scholarly journals Adaptive event-triggered anomaly detection in compressed vibration data

2019 ◽  
Vol 122 ◽  
pp. 480-501 ◽  
Author(s):  
Yang Zhang ◽  
Paul Hutchinson ◽  
Nicholas A.J. Lieven ◽  
Jose Nunez-Yanez
Keyword(s):  
Anomaly Detection ◽  
Vibration Data ◽  
Event Triggered

scholarly journals Deep learning-based anomaly-onset aware remaining useful life estimation of bearings

2021 ◽  
Vol 7 ◽  
pp. e795
Author(s):  
Pooja Vinayak Kamat ◽  
Rekha Sugandhi ◽  
Satish Kumar
Keyword(s):  
Deep Learning ◽  
Anomaly Detection ◽  
Short Term Memory ◽  
Rotating Machinery ◽  
Remaining Useful Life ◽  
Operating Conditions ◽  
Learning Models ◽  
Vibration Data ◽  
Degradation Data ◽  
Useful Life

Remaining Useful Life (RUL) estimation of rotating machinery based on their degradation data is vital for machine supervisors. Deep learning models are effective and popular methods for forecasting when rotating machinery such as bearings may malfunction and ultimately break down. During healthy functioning of the machinery, however, RUL is ill-defined. To address this issue, this study recommends using anomaly monitoring during both RUL estimator training and operation. Essential time-domain data is extracted from the raw bearing vibration data, and deep learning models are used to detect the onset of the anomaly. This further acts as a trigger for data-driven RUL estimation. The study employs an unsupervised clustering approach for anomaly trend analysis and a semi-supervised method for anomaly detection and RUL estimation. The novel combined deep learning-based anomaly-onset aware RUL estimation framework showed enhanced results on the benchmarked PRONOSTIA bearings dataset under non-varying operating conditions. The framework consisting of Autoencoder and Long Short Term Memory variants achieved an accuracy of over 90% in anomaly detection and RUL prediction. In the future, the framework can be deployed under varying operational situations using the transfer learning approach.

scholarly journals Transfer Learning Strategies for Anomaly Detection in IoT Vibration Data

2021 ◽  
Author(s):  
Clemens Heistracher ◽  
Anahid Jalali ◽  
Indu Strobl ◽  
Axel Suendermann ◽  
Sebastian Meixner ◽  
...  
Keyword(s):  
Anomaly Detection ◽  
Learning Strategies ◽  
Transfer Learning ◽  
Training Data ◽  
Sensor Data ◽  
Classification Models ◽  
Vibration Data ◽  
Starting Point ◽  
Wide Range ◽  
Anomaly Classification

<div>Abstract—An increasing number of industrial assets are equipped with IoT sensor platforms and the industry now expects data-driven maintenance strategies with minimal deployment costs. However, gathering labeled training data for supervised tasks such as anomaly detection is costly and often difficult to implement in operational environments. Therefore, this work aims to design and implement a solution that reduces the required amount of data for training anomaly classification models on time series sensor data and thereby brings down the overall deployment effort of IoT anomaly detection sensors. We set up several in-lab experiments using three peristaltic pumps and investigated approaches for transferring trained anomaly detection models across assets of the same type. Our experiments achieved promising effectiveness and provide initial evidence that transfer learning could be a suitable strategy for using pretrained anomaly classification models across industrial assets of the same type with minimal prior labeling and training effort. This work could serve as a starting point for more general, pretrained sensor data embeddings, applicable to a wide range of assets.</div>

scholarly journals Real-Time Anomaly Detection using Average Level-Crossing Rate

2020 ◽  
Vol 9 (4) ◽  
pp. 2693-2698
Keyword(s):  
Anomaly Detection ◽  
Real Time ◽  
Warning Signal ◽  
Rate Of Change ◽  
Percentage Increase ◽  
Level Crossing ◽  
Vibration Signals ◽  
Average Value ◽  
Vibration Data ◽  
Level Crossing Rate

