本研究旨在開發一套結合數位孿生與深度學習之啟動機動力傳動模組品質檢測系統，以取代傳統仰賴人工聽診與動力計判讀的主觀檢測方式。針對實驗資料取得不易之挑戰，本研究建立啟動機之數位孿生模型，模擬正常運作條件下之多維訊號，並大量生成通過樣本以進行非監督式預訓練，進一步提升模型初始權重品質與泛化能力。同時，本研究亦制定數據標準化規範，作為未來智慧檢測應用與資料共享之基礎。
模型設計採用時序型Transformer自編碼器架構，藉由重建誤差進行異常偵測。訓練流程結合模擬資料預訓練與少量實測資料微調，並以皮爾森相關係數與穩態參數驗證模擬資料之合理性。實驗結果顯示，本系統能於有限資料下穩定辨識啟動機異常，驗證結合數位孿生與深度學習之品質檢測策略具實用性與可行性，未來可作為啟動機再製產業導入智慧檢測系統之依據。

This study aims to develop a quality inspection system for starter motor power transmission modules by integrating digital twin technology with deep learning, replacing traditional subjective methods that rely on manual auscultation and dynamometer readings. To address the challenge of limited experimental data, a digital twin model of the starter motor was constructed to simulate multi- dimensional signals under normal operating conditions, enabling the large-scale generation of pass samples for unsupervised pre-training. This approach enhances the quality of the model's initial weights and improves its generalization capability. In addition, a data standardization protocol was established to serve as a foundation for future applications in intelligent inspection and data sharing.
The model adopts a time-series Transformer autoencoder architecture, performing anomaly detection through reconstruction error. The training process combines simulated data pre-training with fine-tuning using a small amount of measured data, while the validity of the simulated data is verified through Pearson correlation coefficients and steady-state parameters. Experimental results demonstrate that the proposed system can reliably identify starter motor anomalies even under limited data conditions, confirming the practicality and feasibility of a quality inspection strategy that integrates digital twin and deep learning. This work can serve as a reference for introducing intelligent inspection systems in the starter motor remanufacturing industry.

[1] M. Ali and S. Evans, "The impact of best practice on energy efficiency in industrial decarbonization policy," Energy Policy, vol. 204, p. 114632, 2025.
[2] J. I. Chowdhury, Y. Hu, I. Haltas, N. Balta-Ozkan, G. J. Matthew, and L. Varga, "Reducing industrial energy demand in the UK: A review of energy efficiency technologies and energy saving potential in selected sectors," Renewable and Sustainable Energy Reviews, vol. 94, pp. 1153–1178, 2018.
[3] R. Saidur, "A review on electrical motors energy use and energy savings," Renewable and Sustainable Energy Reviews, vol. 14, no. 3, pp. 877–898, 2010.
[4] A. Vijaya, J. P. S. Meisterknecht, L. S. Angreani, and H. Wicaksono, " Advancing sustainability in the automotive sector: A critical analysis of environmental, social, and governance (ESG) performance indicators," Cleaner Environmental Systems, vol. 16, 100248, 2025.
[5] P. Rathore, S. Kota, and A. Chakrabarti, "Sustainability through remanufacturing in India: a case study on mobile handsets," Journal of Cleaner Production, vol. 19, no. 15, pp. 1709–1722, 2011.
[6] K. Govindan, K. M. Shankar, and D. Kannan, "Application of fuzzy analytic network process for barrier evaluation in automotive parts remanufacturing towards cleaner production – a study in an Indian scenario," Journal of Cleaner Production, vol. 114, pp. 199–213, 2016.
[7] W. Liu, W. Ma, Y. Hu, M. Jin, K. Li, X. Chang, and X. Yu, "Production planning for stochastic manufacturing/remanufacturing system with demand substitution using a hybrid ant colony system algorithm," Journal of Cleaner Production, vol. 213, pp. 999–1010, 2019.
[8] A. J. Martyr and D. R. Rogers, "Dynamometers: the measurement and control of torque, speed, and power," in Engine Testing, 5th ed., A. J. Martyr and D. R. Rogers, Eds. Oxford, U.K.: Butterworth-Heinemann, 2021, ch. 9, pp. 235–263.
[9] J. Zipfel, F. Verworner, M. Fischer, U. Wieland, M. Kraus, and P. Zschech, "Anomaly detection for industrial quality assurance: A comparative evaluation of unsupervised deep learning models," Computers & Industrial Engineering, vol. 177, 109045, 2023.
[10] F. Kähler, O. Schmedemann, and T. Schüppstuhl, "Anomaly detection for industrial surface inspection: application in maintenance of aircraft components," Procedia CIRP, vol. 107, pp. 246–251, 2022.
[11] B. Jung, H. You, and S. Lee, "Anomaly candidate extraction and detection for automatic quality inspection of metal casting products using high-resolution images," Journal of Manufacturing Systems, vol. 67, pp. 229–241, 2023.
[12] K. Choi, J. Yi, C. Park, and S. Yoon, "Deep learning for anomaly detection in time-series data: Review, analysis, and guidelines," IEEE Access, vol. 9, pp. 120043–120065, 2021
[13] P. Yan et al., "A comprehensive survey of deep transfer learning for anomaly detection in industrial time series: Methods, applications, and directions," IEEE Access, vol. 12, pp. 3768–3789, 2024.
[14] Y. Wei, J. Jang-Jaccard, W. Xu, F. Sabrina, S. Camtepe and M. Boulic, "LSTM-Autoencoder-Based Anomaly Detection for Indoor Air Quality Time-Series Data," in IEEE Sensors Journal, vol. 23, no. 4, pp. 3787-3800, 15 Feb.15, 2023.
