SensorSCAN: Self-Supervised Learning and Deep Clustering for Fault Diagnosis in Chemical Processes

TL;DR

SensorSCAN is an unsupervised deep learning method that effectively detects and diagnoses faults in chemical processes using raw sensor data, reducing the need for expert annotation and performing well even with limited labeled data.

Contribution

We introduce SensorSCAN, a novel self-supervised deep clustering approach for fault diagnosis in chemical processes, outperforming existing methods without requiring extensive labeled data.

Findings

Significantly outperforms existing fault detection methods (+0.2-0.3 TPR at fixed FPR)

Effectively detects most process faults without expert annotations

Achieves near state-of-the-art performance with minimal labeled data

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.