Multi-mode operation of principal component analysis with k-nearest neighbor algorithm to monitor compressors for liquefied natural gas mixed refrigerant processes

Ha, Daegeun; Ahmed, Usama; Pyun, Hahyung; Lee, Chul‐Jin; Baek, Kye Hyun; Han, Chonghun

doi:10.1016/j.compchemeng.2017.05.029

Cited by 48 publications

(16 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In Ref. [5], k-NN has been proved to be a simple and effective method for clustering. ANNs can not only solve complex nonlinear mapping relationships, but also improve the accuracy of classifications, which have shown good results in the qualitative and quantitative identification of harmful gases [6].…”

Section: Introductionmentioning

confidence: 99%

Development of a LeNet-5 Gas Identification CNN Structure for Electronic Noses

Wei

Zhao

et al. 2019

Sensors

130

View full text Add to dashboard Cite

A new LeNet-5 gas identification convolutional neural network structure for electronic noses is proposed and developed in this paper. Inspired by the tremendous achievements made by convolutional neural networks in the field of computer vision, the LeNet-5 was adopted and improved for a 12-sensor array based electronic nose system. Response data of the electronic nose to different concentrations of CO, CH4 and their mixtures were acquired by an automated gas distribution and test system. By adjusting the parameters of the CNN structure, the gas LeNet-5 was improved to recognize the three categories of CO, CH4 and their mixtures omitting the concentration influences. The final gas identification accuracy rate reached 98.67% with the unused data as test set by the improved gas LeNet-5. Comparison with results of Multiple Layer Perceptron neural networks and Probabilistic Neural Network verifies the improvement of recognition rate while with the same level of time cost, which proved the effectiveness of the proposed approach.

show abstract

Section: Introductionmentioning

confidence: 99%

Development of a LeNet-5 Gas Identification CNN Structure for Electronic Noses

Wei

Zhao

et al. 2019

Sensors

130

View full text Add to dashboard Cite

show abstract

“…Its performance was demonstrated in the context of two industrial installations (an electric arc furnace and a wind tunnel) and several public datasets. In the direction of multivariate statistical analysis we find the work of Ha et al [21]. Here, the multi-mode principal component analysis (PCA) was used together with the K-nearest neighbor algorithm for process monitoring and data classification.…”

Section: Related Workmentioning

confidence: 99%

“…A considerable amount of research has focused on the development of intrusion/anomaly detection systems. Different strategies have been suggested by embracing diverse techniques such as classification [19], [20], multivariate statistical analysis, principal component analysis [21], [22], and data fusion [23]- [25].…”

Section: Introductionmentioning

confidence: 99%

Anomaly Detection in Aging Industrial Internet of Things

2019

View full text Add to dashboard Cite

The Industrial Internet of Things (IIoT) have been designed to perform a more agile and efficient automation, control, and orchestration of future industrial systems while improving the energy efficiency in smart factories. Unfortunately, while the benefits of the IIoT are undeniable, their pervasive adoption as key enablers for future industries also paved the way for new security risks. In fact, the damaging effects of exploiting vulnerable IIoT have been repeatedly demonstrated and publicly reported. The Mirai botnet, various reports on hackable and invasive devices, alongside the infamous Stuxnet malware, constitute significant proof on the undisputed and disruptive effect of the malware-targeting IIoT systems. As a response, a plethora of solutions has been developed to address the issue of securing IIoT systems in specific sectors. Nevertheless, we believe that the gradual decay of the IIoT's physical dimension (e.g., the physical process), also called aging, is a natural component of the IIoT's life cycle, which has not received sufficient attention from the scientific community. This paper develops a methodology for detecting abnormal behavior in the context of aging IIoT. The approach leverages multivariate statistical analysis [e.g., principle component analysis (PCA)], alongside the Hotelling's T 2 statistics, and the univariate cumulative sum in order to detect abnormal process events. An innovative feature of the developed approach is the detection of stealth attacks attempting to influence the dataset in each age. The extensive experimental results on a continuous stirred-tank reactor (CSTR) model demonstrate its applicability to the IIoT and its superior performance to the recently reported techniques. INDEX TERMS Anomaly detection systems, Industrial Internet of Things, aging processes, multivariate statistical analysis.

show abstract

“…After mode identification, each mode is monitored by such traditional methods as principal component analysis/independent component analysis (PCA/ ICA) and their extensions and so forth. [8][9][10]. Recently, several efforts have been made on applying the multiple models with Bayesian fusion to multimode monitoring.…”

Section: Introductionmentioning

confidence: 99%

SPA‐Based Modified Local Reachability Density Ratio wSVDD for Nonlinear Multimode Process Monitoring

et al. 2021

View full text Add to dashboard Cite

Many industrial processes are operated in multiple modes due to different manufacturing strategies. Multimodality of process data is often accompanied with nonlinear and non-Gaussian characteristics, which makes data-driven monitoring more complicated. In this paper, statistics pattern analysis (SPA) is introduced to extract low- and high-order statistics from raw process data. Support vector data description (SVDD), which can deal with nonlinear and non-Gaussian problems, is applied to monitor multimode process in this paper. To improve detection performance of SVDD for training multimode data with outliers, modified local reachability density ratio (mLRDR) is proposed as a weight factor to be embedded in the weighted-SVDD (wSVDD) model, in which the local neighbors in terms of both space and time are considered. Finally, the effectiveness and superiority of our proposed method are demonstrated by the Tennessee-Eastman (TE) process and wastewater treatment process (WWTP).

show abstract

Multi-mode operation of principal component analysis with k-nearest neighbor algorithm to monitor compressors for liquefied natural gas mixed refrigerant processes

Cited by 48 publications

References 19 publications

Development of a LeNet-5 Gas Identification CNN Structure for Electronic Noses

Development of a LeNet-5 Gas Identification CNN Structure for Electronic Noses

Anomaly Detection in Aging Industrial Internet of Things

SPA‐Based Modified Local Reachability Density Ratio wSVDD for Nonlinear Multimode Process Monitoring

Contact Info

Product

Resources

About