Stepwise Regression for Engine Abnormal Combustion Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for recognizing abnormal combustion events in internal combustion engines, such as knocking, pre-ignition, and misfire, are not reliable under all operating conditions and require significant calculation capacity, leading to extended execution times in control units.
Innovation Solution
A stepwise regression calculation method is employed, involving signal preparation, feature adaptation, linear model regression, nonlinear model error compensation, and results adaptation to enhance recognition and classification of abnormal combustion events, reducing calculation load and improving robustness against mechanical and electrical interference.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If complex evaluation methods based on linear system theory and signal processing are used for abnormal combustion recognition, then measurement precision is improved, but calculation capacity requirements increase and execution time is extended
Solution Approach 1:
The evaluation method is segmented into distinct processing stages: signal preprocessing, feature extraction, and classification. Each stage handles specific aspects of the analysis, reducing the computational burden on any single component while maintaining overall recognition accuracy.
Solution Approach 2:
Relevant features are extracted from the raw sensor signals before classification. By taking out only the essential characteristics needed for abnormal combustion detection, the method reduces calculation requirements while preserving measurement precision.
2Measurement precision
If complex evaluation methods are used for abnormal combustion recognition, then measurement precision is improved, but execution time is extended
Solution Approach 1:
Signal preprocessing and feature extraction are performed in advance before the actual classification decision is needed. This preliminary action prepares the data structure and identifies key characteristics beforehand, reducing the time required for final recognition while maintaining accuracy.
3Reliability
If existing recognition methods are used, then some abnormal combustion events can be detected, but reliability under all operating conditions cannot be guaranteed
Solution Approach 1:
The evaluation system is designed with universal features and classification rules that can handle multiple types of abnormal combustion events (knocking, pre-ignition, misfire) under various operating conditions. The same core methodology adapts to different combustion anomalies without requiring separate specialized methods for each case.
Data Source
AI summary
A method for evaluating abnormal combustion events of an internal combustion engine of a motor vehicle by regression calculation of a physical reference variable, in which method a recognition variable of the abnormal combustion event is calculated from a measured sensor signal. In a method with which abnormal combustion processes of the internal combustion engine can be reliably recognized and classified at all operating points, a stepwise system is used for regression calculation of the recognition variable, in which system at least one reference variable that corresponds to a measured reference variable of the sensor signal is calculated from the sensor signal.