SCR Failure Detection via NOx Correlation Factor
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing SCR failure detection systems are prone to incorrect determinations due to NOx sensor degradation caused by gain and offset issues, and regulatory requirements demand no gap between best performing unacceptable (BPU) and BPU NO sensor detection capabilities.
Innovation Solution
A failure detection system that calculates upstream and downstream NOx averages and standard deviations, generates a linear correlation factor using Pearson's correlation, and detects a BPU part by comparing this factor to a preprogrammed threshold, while also considering temperature and mass flow ranges, to provide a malfunction alert.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional SCR failure detection methods using upstream and downstream NOx sensors are used, then SCR malfunction can be detected, but sensor degradation causes incorrect determinations of SCR NOx conversion efficiency
Solution Approach 1:
The patent changes the detection parameter from absolute NOx conversion efficiency to the ratio of downstream to upstream NOx concentrations. This ratio metric is less sensitive to sensor degradation and gain/offset variations, thereby maintaining detection reliability even when sensor precision deteriorates over time
Solution Approach 2:
The patent introduces a correlation coefficient as an intermediary metric between the raw sensor measurements and the final failure determination. This intermediary parameter captures the relationship between upstream and downstream NOx readings without being directly affected by absolute sensor calibration errors
2Reliability
If regulatory requirements demand no gap between BPU and WPA SCR detection capability, then detection sensitivity is maximized, but sensor degradation creates detection gaps
Solution Approach 1:
The patent transforms the detection approach by using the ratio of downstream to upstream NOx concentrations rather than absolute values. This parameter transformation ensures that even degraded sensors can maintain the required detection sensitivity and eliminate gaps between BPU and WPA SCR detection capabilities
Solution Approach 2:
Instead of trying to maintain absolute sensor precision, the patent inverts the approach by using relative measurements (ratios and correlations) that are inherently more robust to sensor degradation, thereby maintaining detection capability without relying on high absolute measurement precision
Data Source
AI summary
Selective catalytic reaction (SCR) failure detection systems and methods for propulsion systems. The method includes obtaining (a) an upstream (of the SCR unit) NOx concentration value, and (b) a downstream NOx concentration value, and caching (a) and (b). The obtaining and caching is repeated until N cached values are obtained, where N is a preprogrammed number. Using the N cached values, an upstream average of NOx, an upstream standard deviation, a downstream average of NOx, and a downstream standard deviation are calculated. The calculated values are input for a failure detection algorithm pertaining to nitrogen oxides (NOx) that generates a linear correlation factor. A best performing unacceptable (BPU) part is detected when the linear correlation factor is greater than a preprogrammed fail threshold.