Multi functional sensor system for gas turbine combustion monitoring and control

Patent metadata

Field	Value
Publication number	US-9568378-B2
Application number	US-201314109992-A
Country	US
Kind code	B2
Filing date	Dec 18, 2013
Priority date	Dec 18, 2013
Publication date	Feb 14, 2017
Grant date	Feb 14, 2017

Abstract

Thermoacoustic sensors, such as dynamic pressure sensors, in the combustor measure vibratory responses of the combustor. An integrated monitoring and control system controller correlates sensor readings with combustion thermoacoustic properties in order to identify combustion anomalies by wavelet or Fourier analysis techniques; determine bulk temperature characteristics within the combustor with dominant mode frequency analysis techniques; and optionally determines absolute active path temperatures within the combustor with acoustic pyrometry transmission and time-of-flight analysis techniques. In some embodiments all of the monitoring functions are performed with a commonly shared array of thermoacoustic sensors that function as both combustion dynamics thermoacoustic vibration/wave receivers and acoustic transmitters. The monitored combustion properties are used for controlling gas turbine combustion.

First claim

What is claimed is:

1. A method for detecting and classifying combustion anomalies in a combustor of a gas turbine engine, comprising:
placing in a gas turbine combustor at least two thermoacoustic sensors capable of generating respective sensor output signals indicative of combustion thermoacoustic oscillations in the combustor and coupling the sensors to a controller that is capable of correlating sensor output signals with combustion conditions;
generating dynamic sensor output signals with the thermoacoustic sensors;
performing a wavelet or Fourier spectral analysis of the dynamic sensor output signals with the controller, to determine whether a combustion anomaly has occurred;
monitoring temperature within the combustor with the thermoacoustic sensors by performing one or both of:
bulk temperature frequency analysis of the dynamic sensor output signals; or
active path temperature monitoring of the dynamic sensor output signals by acoustic pyrometry; and
classifying the anomaly with the controller based on the wavelet or the Fourier spectral analysis and the monitored temperature.

2. The method of claim 1 , the wavelet analysis of each respective dynamic sensor output signal is performed in the controller by:
dividing the dynamic sensor output signal into time segments to derive a plurality of data points for each of the time segments;
transforming the sampled dynamic signal by performing a wavelet transform to calculate wavelet coefficients for the data points within the processed time segment, targeting at least one region of interest within the wavelet transformed segment and normalizing amplitude of the wavelet coefficients within each targeted region by a baseline signal; and
determining whether any combustion anomalies have occurred during each of the time segments using the normalized amplitudes of the wavelet coefficients within each targeted region by comparing the respective normalized amplitudes of the wavelet coefficients within each target region to a predetermined threshold amplitude.

3. The method of claim 1 , the bulk temperature monitoring is performed in the controller by:
identifying one or more acoustic frequencies in the dynamic sensor output signal of a thermoacoustic sensor at a first location in the combustor;
determining, for each of the one or more acoustic frequencies, a bulk first temperature value, at the first sensor location that is directly proportional to each one of the one or more acoustic frequencies and a calculated constant value corresponding to each of the one or more acoustic frequencies;
determining a second temperature in the engine with an alternative temperature monitoring method;
based on the second temperature, determining a calculated temperature value at the first location;
comparing the first temperature value determined for each of the one or more frequencies at the first location to the calculated temperature value and, for each of the one or more frequencies, changing the calculated constant values to recalculated constant values based on the comparison; and
determining subsequent bulk first temperature values at the first location based on further identified acoustic frequencies.

4. The method of claim 3 , the determining subsequent bulk first temperature values at the first location also based on further recalculated constant values.

5. The method of claim 3 , the second temperature determined at a location within an exhaust system of the gas turbine.

6. The method of claim 1 , the active path temperature monitoring within the combustor of each respective dynamic sensor output signal is performed in the controller by:
transmitting acoustic signals from at least one thermoacoustic transmitter that is located in the combustor;
receiving the acoustic signals from the acoustic transmitter by at least one thermoacoustic sensors located at a predetermined location within the combustor in a distinct line-of-sound path from the transmitter;
determining a time-of-flight for the signals traveling along each of the line of sound paths; and
processing the time-of-flight for the signals traveling along the line of sound paths to determine the path temperature.

7. The method of claim 1 , the wavelet or Fourier spectral analysis anomaly detection and temperature monitoring performed with common thermoacoustic sensors.

8. The method of claim 1 , the classifying performed with a plurality of monitored temperatures monitored at different locations within the combustor.

Assignees

• Desilva Upul P
• Siemens Energy Inc

Inventors

• Desilva Upul P

Classifications

• F23N2225/08

  temperature · CPC title

• F23N2223/00

  Signal processing; Details thereof · CPC title

• F23N2241/20

  Gas turbines · CPC title

• G01K11/22 Primary

  using measurement of acoustic effects · CPC title

• F23N5/10

  using thermocouples · CPC title