Generic Deep Autoencoder for Time-Series

引用格式

@INPROCEEDINGS{Terb2205:Hybrid, AUTHOR="Anika Terbuch and Paul O'Leary and Peter Auer", TITLE="Hybrid Machine Learning for Anomaly Detection in Industrial {Time-Series} Measurement Data", BOOKTITLE="2022 IEEE International Instrumentation and Measurement Technology Conference (I2MTC) (I2MTC 2022)", ADDRESS="Ottawa, Canada", DAYS=16, MONTH=may, YEAR=2022, KEYWORDS="Hybrid Learning; Outlier Detection; Time Series", ABSTRACT="This paper presents a parallel hybrid machine learning system for the identification of anomalies in large sets of multivariate time-series (MVTS) measurement data. The goal is to achieve a more reliable detection of anomalies in safety relevant applications. Key performance indicators (KPI) are used as a measure for predicted possible sources of error. Whereas, a long short term memory variational autoencoder (LSTM-VAE) is used to implement a data-driven model for the system behavior. Combined in a parallel hybrid manner they provide a more reliable anomaly detection. The proposed structure is validated with a case study relating to a ground improvement process for building foundations. The data consists of large sets of real-time multi-variate time-series sensor data, emanating from the instrumented drilling rig. The performance of the LSTM-VAE is optimized using a genetic algorithm to select the optimal values for the hyperparameters. The implemented framework will also support future research into hybrid learning systems applied to real-time machine data analysis." }