active combustion control for ultra low emissions in aircraft gas-turbine …€¦ · ·...

Controls and Dynamics Branch at Lewis FieldGlenn Research Center

Active Combustion Control for Ultra Low Emissions in

Aircraft Gas-Turbine Engines

John DeLaatControls and Dynamics Branch

Ph: (216) 433-3744email: [email protected]


Team Members• Controls and Dynamics Branch

– Dan Paxson: Dynamic Models– George Kopasakis: Control Methods– Joe Saus: Actuators

• Combustion Branch– Clarence Chang: Combustion Science

• Engineering Directorate– Dan Vrnak: Control Software

• Supersonics Project– Dan Bulzan – Supersonics (and Subsonics) Combustion API

• Other NASA Participants– Sensors, Materials, Combustion and Flow Diagnostics

• Non-NASA Participants– General Electric, Pratt & Whitney, UTRC, Honeywell, Penn State,

Virginia Tech, Georgia Tech


OUTLINE

• Motivation: Low Emissions Combustors• NASA’s Overall Combustors Effort• Thermo-Acoustic Instability => Active Control• Technical Challenges and Approach• Early Results, Enabling Technologies• Recent Results• Modeling – Dan Paxson• Current Efforts, Future Plans• Opportunities for Collaboration


Fundamental Aeronautics, SupersonicsHigh Altitude Emissions

Objectives

• Develop the necessary technologies to enable low emissions (gaseous and particulate) combustion systems to be developed for supersonic cruise applications.

• Develop and validate physics-based models to enable quantitative emissions and performance predictions at supersonic cruise conditions using Combustion CFD simulations.

• Develop and validate high temperature sensors for use in intelligent engines.

Axial Velocity Predictions of Lean Direct Injection Low NOxEmissions Concept

Zero Axial Velocity Contours Side View through center

Also - Fundamental Aeronautics, Subsonics,Combustion•Combustion Chemistry and Turbulence Modeling•Particulates Sampling and Modeling•Alternate Fuels


90% NOx Reduction Combustion:Multi-Point Lean Direct Injection

1. Energetic quick-mixing before auto ignition at high power condition2. Lean and uniform front end makes less CO and NOx initially3. Less CO initially, shorter combustor needed4. Shorter combustor, shorter residence time, less additional NOx5. Multiple injection points allow temporal and spatial fuel/air control

- allows active fuel-shifting control to improve operability


2. Reduced film cooling: reduced damping

3. More uniform temperature and composition

1. Higher-performance fuel injectors: more turbulence

4. No dilution holes: reduced flame-holding

Ultra-Lean-Burning Combustors Are More Susceptible to Thermo-Acoustic Instabilities


Combustion Instability Control StrategyObjective: Suppress combustion thermo-acoustic instabilities when

they occur

Combustor Acoustics

CombustionProcess

SensorControllerActuator

++

Closed-Loop Self-Excited System

Natural feed-back process

Artificial control process

Fuel-airMixture system

Φ’P’


Synergistic Technologies to Enable Ultra-Low Emissions Combustion

Ultra-Low Emissions

Combustion

Fuel

Fuel InjectionDynamics

and Flameholding

Materials

FeedbackSensor

Fuel Actuator

Active Combustion

ControlMethods

ManufacturingProcesses

Combustorand

Fuel SystemDynamics


Active Combustion Control - Technical Challenges• Combustor dynamics largely unmodeled

• Noisy environment

• Liquid fuel – introduces additional unmodeled dynamics including time delay (atomization, vaporization, …)

• Actuation system – enough bandwidth and authority, not just valve (also feedline, injection, …)

• Simplified models needed for control design evaluation

• Control methods that can: – identify instability– suppress instability in presence of large time delay, substantial

noise, unmodeled dynamics

• Realistic experimental testbeds (combustor, actuation system)


Combustor Instrumentation (pressures, temp’s)

Fuel InjectorEmissions Probe

Research combustor rig and models

Fuel delivery system model and hardwareAccumulator

ModulatingValveFuel

In

Fuel NozzleAssy

Modulated Fuel Flow

Active Combustion Instability Control Via Fuel ModulationHigh-frequency fuel valve

Phase ShiftController

Fuel Valve

Fuel lines, Injector& Combustion Σ

AcousticsNL

Flame

White Noise

+++

Filter

Pressure fromFuel Modulation Combustor Pressure

Instability Pressure

Advanced Control Methods


Combustion Dynamics Modeling

Detailed, physics-based dynamic models

Fundamental understanding of combustor dynamics to aid passive,

active instability suppression

vt: -130 -115 -100 -85 -70 -55 -40 -25 -10 5 20 35 50

Penn State Injector Response Model Plot

Simplified Quasi-1D dynamic models

Allow physics-based control method validation

Results from NASA Sectored-1D Model of LPP

Combustor Rig – D. Paxson

0 500 1000 1500 200010

−10

10−8

10−6

10−4

10−2

100

Pow

er D

ensi

ty (

psi*

*2/h

z.)

