Fault detection in industrial processes using canonical variate analysis and dynamic principal component analysis

2000 ◽  
Vol 51 (1) ◽  
pp. 81-93 ◽  
Author(s):  
Evan L. Russell ◽  
Leo H. Chiang ◽  
Richard D. Braatz
Keyword(s):  
Principal Component Analysis ◽  
Fault Detection ◽  
Principal Component ◽  
Component Analysis ◽  
Canonical Variate Analysis ◽  
Industrial Processes ◽  
Canonical Variate ◽  
Variate Analysis ◽  
Dynamic Principal Component Analysis

scholarly journals Morphological Discrimination of Greek Honey Bee Populations Based on Geometric Morphometrics Analysis of Wing Shape

2014 ◽  
Vol 58 (1) ◽  
pp. 75-84 ◽  
Author(s):  
Leonidas Charistos ◽  
Fani Hatjina ◽  
Maria Bouga ◽  
Mica Mladenovic ◽  
Anastasios D. Maistros
Keyword(s):  
Principal Component Analysis ◽  
Geometric Morphometrics ◽  
Honey Bee ◽  
Principal Component ◽  
Component Analysis ◽  
Canonical Variate Analysis ◽  
Procrustes Analysis ◽  
Canonical Variate ◽  
Wing Vein ◽  
Variate Analysis

Abstract Honey bees collected from 32 different localities in Greece were studied based on the geometric morphometrics approach using the coordinates of 19 landmarks located at wing vein intersections. Procrustes analysis, principal component analysis, and Canonical variate analysis (CVA) detected population variability among the studied samples. According to the Principal component analysis (PCA ) of pooled data from each locality, the most differentiated populations were the populations from the Aegean island localities Astypalaia, Chios, and Kythira. However, the populations with the most distant according to the canonical variate analysis performed on all measurements were the populations from Heraklion and Chania (both from Crete island). These results can be used as a starting point for the use of geometric morphometrics in the discrimination of honey bee populations in Greece and the establishment of conservation areas for local honey bee populations.

Hybrid Dynamic Principal Conponent Analysis Approach for Fault Detection in Steel Rolling Process

2011 ◽  
Vol 219-220 ◽  
pp. 1574-1577
Author(s):  
Huai Tao Shi ◽  
Jian Chang Liu ◽  
Long Li ◽  
Yu Zhang
Keyword(s):  
Principal Component Analysis ◽  
Fault Detection ◽  
Computational Efficiency ◽  
Principal Component ◽  
Component Analysis ◽  
Rolling Process ◽  
Divide And Conquer ◽  
Detection Methods ◽  
Steel Rolling ◽  
Dynamic Principal Component Analysis

In traditional dynamic principal component analysis (DPCA) for fault detection, there are some drawbacks such as an excess of the number of principal components (PCs), low computational efficiency, etc. For dealing with the problem, this paper develops a hybrid dynamic principal component analysis (HDPCA) technique, this method can remove spacial and serial correlation by divide-and-conquer algorithm instead of parallel processing strategy, which can detect individual fault accurately and efficiently. The strip breaking fault in steel rolling process is used to demonstrate the improved performance of developed technique in comparison with traditional DPCA fault detection methods. It can be perceived that HDPCA algorithm has the better performance of fault detection and computational efficiency.

Biosystematics of the Cotesia flavipes species complex (Hymenoptera: Braconidae): morphometrics of selected allopatric populations

1997 ◽  
Vol 87 (1) ◽  
pp. 61-66 ◽  
Author(s):  
Susan W. Kimani-Njogu ◽  
William A. Overholt ◽  
James Woolley ◽  
Annette Walker
Keyword(s):  
Principal Component Analysis ◽  
Species Complex ◽  
Principal Component ◽  
Cotesia Flavipes ◽  
Canonical Variate Analysis ◽  
Canonical Variate ◽  
Allopatric Populations ◽  
Putative Species ◽  
Variate Analysis ◽  
China And Japan

AbstractMorphometric studies of allopatric populations of the Cotesia flavipes species complex representing three putative species; C. flavipes Cameron, C. sesamiae (Cameron) and C. chilonis (Matsumura), were conducted. Sixteen characters were measured. Principal component analysis separated the complex into three somewhat overlapping groups that corresponded well with previous concepts of the species. Canonical variate analysis separated the complex into three distinct clusters with populations from Africa together, populations from Asia and the Neotropics forming a second cluster, and material from China and Japan forming a third cluster. The Mahalanobis squared distances between the three clusters were nearly equal. Results support recognition of three species in the C. flavipes complex.

Sign in / Sign up

Export Citation Format

Share Document