Engine Event Location via Knock Sensor Signal Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for determining the location of engine events in combustion engines, such as peak firing pressure, are costly and prone to damage, making them uneconomical for widespread use.
Innovation Solution
A system that uses a knock sensor to transform signals into power spectral density and feature vectors, employing a predictive model to estimate the location of engine events, allowing for accurate prediction without the need for expensive in-cylinder sensors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If in-cylinder sensors are used to determine engine event location, then measurement precision is improved, but device complexity and cost increase
Solution Approach 1:
The patent uses knock sensor signals as a copy or proxy for the actual in-cylinder pressure events. Instead of directly measuring the engine event with an expensive in-cylinder sensor, the system captures vibrations through the engine block using knock sensors and processes these signals to estimate the engine event location, thereby achieving accurate measurement without complex sensor installation
Solution Approach 2:
The patent replaces the mechanical in-cylinder pressure sensor system with an acoustic/vibration-based measurement system. By substituting direct mechanical contact sensors with knock sensors that detect vibrations through the engine block, the system achieves measurement functionality while reducing device complexity and cost
2Measurement precision
If in-cylinder sensors are used to determine engine event location, then measurement precision is improved, but reliability decreases due to susceptibility to damage
Solution Approach 1:
The system uses knock sensor vibrations as a safe copy of the combustion event signals. These sensors are mounted on the external engine block rather than inside the cylinder, protecting them from the harsh in-cylinder environment while still capturing sufficient information to accurately determine engine event location
Solution Approach 2:
The engine block acts as an intermediary medium that transmits vibration signals from the combustion event to the knock sensors. This indirect measurement approach protects the sensors from direct exposure to high-temperature, high-pressure in-cylinder conditions while maintaining signal fidelity for accurate event location detection
3Ease of manufacture
If signal processing algorithms are implemented to estimate engine events, then device complexity increases, but manufacturing cost decreases
Solution Approach 1:
The patent transforms the knock sensor signal from the time domain to the frequency domain using Fast Fourier Transform (FFT), and then applies envelope detection to extract relevant features. These parameter transformations convert complex vibration signals into simplified spectral characteristics that are easier to process and interpret for engine event detection
Solution Approach 2:
The system extracts specific frequency components and envelope characteristics from the complex knock sensor signal. By isolating and analyzing only the relevant spectral features associated with engine events, the system reduces processing complexity while maintaining detection accuracy, focusing computational resources on the most informative signal aspects
Data Source
AI summary
Systems and methods for estimating an engine event location are disclosed herein. In one embodiment, a controller is configured to receive a signal from at least one knock sensor coupled to a reciprocating engine, transform the signal, using a multivariate transformation algorithm, into a power spectral density, transform the power spectral density into a plurality of feature vectors using predictive frequency bands, predict the engine event location using at least the plurality of feature vectors and a predictive model, and adjust operation of the reciprocating engine based on the engine event location.