IP Library Granted Patent US 8,825,214
Granted Patent B2
US 8,825,214 · App. 13/300,110 · Granted Sep 2, 2014

Method of analyzing and controlling a combustion process in a gas turbine and apparatus for performing the method

View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 8,825,214
App. No.
13/300,110
Granted
Sep 2, 2014
Kind
B2
Abstract

A method is provided for the measurement of parameters of a gas present in a gas turbine combustion chamber. The method includes tuning a laser to a range containing the absorption lines of species to be analyzed in the gas, and directing the laser light through the combustion chamber and detecting laser light reflected off boundary walls of the combustion chamber. In order to analyze the absorption spectrum measured at high temperatures and pressures, a signature recognition algorithm is applied to the spectrum. The measured absorption spectrum is cross-correlated with a calibration absorption model spectrum for the absorption lines at several temperatures, pressures, and concentrations generated prior to the measurement. Values for pressure, temperature, and concentrations of selected species in the gas are determined simultaneously allowing direct control of the combustion chamber process. An apparatus for carrying out the method is also provided.

Claims (21)

1. A method for analyzing and controlling a combustion process in a combustion chamber of a gas turbine by measurement of an absorption spectrum, the method comprising:

providing a molecule absorption spectrum including a range of absorption lines for selected molecules of gas involved in the combustion process in the combustion chamber by using a databank containing standards for molecule absorption spectra including the selected molecules of the gas, and generating a calibration model absorption spectrum containing absorption lines for the selected molecules of the gas at a plurality of temperatures, pressures, and concentrations of the selected molecules of the gas using an algorithm;

providing one or more tunable laser sources, which are tunable over the spectrum containing absorption lines of a selected species of the gas in the combustion chamber;

tuning the one or more tunable laser sources to a wavelength range that contains absorption wavelengths of the selected species in the gas;

directing laser light from the one or more tunable laser sources through the gas in the combustion chamber via transmitting optics;

detecting the laser light reflected off a boundary wall of the combustion chamber via receiving optics and a detection device;

generating a measured absorption spectrum for the selected species of the gas in the chamber, based on the detected laser light;

analyzing the measured absorption spectrum using a signature recognition algorithm by applying a cross-correlation between the calibration model absorption spectrum and the measured absorption spectrum;

deriving, simultaneously, from the cross-correlation, values for a gas temperature of the selected molecules of the gas in the chamber by using the same algorithm as was used for the generation of the calibration model absorption spectrum;

generating control values for the combustion process based on the derived values for the gas temperature, and directing the control values to a combustion chamber operation control unit; and

controlling the combustion process using the generated control values.

2. The method according to claim 1 , further comprising: using for the signature recognition algorithm a multivariate algorithm.

3. The method according to claim 2 , wherein the multivariate algorithm includes a partial least squares regression algorithm.

4. The method according to claim 1 , further comprising: using for the signature recognition algorithm one of a wavelet algorithm and a fuzzy logic algorithm.

5. The method according to claim 1 , further comprising: using for the signature recognition algorithm a neural network algorithm.

6. The method according to claim 1 , further comprising: using a distributed feedback (DFB) laser as the one or more tunable laser sources.

7. The method according to claim 6 , further comprising: modulating the DFB laser at a frequency in the range from 0.5-10 kHz.

8. The method according to claim 6 , further comprising: modulating the DFB laser at a frequency in the kHz range.

9. The method according to claim 6 , further comprising: tuning the DFB laser back and forth over a given wavelength range by modulating a laser injection current according to a triangular function to tune the DFB laser over the given wavelength range.

10. The method according to claim 1 , further comprising: tuning each of the tunable lasers to a respectively different wavelength range; modulating each laser at a different frequency, directing their emitted light through transmitting optics through the combustion chamber; detecting the laser light reflected off a combustion chamber wall by heterodyne detection; and analyzing the measured absorption spectrum using the signature recognition algorithm.

11. The method according to claim 1 , further comprising: deriving, simultaneously, from the cross-correlation, values for the gas pressure, or concentrations of the selected species in the gas, by using the same algorithm as was used for the generation of the calibration model absorption spectrum.

Assignments (3)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Feb 11, 2017
From: GENERAL ELECTRIC TECHNOLOGY GMBH
To: ANSALDO ENERGIA SWITZERLAND AG
Reel/Frame 041686/0884 →
CHANGE OF NAME Recorded Mar 22, 2016
From: ALSTOM TECHNOLOGY LTD
To: GENERAL ELECTRIC TECHNOLOGY GMBH
Reel/Frame 038216/0193 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jan 19, 2012
From: HAFFNER, KEN YVES
To: ALSTOM TECHNOLOGY LTD
Reel/Frame 027559/0786 →