Patent Grant
7061609
The present invention relates to an optical wavelength analysis arrangement comprising:
An arrangement as defined in the outset, is known from WO 99/09370, in which such an arrangement is described for usage in Fiber Bragg Grating (FBG) based, structure monitoring applications. In such applications, physical parameters like strain, temperature, pressure and others, are measured by a fiber network, containing a plurality of optical FBG sensors distributed over the structure. FBG sensors are capable of precise and absolute measurement of physical parameters as mentioned above. A FBG sensor installed in a fiber-optic network, reflects light signals that travel through the optical fiber, with a wavelength λ that relates to the FBG periodicity Λ as given by equation [1]:
λ=2nΛ [1],
where λ is the wavelength of the light reflected by the FBG sensor,
and Λ is the periodicity of the FBG sensor, respectively.
Physical parameters that can be measured with FBGs, are related to the reflected wavelength due to the coupling of the physical parameters to the refractive index or the periodicity of the grating.
In structure monitoring applications, measurement of one or more specific physical parameters derived from a signal of an optical sensor in the fiber, is performed at a plurality of locations in the structure (e.g., a fuselage of an aircraft). To identify the origin of signals, each optical sensor generates a signal with a wavelength, specific for that sensor in that location. Thus, each wavelength corresponds to a location in the network. The signal wavelengths are well separated by intervals. The intervals are large enough to prevent overlap of sensor signals, when the response of a sensor changes due to change in a physical parameter, measured at the location of the sensor.
The light beam reflected by the sensors on the fiber network thus comprises a plurality of signals with different wavelengths has to be analysed by e.g., spectrometric means.
In many optical applications like FBG sensor networks, the wavelength of incident light is measured by a spectrometric arrangement with the purpose to determine a physical parameter related to the wavelength.
The method of spectrometry to determine the wavelength of light originated by the optical sensor, is well known. Light, gathered from a source (e.g., an optical sensor), is projected on a grating. Due to the wave characteristics of the light and the periodicity of the grating, the light is diffracted by the grating in one or more orders with their own direction as related to the wavelength(s) of the light, the angle of incidence, and the periodicity of the grating. By measurement of the angle of the diffraction direction(s) in the spectrometer, the wavelength of the light is determined. In spectrometers, as known in the art, the diffracted light is projected on a detector array (e.g., a linear or two-dimensional CCD system). In such an arrangement the position of the projected light on the detector is proportional to the wavelength of the light. The position of the projection is determined by fitting a mathematical model to the intensity data as measured by the detector's elements. The fitting procedure is needed here, since the spatial intensity distribution of the incident light beam that enters the spectrometer's slit is not uniform, because typically, the beam is focussed on the slit to collect as much optical power as possible. The spatial intensity profile of the projected light beam is usually described by a peak-shaped curve. To determine the centroid of the signal, a model describing the shape of the peak is fitted to the measured signal. Due to the non-linearity of the spatial intensity distribution, a useful fit can be accomplished only if a plurality of data points within the profile are measured. To obtain a reasonable accuracy with a resolution higher than the size of a detector element (a “pixel”), the spot projected on the detector array must cover a sufficiently large number of elements in the array, all of which must be sampled in the fitting procedure. Typically, a resolution of approximately 1/10 pixel is possible on a range of 10 pixels.
For a measuring range, which is, for example, 50 times larger, the range on which the light beam is projected must be extended to 50 detector elements. Taking into account a cross-talk separation between signals of about 10 detector elements, in that case a range of more than 60 detector elements on the detector is needed for one signal. Usually, in a spectrometric application, many wavelengths are to be measured simultaneously, which requires that the spectrometer provides a sufficient large detector array. When, for example, 32 signals must be measured simultaneously, the detector array needs approximately 2000 elements.
In WO99/09370 a number of fiber channels each comprising a plurality of signals with different wavelengths are monitored by spectrometric means using a two-dimensional detector array on which the spectra of each fiber channel are projected on elongated regions of the array.
