Isolation and Localization of Unknown Faults Using Neural Network-Based Residuals

Localization of unknown faults in industrial systems is a dif- ficult task for data-driven diagnosis methods. The classification performance of many machine learning methods relies on the quality of training data. Unknown faults, for example faults not represented in training data, can be detected using, for example, anomaly classifiers. However, mapping these unknown faults to an actual location in the real system is a non-trivial problem. In model-based diagnosis, physical-based models are used to create residuals that isolate faults by mapping model equations to faulty system components. De- veloping sufficiently accurate physical-based models can be a time-consuming process. Hybrid modeling methods combining physical-based methods and machine learning is one solu- tion to design data-driven residuals for fault isolation. In this work, a set of neural network-based residuals are designed by incorporating physical insights about the system behavior in the residual model structure. The residuals are trained using only fault-free data and a simulation case study shows that they can be used to perform fault isolation and localization of unknown faults in the system.

[1]  E. S. Page.  CONTINUOUS INSPECTION SCHEMES , 1954 .

[2]  Xin Zhou,et al.  Deep neural networks: A promising tool for fault characteristic mining and intelligent diagnosis of rotating machinery with massive data , 2016 .

[3]  Barak A. Pearlmutter.  Gradient calculations for dynamic recurrent neural networks: a survey , 1995, IEEE Trans. Neural Networks.

[4]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[5]  Sylvain Verron,et al.  A generic framework for decision fusion in Fault Detection and Diagnosis , 2018, Eng. Appl. Artif. Intell..

[6]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[7]  Luca Antiga,et al.  Automatic differentiation in PyTorch , 2017 .

[8]  Gautam Biswas,et al.  Diagnosability Analysis Considering Causal Interpretations for Differential Constraints , 2012, IEEE Transactions on Systems, Man, and Cybernetics - Part A: Systems and Humans.

[9]  Belarmino Pulido Junquera,et al.  State space neural networks and model-decomposition methods for fault diagnosis of complex industrial systems , 2019, Eng. Appl. Artif. Intell..

[10]  Brian C. Williams,et al.  Diagnosing Multiple Faults , 1987, Artif. Intell..

[11]  Daniel Jung,et al.  A Combined Data-Driven and Model-Based Residual Selection Algorithm for Fault Detection and Isolation , 2019, IEEE Transactions on Control Systems Technology.

[12]  Erik Frisk,et al.  A Toolbox for Analysis and Design of Model Based Diagnosis Systems for Large Scale Models , 2017 .

[13]  Charu C. Aggarwal,et al.  Outlier Detection for Temporal Data: A Survey , 2014, IEEE Transactions on Knowledge and Data Engineering.

[14]  P J Webros.  BACKPROPAGATION THROUGH TIME: WHAT IT DOES AND HOW TO DO IT , 1990 .

[15]  Louise Travé-Massuyès,et al.  Bridging control and artificial intelligence theories for diagnosis: A survey , 2014, Eng. Appl. Artif. Intell..

[16]  Charu C. Aggarwal,et al.  Neural Networks and Deep Learning , 2018, Springer International Publishing.

[17]  Mattias Nyberg,et al.  Modeling and inference for troubleshooting with interventions applied to a heavy truck auxiliary braking system , 2012, Eng. Appl. Artif. Intell..

[18]  Jürgen Schmidhuber,et al.  Deep learning in neural networks: An overview , 2014, Neural Networks.

[19]  Maria J. Fuente,et al.  Improving parameter estimation using minimal analytically redundant subsystems , 2011, IEEE Conference on Decision and Control and European Control Conference.

[20]  Mattias Krysander,et al.  An Efficient Algorithm for Finding Minimal Overconstrained Subsystems for Model-Based Diagnosis , 2008, IEEE Transactions on Systems, Man, and Cybernetics - Part A: Systems and Humans.

[21]  Erik Frisk,et al.  Combining model-based diagnosis and data-driven anomaly classifiers for fault isolation , 2018, Control Engineering Practice.

[22]  Michel Kinnaert,et al.  Diagnosis and Fault-Tolerant Control , 2004, IEEE Transactions on Automatic Control.

[23]  Alexander Feldman,et al.  An Efficient Algorithm for Finding Minimal Overconstrained Subsystems for Model-Based Diagnosis , 2013 .

[24]  Carlos Alonso González,et al.  Possible conflicts: a compilation technique for consistency-based diagnosis , 2004, IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics).

[25]  Krishna R. Pattipati,et al.  Incremental Classifiers for Data-Driven Fault Diagnosis Applied to Automotive Systems , 2015, IEEE Access.

[26]  Markus Stumptner,et al.  Diagnosability analysis without fault models 1 , 2006 .

[27]  Sylvain Verron,et al.  Bridging data-driven and model-based approaches for process fault diagnosis and health monitoring: A review of researches and future challenges , 2016, Annu. Rev. Control..