Optical Measurments Webinar

Download as pptx, pdf, or txt
Download as pptx, pdf, or txt
You are on page 1of 46

Enabling the Extraordinary

To Fly To Power To Live

OPTICAL MEASUREMENTS
FOR CONDITION MONITORING

Gianluca Nicchiotti
Enabling the Extraordinary
To Fly To Power To Live
Presentation Outline
Holder exponent vs time
0.25

0.2

0.15

0.1

0.05

-0.05
100 200 300 400 500 600 700
Time [s]

The Green Push Optical Sensing Meggitt Sensor Optical vs Piezo Conclusions

Lean Combustion Fabry-Pérot Working System Architecture Test Campaign Current developments
Principle
Hydrogen Properties and Features Results Take away concepts
Benefits
Questions

3
Meggitt proprietary and confidential. No unauthorised copying or disclosure.
THE GREEN PUSH

4
Green Push and Sensing Technologies

Source Georgia Tech

• Zero Carbon Emission

End User

• Lean Combustion (Reduce LBO Safety


Margins)
Customer • New Fuels: Hydrogen

• Flame Monitoring
• Lean Blow Out(LBO)

Sensing Technology detection/prediction


• FlashBack detection/prediction

5
16 October 2023 Meggitt proprietary and confidential. No unauthorised copying or disclosure.
Way1: Lean Combustion
Motivation and Consequences

New sensing and data processing


methodologies
LBO Management
Reduce LBO
Safety Margin
Lean
Combustion

Emission
standards
Operation
savings

6
Meggitt proprietary and confidential. No unauthorised copying or disclosure.
Lean combustion challenges

Higher reactant flow velocity


Reaction time not keeping the pace of mixing time
Pulsations due to flame-front instability

Consequences

• Lean Blow Out


• Flame instabilities

7
Meggitt proprietary and confidential. No unauthorised copying or disclosure.
Way2 :Hydrogen
Motivation and Consequences

New sensing and data processing


methodologies
Flashback
detection

Hydrogen
Combustion

Emission
standards

8
Meggitt proprietary and confidential. No unauthorised copying or disclosure.
Hydrogen combustion challenges

Higher flame velocity


Thermo-acoustic effect : hydrogen addition is likely to favor
high-frequency instabilities
Change in flame color from visible to IR: at 100% hydrogen
the flame is invisible to the eye

Consequences

• Flashback
• Flame instabilities
• Flame detection

9
Meggitt proprietary and confidential. No unauthorised copying or disclosure.
Sensing Technologies
Rationale Capability and Drawbacks

Burning velocity, flame thickness and flow dynamics govern Chemiluminescence Detector Maintainability
flame processes close to LBO and Flashback producing
Presence and strength of the combustion process
corresponding changes in: Reaction rate and heat release rate

Particle Image Velocimetry (PIV)


Ion Emissions Operating temperature
• Quantitative flow visualization technique
• Instantaneous whole-field fluid velocities
Flow speed
High-speed intensified CCD camera Response Time
Flame Shape
• Flame presence and shape Piezo-Electric

Pressure/temperature
SNR at high temperatures and low
Dynamic Pressure Sensors frequencies
• Temporal rate of change of heat release

10
Meggitt proprietary and confidential. No unauthorised copying or disclosure.
Optical Sensors & Green Challenge

Challenge Green Combustion Conventional Optical


Sensors
Temperature Combustion Temperature All passive head – more
temperatures↑ limitations more headroom upwards
severe
Reliability Higher temperature Active (piezo- All passive head – higher
requirement electric/resistive) reliability expected
Sensitivity Maintain at higher temp ↓ with ↑ temperature Constant with ↑ temperature
Bandwidth Thermo-acoustic Semi-infinite tube Direct mounting
oscillation BW ↑ limits BW
Flame Health Flash Back LowFrequency Immune to twinning
Monitoring Blow out phenomena limitiations pyroelectricity
Noise More complex control Electrical No EMI on fibre link/sensor
systems cabling/shielding

11
Meggitt proprietary and confidential. No unauthorised copying or disclosure.
OPTICAL SENSING

12
Meggitt Positioning

Meggitt’s view on optical sensing technologies trend


– Commercially widely deployed in Medical and O&G
– On-going transition from advanced research stage to implementation on in-service asset in Aerospace
and Power Generation
– Current base of technology providers is predominantly SMEs

Meggitt’s strategic positioning


– Become a world-class integrator of optical sensing technologies for Aerospace and Power Generation

Already active partnerships and projects


– Point sensing (Fabry-Pérot): Pressure, vibration, temperature sensing, …
– Distributed sensing (Fiber Bragg Gratings): Embedded optical fibres to measure strain and temperature