As known to those skilled in the art, in a such FBG sensor network, measurement of wavelengths of optical signals must be carried out at a rate, sufficiently efficient with respect to the number of FBG sensors in the network and the requirements for the type of application e.g., monitoring a structure by the FBG sensor network. By consequence of the method of fitting the peak shapes, a disadvantage of the arrangement of WO 99/09370 to obtain sub-element accuracy is, the rate at which data can be collected and calculated, especially, when high sample rates are required. From the paper of S. Chen et al., “Multiplexing of large-scale FBG arrays using a two-dimensional spectrometer”, SPIE vol. 3330 (1999), p. 245–252, it can be found that with a projection of a signal from a light beam on a field of 7×7 detector elements a resolution of approximately 1/56 of an element can be accomplished.
Due to the large amount of detector elements in an array, in such systems the overall sampling rate is in the order of only 25–100 Hz.
It is an object of the present invention to provide an arrangement and a method to improve the measurement of a signal and the processing of the signal by processing means, in order to determine the wavelength of the signal in a simple, efficient, and fast manner, with high accuracy.
This object is obtained in an optical wavelength analysis arrangement as defined in the outset characterised in that:
With such an arrangement, the spatial intensity profile of a signal is already well known before detection. This implies that to determine the wavelength of the signal as measured by the detection means, the detection means needs a subset of at least one, at most two of detector elements to measure the signal. In case of a uniform distribution and a known constant optical power level of the light beam, only a single detector element, comprising part of the image of a signal, is sufficient. If however, the optical power level is unknown due to e.g. fluctuations, a subset of two adjacent detector elements, that comprise the complete image of a signal, will suffice to determine the wavelength. Due to the small number of detector elements to be sampled, the calculation of the wavelength of a signal in the arrangement of the present invention is strongly simplified. Accordingly, the computation time involved to measure the wavelength of a signal is strongly reduced.
Consequently, the measurement of wavelengths of a plurality of signals in the arrangement of the present invention requires less time than is known from the prior art. Also, since less detection elements are necessary to measure a signal, the detection means in this arrangement can have a smaller number of detection elements than is known from the prior art. Therefore, the total readout time for the detector array is also reduced.
The present invention also relates to a method of optical wavelength analysing comprising the steps of:
The present invention also relates to a computer arrangement comprising processor means and arranged to receive detector output signals from detector means comprising one or more subsets of detector elements, each subset having at least a first detector element for receiving a first signal portion with a first signal portion level of a signal derived from a beam with a spatial uniform intensity distribution, the arrangement being programmed to determine, for each one of the subsets the wavelength of the signal received in dependence on the first signal portion level and a calibration value.
Moreover, the invention relates to a computer program product to be loaded by a computer arrangement comprising processor means and arranged to receive detector output signals from detector means comprising one or more subsets of detector elements, each subset having at least a first detector element for receiving a first signal portion with a first signal portion level of a signal derived from a beam with a spatial uniform intensity distribution, the computer program product, after being loaded by the computer arrangement, providing the computer arrangement with the capability to determine, for each one of the subsets the wavelength of the signal received in dependence on the first signal portion level and a calibration value.
Furthermore, the present invention relates to an optical wavelength analyser comprising:
Finally, the invention relates to a data carrier provided with a computer program product as defined above.
Below, the invention will be explained with reference to some drawings, which are intended for illustration purposes only and not to limit the scope of protection as defined in the accompanying claims.
a and 4b show an illustration of the method to calculate the position and wavelength of a signal, projected on the detection means, and their respective change, in an arrangement of the present invention;
a shows an arrangement of the present invention, in which an alternative set-up of the diffractive means is applied;
b shows an arrangement of the present invention, in which the diffractive means is combined into a single optical element;
c shows an arrangement of the present invention, in which prismatic means are applied;
d shows an arrangement of the present invention, in which a combination of dispersive means are utilised.
The present invention relates to an arrangement and method of simple and fast measurement of a wavelength of a signal from an optical sensor, which may be an FBG sensor, or any other type of optical sensor as known in the art.
In the present invention the signal to be measured is adapted in such a way that the signal has a spatial uniform intensity distribution. The intensity uniformity of the signal simplifies measurement of the wavelength of such a signal. Also, computations relating to the wavelength of the measured signal are simplified by the spatial uniformity of the signal.
