A semi-supervised Laplacian extreme learning machine and feature fusion with CNN for industrial superheat identification

Lei, Yongxiang; Chen, Xiaofang; Min, Mengcan; Xie, Yongfang

doi:10.1016/j.neucom.2019.11.012

Cited by 51 publications

(27 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As listed in Table 6 and Table 7, the process monitoring performance of the EUELM, UELM and KPCA based approches for all the seventeen fault patterns are investigated. According to the Table 6, we find the UELM owns much earlier fault detection times for the faults IDV(1), IDV(2), IDV(5), IDV(6), IDV(8), IDV(10), IDV (13) and IDV(16) ~ IDV (21) than that of the KPCA. For the rest of the faults apart from the faults IDV (7) and IDV (14), both the UELM and KPCA based methods have similar fault detection times.…”

Section: (3) Fault Detection Effect Comparisonmentioning

confidence: 91%

“…To improve the effectiveness in disposing non-Gaussian noises, Yang et al [20] proposed a novel SELM method based on robust regularized correntropy criterion. By means of integrating Laplacian regularization to learn the manifold structure of hole image samples, Lei et al [21] further discussed a modified version of the SELM to classify the superheat degree.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

An Enhanced Unsupervised Extreme Learning Machine Based Method for the Nonlinear Fault Detection

Shao

Kang

et al. 2021

IEEE Access

View full text Add to dashboard Cite

Although the unsupervised extreme learning machine (UELM) based methods have been widely used to diagnosis the nonlinear process faults recently, the UELM algorithm is only designed to preserve the local adjacency similarity of the input dataset instead of mining the intra-class variations. Besides, the determination of the optimal UELM hidden nodes number is a tough issue. In order to deal with these two problems, a novel enhanced UELM (EUELM) based scheme is developed to effectively detect the nonlinear process faults in our work. In the proposed EUELM approach, the UELM algorithm is first improved by naturally incorporating the diversity analysis technique into the original UELM objective function to preserve both the intra-class variation and the local adjacency similarity of the input dataset. Then, to settle the difficult issue of selecting the optimal number of hidden nodes, kernel trick is further employed in the EUELM approach to mine the data strong nonlinearity. Based on the extracted diversity and local similarity low dimensional feature information, the k-nearest neighbor (KNN) principle is applied to derive a monitoring statistic for fault detection. At last, the experiments and comparisons on the monitoring effectiveness of the suggested EUELM based approach are made on a numerical nonlinear system and the benchmark Tennessee Eastman (TE) process. The obtained monitoring results illustrate that the significant improvements can be achieved by the proposed EUELM based fault detection approach compared with other popular and related approaches.

show abstract

Section: (3) Fault Detection Effect Comparisonmentioning

confidence: 91%

Section: Introductionmentioning

confidence: 99%

An Enhanced Unsupervised Extreme Learning Machine Based Method for the Nonlinear Fault Detection

Shao

Kang

et al. 2021

IEEE Access

View full text Add to dashboard Cite

show abstract

“…For feature level fusion, the hand-crafted features and CNN features are directly combined by concatenation, then reduced and utilized to train the classifier [32][33][34]. In [38], for lung nodule classification, a set of hand-crafted features and the CNN feature are fused by the cascade method.…”

Section: Fusion Of Hand-crafted Features and Cnn Featuresmentioning

confidence: 99%

“…The idea of fusing CNN features and hand-crafted features to improve model performance is widely used in the field of existing computer vision, such as breast cancer detection [32], industrial superheat identification [33], ship classification [34], vehicle detection [35], person reidentification [36], osteoporosis diagnoses [37], lung nodule classification [38]. In the textile industry, fabric defect detection is the only task for which researchers have used the feature fusion idea [39].…”

Section: Introductionmentioning

confidence: 99%

“…According to their implementation, the feature fusion frameworks can be categorized into three types, i.e. feature level fusion [32][33][34] with simple concatenation of features; fusion by ensemble [35] with combining multiple classifiers trained by hand-crafted and CNN features in respective; joint level fusion [39] with combining different features in the CNN framework. In general, the feature level fusion fails to consider the difference of the dimension and range between the features, which may result in an unbalanced combination.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Fusing Convolutional Neural Network Features With Hand-Crafted Features for Objective Fabric Smoothness Appearance Assessment

et al. 2020

View full text Add to dashboard Cite

In the textile and apparel industry, it remains a challenging task to evaluate the fabric smoothness appearance objectively. In existing studies, with computer vision technology, researchers use the hand-crafted image features and deep convolutional neural network (CNN) based image features to describe the fabric smoothness appearance. This paper presents an image classification framework to evaluate the fabric smoothness appearance degree. The framework contains a feature fusion module to fuse the handcrafted and CNN features to take both advantages of them. The framework uses the multi-scale spatial masking model and a pre-trained CNN to extract hand-crafted and CNN features of fabric images respectively. In addition, a mislabeled sample filtering module is set in the framework, which helps to avoid the negative impact of mislabeled samples in training. In the experiments, the proposed framework achieves 85.2%, 96.1%, and 100% average evaluation accuracies under errors of 0 degree, 0.5 degree, and 1 degree respectively. The experiments on the feature fusion and mislabeled sample filtering verified their effectiveness in improving the evaluation accuracies and the label noise robustness. The proposed method outperforms the state-of-theart methods for fabric smoothness assessment. Promisingly, this paper can provide novel research ideas for image-based fabric smoothness assessment and other similar tasks.

show abstract

A Robust ELM Algorithm for Compensating the Effect of Node Fault and Weight Noise

Adegoke

Xiao

Leung

et al. 2022

Recent Advances in Soft Computing and Data Mining

View full text Add to dashboard Cite

A semi-supervised Laplacian extreme learning machine and feature fusion with CNN for industrial superheat identification

Cited by 51 publications

References 20 publications

An Enhanced Unsupervised Extreme Learning Machine Based Method for the Nonlinear Fault Detection

An Enhanced Unsupervised Extreme Learning Machine Based Method for the Nonlinear Fault Detection

Fusing Convolutional Neural Network Features With Hand-Crafted Features for Objective Fabric Smoothness Appearance Assessment

A Robust ELM Algorithm for Compensating the Effect of Node Fault and Weight Noise

Contact Info

Product

Resources

About