Vibration data collected from piezoelectric sensors serve as a means for detecting faults in machines that have rotating parts. The sensor output that is sampled at the Nyquist rate is stored for analysis of faults in the traditional condition monitoring system. The massive amount of data makes the analysis very difficult. Very complex procedures are adopted for anomaly detection in standard methods. The proposed system works on the analog output of the sensor and does not require conventional steps like sampling, feature extraction, classification, or computation of the spectrum. It is a simple system that performs real-time detection of anomalies in the bearing of a machine using vibration signals. Faults in the machines usually create an increase in the frequency of the vibration data. The amplitude of the signal also changes in some situations. The increase in amplitude or frequency leads to a corresponding increase in the level-crossing rate, which is a parameter indicating the rate of change of a signal. Based on the percentage increase in the average value of the level-crossing rate (ALCR), a suitable warning signal can be issued. It does not require the data from a faulty machine to set the thresholds. The proposed algorithm has been tested with standard data sets. There is a clear distinction between the ALCR values of normal and faulty machines, which has been used to release accurate indications about the fault. If the noise conditions do not vary much, the pre-processing of the input signal is not needed. The vibration signals acquired with faulty bearings have ALCR values, ranging from 3.48 times to 10.71 times the average value of ALCR obtained with normal bearing. Hence the proposed system offers bearing fault detection with100% accuracy

Transfer Learning Strategies for Anomaly Detection in IoT Vibration Data

2021 ◽  
Author(s):  
Clemens Heistracher ◽  
Anahid Jalali ◽  
Indu Strobl ◽  
Axel Suendermann ◽  
Sebastian Meixner ◽  
...  
Keyword(s):  
Anomaly Detection ◽  
Learning Strategies ◽  
Transfer Learning ◽  
Vibration Data

scholarly journals Transfer Learning Strategies for Anomaly Detection in IoT Vibration Data

2021 ◽  
Author(s):  
Clemens Heistracher ◽  
Anahid Jalali ◽  
Indu Strobl ◽  
Axel Suendermann ◽  
Sebastian Meixner ◽  
...  
Keyword(s):  
Anomaly Detection ◽  
Learning Strategies ◽  
Transfer Learning ◽  
Training Data ◽  
Sensor Data ◽  
Classification Models ◽  
Vibration Data ◽  
Starting Point ◽  
Wide Range ◽  
Anomaly Classification

<div>Abstract—An increasing number of industrial assets are equipped with IoT sensor platforms and the industry now expects data-driven maintenance strategies with minimal deployment costs. However, gathering labeled training data for supervised tasks such as anomaly detection is costly and often difficult to implement in operational environments. Therefore, this work aims to design and implement a solution that reduces the required amount of data for training anomaly classification models on time series sensor data and thereby brings down the overall deployment effort of IoT anomaly detection sensors. We set up several in-lab experiments using three peristaltic pumps and investigated approaches for transferring trained anomaly detection models across assets of the same type. Our experiments achieved promising effectiveness and provide initial evidence that transfer learning could be a suitable strategy for using pretrained anomaly classification models across industrial assets of the same type with minimal prior labeling and training effort. This work could serve as a starting point for more general, pretrained sensor data embeddings, applicable to a wide range of assets.</div>

scholarly journals Experiments on city train vibration anomaly detection Using deep learning approaches

2020 ◽  
Vol 20 (1) ◽  
pp. 329
Author(s):  
Taehee Kim ◽  
Cheolwoo Ro ◽  
Kiho Suh
Keyword(s):  
Deep Learning ◽  
Anomaly Detection ◽  
Real World ◽  
Data Science ◽  
Automated Detection ◽  
Learning Approaches ◽  
Real World Data ◽  
Vibration Data ◽  
Preliminary Research ◽  
Train Vibration

Anomaly detection is widely in demand in the field where automated detection of anomalous conditions in many observation tasks. While conventional data science approaches have shown interesting results, deep learning approaches to anomaly detection problems reveal new perspectives of possibilities especially where massive amount of data need to be handled. We develop anomaly detection applications on city train vibration data using deep learning approaches. We carried out preliminary research on anomaly detection in general and applied our real world data to existing solutions. In this paper, we provide a survey on anomaly detection and analyse our results of experiments using deep learning approaches.

Sign in / Sign up

Export Citation Format

Share Document