In
By means of the control electronics, the detector 8 is connected to a computer 20, which records the spectrum measured by the detector, and calculates the wavelengths of signals in the spectrum.
The memory units shown comprise RAM 22, (E)EPROM 23, ROM 24 and hard disk 18. However, it should be understood that there may be provided more and/or other memory units known to persons skilled in the art. Moreover, one or more of them may be physically located remote from the processor means 21, if required. The processor means 21 are shown as one box, however, they may comprise several processing units functioning in parallel or controlled by one main processor, that may be located remote from one another, as is known to persons skilled in the art. Moreover, other input/output devices than those shown (i.e., 27, 28, 29, 30) may be provided.
The shape of the image 31 is conformal with the rectangular slit opening 4. By design of the projection system, the width B of the image 31 is smaller than the width W of one of the detector elements 32, 33. Thus, the image only partially covers both detector elements in the one (e.g., horizontal) direction as indicated by arrow X. The covered length on the elements 32, 33 is denoted as x32, x33 respectively. In the other perpendicular vertical direction as indicated by arrow Y, the height H of the image is also smaller than the detector element height h. However, the height H of the image may be larger than the detector element height h, thus covering a detector element completely in this direction.
By a calibration procedure as known in the art, the position of the detector elements is translated to a wavelength scale. By means of this calibration procedure the wavelength of a signal can be determined from the position of the signal's image on the detector elements 32, 33. In
a and 4b show an illustration of the method to calculate the position and wavelength of a signal, projected on detector 8, and their respective change, in an arrangement of the present invention.
In
If a spatial uniform intensity of the image is assumed, the optical power 43, 44 measured on a detector element 32, 33 will be proportional to the area H*x32, H*x33 covered on the element, which is directly proportional to the coverage in the direction X, since the coverage in the direction Y is constant.
The difference between the optical power 43, 44 is determined. In order to correct for fluctuations of the optical power of the beam, the difference is normalised by dividing by the sum of optical power 43 and 44.
In this case, the optical power 43, 44 measured is equal on each element, since x32 equals x33. The difference in optical power is zero, which indicates that the image is projected symmetrically on the detector elements, with the centre of the image at the interface between the two elements. Thus, the wavelength of the signal equals λ1.
b shows the projection of image 31 on detector 8 for a signal with a wavelength that differs from λ1 by an amount δλ.
Because the position of the image 31 on the detector 8 is directly proportional (as calibrated) to the wavelength of the projected image, a wavelength difference δλ is directly proportional to a shift δx of the image's centreline. The centreline of the image is projected shifted over a distance δx with respect to the interface between the two elements 32, 33. Since the illuminated width x′32 is not equal to x′33, the normalised difference of optical power 45 and 46 measured on the respective detector elements 32, 33 and divided by the sum of the optical powers 45 and 46, is unequal to zero.
In this way, the wavelength of a signal can be measured advantageously, by determining the normalised difference of the optical power received by two adjacent detector elements. The measurement requires a limited number of detector elements and the calculation requires few and simple computations by processor means 21 with relatively short computation times.
In
In step 51, the processor means 21 enter the procedure by a request to select two adjacent detector elements Ej, Ej+1 for measurement.
In steps 52 and 53, the computer addresses the control electronics 11 to read optical power I(Ej), I(Ej+1) of elements Ej, Ej+1, and to transfer the data in a readable format to the computer.
In step 54, it is checked if a signal is present on the selected elements.
If light is measured, then results are calculated. Otherwise, the following steps 55–58 are skipped.
Step 55 calculates the optical power difference between the elements Ej, Ej+1, normalised by the sum of the optical power I(Ej) and I(Ej+1)
Step 56 calculates the corresponding shift δx on the detector 8.
In step 57 the wavelength shift δλ with respect to the wavelength λ1 corresponding to the centreline between detector elements Ej, Ej+1 is calculated from shift δx by using wavelength calibration data for the detector.
Step 58 calculates the measured wavelength by adding the wavelength shift δλ to the wavelength λ1, corresponding to the position of the detector elements centreline. This wavelength λ1 is derived from the wavelength calibration data for the detector 8.
