The present invention relates to the field of temperature measurement devices and techniques based on optical technology.
In many high temperature processes, it is important to have accurate knowledge of temperature, for example to maximize efficiency. This is true for processes such as materials processing in the metal and glass industries, and is equally true in the measurement of turbine inlet temperatures in jet engines and in stationary gas turbine power plants. However, the maximum temperatures in these processes can reach as high as 1,700 to 2,300° C. Ordinary thermocouples cannot meet the requirements for stable and accurate operation in such high-temperature applications.
It has been shown that temperature sensors based on optical technology may be employed to achieve certain benefits not possessed by conventional thermocouples. An optical thermocouple includes a silica glass fiber, one end of which terminates in a so-called fiber Bragg grating. In one known configuration, the fiber Bragg grating is composed of alternating layers of silicon nitride and silicon-rich silicon nitride. The fiber Bragg grating responds to changes in temperature by corresponding changes in the spectral content of reflected light, specifically by a change in the optical wavelength at which peak reflectivity occurs. This response can be exploited for use in a an optical temperature measurement system.
A measurement system can be built in which broadband optical energy is transmitted along an optical fiber toward one end at which a fiber Bragg grating is formed. The fiber Bragg grating is disposed in an environment whose temperature is to be measured. A broadband optical spectrum analyzer is also coupled to the fiber to receive optical energy reflected from the fiber Bragg grating. By analyzing the output from the optical spectrum analyzer, it is possible to determine the amount of wavelength shift of the peak of the reflectivity characteristic, and then to convert this peak shift into a temperature value.
Optical-based temperature measurement systems such as those described above have several advantages, including the ability to withstand high temperatures and immunity from electrical noise due to their all-dielectric construction. With respect to temperature, however, silica-based fiber and fiber Bragg gratings are generally limited to use at temperatures less than about 1,100° C. It would be desirable to have an optical-based measurement system that permits the measurement of much higher temperatures such as those encountered in the industrial and turbine applications described above.
In accordance with the present invention, a fiber optic temperature sensor and system are disclosed that achieve the benefits of optical temperature sensing at much higher temperatures than have heretofore been possible, thus enabling the accurate measuring of temperature in a variety of high-temperature applications.
The disclosed sensor and system employ optical fiber and fiber Bragg gratings using non-silica materials that can withstand temperature ranges well above the silica-imposed limit of 1,100° C. In one embodiment, the use of sapphire optical fiber enables use of the sensor at temperatures approaching 1,800° C., while an alternative sensor employing yttria-stabilized zirconia is capable of use at temperatures in excess of 2,350° C. These high-temperature fibers are used in conjunction with fiber Bragg gratings made of materials that can also withstand such temperatures. In one case, the grating employs alternating layers of yttria stabilized zirconia, with the percentage of yttria varying in the alternating layers to achieve the desired difference of refractive index. Alternatively, alternating layers of alumina and zirconia can be employed.
The dynamic range of this device is extremely wide, and can be as low as liquid nitrogen temperatures. Unlike black body or pyrometer type devices, there is no dependence upon limiting low photon flux at low temperatures.
Other aspects, features, and advantages of the present invention will be apparent from the Detailed Description that follows.
The invention will be more fully understood by reference to the following Detailed Description of the invention in conjunction with the Drawing, of which:
Extending from the probe 10 is an optical fiber 14. An optical coupler 16 joins the probe fiber 14 to two additional fibers 18, 20. The fiber 18 carries light from a broadband light source 22 to the probe 10 via the coupler 16, and the fiber 20 carries reflected light from the probe 10 to an optical spectrum analyzer (OSA) 24, which may be for example a charge-coupled device (CCD) array. The electrical outputs of the OSA 24 are coupled to a digital processor 26.
The broadband light source 22 can be implemented by a LED or other suitable broadband source. The range of optical wavelengths from the source 22 encompasses a range of reflectance frequencies of a fiber Bragg grating employed within the probe 10, which is described in more detail below.
