A Precise and Reliable Engine Knock Detection Utilizing Meta Classifier
Bibliographic record
Abstract
An increase in temperature and pressure can cause spontaneous ignition of the air-fuel mixture in internal combustion engines, reducing engine efficiency, lifespan, and increasing air pollution. Typically, to predict and detect this effect, a knock sensor is used, which has a low detection accuracy due to the engine vibration noise. In this work, a machine learning model based on a meta-classifier is proposed and implemented for real-time fault detection in combustion engines. First, actual knock sensor data are recorded at diverse engine speeds from our engine test bench. The local dataset is preprocessed and scaled. Then, 30 different features in the time and frequency domains are investigated. Dimensionality of data is reduced employing recursive feature elimination. Then, a stacking classifier is utilized to address the classification problem by combining several classification models through the use of a metaclassifier. To enhance the assessment of the experimental outcomes in knock detection, k-fold cross-validation is utilized to gauge the model's performance with new data. The result shows the proposed method has around 12% higher accuracy during 5 cross folds with least amount of variation. Finally, the model is implemented on an ARM MCU and showed an execution time of 8.9ms, which validates its reliability for real-time operation.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".