TUTCRIS - Tampereen teknillinen yliopisto

TUTCRIS

Condition monitoring of elevator systems using deep neural network

Tutkimustuotosvertaisarvioitu

Standard

Condition monitoring of elevator systems using deep neural network. / Mishra, Krishna; Huhtala, Kalevi.

ICORES 2020 - Proceedings of the 9th International Conference on Operations Research and Enterprise Systems. toim. / Greg H. Parlier; Federico Liberatore; Marc Demange. Vuosikerta 1 SCITEPRESS, 2020. s. 381-387.

Tutkimustuotosvertaisarvioitu

Harvard

Mishra, K & Huhtala, K 2020, Condition monitoring of elevator systems using deep neural network. julkaisussa GH Parlier, F Liberatore & M Demange (toim), ICORES 2020 - Proceedings of the 9th International Conference on Operations Research and Enterprise Systems. Vuosikerta. 1, SCITEPRESS, Sivut 381-387, Valletta, Malta, 22/02/20. https://doi.org/10.5220/0009348803810387

APA

Mishra, K., & Huhtala, K. (2020). Condition monitoring of elevator systems using deep neural network. teoksessa G. H. Parlier, F. Liberatore, & M. Demange (Toimittajat), ICORES 2020 - Proceedings of the 9th International Conference on Operations Research and Enterprise Systems (Vuosikerta 1, Sivut 381-387). SCITEPRESS. https://doi.org/10.5220/0009348803810387

Vancouver

Mishra K, Huhtala K. Condition monitoring of elevator systems using deep neural network. julkaisussa Parlier GH, Liberatore F, Demange M, toimittajat, ICORES 2020 - Proceedings of the 9th International Conference on Operations Research and Enterprise Systems. Vuosikerta 1. SCITEPRESS. 2020. s. 381-387 https://doi.org/10.5220/0009348803810387

Author

Mishra, Krishna ; Huhtala, Kalevi. / Condition monitoring of elevator systems using deep neural network. ICORES 2020 - Proceedings of the 9th International Conference on Operations Research and Enterprise Systems. Toimittaja / Greg H. Parlier ; Federico Liberatore ; Marc Demange. Vuosikerta 1 SCITEPRESS, 2020. Sivut 381-387

Bibtex - Lataa

@inproceedings{9ac1c040b1ec4033bfa27123212619d4,
title = "Condition monitoring of elevator systems using deep neural network",
abstract = "In this research, we propose a generic deep autoencoder model for automatic calculation of highly informative deep features from the elevator data. Random forest algorithm is used for fault detection based on extracted deep features. Maintenance actions recorded are used to label the sensor data into healthy or faulty. In our research, we have included all fault types present for each elevator. The rest of the healthy data is used for validation of the model to prove its efficacy in terms of avoiding false positives. New extracted deep features provide 100{\%} accuracy in fault detection along with avoiding false positives, which is better than statistical features. Random forest was also used to detect faults based on statistical features to compare results. New deep features extracted from the dataset with deep autoencoder random forest outperform the statistical features. Good classification and robustness against overfitting are key characteristics of our model. This research will help to reduce unnecessary visits of service technicians to installation sites by detecting false alarms in various predictive maintenance systems.",
author = "Krishna Mishra and Kalevi Huhtala",
note = "jufoid=87961",
year = "2020",
month = "2",
day = "24",
doi = "10.5220/0009348803810387",
language = "English",
volume = "1",
pages = "381--387",
editor = "Parlier, {Greg H. } and Liberatore, {Federico } and Demange, {Marc }",
booktitle = "ICORES 2020 - Proceedings of the 9th International Conference on Operations Research and Enterprise Systems",
publisher = "SCITEPRESS",

}

RIS (suitable for import to EndNote) - Lataa

TY - GEN

T1 - Condition monitoring of elevator systems using deep neural network

AU - Mishra, Krishna

AU - Huhtala, Kalevi

N1 - jufoid=87961

PY - 2020/2/24

Y1 - 2020/2/24

N2 - In this research, we propose a generic deep autoencoder model for automatic calculation of highly informative deep features from the elevator data. Random forest algorithm is used for fault detection based on extracted deep features. Maintenance actions recorded are used to label the sensor data into healthy or faulty. In our research, we have included all fault types present for each elevator. The rest of the healthy data is used for validation of the model to prove its efficacy in terms of avoiding false positives. New extracted deep features provide 100% accuracy in fault detection along with avoiding false positives, which is better than statistical features. Random forest was also used to detect faults based on statistical features to compare results. New deep features extracted from the dataset with deep autoencoder random forest outperform the statistical features. Good classification and robustness against overfitting are key characteristics of our model. This research will help to reduce unnecessary visits of service technicians to installation sites by detecting false alarms in various predictive maintenance systems.

AB - In this research, we propose a generic deep autoencoder model for automatic calculation of highly informative deep features from the elevator data. Random forest algorithm is used for fault detection based on extracted deep features. Maintenance actions recorded are used to label the sensor data into healthy or faulty. In our research, we have included all fault types present for each elevator. The rest of the healthy data is used for validation of the model to prove its efficacy in terms of avoiding false positives. New extracted deep features provide 100% accuracy in fault detection along with avoiding false positives, which is better than statistical features. Random forest was also used to detect faults based on statistical features to compare results. New deep features extracted from the dataset with deep autoencoder random forest outperform the statistical features. Good classification and robustness against overfitting are key characteristics of our model. This research will help to reduce unnecessary visits of service technicians to installation sites by detecting false alarms in various predictive maintenance systems.

UR - http://www.icores.org/Home.aspx?y=2020

U2 - 10.5220/0009348803810387

DO - 10.5220/0009348803810387

M3 - Conference contribution

VL - 1

SP - 381

EP - 387

BT - ICORES 2020 - Proceedings of the 9th International Conference on Operations Research and Enterprise Systems

A2 - Parlier, Greg H.

A2 - Liberatore, Federico

A2 - Demange, Marc

PB - SCITEPRESS

ER -