Frequency (hz.)

ComputedMeasured

Reduced-order oscillator models

Run fast to allow parametric studies in support of control

system development

Combustor Acoustics

CombustionProcess


High-Bandwidth Fuel Actuator Characterization Testing – J. Saus

PressurizedAir Volume

P

TestValve

P

Pump

Accumulator

Dynamic PressureTransducers

Valve, Feed-system Characterization Rig at NASA GRC


GaTech high-response fuel valve in characterization rig in CE7A

Frequency Response Dynamic Characterization Data

High-Bandwidth Fuel Actuator

Fuel Delivery System Dynamic Response

Stroboscopic Image of Dynamic Fuel Injection

SecondaryFuel

PrimaryFuel

0

0.05

0.1

0.15

0.2

DP23

a/in

put,

psi

/vol

t

-1080

-900

-720

-540

-360

-180

0

180

360

0 500 1000 1500 2000 2500Frequency, Hz

Pha

se, de

g

0ft-1V1ft-1V2ft-1V


Adaptive phase shifting control:“Adaptive Sliding Phasor Averaged Control” – G. Kopasakis

Phase ShiftController

Fuel Valve

Fuel lines, Injector& Combustion Σ

AcousticsNL

Flame

White Noise

++

+

Filter

Pressure fromFuel Modulation Combustor Pressure

Instability Pressure

Pressure from instability

Pressure from Fuel modulation

Boundary of effective stability region

Overall combustor pressure

Boundary of restricted control region


Model-Based Control:“Multi-Scale Extended Kalman Control” – D.K. Le

EK StatesPredictor

ParameterTuning

Time-Scale Averaged Pressure Variance

Sensedcombustion

pressure

Fuelmodulationcommand

Multi-ScaleTones Analysis

Damper

Suppression

Phase Drift Estimation

Phase-AdjustedReconstruction


100 200 300 400 500 600 700 8000

2

4

6

8

10

12

Ampl

itude

, psi

Frequency, Hz

pla1c1, Run 423 and 425, 040527 - 040603

Active Combustion Instability Control Demonstrated Experimentally

Liquid-fueled combustor rig emulates engine observed instability behavior at engine pressures, temperatures, flows

Large amplitude, low-frequency instability suppressed by 90%

High-frequency, low-amplitude instability is identified, while still small, and suppressed almost to the noise floor. •0 •100 •200 •300 •400 •500 •600 •700 •800 •900 •1000

•0

•0.05

•0.1

•0.15

•0.2

•0.25

•0.3

•0.35

•0.4

•Am

plitu

de, p

si•Frequency, Hz

•Open-loop•Adaptive Phase-Shift Control


Additional Test Results:Harmonic-focused control provides over 90% reduction in pressure spectralpeak for large, low-frequency instability

Uncontrolled –vs- Controlled Instability Pressure

97% Reduction

~75% in peak amplitude reduction


0 200 400 600 8000

2

4

6

8

10

12

F re q ue ncy, Hz

P L A1C1psi

Max=11.82 @ 314 .94H z , P S D R MS =9.56

Ru n 425, 040602 , P o in t p t016

0 5 10-15

-10

-5

0

5

10

15

20

psi

T im e , se c

Tim e R MS =9.56 , Mean =-0 .2

0 200 400 600 8000

0.5

1

1.5

2

2.5

3

3.5

F re q u e n cy, Hz

P L A1C1p si

Max=3 .16 @ 100 .1H z , P S D R MS =3.04

Ru n 425, 040602 , P o in t p t023

0 5 10-15

-10

-5

0

5

10

15

psi

T im e , se c

Ti m e R MS =3.07 , Mean =-0 . 3

Uncontrolled

Open-loop perturbations at 100Hz

Additional Test observations: Open-loop 100 Hz fuel valve perturbations shows substantial interference with 315Hz instability


