In this paper, we propose a new algorithm for data extraction from time series data, and furthermore automatic calculation of highly informative deep features to be used in fault detection. In data extraction elevator start and stop events are extracted from sensor data, and a generic deep autoencoder model is also developed for automated feature extraction from the extracted proﬁles. After this, extracted deep features are classiﬁed with random forest algorithm for fault detection. Sensor data are labelled as healthy and faulty based on the maintenance actions recorded. The rest of the 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 outperforms results using existing features. Existing features are also classiﬁed with random forest to compare results. Our developed algorithm provides better results due to the new deep features extracted from the dataset when compared to existing features. This research will help various predictive maintenance systems to detect false alarms, which will in turn reduce unnecessary visits of service technicians to installation sites.
|Title of host publication||33rd Annual European Simulation and Modelling Conference|
|Publication status||Published - 2019|
|MoE publication type||A4 Article in a conference publication|
|Event||European Simulation and Modelling Conference - 33rd Annual European Simulation and Modelling Conference|
Duration: 28 Oct 2019 → 30 Oct 2019
|Conference||European Simulation and Modelling Conference|
|Period||28/10/19 → 30/10/19|