Fault Detection for Batch Processes Based on Segmentation MPCA

2014 ◽  
Vol 1030-1032 ◽  
pp. 1701-1708
Author(s):  
Ning Lv ◽  
Guang Yuan Bai ◽  
Lei Lei Jiang
Keyword(s):  
Fault Detection ◽  
Fermentation Process ◽  
Detection Method ◽  
Piecewise Linear ◽  
Nonlinear Modeling ◽  
Principal Component ◽  
Batch Processes ◽  
Piecewise Linear Approximation ◽  
Process Data ◽  
Speed And Accuracy

Aiming at the dynamic characteristic of batch production process changes fast and the accurate modeling of it is difficult, so this paper proposes an intermittent fault detection method of the principal component analysis based on process segment. According to the different dynamic characteristics of process data, the process is divided into multiple stages, with the method of piecewise linear approximation of nonlinear modeling to model different stages of the process, in order to make up the deficiency of traditional MPCA fault diagnosis methods. Through the fault detection of the beer fermentation process experiments to verify that the method can detect process faults promptly and improve the speed and accuracy of process monitoring.

Application of Phase Division Based on Dissimilarity Index in Batch Process Monitoring

2012 ◽  
Vol 566 ◽  
pp. 134-139 ◽  
Author(s):  
Li Ying Jiang ◽  
Bao Jian Xu ◽  
Jian Hui Xi ◽  
Guo Xiu Fu
Keyword(s):  
Fermentation Process ◽  
Principal Component ◽  
Batch Process ◽  
Batch Processes ◽  
Time Slice ◽  
Monitoring Methods ◽  
Dissimilarity Index ◽  
Process Data ◽  
Batch Process Monitoring ◽  
Simulation Results

An important feature of batch process data is that many batch processes have multiple phases. Many different phased-based monitoring methods had been proposed. The key question of those methods is how to divide the phases of batch process. However, PCA-based methods of phase division that identify phases by extracting the first principal component of each time slice lead easily to high misclassification. In order to overcome the shortcoming of PCA-based methods, a novel phase-division method based on dissimilarity index is proposed. In proposed division method, integral information of each time slice is used to divide phases. The phase-based PCA is built in each phase to monitoring Penicillin fermentation process in order to verify performance of proposed method. The simulation results show that the proposed method is able to detect process faults more prompt and accurate than single MPCA model.

A Principal Component Analysis Based Fault Detection Method in Etch Process of Semiconductor Manufacturing

2012 ◽  
Vol 522 ◽  
pp. 793-798 ◽  
Author(s):  
Jun Gang Yang ◽  
Jie Zhang ◽  
Jian Xiong Yang ◽  
Ying Huang
Keyword(s):  
Principal Component Analysis ◽  
Fault Detection ◽  
Semiconductor Manufacturing ◽  
Detection Method ◽  
Main Idea ◽  
Simple Algorithm ◽  
Principal Component ◽  
Component Analysis ◽  
Training Data ◽  
Detection Model

A Principal Component Analysis based Fault Detection method is proposed here to detect faults in etch process of semiconductor manufacturing. The main idea of this method is to calculate the loading vector and build the fault detection model according to training data. Using this model, the main information of fault data can be obtained immediately and easily. Also the principal component subspace and residual subspace can be constructed. Then, faults are detected by calculating Squared Prediction Error. Finally, an industrial example of Lam 9600 TCP metal etcher at Texas Instruments is used to demonstrate the performance of the proposed PCA-based method in fault detection, and the results show that it has such advantages as simple algorithm and low time cost, thus especially adapts to the real time fault detection of semiconductor manufacturing.

The Fault Diagnosis Model of Beer Fermentation Process Based on Kernel Principal Component Analysis for Constant Value Detection

2014 ◽  
Vol 1030-1032 ◽  
pp. 1822-1827
Author(s):  
Ning Lv ◽  
Guang Yuan Bai ◽  
Lu Qi Yan ◽  
Yuan Jian Fu
Keyword(s):  
Principal Component Analysis ◽  
Computational Complexity ◽  
Fault Detection ◽  
Fermentation Process ◽  
Principal Component ◽  
Component Analysis ◽  
Kernel Principal Component Analysis ◽  
Time Varying ◽  
Beer Fermentation ◽  
Non Linear

In order to overcome the application limitations of principal component analysis fault diagnose model in non-linear time-varying and reduce computational complexity for process monitoring based on non-linear principal component, we introduced kernel transformation theory of nonlinear space to extract data feature extraction and a fault monitoring model based on kernel principal component analysis (KPCA) for constant value detection was proposed. Through the proper selection of kernel function parameter values, the KPCA model can achieve constant value of process fault detection and has lower computational complexity than other non-linear algorithms. The fault detection experiment for beer fermentation process shows that this method is able to detect process faults in a timely manner and has good real-time performance and accuracy in the batch process of slowly time-varying.

Sign in / Sign up

Export Citation Format

Share Document