• Increasing fuel flow increases 530Hz combustion instability, preventing full-power operation

• Continuing research to demonstrate instability control/suppression:

– Apply advanced NASA modeling, control, actuation methods

Recent Results – Lean, Low-Emissions Combustor Instability Characterization

Active Control may extend lean, low-emissions combustor operationActive Control may extend lean, low-emissions combustor operation

Combustor Pressure Oscillations

Amplitude

FrequencySpectra


Simulation of Combustion Instabilities:A Sectored-One-Dimensional Approach

Daniel E. PaxsonNASA Glenn Research Center

Cleveland, Ohio216-433-8334

[email protected]

Propulsion Controls and Diagnostics WorkshopCleveland, Ohio, November 2007


Discussion Outline

• Motivation• Simulation Methodology Description• Results• Comments


Motivation

• Low emission combustors may be susceptible to thermo-acoustic combustion instabilities.

• One possible solution to the this problem is active control.• Successful active control design requires accurate modeling and

simulation.– The essential physical phenomena should be correctly captured (e.g. self-

excitation).– Characterization and a control design necessitate rapid simulation (i.e.

relative simplicity).– Simulation must lend itself to implementing a variety of sensing and actuation

strategies.

For combustor configurations in which the potential instabilities propagate axially, but which contain abrupt changes in cross sectional area, the method to be described can achieve the simulation goals.


Sector 2

Sector 3

Injector Region

Combustor Region

Description-Simplifications

• One-Dimensional• Perfect Gas

Within Each Sector:

Sector 1


Description-Governing Equations

⎥⎥⎥⎥⎥⎥⎥

⎦

⎤

⎢⎢⎢⎢⎢⎢⎢

⎣

⎡

⎟⎟⎠

⎞⎜⎜⎝

⎛+

−

+

=

ρuz2ρu

1pu

ρupρu

F 2

2

γ

γ

⎥⎥⎥⎥⎥

⎦

⎤

⎢⎢⎢⎢⎢

⎣

⎡

+γ

=

ρz2

ρu1)-(

pρuρ

w 2

γ

x),w(Sx

)w(Ftw

=∂

∂+

∂∂

⎥⎥⎥⎥⎥⎥⎥⎥⎥⎥⎥

⎦

⎤

⎢⎢⎢⎢⎢⎢⎢⎢⎢⎢⎢

⎣

⎡

+−⎟⎟⎠

⎞⎜⎜⎝

⎛⎟⎠⎞

⎜⎝⎛∂∂

∂∂

+

++−γ

+⎟⎟⎠

⎞⎜⎜⎝

⎛⎟⎟⎠

⎞⎜⎜⎝

⎛−γ

+∂∂

∂∂

+⎟⎟⎠

⎞⎜⎜⎝

⎛⎟⎠⎞

⎜⎝⎛∂∂

∂∂

=

sst

*t

r

ht0s

0

t

2

*t

0.752*

t

s

zmRxz

ScRexm

QRq1

mT1)Pr(T

2u

xRex

uuxu

Rex

m

x),w(S

ε

ε

ρσε

( )⎭⎬⎫

⎩⎨⎧

<>−

−=igni

igniiign210 TT;0

TT;TT1z)ζρz(ζKR

)T(TQ iinfht −α=

• Reactive Euler Equations with Source Terms

• MacCormack’s Method– Fast– Second Order Accurate

• Artificial Viscosity– Baldwin-MacCormack– Density Instead of Pressure

• Uniform Grid Spacing– Relatively Course– Requires Some “Tuning” (e.g. eddy

viscosity, reaction rates)• Sectors Joined With Compatible

Boundary Conditions– Multi-block Approach– Nothing “Stored” @ Interface


Initial Tests-Conical Pipe Acoustics

0.0

0.5

1.0

0.0 0.2 0.4 0.6 0.8 1.0x/L

A/A

*

Q-1-DSectored

pressuremeasurement

excitation

0.9

1.0

1.1

1.2

0 10 20 30 40 50

Non-Dimensional Time

p/p*

SectoredQ-1-D

• Theoretical resonant frequencies are integer multiples of the fundamental.

• Q-1-D and Sectored models agree• 100 numerical cells.


Validation-Valveless Pulsejet

1.00.56

0.350.22 0.19

0.109 0.144 0.249 0.249 0.249

Tailpipe

CombustionChamber

IntakeFuel Injection

• Coupling between heat release and acoustic modes is self-excited.

