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, physics-based time-domain features, and a compact decision tree trained only on the training bearings. Evaluated it on held-out healthy and faulty bearings, reported catch-rate and false-alarm rate, and justified an alarm threshold with concrete maintenance-cost tradeoffs while identifying the fault as an outer-race/BPFO issue.
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
- Trained a depth-1 decision tree on training bearings only and stated the node count against the embedded budget.
- Used held-out evaluation metrics with both catch-rate and false-alarm rate instead of relying on training performance.
- Wrote a decision memo that ties the threshold to explicit euro costs and identifies the outer-race/BPFO fault from the dominant defect frequency.
- Leakage-Free Model Training
- Honest Held-Out Evaluation
- Cost-Based Threshold Selection
Meine eingereichte Arbeit18 Aufgaben
10 Aufgaben · 6.1k Zeichen · Python · 8 Textantworten · 445 Wörter
- Initial Prediction62 Zeichen
- Load Plot Signal CodePython · 1.8k Zeichen
- Load Plot Signal205 Zeichen
- Example Rms WorkedPython · 697 Zeichen
- Time Features143 Zeichen
- Freq FeaturesPython · 1.2k Zeichen
- Train Classifier363 Zeichen
- Evaluation Heldout288 Zeichen
- Threshold Decision334 Zeichen
- Decision Memo1.1k Zeichen
- Initial Prediction9 Wörter
- Load Plot Signal Code163 Wörter
- Load Plot Signal15 Wörter
- Example Rms Worked72 Wörter
- Time Features10 Wörter
- Freq Features121 Wörter
- Train Classifier31 Wörter
- Evaluation Heldout24 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.