scholarly journals A Novel Feature-Transferred Prediction Network For Remaining Useful Life of Rotating Machinery

2021 ◽  
Author(s):  
Tianjiao Lin ◽  
Huaqing Wang ◽  
Liuyang Song ◽  
Lingli Cui
Keyword(s):  
Neural Network ◽  
Prediction Accuracy ◽  
Rotating Machinery ◽  
Remaining Useful Life ◽  
Operating Conditions ◽  
Cnc Machine Tools ◽  
Data Set ◽  
Public Data ◽  
Target Network ◽  
Useful Life

Abstract With the increasing complexity of CNC machine tools and other rotating machinery, it becomes more and more important to improve the reliability of such machines. It is necessary to estimate the remaining Useful Life (RUL) of the important parts such as bearings of these devices. However, the operating conditions of such parts are often very complicated, and there is a great difference between different devices. Therefore, it is difficult to use the traditional mechanism analysis method, which is not only very hard, but also generally has low prediction accuracy. In order to solve the above problems, this paper proposes a Feature-transferred Prediction Network (FTPN), which can adapt to various working conditions, combining with the neural network method in the field of artificial intelligence, and effectively realizes RUL prediction. Since the existing neural network methods were originally proposed to solve the problems in other fields, such as semantic recognition, this paper uses the feature transfer method to improve them. Firstly, the source network based on Convolutional Neural Network (CNN) is pre-trained for fault recognition, and the fault feature information extracted after training is stored in the fault feature layer. Second, CNN and Gate Recurrent Unit (GRU) are used to build target networks and fit the relationship between time series and remaining longevity. Finally, a special loss function is designed to transfer the features extracted from the source network to the target network to help the target network learn fault features and better predict the mechanical RUL. In order to verify the effectiveness of the proposed method, experiments are carried out using the public data set of accelerated life of bearings, and high prediction accuracy is obtained, which proves that the proposed method has certain generalization. The comparison with the existing methods on the same data set shows that the proposed method has a broad industrial application prospect.

Remaining useful life prediction for lithium-ion batteries using particle filter and artificial neural network

2019 ◽  
Vol 120 (2) ◽  
pp. 312-328 ◽  
Author(s):  
Wei Qin ◽  
Huichun Lv ◽  
Chengliang Liu ◽  
Datta Nirmalya ◽  
Peyman Jahanshahi
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Particle Filter ◽  
Prediction Accuracy ◽  
Lithium Ion ◽  
Remaining Useful Life ◽  
Content Type ◽  
Public Data ◽  
Useful Life ◽  
Artificial Neural

Purpose With the promotion of lithium-ion battery, it is more and more important to ensure the safety usage of the battery. The purpose of this paper is to analyze the battery operation data and estimate the remaining life of the battery, and provide effective information to the user to avoid the risk of battery accidents. Design/methodology/approach The particle filter (PF) algorithm is taken as the core, and the double-exponential model is used as the state equation and the artificial neural network is used as the observation equation. After the importance resampling process, the battery degradation curve is obtained after getting the posterior parameter, and then the system could estimate remaining useful life (RUL). Findings Experiments were carried out by using the public data set. The results show that the Bayesian-based posterior estimation model has a good predictive effect and fits the degradation curve of the battery well, and the prediction accuracy will increase gradually as the cycle increases. Originality/value This paper combines the advantages of the data-driven method and PF algorithm. The proposed method has good prediction accuracy and has an uncertain expression on the RUL of the battery. Besides, the method proposed is relatively easy to implement in the battery management system, which has high practical value and can effectively avoid battery using risk for driver safety.

scholarly journals Remaining Useful Life Prediction Using Temporal Convolution with Attention

AI ◽  
2021 ◽  
Vol 2 (1) ◽  
pp. 48-70
Author(s):  
Wei Ming Tan ◽  
T. Hui Teo
Keyword(s):  
Neural Network ◽  
Time Series ◽  
Time Series Data ◽  
Remaining Useful Life ◽  
Sensor Data ◽  
Series Data ◽  
Multiple Time ◽  
Data Set ◽  
Form Complex ◽  
Useful Life

