Fault Detection of Elevator Systems Using Deep Autoencoder Feature Extraction

Mishra, Krishna Mohan; Krogerus, Tomi; Huhtala, Kalevi

doi:10.1109/rcis.2019.8876984

Cited by 9 publications

(5 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Equipment Fault Parameters Method [40] Motor Bearing Current signal CNN [41] CNC machine Condition Vibrations ANN [42] Motor Operations Current and voltage signal ANN/MLP [43] Pump Condition Multi variables AE [44] CNC machine Mechanical Vibrations signal SAE [45] Motor Operations Stator currents ANN [46] Motor Bearing Vibrations signal LSTM [47] Rotating machinery Bearing Vibrations signal AE+ MLP [48] Rotating machinery Bearing Vibrations signal LSTM [49] Cooling radiator Condition Thermal image CNN [50] Rotating machinery Degradation image Infrared image streams (CNN+LSTM) (LSTM+AE) [51] Compressor Condition Multi variables RNN-LSTM [52] Elevator system Movement Acceleration data AE [53] Motor Condition Current signal EWT-CNN [54] Autoclave sterilizer Pump NTC thermistors LSTM [55] Worm gearboxes Operations Multi variables CNN [56] Rotating machinery Rotor, bearing Vibration signals CNN [57] Railcar factories Wheel bearing Temperature variation ANN [58] Rotating machinery Bearing Accelerometers CNN [59] Motor Bearing Current signal ANN [60] Conveyors system Motor Multi variables CNN [61] Motor Bearing Accelerometer LSTM+RNN [62] Motor Rotor bar Torque control ANN [63] Motor Stator winding stator currents ANN [64] Motor Condition Vibrations signal ANN [65] Rotating machinery Bearing Rotation speed, load levels CNN [66] Motor Stator winding Multi variables MLP+LSTM+CNN [67] Motor Operations Current signal ANN [68] Motor+rotating equipment Bearing Vibrations signal CNN+DNN [69] Motor Bearing Microphone, accelerometer DCNN+CNN-LSTM+LSTM…”

Section: Workmentioning

confidence: 99%

Predictive Maintenance of Electromechanical Systems Using Deep Learning Algorithms: Review

Abood¹,

Nasser²,

Al-Khazraji³

2022

ISI

View full text Add to dashboard Cite

Predictive Maintenance (PM) is a major part of smart manufacturing in the fourth industrial revolution. The classical fault diagnosis approach for complex systems such as an electromechanical system is not effective. Taking the advantage of the successful implementations of PM together with Deep Learning (DL) methods replaces the conventional diagnosis methods with modern diagnosis methods. This study intends to aid experts, engineers, and technicians in different electromechanical systems in comprehending how the PM used DL methods to find the multi-fault diagnosis. In this direction, this paper presents a comprehensive review of recent works of DL techniques that are applied to PM for electromechanical systems by classifying the research according to equipment, fault, parameters, and method. To perform the review, 30 papers that are published in proceedings and journals are reviewed within a time window between the years 2016 to 2022. In the context of the electromechanical system, it is observed that motors are the most equipment selected for PM. Moreover, stator winding faults are found to be less selected than bearing for diagnosis of the unhealthy status of the motor. In terms of DL methods, the study reveals that AE, LSTM, and CNN are mostly used. In addition, poorly mixed models of DL methods are also noticed. Finally, finding the optimal design variable of the DL architecture was not widely explored.

show abstract

Section: Workmentioning

confidence: 99%

Predictive Maintenance of Electromechanical Systems Using Deep Learning Algorithms: Review

Abood¹,

Nasser²,

Al-Khazraji³

2022

ISI

View full text Add to dashboard Cite

show abstract

“…This characteristic of this type of neural network makes it perfect for feature extraction in PdM even with high-dimensional data [36]. Likewise, AEs have been successfully implemented for feature extraction, dimensionality reduction, anomaly detection and time-series prediction [37][38][39]. Given the performance of AEs and convolutional neural networks in handling high-dimensional data and extracting features, it has been decided to use these architectures in the proposed method.…”

