Isomap and Deep Belief Network-Based Machine Health Combined Assessment Model

  title={Isomap and Deep Belief Network-Based Machine Health Combined Assessment Model},
  author={Aijun Yin and Juncheng Lu and Zongxian Dai and Jiang Li and Qi Ouyang},
  journal={Strojniski Vestnik-journal of Mechanical Engineering},
This paper presents a novel combined assessment model (CAM) for machine health assessment, in which 38 original features of the vibration signal were extracted from time domain analysis, frequency domain analysis, and wavelet packet transform (WPT), following which the nonlinear global algorithm Isomap was adopted for dimensionality reduction and extraction of the more representative features. Next, the acquired low-dimensional features array is input into the well trained deep belief… 

Figures from this paper

Integrating Laplacian Eigenmaps Feature Space Conversion into Deep Neural Network for Equipment Condition Assessment
A novel method by integrating Laplacian Eigenmaps (LE) that transforms data features from original high-dimensional space to projected low- dimensional space to extract the more representative features into deep neural network (DNN) for equipment health assessment, in which the bearing run-to-failure data were investigated for validation studies.
A Transient Feature Learning-Based Intelligent Fault Diagnosis Method for Planetary Gearboxes
A novel feature extraction method is put forward, which takes advantage of the transient dynamics and the reconstructed high-dimensional data from the original vibration signal to increase the sensitivity of extracted features fed in ELM.
Machine Learning: Data Pre-processing
In prognostics and health management (PHM), data pre‐processing generally involves the following tasks: data cleansing, normalization, feature discovery, and imbalanced data management. Data
An Improved Fault Diagnosis Using 1D-Convolutional Neural Network Model
The 1D-CNN network architecture is proposed in order to effectively improve the accuracy of the diagnosis of rolling bearing, and the number of convolution kernels decreases with the reduction of the convolution kernel size.
An Improved of Fault Diagnosis Using 1D-Convolutional Neural Network Model
By introducing the dropout operation, the 1D-CNN method obtains high accuracy and improves the generalizing ability and the experimental results show 99.52% of the average accuracy under a single load and 98.26% under different loads.
An Improved Bearing Fault Diagnosis Method using One-Dimensional CNN and LSTM
A large number of academic achievements emerged in the field of bearing fault diagnosis using deep learning algorithms, and an unsupervised fault diagnosis based on a deep belief network (DBN) has been proven.
A comprehensive review of artificial intelligence-based approaches for rolling element bearing PHM: shallow and deep learning
This paper first reviews the fundamentals of prognostics and health management techniques for REBs, and provides overviews of contemporary REB PHM techniques with a specific focus on modern artificial intelligence (AI) techniques (i.e., shallow learning algorithms).
A semi-supervised transferable LSTM with feature evaluation for fault diagnosis of rotating machinery
Compared with the well-known methods, the proposed EGR-STLSTM network significantly improves the diagnostic performance by reusing the pre-trained model and reduces the demand for massive labeled samples under variable working conditions.
Rolling-Element Bearing Fault Diagnosis Using Improved LeNet-5 Network
An end-to-end rolling-element bearing fault diagnosis method based on the improved 1D LeNet-5 network is proposed, which can directly perform 1D convolution and pooling operations on raw vibration signals without any preprocessing.
Multiple regression analysis based approach for condition monitoring of industrial rotating machinery using multi-sensors
This paper proposes a novel approach used for the detection of changes in machine condition using multiple regression, where a novel data description and prediction strategy is developed to make full use of the correlation between different sensors.


Health Condition Monitoring of Machines Based on Hidden Markov Model and Contribution Analysis
  • Jianbo Yu
  • Engineering
    IEEE Transactions on Instrumentation and Measurement
  • 2012
A hidden Markov model (HMM) and contribution-analysis-based method to assess the machine health degradation and a novel machine health assessment indication, HMM-based Mahalanobis distance is proposed to provide a comprehensible indication for quantifying machine health states.
Adaptive Empirical Mode Decomposition for Bearing Fault Detection
Many techniques for bearing fault detection have been proposed. Two of the most effective approaches are using envelope analysis and the empirical mode decomposition method (EMD), also known as
Crack Fault Detection for a Gearbox Using Discrete Wavelet Transform and an Adaptive Resonance Theory Neural Network
In this paper, a new approach using discrete wavelet transform and an adaptive resonance theory neural network for crack fault detection of a gearbox is proposed. With the use of a multi-resolution