• No external, forced excitation.

• Geometry approximates a functional pulsejet.

• Frequency is correct.• 200 cells.

-0.005

0.000

0.005

0.010

0.015

0.020

0.025

0.0 1.0 2.0 3.0 4.0 5.0Non-Dimensional Time

Ray

leig

h In

tegr

al, R

a ∫ ′′=1

00 dxpRqRa


Comparison With ExperimentGRC CE-5 LPP

0.0

5.0

10.0

15.0

20.0

0 20 40 60 80 100

x (inches)

Hyd

raul

ic D

iam

eter

(inc

hes)

Fuel injectionCooling

Tee-section exhaust

Combustor

Pressuretransducer

• Exhaust modeled as mass extraction.

• Cooling spray modeled as a high heat transfer region.

• Flame position adjusted with turbulent diffusivity distribution.

• Noise added at inlet boundary.

• 350 cells, CPU time=1600 integration time .

Constant pressure boundary Wall boundary


Comparison With ExperimentLow Pressure, Low Mass Flow, Low Inlet Temperature

ComputedMeasured


ComputedMeasured

Comparison With ExperimentHigh Pressure, High Mass Flow, High Inlet Temperature


02

46

810

121416

0 5 10 15 20Axial Distance (in.)

Cro

ss S

ectio

nal A

rea

(in.2

) CombustorDiffuser & Plenum

Injector

Comparison With ExperimentGRC CE-9 SNR

constant mass flow boundarychoked boundary

transducer

• Weak oscillations @ approx. 500 hz.• “Pink” noise at injector (momentum source)• Helmoltz-like.

RigSimulation

0

5

10

15

20

25

30

0 10 20 30 40Axial Distance (in.)

Cro

ss S

ectio

nal A

rea

(in.2

)

CombustorDiffuser & Plenum

Injector


transducer

RigSimulation

• Strong oscillations @ approx. 300 hz.

0

5

10

15

20

25

30

0 10 20 30Axial Distance (in.)

Cro

ss S

ectio

nal A

rea

(in.2

)

CombustorDiffuser & Plenum

Injector

RigSimulation

transducer

• Strong oscillations @ approx. 350 hz.• Simulation preceded testing (i.e.

predicitve)



60 seconds of rig data during which f/a ratio increases from 0.028 to 0.03.

2.5 second simulation with linear fuel flow increase corresponding to f/a ratio change from 0.025-0.03.

Area ratio=0.17P0=1.01T0=1.00

ρ’u’=0.005

28.4 in. 35.7 in.

1.34 in.

D=12.0 in.D=4.0 in.

D=2.1 in.

Area ratio=0.093pr=0.545 (choked)

P4DynDn

P4DynUpfuel

air

Comparison With ExperimentGRC CE-5B-Stand1 LEC

• Dump tank cooling spray simulated with a choked (reflective) boundary condition.

• Amplitude variation with f/a matched.


0 100 200 300 400 500 600 700 800 900 10000

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

Am

plitu

de, p

si

Frequency, Hz

Open-loopAdaptive Phase-Shift Control

0 500 10000

2

4

6

8

10

12

14

Frequency, Hz

PLA1C1psi

Max=13.26 @ 317.38Hz, PSD RMS=10.07

Run 423

Comparison With ExperimentGRC CE-9 SNR with Active Control Applied

Simulated Control

High Frequency Instability

Measured Control

Low Frequency Instability


Sectored 1-D Model Concluding Remarks

• The sectored-one-dimensional technique has successfully simulated instabilities in a variety of combustors with complex geometries.

• Simulations run far from real-time, but fast enough for control design.

• Simulated plant responses to control match measured responses.

• Some success shown in prediction of instabilities, but “tuning” of some parameters is still required.

• More work is therefore needed to model complex phenomena in a 1-D compatible fashion.


