Paper Automation Service CEU - Optimization Service_EN_RevUK

© ABB Group Paper Automation Service CEU – Optimization Service_EN_RevC2009-04-28 | Slide 1

Paper Automation Service Central Europe

Optimization Service


Optimization ServiceService Level and Plant Areas


Optimization ServiceDevelopment of Plant Performance

without Service

Time

Per

form

ance

ProjectPhase

Optimization

Maintenance and Service

Application Service

System Availability


Optimization ServicePhases

Diagnose

Measure process

Detailed scope with defined objectives

Forecast performance

Implement

Improve performance

Apply corrective actions

Sustain

Maintain performance

Preventative maintenance

Monitor machine index

Sustain

Implement

Diagnose

Service

1

2

3Diagnose

Implement

Sustain


Optimization ServiceAnalysis

Every Optimization begins with an analysis

Paper MachineFingerprint

PulsationVibration

Study

StockApproach

Transition

Color

Process

Boiler


Optimization ServicePhase 1: Diagnose - Paper Machine Fingerprint

MachineResponse

ProductVariability

Fiber LineStability

Provides information on:• Control behavior (disturbance reduction, responsiveness)• Grade change recovery• Sheet break recovery

Provides information on:• Controllable energy• Mechanical pulsations of vibrations• Process disturbances• Benchmark of machine stability

Provides information on:• Stock approach performance• Controllable energy and tuning quality• Oscillation sources

Provides information on: • Potential of CD controls• CD disturbances • CD control quality and tuning• Mapping problems

-7

-6

-5

-4

-3

-2

-1

0

1

2

3

0 5 10 15 20 25 30 35 40 45 50 55 60 65

Actual vs. Calc. Profiles for 1 inch Zones

Pro

file

Zones

Actual Calculated

ProfilingCapability

Sig

ma

Current Potential

0.000

0.025

0.050

0.075

0.100

0.125

0.150

0.175

0.200

10-4 10-3 10-2 10-1

Pow

er (

varia

nce)

[e-2

]

Frequency (Cycle/Time)

VPA

Diagnose

Implement

Sustain


Paper Machine FingerprintModule 1: VPA Variance Partition Analysis


Paper Machine FingerprintModule 1: VPA

TOTCDMDLMDS

As an initial analysis of the Paper Machine Fingerprint, QCS reel reports of the last 3 month are reviewed.

The results show an overview about the process variances during the last quarter.

Diagnose

Implement

Sustain



This analysis allows conclusions about possible causes of process variation an potentially useful methods for further analysis.

MDS(Machine Direction Short Term Variation)Variance in machine direction with periods

between 0.1 und 60 seconds.

MDL(Machine Direction Long Term Variation)Variance in machine direction with periods

above 60 seconds.

CD(Cross Direction Variation)Variance in cross direction.

TOT(Total Variation)

Variance over all collected measurements

Diagnose

Implement

Sustain



The following limits and distributions are valid for the evaluation of the variances. They can be seen as ideal target values.

Total 2 Sigma (TOT)in % of the average

20%

Variance Cross Direction (CD) in % of Total Variance (TOT)

10%

Variance Long (MDL) in % of Total Variance (TOT)

70%

Variance Short (MDS) in % of Total Variance (TOT)

Basis Weight < 1.7%Conditioned Weight < 1,5%Moisture < 10%Caliper < 1%Ash < 4,5%

Diagnose

Implement

Sustain



VPA results:

Comparison over machines and grades

Trend displays

Bar charts with target regions

90

95

100

105

0.1

0.6

1.1

1.6

2.1

2.6

3.1

01.03.2008 12.03.2008 24.03.2008 02.04.2008 20.04.2008 29.04.2008 11.05.2008 21.05.2008

VPA Average [TB Berichte:PM 1 Sorte XYZ:Atro]

2 S

igm

a

Reel Time Stamp

Average CD MDL MDS TOT

Sorte A

Sorte B

Sorte C

Sorte D

Sorte ASorte B

Sorte DSorte C

Diagnose

Implement

Sustain


Paper Machine FingerprintModule 2: Product Variability

0.000

0.025

0.050

0.075

0.100

0.125

0.150

0.175

0.200

10 -4 10 -3 10 -2 10 -1

Pow

er (

varia

nce)

[e-2

]




Because of continuous production conditions process variations show up frequently in machine direction.

6,50

6,75

7,00

7,25

1,60

1,65

1,70

1,75

1,80

9300 9400 9500 9600 9700 9800 9900 10000 10100 10200 10300

Raw Data

Sample Number, Ts = 5 , Total Samples = 11787

MT01_OperMeas_g.Measurement Application_PM5MD01.MT01_Target_g

PR01_g.Measurement PR01_fb.PID_OperSP

Diagnose

Implement

Sustain



The frequencies can be assigned to different frequency bands.

