Engine Sensor Mismatch Correction via Temperature Uniformity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The accuracy of fault diagnosis in internal combustion engines is degraded due to mismatches in detection characteristics between first and second water temperature sensors, leading to potential erroneous diagnoses.
Innovation Solution
A fault diagnostic system for internal combustion engines that includes a learning unit to correct temperature differences between sensors when stable, using a cooling device with a radiator and cooling water passage, and a diagnostic unit to perform fault diagnosis based on corrected temperature data, while preventing erroneous learning and correction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If fault diagnosis is performed using detected values from first and second water temperature sensors, then fault detection capability is provided, but accuracy in fault diagnosis is degraded due to mismatch between detection characteristics of the sensors
Solution Approach 1:
The patent changes the parameter being measured from raw sensor detected values to corrected values that compensate for sensor mismatch. The ECU learns the difference between detected values when cooling water temperature is uniform (engine stopped) and applies this correction to fault diagnosis, thereby improving measurement precision and diagnosis accuracy simultaneously
Solution Approach 2:
The patent implements a feedback mechanism where the ECU continuously monitors the difference between first and second sensor detected values, learns the mismatch characteristic, and feeds this information back to correct future measurements. This closed-loop approach ensures that diagnosis accuracy is maintained despite sensor characteristic variations
2Measurement precision
If learning of sensor difference is performed continuously, then sensor mismatch correction is improved, but erroneous learning due to external disturbances increases
Solution Approach 1:
The patent performs the learning action preliminarily under specific conditions (when engine is stopped and cooling water temperature is uniform) before actual operation. This ensures that the learned difference value accurately represents sensor mismatch without being contaminated by operational disturbances, thereby preventing erroneous learning
Solution Approach 2:
The patent takes preliminary anti-action by setting specific learning conditions (engine stopped, temperature uniformity) that prevent external disturbances from affecting the learning process. This anticipates and blocks potential sources of learning errors before they can occur
3Reliability
If correction of detected values is applied during engine operation, then fault diagnosis accuracy is improved, but correction errors from improper learning propagate
Solution Approach 1:
The learning process is performed preliminarily when the engine is stopped and cooling water temperature is uniform, establishing an accurate baseline correction value before operation begins. This preliminary calibration ensures that subsequent corrections during operation are based on accurate sensor mismatch characterization
Solution Approach 2:
The ECU continuously compares the learned difference value with actual sensor readings during operation and applies correction only when conditions match the learning state, providing feedback control that prevents propagation of correction errors
Data Source
AI summary
A learning unit learns a difference between a detected value ECT and a detected value RCT when it is determined that the detected value ECT and detected value RCT are stable while an engine is at a stop. A diagnostic unit performs a fault diagnosis for the engine based on the difference between the detected value ECT and detected value RCT having been corrected through use of a learned value learned by the learning unit. Accordingly, a fault diagnostic system for an internal combustion engine and a fault diagnostic method for an internal combustion engine can be achieved in which accuracy in fault diagnosis for the internal combustion engine can be improved and an erroneous diagnosis can be restrained.