Formed at the distal end of the tip optical fiber 34 is a 1/4-wavelength fiber Bragg grating 36, which is used as a wavelength-selective reflector. The grating can be made using different types of ceramic systems. In one scheme, the grating 36 is made using yttria-stabilized zirconia, with alternating layers having different concentrations of yttria to achieve the small difference of refractive index that is required for a narrow reflecting structure. The percentage of yttria doping can be from, typically, 5% to 40%. This structure retains its chemical stability when subjected to temperatures as high as 2400° C. Also, the thermal expansion properties of such layers are well matched, minimizing destructive thermal-induced mechanical strain. This is extremely important.
As an alternative, alternating layers of alumina and zirconia can be employed. It may be desirable to add yttria to the zirconia layers to improve the refractive index matching between the two layers. A layer having 20% yttrium has a refractive index of 1.9, which is close to the refractive index of 1.76 of alumina.
The grating 36 can be formed using a process in which a layer is deposited at the end of the fiber 18 while the reflectance at a particular wavelength is monitored. The reflectance will vary between a maximum and a minimum as each layer is deposited. When a peak or valley of the reflectance is reached during the deposition of one layer, the deposition is stopped and the deposition of the next layer is begun. This process is repeated until the desired number of layers have been deposited.
Additionally, it is possible to form the grating 36 using other combinations of repeating sequences of materials of different refractive indices that will provide high reflectivity over a narrow wavelength region.
As shown in
where X represents the raw spectrum data vector and Y represents the normalized data vector. To facilitate subsequent processing, only the main portion of the spectrum containing the peak is utilized. This vector can be represented as
A=[ai, ai+1, . . . , ai+N)
In step 40, measured spectrum data is obtained at an unknown temperature being measured, and this data is normalized using the same normalization function described above. To facilitate the analysis steps to follow, the normalized measured spectrum data is saved as an array of sub-vectors of the overall vector output of the OSA 24. These can be represented as follows:
where m represents an assumed maximum pixel shift of the measured characteristic spectrum, which corresponds to the highest temperature to be read by the probe 36.
At step 42, the “whole” part h of the spectrum peak shift (in integer number of pixels or CCD elements) is determined using a least squares algorithm on the reference and measured spectrums. This involves computing a measure of the difference between the normalized reference spectrum vector and each of the normalized measured spectrum vectors, and then determining which of the computed difference values is the smallest. This algorithm can be expressed as follows:
1. For k=0 to k=m, calculate:
2. Find the minimum dk, which is denoted dh. The value h is the whole part of the peak shift.
In step 44, the fractional part t of the peak shift is determined. This preferably uses an “extreme value” calculation, which is described with reference to
Finally, in step 46, the spectral shift is calculated as
Sshift=Wpixel*Spixel,
where
Spixel=h+t
and Wpixel is equal to the per-pixel spectral width of the OSA 24. If linearity is assumed, the value Wpixel can be calculated by dividing the total spectral width of the OSA 24 by the number of pixels (CCD elements) in the array.
The value Sshift can then be translated to a temperature using a pre-computed conversion factor obtained during a calibration process. This factor has units of degrees/(nm of wavelength), and thus yields a temperature in degrees when multiplied by Sshift. In one type of calibration process, the steps of
As an example of the use of the conversion factor, if it is assumed that the conversion factor is 15° C. per nm, then a value of Sshift=37.6 yields a measured temperature T of
It will be apparent to those skilled in the art that modifications to and variations of the disclosed methods and apparatus are possible without departing from the inventive concepts disclosed herein, and therefore the invention should not be viewed as limited except to the full scope and spirit of the appended claims.
This application claims priority under 35 U.S.C. §119(e) of U.S. Provisional Patent Application No. 60/428,099 filed Nov. 21, 2002, the disclosure of which is hereby incorporated by reference herein.
This invention was made with Government support under contract no. N00014-97-G011 awarded by the Department of the Navy, and from the Air Force Office of Scientific Research, under contract number ______. The Government has certain rights in the invention.
| Filing Document | Filing Date | Country | Kind | 371c Date |
|---|---|---|---|---|
| PCT/US03/37303 | 11/21/2003 | WO | 11/20/2005 |
| Number | Date | Country | |
|---|---|---|---|
| 60428099 | Nov 2002 | US |