A Handy Tool for Chemometrics

8/12/2019 A Handy Tool for Chemometrics

1/18

The Unscrambler

A Handy Tool for Doing Chemometrics

Prof. Waltraud Kessler

Prof. Dr. Rudolf Kessler

Hochschule Reutlingen, School of Applied Chemistry

Steinbeistransferzentrum Prozesskontrolle und DatenanalyseCamo Process AS


2/18

2

Topics The Unscrambler by Camo

Many possibilities for Analysing Data Examples

NIR-Spectra Fluorescence Exitation Emission Spectra

Life Demonstration 3-way Data Handling


3/18

3

The Unscrambler

Main FeaturesExploratory Analysis

Descriptive statistics

Principal Component Analysis (PCA)Multivariate Regression Analysis

Partial Least Squares regression (PLS)

Principal Component Regression (PCR)

Multiple Linear Regression (MLR)

Prediction

ClassificationSoft Independent Modeling of Class Analogies (SIMCA)

PLS-Discriminant Analysis

Experimental DesignFractional and full factorial designs, Placket-Burmann,Box Behnken, Central Composite, Classical mixturedesigns, D-optimal designs

ANOVA, Response Surface ANOVA, PLS-R


4/18

4

The Unscrambler

Also Features

Raw data checks Data preprocessing

Over 100 pre-defined plots

Automatic outlier detection

Automatic variable selection

and more


5/185

Example: Fiber Board Production

In-situ Measurements of Fibres in Blowpipe

Blowpipe:

~ 180C

~ 5 bar

~ velocity of fibres ~ 20 m/s

NIR FOSS Process

Spectrometer

with fibre bundle and

diffuse reflectance probe

400 - 2200 nm


6/18


7/187

-0.05

0

0.05

-0.10 -0.05 0 0.05 0.10

RESULT8, X-expl: 72%,24%

oRfoRfoRfoRf oRf

oRf

oRf

oRf

oRfoRfoRfoRfoRfoRfoRfoRfoRf

oRfoRf

oRfoRf

oRf

oRf

oRf

oRfoRf oRfoRfoRfoRf

oRg

oRgoRg

oRoRgoRgoRg

oRgoRgoRg

oRg

oRgoRgoRg

oRgoRgoRg

oRgoRg

oRg

oRgoRgoRgoRg

oRg

oRgoRgoRgoRgoRgoRg

oRg

oRg

oRg

oRgoRg

mRf

mRf

mRf

mRf

mmRf

mRf

mRfmRfmRf

mRf

mRf

mRfmRfmRfmRf

mRf

mRfmRfmRf

mRfmRf mRf

mRf

mRf

mRfmRfmRf

mRf

mRf

mRfmRfmRf

mRg

mRg

mRg

mRg

mRg

mRgmRg

mRgmRgmRg

mRgmRg

mRgmRgmRgmRgmRg

mRgmRgmRgmRgmRg

mRg

mRgmRgmRgmRgmRg

mRgmRg

mRgmRg

mRg

mRgmRg

mRgmRg

mRg

PC1

PC2 Scores

Principal Component Analysis

Separate the Overlapping Information

PC2=

Fine

ness

co

arse

fine

PC1 = kind of wood: Spruce Spruce with bark


8/188

-0.1

0

0.1

500 1000 1500 2000

RESULT9, PC(X-expl):1(72%)

X-variables

X-loadings

PC1:Kind of wood

Principal Component Analysis

Scores and Loadings for PC1 and PC2

-0.10

-0.05

0

0.05

06:21:49_13.11. 11:43:52_13.11. 07:05:02_14.11. 14:15:56_14.11.

RESULT2, PC(X-expl): 1(95%)

Samples

Scores

-0.1

0

0.1

500 1000 1500 2000

RESULT9, PC(X-expl):2(24%)

X-variables

X-loadings

-0.05

0

0.05

06:21:49_13.11. 10:19:54_13.11. 14:37:28_13.11. 08:47:24_14.11. 14:06:48_14.11.

RESULT5, PC(X-expl): 2(24%)

Samples

Scores

PC2:FinenessSpruce

with bark

Spruce

fine

coarse


9/189

PLS Regression

Degradation of Lignin for Spruce

-0.02

-0.01

0

0.01

0.02

-0.06 -0.03 0 0.03 0.06

RESULT16, X-expl: 80%,5% Y-expl: 88%,6%

2.22.22.2

2.2

2.5

2.52.5

2.5

2.52.52.52.5

2.52.52.52.5

2.5

2.9 2.92.92.9

2.9

2.9

2.9 2.9

2.92.9

2.92.9

2.9

PC1

PC2 Scores

2.4

2.7

3.0

2.1 2.4 2.7 3.0

RESULT16, (Y-var, PC): (SFC,2)

Elements:

Slope:

Offset:

Correlation:RMSEP:

SEP:

Bias:

30

0.910025

0.237915

0.9416800.085069

0.086517

0.000981

Measured

Predicted Y

0

0.5E-06

0.0000010

0.0000015

0.0000020

10 20 30

RESULT16, PC:2 2

Samples

X-variance Residual Sample Variance

0

0.01

0.02

0.03

0.04

10 20 30

RESULT16, PC:2 2

Sample

Y-variance Residual Sample Variance


10/1810

Analysing Three-Way Data

Mode 2

Mode

1

Mode3

I

L

KMode 2

Mode

1

Mode3

I

L

K

Two different types of modes

are distinguished:

Sample mode -usually first mode

Variable mode -usually second and/or third mode

Sample mode - O

Variable mode - V

OV2 or O2V


11/1811

Substructures in Three-way Arrays

L f rontal slices I horizontal slices

K vertical slices

L f rontal slices I horizontal slices

K vertical slices

Three-way arrays can be divided into different slicesDecide which slices are put together to form a two-dimensional array


12/1812

Samples: 32 fibres from steam treated and ground woodchips

X-Data: Fluorescence Excitation-Emission spectra

(250 - 575 nm) x (300 - 600 nm)

Y-Data: Kind of wood (beech and spruce)

Severity of treatment (a combination of time and temperature)

Age of wood (fresh and old)

Plate gap of grinding ( fine and coarse).

Three-way Data Example:

Fluorescence Excitation Emission Spectra


13/1813

Three-way Data Example:


Beech

Spruce

Treatment: low middle severe


14/18

14


Results of N-PLSR


15/18

15


x1 and x2 Loading Weights


16/18

16

3D Data Import: ASCII, Excel, JCAMP-DX, Matlab

Swapping: toggle freely between the 6 OV2 and 6 O2V

layouts of a 3D table

Matrix plots: Contour and landscape plots of the samples

Variable sets: Create Primary variable sets and

Secondary Variables sets

Possibilities for Three-way Data in

The Unscrambler


17/18

17

Easy to make models Easy to interpret results

High user-friendliness

Less time spent doing data analysis,

more information extracted from your data

Faster decision making

The Unscrambler

Benefits


18/18

18

Fully functioning version

Includes the Unscrambler user manual

Includes 7 tutorial exercises and associated files

Includes 3 demonstration tours

Try The Unscrambler

9.2 for 30 daysFree trial version available on www.camo.com

CAMO Software India Pvt. Ltd.,

14 -15, Krishna Reddy Colony, Domlur Layout,

Bangalore - 560071, INDIA

[email protected]

For details, contact:

A Handy Tool for Chemometrics

Documents

Transcript of A Handy Tool for Chemometrics