Sensor Virtualization for Anomaly Detection of Turbo-Machinery Sensors—An Industrial Application
We apply a Granger causality and auto-correlation analysis to train a recurrent neural network (RNN) that acts as a virtual sensor model. These models can be used to check the status of several hundreds of sensors during turbo-machinery units’ operation. Checking the health of each sensor is a time-...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2023-07-01
|
Series: | Engineering Proceedings |
Subjects: | |
Online Access: | https://www.mdpi.com/2673-4591/39/1/96 |
_version_ | 1797580357292261376 |
---|---|
author | Sachin Shetty Valentina Gori Gianni Bagni Giacomo Veneri |
author_facet | Sachin Shetty Valentina Gori Gianni Bagni Giacomo Veneri |
author_sort | Sachin Shetty |
collection | DOAJ |
description | We apply a Granger causality and auto-correlation analysis to train a recurrent neural network (RNN) that acts as a virtual sensor model. These models can be used to check the status of several hundreds of sensors during turbo-machinery units’ operation. Checking the health of each sensor is a time-consuming activity. Training a supervised algorithm is not feasible because we do not know all the failure modes that the sensors can undergo. We use a semi-supervised approach and train an RNN (LSTM) on non-anomalous data to build a virtual sensor using other sensors as regressors. We use the Granger causality test to identify the set of input sensors for a given target sensor. Moreover, we look at the auto-correlation function (ACF) to understand the temporal dependency in data. We then compare the predicted signal vs. the real one to raise (in case) an anomaly in real time. Results report 96% precision and 100% recall. |
first_indexed | 2024-03-10T22:48:48Z |
format | Article |
id | doaj.art-f7aacaf9b8b3452d809e09f204a61049 |
institution | Directory Open Access Journal |
issn | 2673-4591 |
language | English |
last_indexed | 2024-03-10T22:48:48Z |
publishDate | 2023-07-01 |
publisher | MDPI AG |
record_format | Article |
series | Engineering Proceedings |
spelling | doaj.art-f7aacaf9b8b3452d809e09f204a610492023-11-19T10:31:32ZengMDPI AGEngineering Proceedings2673-45912023-07-013919610.3390/engproc2023039096Sensor Virtualization for Anomaly Detection of Turbo-Machinery Sensors—An Industrial ApplicationSachin Shetty0Valentina Gori1Gianni Bagni2Giacomo Veneri3Baker Hughes, Doddanakundi Industrial Area 2, Bengaluru 560037, IndiaBaker Hughes (Nuovo Pignone Tecnologie), Via Felice Matteucci 2, 50127 Firenze, ItalyBaker Hughes (Nuovo Pignone Tecnologie), Via Felice Matteucci 2, 50127 Firenze, ItalyBaker Hughes (Nuovo Pignone Tecnologie), Via Felice Matteucci 2, 50127 Firenze, ItalyWe apply a Granger causality and auto-correlation analysis to train a recurrent neural network (RNN) that acts as a virtual sensor model. These models can be used to check the status of several hundreds of sensors during turbo-machinery units’ operation. Checking the health of each sensor is a time-consuming activity. Training a supervised algorithm is not feasible because we do not know all the failure modes that the sensors can undergo. We use a semi-supervised approach and train an RNN (LSTM) on non-anomalous data to build a virtual sensor using other sensors as regressors. We use the Granger causality test to identify the set of input sensors for a given target sensor. Moreover, we look at the auto-correlation function (ACF) to understand the temporal dependency in data. We then compare the predicted signal vs. the real one to raise (in case) an anomaly in real time. Results report 96% precision and 100% recall.https://www.mdpi.com/2673-4591/39/1/96virtual sensoranomaly detectiontime series multi-regressionGranger causalityturbo-machinery |
spellingShingle | Sachin Shetty Valentina Gori Gianni Bagni Giacomo Veneri Sensor Virtualization for Anomaly Detection of Turbo-Machinery Sensors—An Industrial Application Engineering Proceedings virtual sensor anomaly detection time series multi-regression Granger causality turbo-machinery |
title | Sensor Virtualization for Anomaly Detection of Turbo-Machinery Sensors—An Industrial Application |
title_full | Sensor Virtualization for Anomaly Detection of Turbo-Machinery Sensors—An Industrial Application |
title_fullStr | Sensor Virtualization for Anomaly Detection of Turbo-Machinery Sensors—An Industrial Application |
title_full_unstemmed | Sensor Virtualization for Anomaly Detection of Turbo-Machinery Sensors—An Industrial Application |
title_short | Sensor Virtualization for Anomaly Detection of Turbo-Machinery Sensors—An Industrial Application |
title_sort | sensor virtualization for anomaly detection of turbo machinery sensors an industrial application |
topic | virtual sensor anomaly detection time series multi-regression Granger causality turbo-machinery |
url | https://www.mdpi.com/2673-4591/39/1/96 |
work_keys_str_mv | AT sachinshetty sensorvirtualizationforanomalydetectionofturbomachinerysensorsanindustrialapplication AT valentinagori sensorvirtualizationforanomalydetectionofturbomachinerysensorsanindustrialapplication AT giannibagni sensorvirtualizationforanomalydetectionofturbomachinerysensorsanindustrialapplication AT giacomoveneri sensorvirtualizationforanomalydetectionofturbomachinerysensorsanindustrialapplication |