scholarly journals Sparse Optimistic Based on Lasso-LSQR and Minimum Entropy De-Convolution with FARIMA for the Remaining Useful Life Prediction of Machinery

Entropy ◽  
2018 ◽  
Vol 20 (10) ◽  
pp. 747 ◽  
Author(s):  
Bo Wu ◽  
Yangde Gao ◽  
Songlin Feng ◽  
Theerasak Chanwimalueang
Keyword(s):  
Health Monitoring ◽  
Hybrid Method ◽  
Health Indicators ◽  
Remaining Useful Life ◽  
Least Square ◽  
Qr Factorization ◽  
Maintenance Cost ◽  
Minimum Entropy ◽  
Useful Life

To reduce the maintenance cost and safeguard machinery operation, remaining useful life (RUL) prediction is very important for long term health monitoring. In this paper, we introduce a novel hybrid method to deal with the RUL prediction for health management. Firstly, the sparse reconstruction algorithm of the optimized Lasso and the Least Square QR-factorization (Lasso-LSQR) is applied to compressed sensing (CS), which can realize the sparse optimization for long term health monitoring data. After the sparse signal is reconstructed, the minimum entropy de-convolution (MED) is used to identify the fault characteristics and to obtain significant fault information from the machinery operation. Health indicators with Skip-over, sample entropy and approximate entropy are then performed to track the degradation of the machinery process. The performance analysis of the Skip-over is superior to other indicators. Finally, Fractal Autoregressive Integrated Moving Average model (FARIMA) is employed to predict the Skip-over using the R/S method. The analysis results evidence that the novel hybrid method yields a good performance, and such method can achieve highly accurate RUL prediction and safeguard machinery operation for long term monitoring.

scholarly journals Adapting Approximate Entropy as a Health Indicator of Rotating Machinery for Estimation of Remaining Useful Life

2019 ◽  
Vol 11 (1) ◽  
Author(s):  
Cody Walker
Keyword(s):  
Health Monitoring ◽  
Health Indicators ◽  
Approximate Entropy ◽  
Remaining Useful Life ◽  
Frequency Content ◽  
Health Indicator ◽  
Vibration Signals ◽  
Early Results ◽  
Useful Life ◽  
Significant Attention

Despite significant attention to online health monitoring and prognostics of bearings, many common health indicators are not sensitive to early stages of degradation. This research investigates the use of approximate entropy (ApEn), previously developed for fault diagnostics, as a health indicator for prognostics. ApEn quantifies the regularity of a signal; as bearings degrade, the frequency content of vibration signals changes and affects the ApEn as the vibration becomes more chaotic. Early results suggest ApEn supports earlier degradation detection and more predictable progression from fault to failure. This research focuses on optimizing parameters of the ApEn calculation to provide guidance across a variety of bearing types, sizes, and geometries in both steady-state and transient operation.

scholarly journals A Novel Deep Learning Approach for Machinery Prognostics Based on Time Windows

2019 ◽  
Vol 9 (22) ◽  
pp. 4813 ◽  
Author(s):  
Hanbo Yang ◽  
Fei Zhao ◽  
Gedong Jiang ◽  
Zheng Sun ◽  
Xuesong Mei
Keyword(s):  
Time Windows ◽  
Remaining Useful Life ◽  
Sensor Data ◽  
Time Interval ◽  
Kernel Module ◽  
Initial Stage ◽  
Research Task ◽  
Consecutive Time ◽  
Useful Life

Remaining useful life (RUL) prediction is a challenging research task in prognostics and receives extensive attention from academia to industry. This paper proposes a novel deep convolutional neural network (CNN) for RUL prediction. Unlike health indicator-based methods which require the long-term tracking of sensor data from the initial stage, the proposed network aims to utilize data from consecutive time samples at any time interval for RUL prediction. Additionally, a new kernel module for prognostics is designed where the kernels are selected automatically, which can further enhance the feature extraction ability of the network. The effectiveness of the proposed network is validated using the C-MAPSS dataset for aircraft engines provided by NASA. Compared with the state-of-the-art results on the same dataset, the prediction results demonstrate the superiority of the proposed network.

