An Encoder-Decoder Based Approach for Anomaly Detection with Application in Additive Manufacturing
This addresses anomaly detection for industrial manufacturing processes, offering an incremental improvement with unsupervised learning and predictive capabilities.
The paper tackles anomaly detection in sequential sensor data from industrial manufacturing by proposing an unsupervised encoder-decoder deep learning approach that detects anomalies and predicts future process steps, demonstrating its ability to identify injected anomalies and reveal previously unknown temperature non-uniformity in a real-world testbed dataset.
We present a novel unsupervised deep learning approach that utilizes the encoder-decoder architecture for detecting anomalies in sequential sensor data collected during industrial manufacturing. Our approach is designed not only to detect whether there exists an anomaly at a given time step, but also to predict what will happen next in the (sequential) process. We demonstrate our approach on a dataset collected from a real-world testbed. The dataset contains images collected under both normal conditions and synthetic anomalies. We show that the encoder-decoder model is able to identify the injected anomalies in a modern manufacturing process in an unsupervised fashion. In addition, it also gives hints about the temperature non-uniformity of the testbed during manufacturing, which is what we are not aware of before doing the experiment.