Slow Loopstau>20 sec Fast Loops

tau<20 secMechanicalConditions

Vibrations

Surges

Pulsations

Dilute Stock Circuit

Headbox / Former

StockPrep

0.0001 0.001 0.01 0.1 1.0 10.0 100.0

Frequency [Cycles/Second]

Decade 51 s to 0.1 s

Decade 60.1 s to 0.01 s

Decade 410 s to 1 s

Decade 32 m to 10 s

Decade 13hrs to 20 m

Decade 220 m to 2 m

Diagnose

Implement

Sustain



Raw data are logged in 3 steps

1.2000 Samples/sec.

410000 points3.4 minutes

at 3 positions

2.200 Samples/sec.

410000 points34 minutes

at one position

3.5 Sekunden Samples.

Über ca. 8 hoursIn scan mode

2 kHz 200 Hz 5 s

AGP 400 5sec. TBM from QCS

Diagnose

Implement

Sustain



High frequency data will be collected directly from the frame via Datalogger AGP400™.

Additional process data can be logged

Diagnose

Implement

Sustain



The collected data will be compared and analyzed in different ways

6,50

6,75

7,00

7,25

1,60

1,65

1,70

1,75

1,80

9300 9400 9500 9600 9700 9800 9900 10000 10100 10200 10300

Raw Data

Sample Number, Ts = 5 , Total Samples = 11787

MT01_OperMeas_g.Measurement Application_PM5MD01.MT01_Target_g

PR01_g.Measurement PR01_fb.PID_OperSP

0.000

0.025

0.050

0.075

0.00

0.01

0.02

0.03

0.000

0.025

0.050

10-2 10-1 100 101 102

MO11 1k Power Spectrum Comparison

MO

11 T

1kM

O11

M1k

MO

11 D

1k


A frequency analysis (FFT) determines frequencies contained in the signals and localizes causes of disturbances.

Diagnose

Implement

Sustain



0

5

10

15

20

25

30

35

40

Com

paris

on

CD Decade 1 Decade 2 Decade 3 Decade 4 Decade 5 Decade 6

Weight Moisture

Weight 18.2127838 3.3569510 1.3546985 6.5347900 8.4648914 21.9569969 40.1188774Moisture 4.2681885 1.1965742 1.4669303 5.4170699 34.6394196 38.3444977 14.6673136

QCSPotentials

DCSPotentials

MechanicalPotentials

MDSMDLCD

As a result an index-based diagram of the variances divided into 6 decades is compiled

Diagnose

Implement

Sustain


Paper Machine FingerprintModule 3: Machine Response



Based on bump tests of the QCS controls in Manual and Auto time constants are determined and compared.

The relation between Manual and Auto represents an index that indicates the performance of the control. A Value of 1.5 is the target based on experience.

- Setpoint - Measurement

Bump in Manual


Bump in AutoDiagnose

Implement

Sustain



Results of the machine response analysis:

Comparison of the responses in Manual and Auto

Additional findings:

Wrong process model parameters in the QCS

Improvements with or without feed-forward controls

Bad control parameters in QCS

Performance of DCS loops (Level 1)

Diagnose

Implement

Sustain

0

2

4

6

8

10

12

14

Moisture Total Head WeightTarget: <1.5; Actual: 12.7; 6.0; 2.8

Ind

ex

Target


Paper Machine FingerprintModule 4: Profile Capability



The controllable frequencies and with that a residual goal plot are determined from the raw profiles with Fourier analysis (FFT)

Raw Profile Plot

Residual Profile Plot

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09

Power SpectrumProfile: Moisture, 300 Cycles

Pow

er (v

aria

nce)

[e-2

]

Frequency (Cycle/Length)

Sig

ma

Current Potential

Diagnose

Implement

Sustain



Additional possible findings from machine response module:

Edge problems (deckle waves, wire curls, edge flows)

Global and local mapping issues

Model mismatch

Tuning issues (I versus PI)

Actuator tracking problems

Influences from wire and felt cleaning

Startup or sheet break recovery issues

Grade dependent CD tuning

etc.

Diagnose

Implement

Sustain


Paper Machine FingerprintModule 5: Fiber Line Stability



The machine chest is the interface between stock approach and machine

Process variances in this area are directly affecting paper quality

Paper MachineStock Approach

Machine Chest

Diagnose

Implement

Sustain



The following controls and measurements are analyzed:

Stock Flow

Consistency

LevelMachine Schest

TotalHead

Machine Chest

Stock Valve Fan Pump

White Water 1

Headbox

Diagnose

Implement

Sustain



The difference between Measurement and Setpoint is calculated. The error is accumulated. Based on the scaling of the measurement an index is calculated.