13
Meggitt proprietary and confidential. No unauthorised copying or disclosure.
Fabry-Pérot Point Sensing
Working principle

Pressure diaphragm moves


in response to pressure
variations
Optoelectronic interrogator Optical Fiber Cable: transmits light to
 emits a light signal and from the optical interferometer
 captures the reflected signal
 measures the sensor gap Interferometer
 converts the gap to pressure
Gap

Optical interferometer
inside a sealed cavity

14
Meggitt proprietary and confidential. No unauthorised copying or disclosure.
Fabry-Pérot Point Sensing
Interferometer Characteristics

Low Finesse i.e., low mirrors reflectivity

Wideband light source (50 nm FWHM)

15
Meggitt proprietary and confidential. No unauthorised copying or disclosure.
Fabry-Pérot Point Sensing
Interferometer: Basic Description

Fabry–Pérot interferometer is a cavity made by two parallel reflecting surfaces. The reflected light is frequency
modulated as a function of the cavity size.

The spectral response of a Fabry-Pérot is determined by interference between the light reflected from different
surfaces

 2 
At every reflection phase increases of   2nl cos( )
  


F sin 2 ( )
Reflection coefficient is R ( )  2

1  F sin 2 ( )
2

Where F is the finesse 4r


F
(1  r ) 2

16
Meggitt proprietary and confidential. No unauthorised copying or disclosure.
Meggitt Optical Sensing Technology
Optical Technology : Benefits

Enhanced sensitivity & accuracy

Better sensitivity @ high & low frequencies

Insensitive to EMI perturbations

Inherently safe

Extremely compact size (sensor head)

Suitable for multiplexing

Multiple sensing capabilities (pressure, temperature, strain,…)

17
Meggitt proprietary and confidential. No unauthorised copying or disclosure.
Fabry-Pérot Point Sensing
Versatility

The dimension of the Fabry-Perot gap modulates the reflected spectrum

The measured spectrum contains the information of the gap dimension =>

Any physical quantity modifying the FP cavity dimension can be related to the
reflected spectrum

Different cavity or processing designs allow to measure either:


“Dynamic” Pressure
Static Pressure
Temperature
Displacement , Vibration, etc.

18
Meggitt proprietary and confidential. No unauthorised copying or disclosure.
MEGGITT SENSOR

19
Optical Sensing System
System Architecture
System Boundary

Front End
Susbystem
Installation Optoelectronic Data processing
Combustion Sensor Optical cable
Interface module algorithms
Chamber Back End
Subsystem

Control and
monitoring
algorithms

Sensor

Interrogator
20
16 October 2023 Meggitt proprietary and confidential. No unauthorised copying or disclosure.
Optical Pressure Sensor
Working principle

Combustor Side

Interrogator

Pressure Measurement

Control Center Side

21
Meggitt proprietary and confidential. No unauthorised copying or disclosure.
Fabry-Pérot Point Sensing 2𝜋
𝛿= 𝑥 tan ⁡(𝜃)
The Fizeau Wedge Role 𝜆

• Fizeau wedge consists of two surfaces intersecting at a very small angle


• Transmitted light peaks occur
We T ( ) 
1
1  F sin 2 ( )
dge

When the thickness of the FP gap changes with diaphgram deflection, the
peak the location of maximally transmitted light will shift a corresponding
Pro amount to a new position along the fizeau wedge
ces
s
• The CCD array placed behind the Fizeau wedge receives the spatially
spread light signal At each pixel corresponds a unique Fabry Perot gap
CC thickness
D
Fizeau wedge converts a spectral modulation into a spatial modulation

22
Meggitt proprietary and confidential. No unauthorised copying or disclosure.
Optical Components
Key equations

2 
F sin ( )
R ( )  2
2 
Fabry-Pérot Reflectivity

1  F sin ( )
2

1
T ( ) 
Wedge Transmissivity
1  F sin ( )
2

23 EVI-GTI 2021
16 October 2023 Meggitt proprietary and confidential. No unauthorised copying or disclosure.
Optical Pressure Sensor
Design goals and main features

Improve piezo performances:


• Better SNR for frequencies lower than 50 Hz at high temperature
• Insensitivity to vibrations
• Insensitivity to EMI
Parameter Feature

Dynamic pressure range Full 0.2 to 3 bar pk-pk


Scale
Frequency response 2 to 20000 Hz
Precision 0.2 mbar pk

Accuracy < ±7%

Operating temp. range -55°C to 600°C (survival 650°C)

Operating pressure range to 30 bar (survival 69 bar)

