Wayside Condition Monitoring of Metro Wheelsets Using Vibration and Acoustic Sensors
Bibliographic record
Abstract
This paper presents an efficient wayside acoustic and vibration-based detection for wheelset faults on metro trains, which is crucial for the safety of the run. The proposed condition monitoring scheme includes four main steps: data acquisition, signal segmentation by one-period analysis, feature extraction; Time-Domain Features (TDF), Wavelet Packet Energy (WPE) features, and Linear Configuration Pattern Kurtograms (LCP-K), which applies a location invariant textural descriptor to Kurtogram images of the signal, and classification with state-of-art; Fisher’s Linear Discriminant Analysis (FLDA), Support Vector Machine (SVM), Decision Tree (Dec. Tree) and Linear Perceptron classifiers alongside classifier combination techniques. During the research, results are obtained on both measured and boosted data. Thus, two databases (A1 and A2), each of which consists of measured vibration and acoustic signals belonging to healthy and faulty cases of the wheelsets of Prague metros, are established. Due to a limited number of faulty instances, features are augmented with Adaptive Synthetic Sampling (ADASYN), and larger vibration and acoustic databases SA1 and SA2 are established to validate methods. Obtained results show that TDF with Dec. Tree classifier can detect wheelset faults by 100% with vibrations signals (A1), and the novel LCP-K algorithm outperforms both acoustic databases (A2 and SA2) up to 93%, and finally, WPE features via combined classifies, reaches a 100% fault detection performance. The proposed framework provides cost-effective maintenance, which can aid metro train specialists, with potential further applicability to other types of railways.
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 | 0.000 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.001 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.001 | 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".