Error- Setpoint - Measurement

Diagnose

Implement

Sustain



Big Index


Small Index


Diagnose

Implement

Sustain



Limit

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.40

Co

mp

aris

on

[e-2

]

1 2 3 40.0299000

0.09970000.2298000

0.4104000

Total Head Level Schest Stock Flow Consistency

The indices are compared with their limits

Optimization potentials are revealed

Diagnose

Implement

Sustain


Paper Machine FingerprintSummary

Results of the Paper Machine Fingerprints:

Daily detailed report of the accomplished analysis

Index-based summary of KPI‘s

Summary of critical areas

Implementation plan improve the machine performance

Return on investment (ROI) forecasts

MachineResponse

ProductVariability

Fiber LineStability

-7

-6

-5

-4

-3

-2

-1

0

1

2

3

0 5 10 15 20 25 30 35 40 45 50 55 60 65


Pro

file

Zones

Actual Calculated

ProfilingCapability

Sig

ma

Current Potential

0.000

0.025

0.050

0.075

0.100

0.125

0.150

0.175

0.200

10-4 10-3 10-2 10-1

Pow

er (

vari

ance

)[e-

2]


VPA

Diagnose

Implement

Sustain


Optimization Service Phase 2: Implement

Implementation of the proposed changes from the five analysis modules, e.g.:

Control tuning for MD and/or CD

Adaption and extension of control strategies

Grade-dependent tuning setup

Assignment of detected frequencies to the process environment in order to find causes of product variability

Change of setpoint targets after reduction of the product variability in order to increase production and reduce energy consumption

Diagnose

Implement

Sustain



Typical savings for different paper machines:

Every Dollar invested in the reduction of product variability results in a return of 5 to 10 times.

Mill Investment Annual Savings

Ratio

Medium 65k 325k 5 to 1

Fine Writing 95k 610k 6.4 to 1

Newsprint 20k 250k 12.5 to 1

Tissue 70k 650k 9.3 to 1

Towel 75k 560k 7.5 to 1

Pulp 120k 2,270k 18.9 to 1

Average 74k 778k 10.5 to 1

Diagnose

Implement

Sustain



Example for proposed improvements for a medium mill:

Target shift: Weight

A 0.1lb shift in basis weight can be made. This results in savings of $116.500.

Target shift: Moisture

A 0.1% moisture target shift (i.e. 8.9 to 9.0) can be made. If speed limited the savings would be $40.600.

Sheet break recovery, number of breaks and faster startup

A 5 min improvement after a break and less breaks represent $264.300 in savings.

Reduced rejects and customer returns

Improvement in quality results in savings of $4,180.

Total benefits

If the recommendations are followed, the savings as a result of the project will be $425.580.

Diagnose

Implement

Sustain


Goal: Sustain the reached performance improvements

Adapt service procedures

Change operation strategy

Remote monitoring

Optimization Service Phase 3: Sustain

Availability

Optimization

Maintenance

Continuous Improvement

Sustain PhaseImplement

Utilization

Diagnose

Implement

Sustain


Optimization Service Phase 3: Sustain

Remote

Fiber Line Stability

Product Variability

Machine Response

Profile Capability

VPA Analyis

KPI

KPI

KPI

KPI

KPI

Local

RD

S

OPCSQL

MachineResponse

ProductVariability

Stock ApproachStability

-7

-6

-5

-4

-3

-2

-1

0

1

2

3

0 5 10 15 20 25 30 35 40 45 50 55 60 65


Pro

file

Zones

Actual Calculated

ProfileCapability

Sig

ma

Current Potential

0.0000.025

0.050

0.075

0.100

0.125

0.150

0.175

0.200

10-4 10-3 10-2 10-1

Pow

er (

vari

ance

)[e-

2]


VPA

Combination of Optimization Service and Remote Service to continuously remote monitor important performance indices (KPI)

Diagnose

Implement

Sustain


Optimization Services European Network

Robert SteinhauserManager

OS Central

Magnus VestbergManagerOS North

Jonas WarnqvistBusiness Driver

North+

OS NetworkCoordinator

Uwe BeekhuisBusiness Driver

Central

Fabien LaleufManager

OS South 2

Paul KellyBusiness Driver

UK

John FraserManagerOS UK

FranceSpainPortugalNorth Africa

GermanyBelgiumNetherlandsAustriaSwitzerlandChech Rep.SlovakiaSloveniaHungaryPoland South U.K.

SwedenNorwayDenmarkFinlandPoland NorthRussia

Pool of specialists available for OS in Europe

Luciano BianchiManager

OS South 1

ItalyMEA

Tadeuz RomanowskiBusiness Driver

South 2

Michele StrepparolaBusiness Driver

South 1

Andreas Zehnpfund

André RöppnackJonas Warnqvist

Jonas Berggren

Carl-Magnus Tjeder

Niclas Alfsvåg

Hans Lundberg

Pontus Johnsson

Nicklas Ljungberg

Kent Norman

John-Helge Askeland

Paul Kelly

Joe Bell (Ireland)

Christophe MicloMarco Nieri

Alessandro Camoirano

Dario Fenini

Michele Strepparola

Cees Funcke Patrick Laweless (Ireland)

Jonas Warnqvist team leader.

JC Benedic (Color Control)

P.Beuste & D.Trouche Uwe Beekhuis

Paper Automation Service CEU - Optimization Service_EN_RevUK

Documents

paper machine

vpadiagnose

machine direction

total variance

sustain

implement

service

analysis