SensorSCAN: Self-Supervised Learning and Deep Clustering for Fault Diagnosis in Chemical Processes (2022-08-17T00:00:00.000000Z)

TL;DR

This paper proposes SensorSCAN, a novel method for unsupervised fault detection and diagnosis, designed for industrial chemical process monitoring, and demonstrates the model's performance on two publicly available datasets of the Tennessee Eastman Process with various faults.

Abstract

Modern industrial facilities generate large volumes of raw sensor data during the production process. This data is used to monitor and control the processes and can be analyzed to detect and predict process abnormalities. Typically, the data has to be annotated by experts in order to be used in predictive modeling. However, manual annotation of large amounts of data can be difficult in industrial settings. In this paper, we propose SensorSCAN, a novel method for unsupervised fault detection and diagnosis, designed for industrial chemical process monitoring. We demonstrate our model's performance on two publicly available datasets of the Tennessee Eastman Process with various faults. The results show that our method significantly outperforms existing approaches (+0.2-0.3 TPR for a fixed FPR) and effectively detects most of the process faults without expert annotation. Moreover, we show that the model fine-tuned on a small fraction of labeled data nearly reaches the performance of a SOTA model trained on the full dataset. We also demonstrate that our method is suitable for real-world applications where the number of faults is not known in advance. The code is available at https://github.com/AIRI-Institute/sensorscan.

Authors

Maksim Golyadkin

1 papers

Vitaliy Pozdnyakov

1 papers

L. Zhukov

1 papers

TL;DR

Abstract

Authors

References104 items

Long–short-term memory encoder–decoder with regularized hidden dynamics for fault detection in industrial processes

One step forward for smart chemical process fault detection and diagnosis

High-fidelity positive-unlabeled deep learning for semi-supervised fault detection of chemical processes

A convolutional neural network method based on Adam optimizer with power-exponential learning rate for bearing fault diagnosis

Deep Learning-based Fault Detection in the Tennessee Eastman Process

A novel orthogonal self-attentive variational autoencoder method for interpretable chemical process fault detection and identification

Time-Series Representation Learning via Temporal and Contextual Contrasting

Towards robust and understandable fault detection and diagnosis using denoising sparse autoencoder and smooth integrated gradients.

CoAtNet: Marrying Convolution and Attention for All Data Sizes

Semi-Supervised Time Series Classification by Temporal Relation Prediction

An extended Tennessee Eastman simulation dataset for fault-detection and decision support systems

Deep unLSTM network: Features with memory information extracted from unlabeled data and their application on industrial unsupervised industrial fault detection

Fault detection in Tennessee Eastman process with temporal deep learning models

A Review on Data‐Driven Learning Approaches for Fault Detection and Diagnosis in Chemical Processes

Deep Semi-supervised Learning for Time Series Classification

Improving Unsupervised Image Clustering With Robust Learning

Time Series Change Point Detection with Self-Supervised Contrastive Predictive Coding

A Transformer-based Framework for Multivariate Time Series Representation Learning

Multiscale intelligent fault detection system based on agglomerative hierarchical clustering using stacked denoising autoencoder with temporal information

An optimized long short-term memory network based fault diagnosis model for chemical processes

An unsupervised fault diagnosis method for rolling bearing using STFT and generative neural networks

Deep Semantic Clustering by Partition Confidence Maximisation

Language Models are Few-Shot Learners

Learning To Classify Images Without Labels

TapNet: Multivariate Time Series Classification with Attentional Prototypical Network

A new unsupervised data mining method based on the stacked autoencoder for chemical process fault diagnosis

Applications of machine learning to machine fault diagnosis: A review and roadmap

Fault detection and diagnosis in process system using artificial intelligence-based cognitive technique

GATCluster: Self-Supervised Gaussian-Attention Network for Image Clustering

A Simple Framework for Contrastive Learning of Visual Representations

PyTorch: An Imperative Style, High-Performance Deep Learning Library

Fault Detection and Diagnosis Using Combined Autoencoder and Long Short-Term Memory Network

A novel process monitoring approach based on variational recurrent autoencoder

Design teacher and supervised dual stacked auto-encoders for quality-relevant fault detection in industrial process

Robust Anomaly Detection for Multivariate Time Series through Stochastic Recurrent Neural Network

When Does Label Smoothing Help?

Using Labeled Autoencoder to Supervise Neural Network Combined with k-Nearest Neighbor for Visual Industrial Process Monitoring

Fault diagnosis for distillation process based on CNN–DAE

Generative Adversarial Network Based Anomaly Detection on the Benchmark Tennessee Eastman Process

Metrics and Methods to Assess Building Fault Detection and Diagnosis Tools

MAD-GAN: Multivariate Anomaly Detection for Time Series Data with Generative Adversarial Networks

Unsupervised fault diagnosis of rolling bearings using a deep neural network based on generative adversarial networks

TIME SERIES

UMAP: Uniform Manifold Approximation and Projection

Invariant Information Clustering for Unsupervised Image Classification and Segmentation

Representation Learning with Contrastive Predictive Coding

Deep convolutional neural network model based chemical process fault diagnosis

A novel method for intelligent fault diagnosis of rolling bearings using ensemble deep auto-encoders

Fault Detection for Non-Gaussian Processes Using Generalized Canonical Correlation Analysis and Randomized Algorithms

A deep belief network based fault diagnosis model for complex chemical processes

Issues and Advances in Anomaly Detection Evaluation for Joint Human-Automated Systems

Attention is All you Need

Data augmentation of wearable sensor data for parkinson’s disease monitoring using convolutional neural networks

Discriminatively Boosted Image Clustering with Fully Convolutional Auto-Encoders

On Generative Topographic Mapping and Graph Theory combined approach for unsupervised non-linear data visualization and fault identification

Learning Discrete Representations via Information Maximizing Self-Augmented Training

Towards K-means-friendly Spaces: Simultaneous Deep Learning and Clustering

Benchmark Datasets for Fault Detection and Classification in Sensor Data

An improved SVM integrated GS-PCA fault diagnosis approach of Tennessee Eastman process

Rethinking the Inception Architecture for Computer Vision

A Hierarchical Deep Neural Network for Fault Diagnosis on Tennessee-Eastman Process

Unsupervised Deep Embedding for Clustering Analysis

Unsupervised and Semi-supervised Learning with Categorical Generative Adversarial Networks

An extensive reference dataset for fault detection and identification in batch processes

Adaptive total PLS based quality-relevant process monitoring with application to the Tennessee Eastman process

Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift

The development of an industrial-scale fed-batch fermentation simulation.

Adam: A Method for Stochastic Optimization

Deep Embedding Network for Clustering

Fault detection in the Tennessee Eastman benchmark process using dynamic principal components analysis based on decorrelated residuals (DPCA-DR)

A support vector clustering‐based probabilistic method for unsupervised fault detection and classification of complex chemical processes using unlabeled data

Fault diagnosis of Tennessee Eastman process with multi- scale PCA and ANFIS

A new integrated on-line fuzzy clustering and segmentation methodology with adaptive PCA approach for process monitoring and fault detection and diagnosis

Weighted and constrained possibilistic C-means clustering for online fault detection and isolation

Locally Consistent Concept Factorization for Document Clustering

A New Fault Diagnosis Method Using Fault Directions in Fisher Discriminant Analysis