A data-driven model for milling tool remaining useful life prediction with convolutional and stacked LSTM network

An, Qinglong; Tao, Zhengrui; Xu, Xingwei; Mansori, Mohamed El; Chen, Ming

doi:10.1016/j.measurement.2019.107461

Cited by 173 publications

(89 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Next, the stacked LSTM is employed as a predictor for forecasting, which is the raw value based on the defined time window [ 24 , 25 , 26 ]. The LSTM model was proposed in 1997, and it shows certain advantages in dealing with long-term time-series data [ 27 , 28 , 29 ].…”

Section: Framework and Methodsmentioning

confidence: 99%

Data-Driven Anomaly Detection Approach for Time-Series Streaming Data

Zhang

Guo

et al. 2020

Sensors

View full text Add to dashboard Cite

Recently, wireless sensor networks (WSNs) have been extensively deployed to monitor environments. Sensor nodes are susceptible to fault generation due to hardware and software failures in harsh environments. Anomaly detection for the time-series streaming data of sensor nodes is a challenging but critical fault diagnosis task, particularly in large-scale WSNs. The data-driven approach is becoming essential for the goal of improving the reliability and stability of WSNs. We propose a data-driven anomaly detection approach in this paper, named median filter (MF)-stacked long short-term memory-exponentially weighted moving average (LSTM-EWMA), for time-series status data, including the operating voltage and panel temperature recorded by a sensor node deployed in the field. These status data can be used to diagnose device anomalies. First, a median filter (MF) is introduced as a preprocessor to preprocess obvious anomalies in input data. Then, stacked long short-term memory (LSTM) is employed for prediction. Finally, the exponentially weighted moving average (EWMA) control chart is employed as a detector for recognizing anomalies. We evaluate the proposed approach for the panel temperature and operating voltage of time-series streaming data recorded by wireless node devices deployed in harsh field conditions for environmental monitoring. Extensive experiments were conducted on real time-series status data. The results demonstrate that compared to other approaches, the MF-stacked LSTM-EWMA approach can significantly improve the detection rate (DR) and false rate (FR). The average DR and FR values with the proposed approach are 95.46% and 4.42%, respectively. MF-stacked LSTM-EWMA anomaly detection also achieves a better F2 score than that achieved by other methods. The proposed approach provides valuable insights for anomaly detection in WSNs by detecting anomalies in the time-series status data recorded by wireless sensor nodes.

show abstract

Section: Framework and Methodsmentioning

confidence: 99%

Data-Driven Anomaly Detection Approach for Time-Series Streaming Data

Zhang

Guo

et al. 2020

Sensors

View full text Add to dashboard Cite

show abstract

“…Despite the causal and dilated convolution used for the TCN, the model may sometimes encounter problems such as gradient disappearance. To address this issue, the TCN structure is made to be generic, motivated by the residual structure presented in ResNet (An et al, 2020). In this paper, the residual convolution takes X series of input, transforms them, and the results are concatenated with the input.…”

Section: Methodsmentioning

confidence: 99%

“…Xia et al (2020) presented an ensemble framework with convolutional bi-directional LSTM for RUL prediction which could adaptively select trained base models for ensemble and further predicting RUL. An et al (2020) utilized convolutional stacked LSTM for RUL prediction of milling tools where time-domain and frequency-domain features were combined, encoded and denoised through unidirectional LSTM.…”

Section: Introductionmentioning

confidence: 99%

Remaining Useful Life Prediction Based on Improved Temporal Convolutional Network for Nuclear Power Plant Valves

Wang

Peng

et al. 2020

Front. Energy Res.

View full text Add to dashboard Cite

Proper risk assessment and monitoring of critical component is crucial to the safe operation of Nuclear Power Plants. One of the ways to ensure real-time monitoring is the development of Prognostics and Health Management systems for safety-critical equipment. Recently, the remaining useful life prediction (RUL) has been found to be important in ensuring predictive maintenance and avoiding critical component failure. With the development of artificial intelligent techniques, deep learning algorithms are becoming popular for RUL prediction. Consequently, this paper presents RUL prediction techniques for nuclear plant electric gate valves with a temporal convolution network (TCN). The main advantage of using TCN is its ability to capture and process useful information in short-term sensor measurement changes. Moreover, the efficiency of the proposed TCN is enhanced by incorporating a convolution auto-encoder as a preprocessing layer in its structure, which greatly improved the residual convolution mode. The proposed method is verified on the electric gate valves experimental dataset that represents the real-world operation of the valve, and the result obtained is compared with other conventional data-driven approaches. The evaluation result shows impressive performance of the proposed model in predicting the remaining service life of the gate valves used in the nuclear reactor control system. Moreover, the generalization of the proposed model is evaluated on the turbofan engine benchmark dataset. The evaluation result also shows improved performance in the predicted RUL. Broader application of the proposed TCN is envisaged for critical components in other industries.

show abstract

“…Excessive tool wear can even lead to tool breakages. The diagnoses of tool status were proposed by on-line and off-line monitoring [ 27 , 28 , 29 , 30 , 31 ]. For off-line monitoring, the tools are dismounted to measure the worn area.…”

Section: Introductionmentioning

confidence: 99%

Deep Learning Approach for Vibration Signals Applications

Chen

Lee

2021

Sensors

View full text Add to dashboard Cite

This study discusses convolutional neural networks (CNNs) for vibration signals analysis, including applications in machining surface roughness estimation, bearing faults diagnosis, and tool wear detection. The one-dimensional CNNs (1DCNN) and two-dimensional CNNs (2DCNN) are applied for regression and classification applications using different types of inputs, e.g., raw signals, and time-frequency spectra images by short time Fourier transform. In the application of regression and the estimation of machining surface roughness, the 1DCNN is utilized and the corresponding CNN structure (hyper parameters) optimization is proposed by using uniform experimental design (UED), neural network, multiple regression, and particle swarm optimization. It demonstrates the effectiveness of the proposed approach to obtain a structure with better performance. In applications of classification, bearing faults and tool wear classification are carried out by vibration signals analysis and CNN. Finally, the experimental results are shown to demonstrate the effectiveness and performance of our approach.

show abstract

A data-driven model for milling tool remaining useful life prediction with convolutional and stacked LSTM network

Cited by 173 publications

References 11 publications

Data-Driven Anomaly Detection Approach for Time-Series Streaming Data

Data-Driven Anomaly Detection Approach for Time-Series Streaming Data

Remaining Useful Life Prediction Based on Improved Temporal Convolutional Network for Nuclear Power Plant Valves

Deep Learning Approach for Vibration Signals Applications

Contact Info

Product

Resources

About