Engine Knock Diagnostic System Using FFT Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing engine knock detection systems face challenges in accurately distinguishing between genuine knock and false vibrations caused by malfunctioning sensors or external noise, leading to potential misdiagnosis and ineffective remedial actions.
Innovation Solution
A diagnostic system that analyzes knock data for excessive knock, insufficient variation, and abnormal range values by using vibration sensors, Fast Fourier Transform (FFT) analysis, and statistical methods to identify system faults, generating Diagnostic Trouble Codes (DTCs) for remedial actions such as adjusting spark timing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional knock detection systems use vibration sensors and FFT analysis to detect engine knock, then knock detection capability is provided, but false vibrations from malfunctioning sensors or external noise cause misdiagnosis
Solution Approach 1:
The patent segments the knock detection process into distinct phases: a learning phase where the system establishes baseline vibration patterns without knock, and a diagnostic phase where deviations from this baseline are analyzed. This segmentation allows the system to distinguish between normal engine vibrations and actual knock events, reducing false positives from sensor malfunctions or external noise.
Solution Approach 2:
The system performs preliminary action by conducting a learning phase before actual knock detection begins. During this phase, the system collects and analyzes vibration data to establish reference patterns for normal engine operation. This preliminary characterization of baseline vibrations enables more accurate distinction between genuine knock and false vibrations later during diagnostic operations.
2Difficulty of detecting and measuring
If the system analyzes vibration frequency content to detect knock, then knock detection is enabled, but distinction between genuine knock and false vibrations becomes difficult
Solution Approach 1:
The patent applies local quality by analyzing specific frequency ranges and temporal patterns characteristic of genuine knock events. Rather than treating all vibrations equally, the system focuses on localized frequency bands and time-domain characteristics that distinguish knock from other vibrations. This targeted analysis improves the ability to identify genuine knock sources while filtering out false vibrations.
Solution Approach 2:
The system incorporates feedback mechanisms where detection results are continuously compared against the learned baseline patterns. When vibrations are detected, the system feedback-loops the analysis to determine whether the pattern matches genuine knock characteristics or represents false vibrations. This feedback-driven discrimination improves measurement precision in identifying true knock events.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The system effectively detects and diagnoses engine knock issues, preventing misdiagnosis and ensuring accurate remedial actions, thereby improving engine performance and reliability.
Implementation Method 1
In various implementations, the vibration sensor 106 may include a piezoelectric accelerometer.
Implementation Method 2
The DSP 114 performs a Fast Fourier Transform (FFT) on the digitized output. The frequency content of the digitized output is transmitted to a knock detection module 116.
Data Source
AI summary
A knock diagnostic module having a knock module that increments a sample count when a cylinder firing signal corresponding to a first cylinder is received and selectively increments a knock count based on a knock detection signal that corresponds to the cylinder firing signal of the first cylinder A knock analysis module analyzes the knock count of the first cylinder when the sample count of the first cylinder reaches a predetermined value and selectively generates an excessive knock signal when the knock count exceeds a predetermined threshold. A remedial action module selectively performs a remedial action based on the excessive knock signal.