[15] J. Kim, H. Kang, and P. Kang, "Time-series anomaly detection with stacked Transformer representations and 1D convolutional network," Eng. Appl. Artif. Intell., vol. 120, p. 105964, 2023.
[16] C. Tan, F. Sun, T. Kong, W. Zhang, C. Yang, and C. Liu, "A survey on deep transfer learning," arXiv preprint arXiv:1808.01974, 2018.
[17] C. M. A. Roelofs, C. Gück, and S. Faulstich, "Transfer learning applications for autoencoder-based anomaly detection in wind turbines," Energy and AI, vol. 17, 100373, 2024.
[18] C. M. A. Roelofs, C. Gück, and S. Faulstich, "Transfer learning applications for autoencoder-based anomaly detection in wind turbines," Energy and AI, vol. 17, 100373, 2024.
[19] S. Shao, S. McAleer, R. Yan and P. Baldi, "Highly Accurate Machine Fault Diagnosis Using Deep Transfer Learning," in IEEE Transactions on Industrial Informatics, vol. 15, no. 4, pp. 2446-2455, April 2019.
[20] E. H. Glaessgen and D. S. Stargel, "The digital twin paradigm for future NASA and U.S. Air Force vehicles," in Proc. 53rd AIAA/ASME/ASCE/AHS/ASC Structures, Structural Dynamics and Materials Conf.—Special Session on the Digital Twin, Honolulu, HI, USA, Apr. 2012, AIAA Paper 2012-1818.
[21] E. Glaessgen and D. Stargel, "The digital twin paradigm for future NASA and U.S. air force vehicles," April. 2012.
[22] "Digital Twins and AI: Transforming Industrial Operations". Retrieved March 5, 2025, from https://www.reliableplant.com/Read/31897/digital-twins-ai.
[23] J. Xia, R. Huang, J. Li, Z. Chen, and W. Li, "Digital twin-assisted fault diagnosis of rotating machinery without measured fault data," IEEE Transactions on Instrumentation and Measurement, vol. 73, Art. no. 3531210, pp. 1–10, 2024.
[24] Y. Xiao, H. Shao, S. Han, Z. Huo, and J. Wan, "Novel joint transfer network for unsupervised bearing fault diagnosis from simulation domain to experimental domain," IEEE/ASME Transactions on Mechatronics, vol. 27, no. 6, pp. 5254–5263, Dec. 2022.
[25] Y. Xu, Y. Sun, X. Liu, and Y. Zheng, "A digital-twin-assisted fault diagnosis using deep transfer learning," IEEE Access, vol. 7, pp. 19990–19999, 2019.
[26] K. Feng, J. C. Ji, Y. Zhang, Q. Ni, Z. Liu, and M. Beer, "Digital twin-driven intelligent assessment of gear surface degradation," Mechanical Systems and Signal Processing, vol. 186, 109896, 2023.
[27] Y. Zhang, J. C. Ji, Z. Ren, Q. Ni, F. Gu, K. Feng, K. Yu, J. Ge, Z. Lei, and Z. Liu, "Digital twin-driven partial domain adaptation network for intelligent fault diagnosis of rolling bearing," Reliability Engineering & System Safety, vol. 234, 109186, 2023.
[28] M. Xia, H. Shao, D. Williams, S. Lu, L. Shu, and C. W. de Silva, "Intelligent fault diagnosis of machinery using digital twin-assisted deep transfer learning," Reliability Engineering & System Safety, vol. 215, 107938, 2021.
[29] "Two-port network," Wikipedia. Apr. 23, 2024. Retrieved March 5, 2025, from https://en.wikipedia.org/w/index.php?title=Two-port_network&oldid=1220308047.
[30] "Starter Motor Teardown and Feature Study". Retrieved March 5, 2025, from https://advancedstructures.in/starter-motor-teardown-and-feature-study/.
[31] "How an Engine Starter Motor Works". Retrieved March 5, 2025, from https://youtu.be/kFsl5r34lCI?si=03om5ei4yUAANhRR.
[32] 吳建諭，「永磁馬達特性快速檢測系統設計」，碩士論文，國立成功大學機械工程研究所，2010。
[33] 黃子豪，「方塊圖解析三端口傳動機構之運動學關係」，碩士論文，國立成功大學機械所，2021。
[34] 鄭博文，「數位孿生與熱影像於啟動機動力傳動模組健康評估」，碩士論文，國立成功大學機械所，2025。
[35] 究準科技股份有限公司。Retrieved March 5, 2025, from https://www.join-precision.com/Motor-Test-System.
[36] A. Vaswani, N. Shazeer, N. Parmar, J. Uszkoreit, L. Jones, A. N. Gomez, L. Kaiser, and I. Polosukhin, "Attention is all you need," CoRR, vol. abs/1706.03762, 2017.
[37] Z. Zhang, G. Qiu, Y. Cheng, and M. Wang, "Densely‑connected Decoder Transformer for unsupervised anomaly detection of power electronic systems," J. Autom. Intell., to appear, available online May 30, 2025.
[38] M. Rodríguez, D. P. Tobón, and D. Múnera, "A framework for anomaly classification in Industrial Internet of Things systems," Internet Things, vol. 29, art. no. 101446, 2025.
[39] A. Altınörs, F. Yol, and O. Yaman, "A sound based method for fault detection with statistical feature extraction in UAV motors," Appl. Acoust., vol. 183, p. 108325, 2021.
[40] M. Salem, S. Taheri and J. S. Yuan, "Anomaly Generation Using Generative Adversarial Networks in Host-Based Intrusion Detection," 2018 9th IEEE Annual Ubiquitous Computing, Electronics & Mobile Communication Conference (UEMCON), New York, NY, USA, 2018, pp. 683-687.