24
Meggitt proprietary and confidential. No unauthorised copying or disclosure.
OPTICAL VS PIEZO

25
Piezoelectric sensors
Low frequency response limitations

Increase of temperature causes the rapid drop of


sensor resistance (x10 per each 100°C) that
significantly increases the noise levels that may
masks the low amplitude pressure changes at the
low frequencies

26
Meggitt proprietary and confidential. No unauthorised copying or disclosure.
Test Campaign
Objective

Compare the optical system with its piezo counterpart


within several realistic scenarios representative of gas turbine
combustion
in a controlled, precise and repeatable manner
Focus on
– Vibration perturbation Electromagnetic Interference
– Capability to detect combustion instabilities
– Combustion event detection
 Ignition
 Flash-back
 Change in regime
 LBO

27 EVI-GTI 2021
16 October 2023 Meggitt proprietary and confidential. No unauthorised copying or disclosure.
Test Setup
Test rig scheme and sensor arrangement

28
23/06/2020 Meggitt proprietary and confidential. No unauthorised copying or disclosure.
Test Campaign
Hot flow experiments

Test Method Analysis

Ignition EMI Spark ignition with and Coherence with accelerometer


without gas
FlashBack Main air cooling air ratio Pressure low bands
Premixed to Varying burning air Coherence Chemiluminescence
Diffusion Flame
Combustion Koelner Dom Check amplitude and phase at
instabilities eigenfrequencies
Imminent LBO Increase mass flow Time/frequency domain

29
Meggitt proprietary and confidential. No unauthorised copying or disclosure.
Test Campaign
Results I

Ignition EMI Premixed/diffusion transition


At ignition, the optical probes are much less disturbed by EMI Both measurement principles qualify for the detection of diffusion
than the piezoelectric transducers and accelerometers flame matching the chemiluminescence method

30
Meggitt proprietary and confidential. No unauthorised copying or disclosure.
Test Campaign
Results II

Combustion instabilities
The optical sensor and piezoelectric sensors show a good agreement in amplitude and phase response

31
Meggitt proprietary and confidential. No unauthorised copying or disclosure.
Test Campaign
Results III

Flashback
• In the low frequency band of the optical probes [2-5 Hz], the double-peak structure at each cycle coincides at fixed frequency
with the flashback events.
• This analysis cannot be performed with piezo and provides a more reliable (and physically meaningful) flashback detection.
• The low frequency band analysis may help to identify flashback precursors

32
Meggitt proprietary and confidential. No unauthorised copying or disclosure.
Test Campaign
Results IV: LBO and Low frequency characteristic

Scalograms in proximity of the LBO event for both the optical (top) and piezo-electric (bottom) sensors; arrows indicate the LBO time event
at t = 9s

Optical

Piezo-electric

33
23/06/2020 Meggitt proprietary and confidential. No unauthorised copying or disclosure.
Optical sensing
Low Frequency Analysis

Optical sensing offers a reliable foundation for methodologies requiring good quality low frequency data

Wavelet Analysis Fractal Analysis

34
Meggitt proprietary and confidential. No unauthorised copying or disclosure.
LBO precursor detection
Wavelet Indicators

35
Meggitt proprietary and confidential. No unauthorised copying or disclosure.
LBO precursors
Comparison optical vs piezo

36
Meggitt proprietary and confidential. No unauthorised copying or disclosure.
Fractal Analysis
Vs equivalence ratio

Decreasing the equivalence ratio by increasing air flow rate approaches the flame to extinction

37
Meggitt proprietary and confidential. No unauthorised copying or disclosure.
Fractal analysis
Evolution towards LBO

38
Meggitt proprietary and confidential. No unauthorised copying or disclosure.
Fractal Analysis
PE vs optical

The Holder exponent, derived from optical sensor measurements at low frequency, approaching the LBO event; pressure values in
arbitrary units are plotted below (dotted line)

Piezo-electric Optical

39
23/06/2020 Meggitt proprietary and confidential. No unauthorised copying or disclosure.
CONCLUSIONS

40
Recap
Sensing System

41
Meggitt proprietary and confidential. No unauthorised copying or disclosure.
Take away concepts

• The need to reduce emissions leads to the adoption of new strategies for energy production.
• Decarbonisation of gas turbines present challenges for combustion control that current sensors only partially meet.
• Optical sensors have characteristics that appear to be better suited to applications (temperature, EMI, SNR)
• Meggitt aims at becoming a world-class integrator of optical sensing technologies for Aerospace and Power Generation
• Meggitt system solution for dynamic pressure measurement is based on a combination of two interferometers and signal processing
algorithms
• Meggitt optical interferometric pressure sensing system has been validated in operational environment
• Tests on novel Meggitt optical sensors had been conducted in an “ad hoc” campaign
• Pressure sensing method based on novel optical sensors matches the performances of the more mature piezoelectric sensing in most
case however..
• The higher SNR of optical sensors at low frequencies provides a reliable foundation to monitor phenomena (LBO, Flashback) critical
to transform gas turbines into low or zero-carbon emitting systems

42
Meggitt proprietary and confidential. No unauthorised copying or disclosure.
Current Developments

HT Dynamic pressure sensor prototypes (550°C) for Energy applications


– Tests on turbines since H1 2020.
– Industrialisation phase to be completed by end 2021
– Entry to market in 2022.