In step 59 the procedure ends. The processor means 21 return to the procedure where the request for step 51 originated from, with the value of the measured wavelength, or if no optical power was measured on the detector elements, with a predetermined value e.g., zero to signal this state.
Change of wavelength as a function of time can be measured by repeating the procedure as shown in
A monochromatic light source of which the wavelength changes directly proportional with time, is projected as a parallel beam on the slit opening 4 of the spectrometer. The intensity distribution of the light beam at the slit opening 4 has a spatial uniform intensity distribution. In
The sub-detector element accuracy in such arrangement depends on the signal to noise ratio of the detector elements. In this arrangement, using commercially available detectors an accuracy of 1/500 can be obtained, at an overall sampling rate in the order of a several kHz.
It should be understood that measuring wavelength change using a light beam with a spatial uniform intensity distribution can even be done by measurement of the diffracted signal on only a single detector element, without the use (or need) of a second detector element. As illustrated by
If the light beam 3 has an optical power level which is time invariant, i.e. the intensity of the beam does not change over time, no normalisation step is necessary and it will be sufficient to measure the optical power level of a diffracted signal impinging on a single detector element 32. In that case, the shift of the wavelength can be calculated from the difference of the measured optical power level on the single detector element 32 and the optical power level of a reference diffracted signal, which has been calibrated with respect to its location on the detector element 32. Such calibration procedures to obtain the latter optical power level are known to those versed in the art. In fact, in an embodiment for this type of measurement with a single detector element, the second element 33 in the detector element pair 32, 33 may even be omitted.
Therefore, depending on the application, it may be possible to have only one single detector element, or a few single detector elements set-up separately at various locations in the arrangement to measure wavelength(s).
Also, if the optical power level of the beam is monitored constantly, for example, by a separate detector, measurement on only a single detector element is sufficient to determine the wavelength of a diffracted signal. Here, the shift of the wavelength can be calculated from the quotient of the measured optical power level on the single detector element 32 and the optical power level of a reference signal, used for calibration. In such an embodiment, the optical power level of the reference signal is to be corrected by dividing by the actual optical power level of the diffracted signal.
In
The detector elements 33d, 34, 35, 35a, 35b, 35c, 35d, 36, 37, 37a, of the detector 8 are arranged in a row. On the detector 8 a signal's image 31 of the slit opening 4 is projected by the optics of the spectrometer 2 as shown in
The shape of the image 31 is conformal with the rectangular slit opening 4. By design of the projection system, the height H of the image 31 is smaller than the height h of the detector elements 33d, 34, 35, 35a, 35b, 35c, 35d, 36, 37, 37a. Thus, in the one (e.g., horizontal) direction as indicated by arrow X, the image 31 covers a plurality of N detector elements 35, 35a, 35b, 35c, 35d, 36 of which the two exterior detector elements 35, 36 are only partially covered.
However, the height H of the image may be larger than the detector element height h, thus covering a detector element completely in this direction.
In this second preferred embodiment the center of the image 31 is determined by weighing of the signals received by the individual detector elements. By weighing, the “center of gravity” of the image intensity distribution can be calculated. The position of the “center of gravity” will depend on the actual intensity distribution. In case of a spatially uniform intensity distribution the result of weighing will be exactly the center position of the image 31.
Assuming the image 31 covers N detector elements Ej . . . Ek(35, 35a, 35b, 35c, 35d, 36) on the detector, each element Ei within that range Ej . . . Ek measuring an intensity I(Ei), the (horizontal) center position C of image 31 can be calculated by:
In case of a uniform intensity distribution of the light beam 3, the intensity I measured on the detector elements 35a, 35b, 35c, 35d (i.e., N−2 elements Ej+1 . . . Ek−1) will be identical for all detector elements Ej+1 . . . Ek−1. Only the intensity on the exterior two elements Ej, Ek (35, 36) will depend on the actual coverage of respective element Ej, and Ek by the image 31. The center position C can then be calculated by:
If the light beam 3 has an optical power level which is time invariant, the signals of the detector elements 35a, 35b, 35c, 35d will be identical for all these elements 35a, 35b, 35c, 35d and also constant over time. In that case, the term (N−2)*I is equal to a constant Q. The center position C can then be calculated by:
Using a calculation scheme according to one of the latter two equations, the center position C of the signal on the detector 8 can be obtained relatively simply.
