Incipient fault detection for nonlinear processes based on dynamic multi-block probability related kernel principal component analysis

Cai, Peipei; Deng, Xiaogang

doi:10.1016/j.isatra.2020.05.029

Cited by 54 publications

(23 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Jiang and Yan 10 proposed a parallel PCA-KPCA monitoring method for a process with linearly correlated and nonlinearly related variables. In order to detect the incipient faults of nonlinear industrial processes effectively, an enhanced KPCA method is proposed in ref ( 11 ). Mansouri et al ( 12 ) introduced a generalized likelihood ratio test into the KPCA method for fault detection of nonlinear processes.…”

Section: Introductionmentioning

confidence: 99%

Fault Detection and Isolation of Non-Gaussian and Nonlinear Processes Based on Statistics Pattern Analysis and the k-Nearest Neighbor Method

et al. 2022

View full text Add to dashboard Cite

Only low-order information of process data (i.e., mean, variance, and covariance) was considered in the principal component analysis (PCA)-based process monitoring method. Consequently, it cannot deal with continuous processes with strong dynamics, nonlinearity, and non-Gaussianity. To this aim, the statistics pattern analysis (SPA)-based process monitoring method achieves better monitoring results by extracting higher-order statistics (HOS) of the process variables. However, the extracted statistics do not strictly follow a Gaussian distribution, making the estimated control limits in Hotelling- T 2 and squared prediction error (SPE) charts inaccurate, resulting in unsatisfactory monitoring performance. In order to solve this problem, this paper presents a novel process monitoring method using SPA and the k -nearest neighbor algorithm. In the proposed method, first, the statistics of process variables are calculated through SPA. Then, the k -nearest neighbor (kNN) method is used to monitor the extracted statistics. The kNN method only uses the paired distance of samples to perform fault detection. It has no strict requirements for data distribution. Hence, the proposed method can overcome the problems caused by the non-Gaussianity and nonlinearity of statistics. In addition, the potential of the proposed method in early fault detection or safety alarm and fault isolation is explored. The proposed method can isolate which variable or its statistic is faulty. Finally, the numerical examples and Tennessee Eastman benchmark process illustrate the effectiveness of the proposed method.

show abstract

Section: Introductionmentioning

confidence: 99%

Fault Detection and Isolation of Non-Gaussian and Nonlinear Processes Based on Statistics Pattern Analysis and the k-Nearest Neighbor Method

et al. 2022

View full text Add to dashboard Cite

show abstract

“…There are three stages in this kind of method: the first is extracting discriminant features, and the second is training machine learning models by using historical feature datasets, the final is applying the trained model on real-time data to predict tool state and RUL. Therefore, the premise of data-driven methods is extracting discriminant tool features from input signals to train machine learning models [5][6][7][8]. Zhang et al [5] utilized wavelet packet to extract the features of nonstationary vibrations in three vertical directions.…”

Section: Introductionmentioning

confidence: 99%

“…Zhu et al [6] discovered that the Holder exponents extracted from wavelet transform modulus maxima can reflect tool wear state. After extracting features from original signals, in order to enhance feature space and reduce input dimension, principal component analysis [7] and factor analysis [8] were adopted. Finally, the extracted features are utilized as input to different models, such as auto-regression [9], manifold learning [10], hidden Markov [11], sparse decomposition [12], and deep learning [13].…”

Section: Introductionmentioning

confidence: 99%

Multi-Step-Ahead Tool State Monitoring Using Clustering Feature-Based Recurrent Fuzzy Neural Networks

Yao

Zhang

2021

IEEE Access

View full text Add to dashboard Cite

Reliable and precise multi-step-ahead tool wear state prediction is significant to modern industries for maintaining part quality and reducing cost. This study proposes a Clustering Feature-based Recurrent Fuzzy Neural Network (CFRFNN) for tool wear state monitoring and remaining useful life (RUL) prediction based on K-means Clustering, Recurrent Fuzzy Neural Network (RFNN) and Genetic Algorithm (GA). K-means Clustering method is utilized to realize tool wear state definition and input signal division, which reduces the dependence on the prior knowledge of tool wear degree and improves the prediction accuracy. Then, an enhanced RFNN model is designed and applied on the clustered features to predict tool wear state. The optimized GA technique is helpful for adaptive optimization of model parameters, which significantly improves convergence rate and prediction accuracy. The experiments on tool state prediction are performed to validate superiority of CFRFNN, and the results demonstrate that the proposed network could reasonably configure the complex non-stationary tool wear process and have high prediction accuracy of tool wear state.INDEX TERMS Multi-step-ahead tool state prediction, remaining useful life (RUL), recurrent fuzzy neural network (RFNN), tool wear monitoring.

show abstract

“…In a physicochemical system under investigation that contains multiple degrees of freedom, e.g., several concentrations range and absorbed doses for a given sample, it is possible to apply the Kernel Principal Component Regression (KPCR) method to determine the possible linearity behavior of the system. KPCR is used to handle the multicollinearity effect among the independent variables from the regression data; this method has been used successfully for this objective [41][42][43][44].…”

Section: Introductionmentioning

confidence: 99%

A new natural detector for irradiations with blue LED light source in photodynamic therapy measurements via UV–Vis spectroscopy

Oliveira

Nascimento

Caldas

2021

Photochem Photobiol Sci

View full text Add to dashboard Cite

Photodynamic therapy has been recently studied, bringing innovations regarding the reduction of exposure time to light by the patient. This work aimed to investigate the feasibility of using Coutarea hexandra (Jacq.) K. Schum (CHS) as a detector in photodynamic therapy measurements. For this, an irradiator containing a blue LED bulb lamp was utilized. The CHS samples were irradiated with ten doses from 0.60 up to 6.0 kJ/cm 2 , and six concentrations were prepared (1, 2, 3, 4, 5, and 6 mg/ml) for the CHS detector samples. After irradiation, the detector samples were evaluated using UV-Vis spectrophotometry. The results showed the behavior of the CHS detector with doses and concentrations, its sensitivity, and its linearity was also evaluated both by Wavelength Method (WM) and the Kernel Principal Component Regression (KPCR) Statistical Method. The values obtained indicate that this method can be applied to the CHS sample detector. In conclusion, the CHS is a promising detector in the field of photodynamic therapy.

show abstract

Incipient fault detection for nonlinear processes based on dynamic multi-block probability related kernel principal component analysis

Cited by 54 publications

References 43 publications

Fault Detection and Isolation of Non-Gaussian and Nonlinear Processes Based on Statistics Pattern Analysis and the k-Nearest Neighbor Method

Fault Detection and Isolation of Non-Gaussian and Nonlinear Processes Based on Statistics Pattern Analysis and the k-Nearest Neighbor Method

Multi-Step-Ahead Tool State Monitoring Using Clustering Feature-Based Recurrent Fuzzy Neural Networks

A new natural detector for irradiations with blue LED light source in photodynamic therapy measurements via UV–Vis spectroscopy

Contact Info

Product

Resources

About