Prognostic techniques attempt to predict the Remaining Useful Life (RUL) of a subsystem or a component. Such techniques often use sensor data which are periodically measured and recorded into a time series data set. Such multivariate data sets form complex and non-linear inter-dependencies through recorded time steps and between sensors. Many current existing algorithms for prognostic purposes starts to explore Deep Neural Network (DNN) and its effectiveness in the field. Although Deep Learning (DL) techniques outperform the traditional prognostic algorithms, the networks are generally complex to deploy or train. This paper proposes a Multi-variable Time Series (MTS) focused approach to prognostics that implements a lightweight Convolutional Neural Network (CNN) with attention mechanism. The convolution filters work to extract the abstract temporal patterns from the multiple time series, while the attention mechanisms review the information across the time axis and select the relevant information. The results suggest that the proposed method not only produces a superior accuracy of RUL estimation but it also trains many folds faster than the reported works. The superiority of deploying the network is also demonstrated on a lightweight hardware platform by not just being much compact, but also more efficient for the resource restricted environment.

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 A New Data-Driven Approach for Power IGBT Remaining Useful Life Estimation Based On Feature Reduction Technique and Neural Network

Electronics ◽  
2020 ◽  
Vol 9 (10) ◽  
pp. 1571
Author(s):  
Adla Ismail ◽  
Lotfi Saidi ◽  
Mounir Sayadi ◽  
Mohamed Benbouzid
Keyword(s):  
Neural Network ◽  
Accelerated Aging ◽  
Principal Component ◽  
Life Time ◽  
Time Domain Analysis ◽  
Energy System ◽  
Remaining Useful Life ◽  
Feature Reduction ◽  
Operating Conditions ◽  
Useful Life

The insulated gate bipolar transistor (IGBT) is a crucial component of power converters (PCVs) and is commonly used in several PCVs topologies. On the other hand, the investigation and the study of the IGBT component show several changes within its behavior and lifetime, while this component is highly influenced by the operating conditions. Indeed, the monitoring of this component is necessary to minimize unexpected downtime of the wind energy system (WES). However, an accurate prediction of IGBTs remaining useful life (RUL) is the key enabler for life-time-optimized operation. Consequently, this work proposes a new prognostic approach for online IGBTs monitoring that adopts the time-domain analysis to extract useful information that is used as an input in the generation of the health indicator. Moreover, this approach is based on combining both of principal component analysis (PCA) technique and the feedforward neural network (FFNN) technique. PCA is used to reduce features extracted from IGBTs and the FFNN is implemented to achieve online regression of the trend parameter obtained from the PCA technique. To investigate and evaluate the performance of our idea we used the NASA Ames Laboratory Prognostics Center of Excellence IGBTs accelerated aging database. Finally, the achieved results clearly show the strength of the new trend parameter for IGBTs RUL prediction. The most notable strong correlation within the proposed approach is in relation to accuracy value, with an acceptable average accuracy rate of 60.4%.

scholarly journals Quantum Gene Chain Coding Bidirectional Neural Network for Residual Useful Life Prediction of Rotating Machinery

2020 ◽  
Author(s):  
Feng Li ◽  
Yang-Yang Cheng ◽  
Bao-Ping Tang ◽  
Xue-Ming Zhou ◽  
Rui-Ping Xiong
Keyword(s):  
Neural Network ◽  
Time Series ◽  
Prediction Accuracy ◽  
Gradient Descent ◽  
Nonlinear Approximation ◽  
Descent Method ◽  
Rotating Machinery ◽  
Gradient Descent Method ◽  
Chain Coding ◽  
Useful Life

