Approach 1:
The patent extracts and eliminates the complex limit test logic from the conventional approach. Instead of performing multiple limit tests and determining failure channels through complex comparison logic, the system simply calculates the difference between the middle value and the average, and uses a single threshold to determine whether to use median or average mode. This extraction of essential functionality reduces complexity while maintaining precision.
Solution Approach 2:
The patent creates a universal voting algorithm that handles both fault isolation and noise reduction cases through a single unified approach. By using the difference between the middle value and average as a universal indicator of signal quality, the system can adapt to different failure modes and noise conditions without requiring separate specialized logic for each scenario.
3Measurement precision
If complex data processing is used for stable environments, then measurement precision is improved, but processing time increases
Solution Approach 1:
The patent applies partial action by performing only the necessary calculations based on signal conditions. Instead of always performing complex multi-test logic, the system performs simple median calculation when in fault isolation mode and simple averaging when in stability mode. This partial execution of processing operations reduces time loss while maintaining precision through conditionally appropriate algorithms.
Data Source
AI summary
An embodiment described herein includes a method for input channel voting. The method includes steps for sorting the three input values by magnitude into a high input value, a middle input value, and a low input value. The three input values are then tested. The testing includes subtracting the middle input value form the high input value to calculate a first difference, and comparing an absolute value of the first difference to a failure threshold. A first intermediate value is determined based on the comparison of the first difference to the failure threshold. The testing also includes subtracting the middle input value from the low input value to calculate a second difference, and comparing the second difference to the failure threshold. A second intermediate value is determined based on this comparison. The first intermediate value and second intermediate value are averaged to produce an output based on the three input values.