In
The spatially uniform signal is projected as an image 31 on a plurality of N detector elements 35, 35a, 35b, 35c, 35d, 36 of the detector 8. The two exterior detector elements 35, 36 are only partially covered. The signal of the imaginary detector elements A and B is the sum of the intensities measured on the respective detector elements belonging to imaginary detector element A, and imaginary detector element B, respectively.
The calculation scheme as explained in
The arrangement as shown in
In
In the arrangement as shown in
A plurality of spectra is projected adjacent to each other on the array. When properly calibrated, all spectra can be analysed simultaneously using the corresponding procedure for that particular embodiment.
Finally,
a shows an arrangement of the present invention in which the lenses 6, 9 of the diffractive means are combined in a single lens 81.
As is known to those skilled in the art, the lens 81 may also be combined with dispersive means 7 into a single optical element as a concave grating 82 with the same functionality as defined by the diffractive means comprising separate projection means 6, 9 and dispersive means 7. The latter arrangement is shown in
c shows an arrangement of the present invention, in which prismatic means is applied. Instead of a grating 7 as dispersive element, a prism 83 is used to redirect each signal in a direction depending on the wavelength λ1, λ2, λ3 of the respective signal.
An unproved sensitivity can be achieved by replacing the grating 7 by a combination of two or more dispersive elements. An example of such an arrangement is shown in
d shows an arrangement of the present invention, in which a combination of dispersive means are utilised. A light beam encompassing a plurality of signals with wavelength λ1, λ2, λ3, respectively, entering through the entrance slit 4 is projected by lens 6 on the grating 7. Here the light beam is diffracted in directions depending on the wavelengthλ1, λ2, λ3 of the respective signals. The diffracted beams are projected on a second dispersive element 83, e.g., a prism. The prism redirects each diffracted beam in a direction depending on its wavelength. By dispersing the incoming light beam by more than one dispersive element, a larger separation between individual signals can be achieved. Typically, by this arrangement a higher sensitivity of the instrument can be obtained.
It will be appreciated that other combinations of dispersive elements are possible with similar improved sensitivity.
Moreover, as known to persons skilled in the art, lenses 6, 9 may also be replaced by reflective optical elements, like concave mirrors (not shown), without changing the functionality of the optical wavelength analysis arrangement. It is also noted that in the present invention a holographic element may provide the same functionality as a prism or grating.
| Number | Date | Country | Kind |
|---|---|---|---|
| PCT/NL00/00281 | May 2000 | WO | international |
| Filing Document | Filing Date | Country | Kind | 371c Date |
|---|---|---|---|---|
| PCT/NL01/00327 | 5/1/2001 | WO | 00 | 2/27/2003 |
| Publishing Document | Publishing Date | Country | Kind |
|---|---|---|---|
| WO01/84097 | 11/8/2001 | WO | A |
| Number | Name | Date | Kind |
|---|---|---|---|
| 3578837 | Brooks | May 1971 | A |
| 4636074 | Levy et al. | Jan 1987 | A |
| 4669880 | Nelson et al. | Jun 1987 | A |
| 4687329 | Schultz | Aug 1987 | A |
| 4932779 | Keane | Jun 1990 | A |
| 5570226 | Ota | Oct 1996 | A |
| 5796479 | Derickson et al. | Aug 1998 | A |
| 5867264 | Hinnrichs | Feb 1999 | A |
| 5933235 | Sampei et al. | Aug 1999 | A |
| 6144025 | Tei et al. | Nov 2000 | A |
| 6154551 | Frenkel | Nov 2000 | A |
| 6424416 | Gross et al. | Jul 2002 | B1 |
| 20020018203 | Battle et al. | Feb 2002 | A1 |
| 20030007147 | Johnson | Jan 2003 | A1 |
| Number | Date | Country |
|---|---|---|
| 0 431 368 | Jun 1991 | EP |
| Number | Date | Country | |
|---|---|---|---|
| 20030156287 A1 | Aug 2003 | US |