New aerospace development of a HT (550°C) static pressure and temperature sensor and
interrogator
– Second Engine Test H1 2022

43
Meggitt proprietary and confidential. No unauthorised copying or disclosure.
References

Lean Blowout Sensing and Processing via Optical Interferometry


and Wavelet Analysis of Dynamic Pressure Data

Gianluca.Nicchiotti1, Krzysztof Solinski2, Fabrice Giuliani3

1,2
Meggitt SA, Energy Sensing & Controls, Fribourg, 1700, Switzerland
[email protected]
[email protected]
3
Combustion Bay One e.U. advanced combustion management, Graz, Austria
[email protected]

ABSTRACT known wavelets time/frequency analysis capability to carry


out an investigation of signal variations at a lower frequency
Dynamic pressure transducers, based on the piezoelectric range up to 40 Hz. The second uses wavelets to extract non-
effect, represent the most widely adopted sensing method linear characteristics of the signal related to the fractal
used for gas turbine combustion monitoring. However, piezo- dimension of the signal itself. The two algorithmic strategies
electric technology shows some drawbacks when low have been applied to signals acquired with both piezoelectric
frequency measurements are requested at temperatures and optical measurement chains.
exceeding 450°C. Because of more stringent environmental
requirements, development efforts are nowadays focused on The flame health indicators computed on data acquired by the
reducing NOx emissions. These resulted in advanced optical sensor, reacted to changes in combustion dynamics
combustor designs that have been found to be particularly preceding the blow out event. This was not the case with the
prone to trigger combustion instabilities and susceptible to data set acquired from the piezoelectric technology. The
blowout. combination of optical sensing and wavelet analysis allows
to define quantities that can be associated to the health of the
Observations and numerical simulations show that, close to flame and give hints about the imminence of the flame
flame extinction, pressure oscillations are characterized by extinction.
high amplitude, low frequency fluctuations caused by a
sequence of flame blowout and re-ignition events. Hence 1. INTRODUCTION
monitoring such phenomena requires good quality, low
frequency data. It appears from early tests that optical sensing
is positively addressing this aspect. Increasingly restrictive emission standards compel power
The pressure sensors based on optical interferometry, briefly generation systems to operate within lean combustion limits
described in this paper, are expected to provide by design in order to reduce the formation of NOx. Under lean
more accurate measurements at low frequency and high conditions, the combustion process becomes more vulnerable
temperature with respect to traditional piezoelectric sensors. to small dynamic pressure perturbations caused by load
By acting on combustion parameters, the test rig, described changes, variations in air temperature or humidity. When the
in the first part of the paper, has allowed to generate imminent air-fuel ratio shifts towards a lean mixture the flame speed
blowout conditions in a controlled manner. gradually decreases from its value at stoichiometric
conditions. Once the flow velocity of the fuel mixture
The data collected using the optical sensors in the vicinity of exceeds the local flame speed of the reacting components, the
the flame extinction were analyzed with two wavelet-based flame becomes unstable and can be swept away by the flow
signal processing methods designed to represent flame health from the unburned reactants and blowout can take place. A
indicators. The first algorithm simply exploits the well- blowout event, i.e., the disappearance of the flame, caused by
an excessive leanness of the reacting mixture is generally
referred to as a lean blowout (LBO) and may occur due to a
flame blow-off or flame extinction. LBO represents the main

44
Meggitt proprietary and confidential. No unauthorised copying or disclosure.
FEEDBACK, COMMENTS & QUESTIONS ?

GO TO WWW.MENTI.COM AND USE CODE 7876 9797

THANK YOU

45
Disclaimer

Meggitt SA
Route de Moncor 4
1701 Fibourg
Switzerland

Information contained in this document may be subject to export control regulations of the United
Kingdom, European Union, United States or other national jurisdictions, including the US
International Traffic in Arms Regulations and/or Export Administration Regulations.
Each recipient of this document is responsible for ensuring that transfer or use of any information
contained herein complies with all relevant Export Control Regulations.
© Meggitt 2019. All rights reserved.

46
Meggitt proprietary and confidential. No unauthorised copying or disclosure.

You might also like