A Deep Learning Solution for Fault Detection and Diagnosis Applied to Internal Combustion Engines
Bibliographic record
Abstract
In today's competitive manufacturing environment, special attention is given to the quality and reliability of manufactured products. Condition monitoring and more precisely Fault Detection and Diagnosis (FDD) are aimed at addressing that attention for increased customer satisfaction. The economic implications of FDD are highly valued in the industry, and academia is leveraged to provide smart responses. The focus of this research is the development of an FDD algorithm for internal combustion engine faults via engine block vibration using deep learning. The FDD solution would have to be implemented in software where it could operate in the absence of human intervention. The proposed solution includes two elements namely: input feature construction and fault classification. Short-time Fourier Transform (STFT) and Convolutional Neural Networks (CNNs) perform the aforementioned elements. The FDD solution detects and diagnoses fault signatures from 4 different knock sensors mounted on a V8-type Ford engine. The solution comprises the STFT which converts the knock sensors’ signal from the time domain to the crank angle-frequency domain, hence providing features to be used for diagnosis. These features are then used as input to a CNN, which can learn the crank angle-frequency patterns found in the input data and subsequently perform classification. Transfer learning is used in the proposed solution to circumvent domain shift and improve generalization. This gives the FDD solution advantages such as high diagnosis accuracy, robustness against perturbations in data quality and no need for human intervention.
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 | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.000 |
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".