Maximizing Operating Effectiveness with Integrated ... Library/Research/Coal/major... · Maximizing Operating Effectiveness with Integrated Combustion and Soot Blowing Optimization

Power-Gen InternationalNovember 29, 2006

Maximizing Operating Effectiveness withIntegrated Combustion and Soot Blowing Optimization

Presented by: Joe Naberhaus

Dynegy Inc., Midwest Fleet Operations

1

AgendaAgenda

Dynegy OverviewMotivations for OptimizationCCPI Project OverviewSoot Cleaning OptimizationLooking Ahead

2

About DynegyAbout Dynegy

NEPOOL

NEPOOL

FRCCFRCC

MAPPMAPP

SPPSPP

MAINMAIN

ERCOTERCOT

WECCWECC

SERCSERC

West

43 MW Gas

Texas

610 MW Gas

Midwest

3,316 MW Coal/Oil442 MW Gas

Southeast

815 MW Gas

Northeast

1,507 MW Gas/Oil370 MW Coal

1,137 MW Gas49 MW Hydro

Midwest-Peakers

2,970 MW Gas

U.S. PORTFOLIO11,259 net MW

3

DynegyDynegy’’s Baldwin Energy Complexs Baldwin Energy Complex

3 - 600 MW Units1970-1975

Units 1 & 2 Cyclone Fired14 Cyclones/Unit

Unit 3 T-Fired6 Mills

PRB CoalConversions1999 & 2000

4

Motivations for OptimizationMotivations for Optimization

Full Load PRB operation requires tight control

Loss of spare fuel delivery capacitySmall process changes have significant effectsSeasonal impacts to heat rate

ExpectationsAbility to control key parameters on consistent basisAbility to compensate for changes in coal qualityImproved understanding of available data and its use for improved operationsAbility to optimize controls to meet plant objectives.

5

NeuCoNeuCo’’s CCPI Project @ Baldwins CCPI Project @ Baldwin

About the Clean Coal Power Initiative (CCPI)$1.3 B initiative to demonstrate clean coal technologies in the fieldSponsored by DOE’s National Energy Technology LaboratoryNeuCo’s project at Baldwin selected as Round 1 winner in 2004

Five integrated optimization modules, parallel development

SCRCombustion Soot blowingPerformance Maintenance

Products developed iteratively with multiple releases

6

Why Optimize Soot Cleaning?Why Optimize Soot Cleaning?

Cleaning actions (or lack thereof) affect many plant parameters:

Slagging/fouling impacts heat transferabilityCapacity: Steam and gas temperatures, spray flows, differential pressures, fan limitsPerformance: Boiler efficiency, heat rateEmissions: NOx, Opacity, LOI, COAvailability/Reliability: Waterwall/tube longevity, EFOR, equipment wear-and-tear

Operational complexities:Fuel and equipment variationsSCR/SNCR systemsLOI control objectives

Bottom line economic impact - $$$$

7

SootOptSootOpt

BoilerEfficiency

Steam Temperature OF

950OF 1010OF1005OF

NOx

Attemperation

Spraying

Typical operating range

Desired Setpoint

SootOpt, because of its neural model, can operate in this narrower range

8

What SootOpt DoesWhat SootOpt Does

Optimizes boiler cleaning based on unit-specific objectives:Improves emissions control (NOx, opacity, CO)

Improves Heat Rate including Reheat & Superheat steam temperature control

Balances tradeoffs between furnace/backpass absorption

Reduces O&M costs by avoiding unnecessary boiler cleaning actions and reducing tube wear and thermal stressing

Compensates for off-design fuels and operations

Leverages existing soot cleaning instrumentation, models, equipment and control systems

9

How SootOpt WorksHow SootOpt Works

Hybrid Expert

Adaptive System

Expert Fuzzy Component

Heuristic Set 1 SH < 990 990 < SH < 995 995 < SH < Tset SH > TsetRH < 990 A A A A990 < RH < 995 A B C C995 < RH < Tset A B B or C CRH > Tset A B C B or C

.

