Uncertainty-aware deep learning for digital twin-driven monitoring: Application to fault detection in power lines

Deep neural networks (DNNs) are often coupled with physics-based models or data-driven surrogate models to perform fault detection and health monitoring of systems in the low data regime. These models serve as digital twins to generate large quantities of data to train DNNs which would otherwise be difficult to obtain from the real-life system. However, such models can exhibit parametric uncertainty that propagates to the generated data. In addition, DNNs exhibit uncertainty in the parameters learnt during training. In such a scenario, the performance of the DNN model will be influenced by the uncertainty in the physics-based model as well as the parameters of the DNN. In this article, we quantify the impact of both these sources of uncertainty on the performance of the DNN. We perform explicit propagation of uncertainty in input data through all layers of the DNN, as well as implicit prediction of output uncertainty to capture the former. Furthermore, we adopt Monte Carlo dropout to capture uncertainty in DNN parameters. We demonstrate the approach for fault detection of power lines with a physics-based model, two types of input data and three different neural network architectures. We compare the performance of such uncertainty-aware probabilistic models with their deterministic counterparts. The results show that the probabilistic models provide important information regarding the confidence of predictions, while also delivering an improvement in performance over deterministic models.

[1]  G. Sansavini,et al.  Simulation-driven deep learning for locating faulty insulators in a power line , 2022, Reliability Engineering & System Safety.

[2]  Dong Liu,et al.  Digital Twin and Data-Driven Quality Prediction of Complex Die-Casting Manufacturing , 2022, IEEE Transactions on Industrial Informatics.

[3]  R. Otis Uncertainty reduction and quantification in computational thermodynamics , 2022, Computational Materials Science.

[4]  Michael D. Todd,et al.  A comprehensive review of digital twin—part 2: roles of uncertainty quantification and optimization, a battery digital twin, and perspectives , 2022, Structural and Multidisciplinary Optimization.

[5]  Yichuan Wang,et al.  Digital twins supported equipment maintenance model in intelligent water conservancy , 2022, Computers and Electrical Engineering.

[6]  Tae Hoon Oh,et al.  Integration of Reinforcement Learning and Model Predictive Control to Optimize Semi‐batch Bioreactor , 2022, AIChE Journal.

[7]  Joon Huang Chuah,et al.  Structural crack detection using deep convolutional neural networks , 2022, Automation in Construction.

[8]  Huan X. Nguyen,et al.  Cloud-Based Digital Twinning for Structural Health Monitoring Using Deep Learning , 2021, IEEE Transactions on Industrial Informatics.

[9]  Xiaohua Zhai,et al.  Revisiting the Calibration of Modern Neural Networks , 2021, NeurIPS.

[10]  Rodolfo S. M. Freitas,et al.  Parametric and model uncertainties induced by reduced order chemical mechanisms for biogas combustion , 2020 .

[11]  Zequn Wang,et al.  Surrogate model uncertainty quantification for reliability-based design optimization , 2019, Reliab. Eng. Syst. Saf..

[12]  Cesare Alippi,et al.  Deep Learning for Time Series Forecasting: The Electric Load Case , 2019, CAAI Trans. Intell. Technol..

[13]  Seongok Ryu,et al.  A Bayesian graph convolutional network for reliable prediction of molecular properties with uncertainty quantification , 2019, Chemical science.

[14]  Paris Perdikaris,et al.  Physics-Constrained Deep Learning for High-dimensional Surrogate Modeling and Uncertainty Quantification without Labeled Data , 2019, J. Comput. Phys..

[15]  Tanmoy Bhattacharya,et al.  The need for uncertainty quantification in machine-assisted medical decision making , 2019, Nat. Mach. Intell..

[16]  William Yang Wang,et al.  Quantifying Uncertainties in Natural Language Processing Tasks , 2018, AAAI.

[17]  Marta Z. Kwiatkowska,et al.  Evaluating Uncertainty Quantification in End-to-End Autonomous Driving Control , 2018, ArXiv.

[18]  S. Roth,et al.  Lightweight Probabilistic Deep Networks , 2018, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.

[19]  Nicholas Zabaras,et al.  Bayesian Deep Convolutional Encoder-Decoder Networks for Surrogate Modeling and Uncertainty Quantification , 2018, J. Comput. Phys..

[20]  G. Tapia,et al.  Bayesian Calibration and Uncertainty Quantification for a Physics-Based Precipitation Model of Nickel–Titanium Shape-Memory Alloys , 2017 .

[21]  Kilian Q. Weinberger,et al.  On Calibration of Modern Neural Networks , 2017, ICML.

[22]  Alex Kendall,et al.  What Uncertainties Do We Need in Bayesian Deep Learning for Computer Vision? , 2017, NIPS.

[23]  Zoubin Ghahramani,et al.  Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learning , 2015, ICML.

[24]  K. Gryllias,et al.  A perspective survey on deep transfer learning for fault diagnosis in industrial scenarios: Theories, applications and challenges , 2022, Mechanical Systems and Signal Processing.

[25]  Uncertainty Estimation Using a Single Deep Deterministic Neural Network-ML Reproducibility Challenge 2020 , 2021 .

[26]  H. D. Macedo,et al.  Uncertainty Quantification and Runtime Monitoring Using Environment-Aware Digital Twins , 2020, ISoLA.

[27]  Tom Minka,et al.  A family of algorithms for approximate Bayesian inference , 2001 .