A Digital-Twin-Assisted Fault Diagnosis Using Deep Transfer Learning

By Yan Xu, Yanming Sun, Xiaolong Liu, Yonghua Zheng. IEEE Access, 2019

415 citations recorded by the source corpus.

Digital twin is a significant way to achieve smart manufacturing, and provides a new paradigm for fault diagnosis. Traditional data-based fault diagnosis methods mostly assume that the training data and test data are following the same distribution and can acquire sufficient data to train a reliable diagnosis model, which is unrealistic in the dynamic changing production process. In this paper, we present a two-phase digital-twin-assisted fault diagnosis method using deep transfer learning (DFDD), which realizes fault diagnosis both in the development and maintenance phases. At first, the potential problems that are not considered at design time can be discovered through front running the ultra-high-fidelity model in the virtual space, while a deep neural network (DNN)-based diagnosis model will be fully trained. In the second phase, the previously trained diagnosis model can be migrated from the virtual space to physical space using deep transfer learning for real-time monitoring and predictive maintenance. This ensures the accuracy of the diagnosis as well as avoids wasting time and knowledge. A case study about fault diagnosis using DFDD in a car body-side production line is presented. The results show the superiority and feasibility of our proposed method.

Subfields: Industrial and Manufacturing Engineering

View Paper: https://www.semanticscholar.org/paper/22f57a41e84ecf1215d036ac3d1d9f2e651de93c

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Industrial and Manufacturing Engineering

Citations in the corpus

The Role of AI, Machine Learning, and Big Data in Digital Twinning: A Systematic Literature Review, Challenges, and Opportunities (2021, methodology): “[45] highlight that, in production sys-”
Digital Twins: A Survey on Enabling Technologies, Challenges, Trends and Future Prospects (2022, background): “Of course, for such models to be able to generalise well on real data, the distribution of the synthetic data they were trained with has to closely resemble the distribution of the real data from the test set [31].”
A Review of Deep Transfer Learning and Recent Advancements (2022, methodology): “[25] 2019 A Digital-Twin-Assisted Fault Diagnosis Using Deep Transfer Learning Tabular/ bigdata no Finetuning no none Mechanic”
Holistic Network Virtualization and Pervasive Network Intelligence for 6G (2023, background): “[99] Smart manufacturing Manufacturing systems Predicting and verifying the system performance Increasing autonomy and enhancing fault diagnosis”
Digital Twin: Vision, Benefits, Boundaries, and Creation for Buildings (2019, methodology): “Supporting technologies for digital twin are 3D CAD modeling, WSNs, machine learning algorithms [32], and data analytics.”
Predictive maintenance using digital twins: A systematic literature review (2022, methodology): “[42] and [62] used a Stacked Sparse Autoencoder (SSA) with a Softmax activation function to classify a machine’s health condition.”
A state-of-the-art survey of Digital Twin: techniques, engineering product lifecycle management and business innovation perspectives (2019, background): “For example, small businesses as highlighted by Lu and Xu (2019) and Park et al. (2019) were able to adopt new revenue and resource models with the aid of DT, proving that DT-enabled BMs are viable.”
Digital Twin for 5G and Beyond (2021, background): “” Only as recently as 2017 has the DT become one of the top strategic technology trends, widely investigated in many industries, including manufacturing, energy, industrial assets, and structures, such as a dual fault diagnosis method based on DT in [5] for high diagnosis accuracy in predicting the ”
A Survey on AI-Driven Digital Twins in Industry 4.0: Smart Manufacturing and Advanced Robotics (2021, background): “, GMM (Gaussian mixture model) [147], SSAE (stacked sparse autoencoder) [148,149], and GAN (generative adversarial network) [150], were investigated in order to reduce the reliance on historical failure data in terms of prognostics and health management (PHM).”
Digital twin paradigm: A systematic literature review (2021, background): “The main scopes in using neural networks and deep learning in digital twin (Lee et al., 2013) are health assessment, performance prediction (Jain and Bhatnagar, 2020), fault diagnosis (Xu et al., 2019).”
A Review on Digital Twin Technology in Smart Grid, Transportation System and Smart City: Challenges and Future (2023, background): “Also, semantic web technologies provide a certain level of abstraction for energy system simulation and modeling, which can be coordinated and adapted to different energy systems [34].”
Digital Twin: Enabling Technologies, Challenges and Open Research (2020, background): “[24] proposes a AI focused solution for fault diagnosis in a smart manufacturing environment.”