TUTCRIS - Tampereen teknillinen yliopisto

TUTCRIS

Fault detection of elevator systems using multilayer perceptron neural network

Tutkimustuotosvertaisarvioitu

Yksityiskohdat

AlkuperäiskieliEnglanti
Otsikko24th IEEE Conference on Emerging Technologies and Factory Automation
AlaotsikkoETFA 2019, September 10-13, 2019 in Zaragoza, Spain
ToimittajatAndrés Nogueiras
JulkaisupaikkaZaragoza, Spain
KustantajaIEEE
Sivut904-909
Sivumäärä6
Vuosikerta24
Painos2019
ISBN (elektroninen)978-1-7281-0303-7
ISBN (painettu)978-1-7281-0302-0
DOI - pysyväislinkit
TilaJulkaistu - 13 syyskuuta 2019
OKM-julkaisutyyppiA4 Artikkeli konferenssijulkaisussa
TapahtumaIEEE International Conference on Emerging Technologies and Factory Automation -
Kesto: 1 tammikuuta 2014 → …

Julkaisusarja

NimiIEEE Conference on Emerging Technologies and Factory Automation
KustantajaIEEE
ISSN (painettu)1946-0740
ISSN (elektroninen)2379-9560

Conference

ConferenceIEEE International Conference on Emerging Technologies and Factory Automation
Ajanjakso1/01/14 → …

Tiivistelmä

In this research, we propose a generic multilayer perceptron (MLP) neural network model based on deep learning algorithm for automatic calculation of highly informative deep features from the elevator time series data and based on extracted deep features faults are detected. Sensor data are labelled as healthy or faulty based on the maintenance actions recorded. The remaining healthy data are used for validation of the model to prove its efficacy in terms of avoiding false positives. We have achieved nearly 100% accuracy in fault detection along with avoiding false positives based on new extracted deep features, which outperform the results using existing features. Existing features are also classified with random forest (RF) to compare results. Multilayer perceptron neural network model based on deep learning approach provides better results due to the new deep features extracted from the dataset compared to existing features. Cross-validation method used with multilayer perceptron plays a significant role in improving accuracy of fault detection. Our model provides good classification and is robust against overfitting characteristics. This research will help various predictive maintenance systems to detect false alarms, which will reduce unnecessary visits of service technicians to installation sites.

!!ASJC Scopus subject areas

Julkaisufoorumi-taso

Tilastokeskuksen tieteenalat