EGR Cooler Sooting Prediction Using Multi-Sensor Engine Diagnostics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for detecting sooting in exhaust gas recirculation coolers of diesel engines provide only binary statements about clogging, failing to offer information on intermediate sooting states or future system status, which can lead to unexpected engine downtime and emission issues.
Innovation Solution
A method using artificial neural networks and time series analysis to continuously monitor and predict the degree of sooting in exhaust gas recirculation coolers by measuring key variables like EGR actuator position, Venturi differential pressure, and temperature, allowing for early maintenance and emission control.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional binary diagnostic methods are used to detect EGR cooler sooting, then the system structure remains simple, but the measurement precision and information completeness about sooting levels are insufficient
Solution Approach 1:
The patent segments the sooting detection into multiple measurement dimensions by using three different sensors (EGR actuator position, Venturi differential pressure, and temperature sensor) to capture different aspects of sooting progression. This allows precise characterization of sooting levels from 0% to 100% through multiple independent measurements rather than a single binary indicator.
Solution Approach 2:
The patent makes existing sensors serve multiple functions: the EGR actuator position sensor not only controls EGR flow but also indicates sooting level; the Venturi differential pressure sensor not only measures flow rate but also detects cross-section reduction due to sooting; the temperature sensor not only monitors exhaust temperature but also indicates cooling efficiency degradation from sooting.
2Reliability
If conventional binary diagnostic methods are used, then the device complexity remains low, but the ability to predict future system status and plan maintenance is lost
Solution Approach 1:
The patent implements preliminary action by continuously monitoring sooting progression and calculating remaining service life before actual failure occurs. The system predicts when the EGR cooler will reach 100% sooting and blocks the engine proactively, allowing maintenance to be scheduled in advance rather than reacting to sudden failures.
Solution Approach 2:
The patent establishes feedback loops where sensor measurements are continuously compared against reference values from a non-sooty EGR cooler, deviations are calculated and fed back to update sooting level estimates, which then feed back into maintenance scheduling decisions. This closed-loop system enables continuous assessment and prediction of EGR cooler condition.
3Loss of information
If continuous monitoring of multiple parameters is implemented, then the information completeness about EGR cooler condition is improved, but the data processing complexity increases
Solution Approach 1:
The patent extracts only the essential information needed for sooting assessment from the multiple sensor measurements. Instead of processing all raw sensor data, the system calculates three specific deviation metrics (actuator position difference, differential pressure difference, temperature difference) that directly correlate with sooting levels, filtering out unnecessary data processing complexity.
Solution Approach 2:
The patent transforms multiple physical parameters (actuator position, pressure differential, temperature) into a unified sooting level parameter expressed as a percentage from 0% to 100%. This parameter transformation simplifies the interpretation of multiple measurements by consolidating them into a single comprehensible metric that directly indicates sooting severity.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method for detecting and predicting the setting of an exhaust-gas recirculation cooler of a diesel internal combustion engine; wherein reference values are determined on a test bench, wherein the determined reference data is stored in a series-production engine control unit; wherein an actual state description, and a future state description, of the setting of the EGR cooler are determined continuously during the normal, running engine mode using the determined reference values, and wherein the need for servicing or for the EGR cooler to be replaced is indicated using a screen or fault memory.