Engine Control Computer Failure Detection Method
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for protecting engine control computers in automotive vehicles fail to detect failures at the output stages early enough, leading to potential destruction of the computer, especially in specific and random configurations.
Innovation Solution
A method for detecting failures in engine control computers involves a measurement time window where variables related to command signal durations and amplitudes are measured and compared to reference values, allowing for timely reinitialization of the software and hardware layers to prevent destruction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If protection means are installed to monitor output stages, then the engine control computer is protected from failures, but failures are not detected early enough in specific configurations leading to destruction
Solution Approach 1:
The patent implements preliminary action by measuring command signal characteristics (duration, amplitude) during normal operation and comparing them against reference values before failures cause destruction. The system establishes baseline measurements during activation phases and continuously monitors for deviations, enabling early detection of abnormal conditions before they lead to catastrophic failure.
Solution Approach 2:
The patent employs feedback mechanisms by continuously monitoring command signal parameters and comparing them with reference values. When deviations exceed thresholds, the system triggers reinitialization sequences. This closed-loop feedback enables real-time detection and response to emerging failures, correcting the insufficient early detection problem of prior art.
2Reliability
If monitoring is performed continuously, then early failure detection is achieved, but system complexity increases
Solution Approach 1:
The patent extracts only the essential monitoring functions needed for early failure detection - specifically measuring command signal duration and amplitude during activation phases. Rather than implementing comprehensive continuous monitoring of all system parameters, the invention focuses on extracting and monitoring only the critical signal characteristics that indicate impending failures, thereby reducing system complexity while maintaining effective early detection.
Solution Approach 2:
The patent applies partial action by implementing monitoring during specific critical phases (activation phases with command signals) rather than continuous monitoring of all operations. The system performs detailed measurements only when command signals are active and compares them against reference values, reducing overall system complexity while maintaining effective failure detection during the most critical operational windows.
Data Source
AI summary
A method of detecting failures of an engine control computer including at least: a hardware layer, a software layer comprising at least one activation cycle of the load m phase(s) of activation of said load with: actuation of the load, first timeout, deactivation of the load, second timeout. The method of detecting failures includes a measurement time window during which the following are carried out: an initialization of the measurement variables, detection of an actuation of the load, an action of reinitializing the engine control computer if at least one of the following conditions is valid: ΔT1>ΔT1_ref, ΔT2<ΔT2_ref, ΔT3>ΔT3_ref, ΔT4<ΔT4_ref.