.

.

.

.

.

Boiler StateCF-SHDivCF-SHFin

CF-SHPrimCF-RH

Flux 1-16Freq IR -1Freq IR -2

NOxSoot Blowing Control PLC

10

SootOpt at Baldwin Unit 3SootOpt at Baldwin Unit 3

630 MW, Base-Loaded, T-FiredSOFA, Low NOx BurnersHigh variability in PRB coalHeat Flux Sensors and Water CannonsDiamond SentrySeries 1500 control system with locally intelligent controlsPrecisionClean and standard IK’s in convection passAlso thermocouples and FEGTPrevailing sootblowing guidelines:

ISB preset flux targets in the furnace, operators interveneOperator initiated in the convection pass

11

SootOpt - Diamond SentrySeries 1500 Interface

12

Optimization Benchmarks

Optimization AlertsCurrent Events

13

14

SootOpt Sensitivities: Net HR

15

SootOpt Causality Profile: RH Attemperation Sprays

17

18

19

20

21

SootOpt at Baldwin: Results Thus FarSootOpt at Baldwin: Results Thus Far

Operating and Producing Results on Unit 3Initially operated only on Water Cannons for furnace cleaningNow operating on furnace and all convection pass regionsNow being integrated with CombustionOpt

Being installed on Unit 2 nowUnique opportunity to quantify contributions of individual ISB control and instrumentation components

22

RH Temps with SootOpt (Furnace Only)RH Temps with SootOpt (Furnace Only)

Hot RH Outlet Temp

4/18

/05

5/8/

05

5/28

/05

6/17

/05

7/7/

05

7/27

/05

8/16

/05

9/5/

05

DateTime

960

970

980

990

1000

1010

1020

HO

T R

EHE

AT

OU

TLE

T TE

MP

23

Net Unit Heat Rate

500 400 300 200 100 0

0

2000

4000

6000

8000

1000

0

1200

0

1400

0

1600

0

1800

0

2000

0

2200

0

2400

0

DataIndex

50

55

60

65

70

75

80

85

90

95

100

Wet

Bul

b Te

mp

Heat Rate Heat Rate vsvs Wet Bulb over TimeWet Bulb over Time

24

Heat Rate Heat Rate vsvs CondCond Press over TimePress over Time

Net Unit Heat Rate

0

2000

4000

6000

8000

1000

0

1200

0

1400

0

1600

0

1800

0

2000

0

2200

0

2400

0

DataIndex

1.8

2.0

2.2

2.4

2.6

2.8

3.0

3.2

3.4

3.6

3.8

4.0

4.2

CO

ND

3A

PRE

SS

25

NOx Model with CombustionOpt & SootOpt MVs as Inputs

NOx Model with only CombustionOptMVs as Inputs

Integration Improves Model Accuracy

26

SootOpt MVs added to CombustionOpt NOx Model

Integration Increases Benefits

27

28

29

SootOpt SootOpt –– SummarySummary

SootOpt improves control of heat-transfer process

Relates impacts to global objectives. Searches for global optimum across multiple sometimes competing objectives

SootOpt performance can be changed by changing relative priority of objectives, e.g., Steam Temperature Max vs. Heat Rate, or all can be given equal priority

Lets user assess where priority should be given based on bottom line impact.

The highly inter-related processes of combustion and boiler cleaning confirms the need for integration to achieve maximum benefits

30

Looking AheadLooking Ahead

Complete Integration of CombustionOpt and SootOpt and at Unit 3Complete installation of SootOpt on Unit 2Integrate CombustionOpt, SCR-Opt, SootOpt, PerformanceOpt and MaintenanceOpt at Unit 2Further refine CombustionOpt, PerformanceOpt & MaintenanceOpt at Unit 1Further refine Unit and Plant AdvisorsSubsequent refinement and releases based on feedback from Baldwin and other NeuCo Showcase sites