Active Combustion Control for Ultra Low Emissions in

Aircraft Gas-Turbine Engines

Current Directions and Future Plans

John DeLaatControls and Dynamics Branch

Ph: (216) 433-3744email: [email protected]


Current Directions and Future Plans

• Penn State and Virginia Tech– Active fuel nozzle, flame transfer

function

• Georgia Tech– Integrated control of:

• Dynamic stability margin• Static stability margin• Dynamic stability mitigation

quartz combustor

Combustor test rig for flame response measurements

fuelfuel modulation actuator

LDI injector

2-D Spray Image

quartz combustor

Combustor test rig for flame response measurements

fuelfuel modulation actuator

LDI injector

2-D Spray Image

1 5 2 0 2 5 3 0 3 5 4 00

0 . 1

0 . 2

0 . 3

0 . 4

Slo

pe A

utoc

orre

latio

n C

urve

(o)

Pre

ssur

e A

mpl

itude

(+)

1

2

3

0

4

In le t V e lo c ity ( m /s )

stable

NRA Cooperative Agreements


Current Directions and Future Plans• Current platform - lean combustor concept (not LDI)

– Actuator research for small “pilot” flows– Dynamic model validation– Instability control demonstration

• Future platform - LDI Multi-point injection– Fundamentals rig in CE7C– High pressure testing in CE5, ASCR– Control methods that exploit multipoint injection– Multidimensional models

• Incorporate technologies from NRA’s

• Harmonic, sub-harmonic models and control


Opportunities for Collaboration

• NRA’s, SBIR – Watch solicitations. Future topics TBD• SAA’s – Have one in place, others welcome

• Requirements definition and feedback (engine, HW mfrs)• Realistic testbeds for technology transfer• Control methods integration and field testing• Modeling methods field testing• Multidimensional models development• Actuator systems, associated models development, field testing


Concluding Remarks - Long Term Goal for Active Combustion Control

• Improve fundamental understanding of the combustor processes

in order to…• More effectively integrate multi-point combustor

design, controls, sensor, and actuator technologies to provide…

– An intelligent fuel/air management system with temporal and spatial fuel modulation for

• Instability avoidance/suppression– Thermoacoustics, blowout

• Pattern factor control• Emissions minimization

to enable…Combustors with extremely low emissions throughout the engine operating envelope


References• Paxson, D.: “A Sectored-One-Dimensional Model for Simulating Combustion Instabilities in Premix Combustors,”

presented at the 38th Aerospace Sciences Meeting & Exhibit. AIAA-2000-0313, NASA TM-1999-209771, January 2000.

• Cohen, J.M. et al: "Experimental Replication of an Aeroengine Combustion Instability," International Gas Turbine & Aeroengine Congress & Exhibition, Munich, Germany, ASME Paper 2000-GT-0093, May 2000.

• DeLaat, J.C.; Breisacher, K.J.; Saus, J.R.; Paxson, D.E.: “Active Combustion Control for Aircraft Gas Turbine Engines.” Presented at the 36th Joint Propulsion Conference and Exposition, Huntsville, Alabama, July 17-19, 2000. NASA TM 2000-210346, AIAA –2000-3500.

• Le, D.K.; DeLaat, J.C.; Chang, C.T.; Vrnak, D.R.: “Model-Based Self-Tuning Multiscale Method for Combustion Control.” Presented at the 41st Joint Propulsion Conference and Exhibit cosponsored by the AIAA, ASME, SAE, and ASEE, Tucson, Arizona, July 10-13, 2005. AIAA-2005-3593.

• Kopasakis, G.; DeLaat, J.; Chang, C.: “Validation of an Adaptive Combustion Instability Control Method for Gas-Turbine Engines,” 40th Joint Propulsion Conference and Exhibit, Ft. Lauderdale, FL, AIAA-2004-4028, NASA TM-2004-213198, October 2004.

• DeLaat, J.C.; Chang, C.T.: "Active Control of High Frequency Combustion Instability in Aircraft Gas-Turbine Engines," 16th International Symposium on Airbreathing Engines, Cleveland, OH, ISABE-2003-1054, NASA TM-2003-212611, September 2003.

• Cohen, J.M.; Proscia, W; and DeLaat, J.C.: “Characterization and Control of Aeroengine Combustion Instability: Pratt & Whitney and NASA Experience.” In “Combustion Instabilities in Gas Turbine Engines, Operational Experience, Fundamental Mechanisms, and Modeling”, AIAA Progress in Astronautics and Aeronautics series, Tim Lieuwen, Vigor Yang editors, Chapter 6, p. 113-145, October 2005.

• Okojie, R.S.; DeLaat, J.C.; Saus, J.R.: “SiC Pressure Sensor for Detection of Combustor Thermo-Acoustic Instabilities.” Presented at the 13th International Conference on Solid-State Sensors, Actuators and Microsystems, Seoul, Orea, June 5-9, 2005. Volume 1, p. 470-473.

active combustion control for ultra low emissions in aircraft gas-turbine …€¦ · ·...

Documents