Abstract In classical recurrent neural networks, the pre- and post-relationships of time series tend to be neglected so that long-term overall memory is generally inaccessible; meanwhile, the weights are transferred and updated mainly by the gradient descent method, which leads to their low prediction accuracy and high computation cost in the application of residual useful life (RUL) prediction of rotating machinery (RM). In view of this, a quantum gene chain coding bidirectional neural network (QGCCBNN) is proposed to predict RUL of RM in this paper. In our proposed QGCCBNN, the quantum bidirectional transmission mechanism is designed to establish the pre- and post-relationships of time series for readjusting the weight parameters according to the feedback from the output layer, so that higher consistency between the input information and the overall memory of the network can be realized, thus endowing QGCCBNN with better nonlinear approximation ability. Moreover, in order to improve the global optimization ability and convergence speed, the quantum gene chain coding instead of gradient descent method is constructed to transmit and update data, in which the qubit probability amplitude real number coding is adopted and the cosine and sinusoidal qubit probability amplitudes corresponding to the minimum loss function are compared with those of the current time by the phase selection matrix for the directional parallel updating of the weight parameters. On this basis, a new RUL prediction method for RM is proposed, and higher prediction accuracy as well as desirable efficiency can be obtained due to the advantages of QGCCBNN in nonlinear approximation ability and convergence speed. The experimental example for RUL prediction of a double-row roller bearing demonstrates the effectiveness of our proposed method.

An improved deep convolution neural network for predicting the remaining useful life of rolling bearings

2020 ◽  
pp. 1-9
Author(s):  
Yiming Guo ◽  
Hui Zhang ◽  
Zhijie Xia ◽  
Chang Dong ◽  
Zhisheng Zhang ◽  
...  
Keyword(s):  
Neural Network ◽  
Degradation Process ◽  
Rolling Bearing ◽  
Remaining Useful Life ◽  
Convolution Neural Network ◽  
Dual Channel ◽  
Time Domain Data ◽  
Channel Input ◽  
Deep Convolution Neural Network ◽  
Useful Life

The rolling bearing is the crucial component in the rotating machinery. The degradation process monitoring and remaining useful life prediction of the bearing are necessary for the condition-based maintenance. The commonly used deep learning methods use the raw or processed time domain data as the input. However, the feature extracted by these approaches is insufficient and incomprehensive. To tackle this problem, this paper proposed an improved Deep Convolution Neural Network with the dual-channel input from the time and frequency domain in parallel. The proposed methodology consists of two stages: the incipient failure identification and the degradation process fitting. To verify the effectiveness of the method, the IEEE PHM 2012 dataset is adopted to compare the proposed method and other commonly used approaches. The results show that the improved Deep Convolution Neural Network can effectively describe the degradation process for the rolling bearing.

scholarly journals Predicting Remaining Useful Life of Rolling Bearings Based on Reliable Degradation Indicator and Temporal Convolution Network with the Quantile Regression

2021 ◽  
Vol 11 (11) ◽  
pp. 4773
Author(s):  
Qiaoping Tian ◽  
Honglei Wang
Keyword(s):  
Quantile Regression ◽  
Probability Density ◽  
Degradation Process ◽  
Remaining Useful Life ◽  
Health State ◽  
Rolling Bearings ◽  
Data Set ◽  
Time Frequency ◽  
Useful Life ◽  
Degradation Indicator

High precision and multi information prediction results of bearing remaining useful life (RUL) can effectively describe the uncertainty of bearing health state and operation state. Aiming at the problem of feature efficient extraction and RUL prediction during rolling bearings operation degradation process, through data reduction and key features mining analysis, a new feature vector based on time-frequency domain joint feature is found to describe the bearings degradation process more comprehensively. In order to keep the effective information without increasing the scale of neural network, a joint feature compression calculation method based on redefined degradation indicator (DI) was proposed to determine the input data set. By combining the temporal convolution network with the quantile regression (TCNQR) algorithm, the probability density forecasting at any time is achieved based on kernel density estimation (KDE) for the conditional distribution of predicted values. The experimental results show that the proposed method can obtain the point prediction results with smaller errors. Compared with the existing quantile regression of long short-term memory network(LSTMQR), the proposed method can construct more accurate prediction interval and probability density curve, which can effectively quantify the uncertainty of bearing running state.

Sign in / Sign up

Export Citation Format

Share Document