Data Augmentation for Roller Bearing Health Indicator Estimation Using Multi-Channel Frequency Data Representations

2021 ◽  
Author(s):  
Jacob Hendriks ◽  
Patrick Dumond
Keyword(s):  
Data Augmentation ◽  
Roller Bearing ◽  
Health Indicators ◽  
Remaining Useful Life ◽  
Failure Data ◽  
Data Representations ◽  
Masking Noise ◽  
Augmentation Techniques ◽  
Added Benefit ◽  
Useful Life

Abstract This paper demonstrates various data augmentation techniques that can be used when working with limited run-to-failure data to estimate health indicators related to the remaining useful life of roller bearings. The PRONOSTIA bearing prognosis dataset is used for benchmarking data augmentation techniques. The input to the networks are multi-dimensional frequency representations obtained by combining the spectra taken from two accelerometers. Data augmentation techniques are adapted from other machine learning fields and include adding Gaussian noise, region masking, masking noise, and pitch shifting. Augmented datasets are used in training a conventional CNN architecture comprising two convolutional and pooling layer sequences with batch normalization. Results from individually separating each bearing’s data for the purpose of validation shows that all methods, except pitch shifting, give improved validation accuracy on average. Masking noise and region masking both show the added benefit of dataset regularization by giving results that are more consistent after repeatedly training each configuration with new randomly generated augmented datasets. It is shown that gradually deteriorating bearings and bearings with abrupt failure are not treated significantly differently by the augmentation techniques.

Hybrid method for remaining useful life prediction in wind turbine systems

2018 ◽  
Vol 116 ◽  
pp. 173-187 ◽  
Author(s):  
M.A. Djeziri ◽  
S. Benmoussa ◽  
R. Sanchez
Keyword(s):  
Wind Turbine ◽  
Hybrid Method ◽  
Life Prediction ◽  
Remaining Useful Life ◽  
Useful Life

Durability of Embedded PZTs in Structural Health Monitoring Systems under Cyclic Loading

2014 ◽  
Vol 891-892 ◽  
pp. 1255-1260 ◽  
Author(s):  
Sanghyun Yoo ◽  
Akbar Afaghi Khatibi ◽  
Everson Kandare
Keyword(s):  
Cyclic Loading ◽  
Structural Health Monitoring ◽  
Lead Zirconate Titanate ◽  
Health Monitoring ◽  
Low Cycle Fatigue ◽  
Maintenance Cost ◽  
Clear Understanding ◽  
Structural Health ◽  
Service Conditions

Structural Health Monitoring (SHM) systems are developed to decrease the maintenance cost and increase the life of engineering structures by fundamentally changing the way structural inspections are performed. However, this important objective can only be achieved through the consistent and predictable performance of a SHM system under different service conditions. The capability of a Piezoelectric lead Zirconate Titanate (PZT)-based SHM system in detecting structural flaws strongly depends on the sensor signals as well as actuator performance. But service conditions can change the behaviour of transducers, raising questions about long term SHM system capability. Although having a clear understanding of the reliable sensor life is important for surface mounted systems, however, this is particularly critical for embedded sensors. This is due to the fact that opportunity for replacement of sensors exists for surface bonded transducers while for the embedded systems, sensor replacement is not straightforward. Therefore, knowledge of the long term behaviour of embedded-SHM systems is critical for their implementation. This paper reports a study on the degradation of embedded PZT transducers under cyclic loadings. Carbon/epoxy laminates with an embedded PZT were subjected to fatigue loading and their performance was monitored using Scanning Laser Vibrometery (SLV). The functionality of PZT transducers under sensing and actuating modes were studied. High and low cycle fatigue tests were performed to establish strain-voltage relationships which can be used to identify critical cyclic loading parameters (number of cycles and R value) under sensing and actuating modes.

Propagative broad learning for nonparametric modeling of ambient effects on structural health indicators

2020 ◽  
pp. 147592172091692 ◽  
Author(s):  
Sin-Chi Kuok ◽  
Ka-Veng Yuen ◽  
Stephen Roberts ◽  
Mark A Girolami
Keyword(s):  
Structural Health Monitoring ◽  
Health Monitoring ◽  
Health Indicators ◽  
Operating Conditions ◽  
Learning Approach ◽  
Nonparametric Modeling ◽  
Structural Health ◽  
Model Free ◽  
Free Data

In this article, a novel propagative broad learning approach is proposed for nonparametric modeling of the ambient effects on structural health indicators. Structural health indicators interpret the structural health condition of the underlying dynamical system. Long-term structural health monitoring on in-service civil engineering infrastructures has demonstrated that commonly used structural health indicators, such as modal frequencies, depend on the ambient conditions. Therefore, it is crucial to detrend the ambient effects on the structural health indicators for reliable judgment on the variation of structural integrity. However, two major challenging problems are encountered. First, it is not trivial to formulate an appropriate parametric expression for the complicated relationship between the operating conditions and the structural health indicators. Second, since continuous data stream is generated during long-term structural health monitoring, it is required to handle the growing data efficiently. The proposed propagative broad learning provides an effective tool to address these problems. In particular, it is a model-free data-driven machine learning approach for nonparametric modeling of the ambient-influenced structural health indicators. Moreover, the learning network can be updated and reconfigured incrementally to adapt newly available data as well as network architecture modifications. The proposed approach is applied to develop the ambient-influenced structural health indicator model based on the measurements of 3-year full-scale continuous monitoring on a reinforced concrete building.

Sign in / Sign up

Export Citation Format

Share Document