Selbstgesteuertes Projekt
Predictive Maintenance: Industrial ML for Fault Detection
Condition-Monitoring Engineer · KI / ML · Juli 2026
Eine ProoV-Fallstudie · Lernprojekt, kein Arbeitsverhältnis
Built an end-to-end rolling-bearing fault detector using vibration snapshots, time-domain statistics, envelope-spectrum defect-band analysis, and a compact decision tree trained only on the training bearings. Evaluated it on a held-out bearing set with reported catch-rate and false-alarm rate, then defended a 0.53 alarm threshold in a maintenance memo that named the outer-race fault from the dominant BPFO frequency.
Bewertet nach
- Feature engineering (signal to honest numbers)25%
- Held-out evaluation (honest measurement)25%
- Cost-based threshold decision + memo25%
- Compact model + embedded-budget justification15%
- Plain-language explanation10%
Bestanden · Bestehensgrenze 60/100
Was herausstach6
- Correctly identified kurtosis as the strongest time-domain fault indicator
- Trained a shallow decision tree on training bearings only and stated depth/node count against the embedded budget
- Reported held-out catch rate and false-alarm rate rather than evaluating on training data
- Wrote a memo that ties the threshold choice to concrete catch-rate, false-alarm, and cost figures
- Leakage-Free Feature Engineering
- Honest Model Evaluation
Meine eingereichte Arbeit19 Aufgaben
11 Aufgaben · 7.7k Zeichen · Python · 8 Textantworten · 487 Wörter
- Initial Prediction62 Zeichen
- Example Load Plot HealthyPython · 859 Zeichen
- Load Plot Signal CodePython · 1.8k Zeichen
- Load Plot Signal205 Zeichen
- Example Rms WorkedPython · 698 Zeichen
- Time Features143 Zeichen
- Freq FeaturesPython · 1.1k Zeichen
- Train Classifier363 Zeichen
- Evaluation Heldout288 Zeichen
- Threshold Decision334 Zeichen
- Decision Memo1.9k Zeichen
- Initial Prediction9 Wörter
- Example Load Plot Healthy81 Wörter
- Load Plot Signal Code163 Wörter
- Load Plot Signal15 Wörter
- Example Rms Worked72 Wörter
- Time Features10 Wörter
- Freq Features106 Wörter
- Train Classifier31 Wörter
Bewusst zusammengefasst — der eingereichte Code und die Texte bleiben privat, damit diese Seite nicht als Lösungsvorlage dient. Die vollständige Einreichung steht hinter dem verifizierten Zertifikat.