A data-driven prognostics framework for tool remaining useful life estimation in tool condition monitoring

Zhang, Chong; Hong, G.Y.; Xu, Huan; Tan, Kay Chen; Zhou, Jun; Chan, Hian Leng; Li, Haizhou

doi:10.1109/etfa.2017.8247659

Cited by 17 publications

(14 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The imbalance ratio (IR) of this dataset is 10. The data preprocessing and time window process are the same with [5].…”

Section: B Experimental Setupmentioning

confidence: 99%

“…C LASS imbalance with disproportionate number of class instances commonly affects the quality of learning algorithms. Multifarious imbalanced data problems exist in numerous real-world applications, such as fault diagnosis [1], recommendation systems, fraud detection [2], risk management [3], tool condition monitoring [4], [5], [6] and medical diagnosis [7], brain computer interface (BCI) [8], [9], data visualization [10], etc. As a result of the equal misclassification costs or balanced class distribution assumption, the traditional learning algorithms are prone to the majority class when dealing with complicated classification problems that have skewed class distribution.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Cost-Sensitive Deep Belief Network for Imbalanced Classification

Zhang

Tan

et al. 2019

IEEE Trans. Neural Netw. Learning Syst.

Self Cite

228

View full text Add to dashboard Cite

Imbalanced data with a skewed class distribution are common in many real-world applications. Deep Belief Network (DBN) is a machine learning technique that is effective in classification tasks. However, conventional DBN does not work well for imbalanced data classification because it assumes equal costs for each class. To deal with this problem, cost-sensitive approaches assign different misclassification costs for different classes without disrupting the true data sample distributions. However, due to lack of prior knowledge, the misclassification costs are usually unknown and hard to choose in practice. Moreover, it has not been well studied as to how cost-sensitive learning could improve DBN performance on imbalanced data problems. This paper proposes an evolutionary cost-sensitive deep belief network (ECS-DBN) for imbalanced classification. ECS-DBN uses adaptive differential evolution to optimize the misclassification costs based on the training data that presents an effective approach to incorporating the evaluation measure (i.e., G-mean) into the objective function. We first optimize the misclassification costs, and then apply them to DBN. Adaptive differential evolution optimization is implemented as the optimization algorithm that automatically updates its corresponding parameters without the need of prior domain knowledge. The experiments have shown that the proposed approach consistently outperforms the state of the art on both benchmark data sets and real-world data set for fault diagnosis in tool condition monitoring.

show abstract

“…The imbalance ratio (IR) of this dataset is 10. The data preprocessing and time window process are the same with [5].…”

Section: B Experimental Setupmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

A Cost-Sensitive Deep Belief Network for Imbalanced Classification

Zhang

Tan

et al. 2019

IEEE Trans. Neural Netw. Learning Syst.

Self Cite

228

View full text Add to dashboard Cite

show abstract

“…However, the aging mechanism of real device systems is usually non-linear, randomized and dynamic, which causes difficulty to obtain accurate results by an analytical model [5]. The data-driven approach is designed to transform the device's detection and operational data into the degradation information of the device, which reveals the system operational status and corresponding degradation mechanism model [6]. Such methods exploit Artificial Intelligence (AI) and statistical methods to learn the degradation patterns of devices and predict the remaining useful life (RUL) of the device [7].…”

Section: Introductionmentioning

confidence: 99%

Empirical Mode Decomposition and Temporal Convolutional Networks for Remaining Useful Life Estimation

Yang

Yao

et al. 2019

Int J Parallel Prog

View full text Add to dashboard Cite

Remaining useful life (RUL) prediction plays an important role in guaranteeing safe operation and reducing maintenance cost in modern industry. In this paper, we present a novel deep learning method for RUL estimation based on time empirical mode decomposition (EMD) and temporal convolutional networks (TCN). The proposed framework can effectively reveal the non-stationary characteristics of bearing degradation signals and acquire time-series degradation signals which namely intrinsic mode functions through empirical mode decomposition. Furthermore, the feature information is used as the input to convolution layer and trained by TCN to predict remaining useful life. The proposed EMD-TCN model structure maintains a superior result compared to several state-of-the-art convolutional algorithms on public data sets. Experimental results show that the average score of EMD-TCN model is improved by 10-20% than traditional convolutional algorithms.

show abstract

“…However, the aging mechanism of real device systems is usually non-linear, randomized and dynamic, which causes difficulty to obtain accurate results by an analytical model. [6] The data-driven approach is designed to transform the device's detection and operational data into the degradation information of device, which reveals the system operational status and corresponding degradation mechanism model [7]. Such methods exploit Artificial Intelligence (AI) and statistical methods to learn the degradation patterns of devices, and predict the Remaining Useful Life (RUL) of the device [8].…”

Section: Introductionmentioning

confidence: 99%

Remaining Useful Life Estimation by Empirical Mode Decomposition and Ensemble Deep Convolution Neural Networks

Yao

Yang

et al. 2019

2019 IEEE International Conference on Prognostics and Health Management (ICPHM)

View full text Add to dashboard Cite

Bearing remaining useful life (RUL) prediction plays a key role in guaranteeing safe operation and reducing maintenance costs. In this paper, we present a novel deep learning method for RUL estimation approach through time Empirical Mode Decomposition (EMD) and Convolutional Neural Network (CNN). EMD can reveal the nonstationary property of bearing degradation signals effectively. After acquiring time-series degradation signals, namely Intrinsic Mode Functions (IMF), we can utilize the featured information as the input of Convolution layer of models. Here, we introduce an EMD-CNN model structure, which keeps the global and local information synchronously compared to a traditional CNN. In order to get a more accurate prediction, an ensemble model with several weighting methods are proposed, where the experiment indicates an improvement of performance.

show abstract

A data-driven prognostics framework for tool remaining useful life estimation in tool condition monitoring

Cited by 17 publications

References 26 publications

A Cost-Sensitive Deep Belief Network for Imbalanced Classification

A Cost-Sensitive Deep Belief Network for Imbalanced Classification

Empirical Mode Decomposition and Temporal Convolutional Networks for Remaining Useful Life Estimation

Remaining Useful Life Estimation by Empirical Mode Decomposition and Ensemble Deep Convolution Neural Networks

Contact Info

Product

Resources

About