{
  "id": "2411.01360",
  "version": "v1",
  "published": "2024-11-02T20:35:13.000Z",
  "updated": "2024-11-02T20:35:13.000Z",
  "title": "Use Digital Twins to Support Fault Diagnosis From System-level Condition-monitoring Data",
  "authors": [
    "Killian Mc Court",
    "Xavier Mc Court",
    "Shijia Du",
    "Zhiguo Zeng"
  ],
  "comment": "6 pages, 4 figure. Paper submitted to 2025 22nd International Multi-Conference on Systems, Signals & Devices (SSD)",
  "categories": [
    "cs.LG",
    "cs.RO"
  ],
  "abstract": "Deep learning models have created great opportunities for data-driven fault diagnosis but they require large amount of labeled failure data for training. In this paper, we propose to use a digital twin to support developing data-driven fault diagnosis model to reduce the amount of failure data used in the training process. The developed fault diagnosis models are also able to diagnose component-level failures based on system-level condition-monitoring data. The proposed framework is evaluated on a real-world robot system. The results showed that the deep learning model trained by digital twins is able to diagnose the locations and modes of 9 faults/failure from $4$ different motors. However, the performance of the model trained by a digital twin can still be improved, especially when the digital twin model has some discrepancy with the real system.",
  "revisions": [
    {
      "version": "v1",
      "updated": "2024-11-02T20:35:13.000Z"
    }
  ],
  "analyses": {
    "keywords": [
      "digital twin",
      "system-level condition-monitoring data",
      "support fault diagnosis",
      "data-driven fault diagnosis model",
      "developing data-driven fault diagnosis"
    ],
    "tags": [
      "conference paper"
    ],
    "note": {
      "typesetting": "TeX",
      "pages": 6,
      "language": "en",
      "license": "arXiv",
      "status": "editable"
    }
  }
}