Section: Introductionmentioning

confidence: 99%

“…In fact, to the best of our knowledge, there are currently no models which try to deal with PdM without R2F data in an optimized way as proposed in this paper. The previous studies targeting PdM, normally consider a complete data set of R2F data [37]. Similarly, some other are only limited to feature extraction and anomaly detection of their system in hand [40].…”

Section: Introductionmentioning

confidence: 99%

Predictive Maintenance: An Autoencoder Anomaly-Based Approach for a 3 DoF Delta Robot

Fathi

Venn

Honegger

2021

Sensors

View full text Add to dashboard Cite

Performing predictive maintenance (PdM) is challenging for many reasons. Dealing with large datasets which may not contain run-to-failure data (R2F) complicates PdM even more. When no R2F data are available, identifying condition indicators (CIs), estimating the health index (HI), and thereafter, calculating a degradation model for predicting the remaining useful lifetime (RUL) are merely impossible using supervised learning. In this paper, a 3 DoF delta robot used for pick and place task is studied. In the proposed method, autoencoders (AEs) are used to predict when maintenance is required based on the signal sequence distribution and anomaly detection, which is vital when no R2F data are available. Due to the sequential nature of the data, nonlinearity of the system, and correlations between parameter time-series, convolutional layers are used for feature extraction. Thereafter, a sigmoid function is used to predict the probability of having an anomaly given CIs acquired from AEs. This function can be manually tuned given the sensitivity of the system or optimized by solving a minimax problem. Moreover, the proposed architecture can be used for fault localization for the specified system. Additionally, the proposed method can calculate RUL using Gaussian process (GP), as a degradation model, given HI values as its input.

show abstract

“…This characteristic of this type of neural network makes it perfect for feature extraction in PdM even with high dimensional data [25]. Likewise, AEs have been successfully implemented for feature extraction, dimensionality reduction, anomaly detection and time-series prediction [26][27][28]. Given the performance of AEs and convolutional neural networks in handling high dimensional data and extracting features, it has been decided to use these architectures in the proposed method.…”

Section: Introductionmentioning

confidence: 99%

“…What is generally missing from conducted studies is a hybrid architecture that benefits from different capabilities of the available data-driven models for anomaly detection and using this information for calculating RUL given that no R2F data is available. The previous studies targeting PdM, normally consider a complete data set of R2F data [26]. However, some other are only limited to feature extraction and anomaly detection of their system in hand [29].…”

Section: Introductionmentioning

confidence: 99%

Predictive Maintenance: An Autoencoder Anomaly-based Approach for a 3 DoF Delta Robot

Fathi¹,

Venn²,

Honegger³

2021

Preprint

View full text Add to dashboard Cite

Performing predictive maintenance (PdM) is challenging for many reasons. Dealing with large datasets which may not contain run-to-failure data (R2F) complicates PdM even more. When no R2F data are available, identifying condition indicators (CIs), estimating the health index (HI), and thereafter, calculating a degradation model for predicting the remaining useful lifetime (RUL) are merely impossible using supervised learning. In this paper, a 3 dof delta robot used for pick and place task is studied. In the proposed method, autoencoders (AEs) are used to predict when maintenance is required based on the signal sequence distribution and anomaly detection, which is vital when no R2F data is available. Due to the sequential nature of the data, non-linearity of the system, and correlations between parameter time series, convolutional layers are used for feature extraction. Thereafter, a sigmoid function is used to predict the probability of having an anomaly given CIs acquired from AEs. This function can be manually tuned given the sensitivity of the system or optimized by solving a minimax problem. Moreover, the proposed architecture can be used for fault localization for the specified system. Additionally, the proposed method is capable of calculating RUL using Gaussian process (GP), as a degradation model, given HI values as its input.

show abstract

Fault Detection of Elevator Systems Using Deep Autoencoder Feature Extraction

Cited by 9 publications

References 27 publications

Predictive Maintenance of Electromechanical Systems Using Deep Learning Algorithms: Review

Predictive Maintenance of Electromechanical Systems Using Deep Learning Algorithms: Review

Predictive Maintenance: An Autoencoder Anomaly-Based Approach for a 3 DoF Delta Robot

Predictive Maintenance: An Autoencoder Anomaly-based Approach for a 3 DoF Delta Robot

Contact Info

Product

Resources

About