Engine Control System Sensor Drift Detection via Rationality Checks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current engine controller software fails to detect impending sensor failures that are within the normal range but no longer accurate due to sensor drift or deterioration, which can adversely affect emissions and are not addressed by existing circuit diagnostics.
Innovation Solution
Implementing rationality checks in the engine control system to monitor sensor data, compare operating conditions, and log impending failures, initiating warnings and remedial actions when predetermined failure criteria are met, using an electronic control unit (ECU) with memory to track faults and their occurrences.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If circuit diagnostics are used to detect sensor failures, then hard circuit failures (short to battery, short to ground) are detected, but in-range failures due to sensor drift or deterioration are not detected
Solution Approach 1:
The system performs preliminary rationality checks by comparing sensor readings against expected physical relationships and operating conditions before failures become severe. This proactive detection identifies in-range drift early, allowing preventive maintenance before emissions are adversely affected.
Solution Approach 2:
The system continuously monitors sensor readings and provides feedback by comparing them against rationality criteria based on engine operating conditions and physical relationships between parameters. This closed-loop feedback mechanism detects deviations indicating sensor drift while the sensor is still within its operational range.
2Reliability
If rationality checks are implemented to detect in-range sensor failures, then emissions compliance is improved, but system complexity increases
Solution Approach 1:
The existing engine control unit's processor is made multi-functional by adding rationality check capabilities to its existing diagnostic functions. The same hardware resources are utilized for both traditional circuit diagnostics and the new in-range failure detection, avoiding additional dedicated hardware and reducing overall system complexity.
Solution Approach 2:
The control system performs self-diagnosis by automatically comparing sensor readings against pre-programmed rationality criteria and physical relationships. The system monitors itself without requiring external diagnostic equipment, reducing the need for additional monitoring hardware and simplifying the overall system architecture.
3Measurement precision
If sensor monitoring is continuous to detect impending failures, then detection accuracy is improved, but processing time and computational load increase
Solution Approach 1:
Rationality checks are performed periodically at defined intervals rather than continuously processing every sensor reading. This periodic monitoring approach maintains detection accuracy for in-range failures while significantly reducing the computational load and processing time requirements compared to continuous analysis of all sensor data.
Solution Approach 2:
The system applies rationality checks selectively to specific sensor pairs and operating conditions rather than uniformly to all sensors at all times. By focusing computational resources on critical comparisons based on current engine state, the system achieves adequate detection accuracy while minimizing unnecessary processing overhead.
Data Source
AI summary
In one aspect, the present invention is directed to a method for operating an electronically controlled internal combustion engine in a vehicle to perform at least one rationality check on at least one sensor to detect impending sensor failure and verify component functionality. The engine is equipped with an engine control system (ECS) having a memory and in electronic communication with various sensors. The sensors transmit data signals to the ECS indicative of associated engine and vehicle component functionality. The method comprises determining a measured operating condition of a first component from sensor data signals indicative of first component functionality, determining a measured operating condition of a second component from sensor data signals indicative of second component functionality, comparing the measured condition of the first component to the measured operating condition of the second component to determine whether sensor readings from the first component are indicative of a component operating within a normal range but indicative of impending sensor or component failure, logging an indication of an impending sensor or component failure as a fault in MPU memory if it occurs for a predetermined period of time and for more than a predetermined number of times and drive cycles, and activating a warning alert to an operator and initiating remedial actions responsive to the indication of impending component or sensor failure. The data logged includes time of fault, type of fault, number of occurrences of the fault, number of drive cycles where the fault occurred and distance traveled with the fault logged.