High spatial resolution distributed temperature sensing method and device based on deconvolution algorithm

Abstract

The present invention provides a high spatial resolution distributed temperature sensing method and device based on a deconvolution algorithm, comprising dividing a temperature sensing optical fiber into a plurality of optical fiber segments, collecting the temperature traces of each optical fiber segment on the temperature sensing optical fiber in real time; for each optical fiber segment, taking into account the pulse broadening that occurs after the input optical signal is transmitted to each optical fiber segment, collecting the actual optical pulse data of the optical fiber segment; taking into account the power loss in the optical transmission process, for each optical fiber segment, determining the inverse of the optimal signal-to-noise ratio corresponding to the optical fiber segment; performing edge fuzzification processing on the temperature traces demodulated corresponding to each optical fiber segment to offset the ringing effect of the demodulated temperature traces; for each optical fiber segment, using a deconvolution algorithm, according to the actual optical pulse data corresponding to the optical fiber segment, the inverse of the optimal signal-to-noise ratio and the temperature trace after edge fuzzification processing, determining the actual temperature trace of the optical fiber segment. The present invention can improve the spatial resolution while ensuring the signal-to-noise ratio.

Description

Technical Field

[0001] The invention belongs to the field of distributed temperature sensing, and in particular relates to a high spatial resolution distributed temperature sensing method based on a deconvolution algorithm. Background Art

[0002] Existing distributed temperature sensors mainly include sensors based on Rayleigh scattered light, Brillouin scattered light, reflected Raman light based on optical frequency domain, and reflected Raman light based on optical time domain. For distributed temperature sensors based on Rayleigh scattered light, the light intensity of its Rayleigh scattered light is about one thousandth of the incident light, while the light intensity of Brillouin scattered light and Raman light is much smaller than this ratio. Therefore, the signal-to-noise ratio of the temperature signal based on Rayleigh scattered light is higher than that of the temperature signal based on the other two types of light. Therefore, the distributed temperature sensor based on Rayleigh scattered light has the characteristics of better dynamic range, temperature measurement time and system stability. However, the disadvantage is that the temperature sensitivity of Rayleigh scattered light is low and the temperature accuracy is insufficient. In addition, the current method of measuring temperature based on Rayleigh scattered light is mainly through frequency domain demodulation, which is relatively expensive. Generally, this technology is not used to measure temperature in ordinary occasions. Therefore, sensors based on Rayleigh scattered light are currently used for power failure monitoring, vibration measurement, etc.

[0003] For distributed temperature sensing based on Brillouin scattered light, since its light intensity and temperature sensitivity are between Rayleigh scattered light and Raman light, both the dynamic range and temperature accuracy are moderate. In addition, it has the advantage of synchronous measurement of stress and temperature. Therefore, there are mature distributed temperature sensing products based on Brillouin scattered light on the market, which is also a research hotspot. However, in practical applications, temperature measurement is easily disturbed by stress caused by the external environment, introducing random errors, and the decoupling of temperature and strain requires additional engineering technology support. Brillouin-based measurement technology is limited in many occasions due to the problem of cross-sensitivity between temperature and strain.

[0004] For distributed temperature sensing based on optical frequency domain reflectometry Raman light, since the spatial resolution of the frequency domain reflectometry sensing system does not depend on the pulse width, optical pulses with longer pulse widths can be used to ensure the signal-to-noise ratio while achieving a spatial resolution of the order of centimeters. However, optical frequency domain reflectometry requires the use of linear tunable lasers, and the system and temperature demodulation algorithm used are relatively expensive and complex, making them unsuitable for temperature sensing in general situations.

[0005] For distributed temperature sensing of Raman light based on optical time domain reflectometry. Since anti-Stokes Raman scattered light has the highest sensitivity to temperature among the three types of scattered light, the theoretical temperature accuracy of the temperature measurement system based on Raman light can be very high, and it is insensitive to external interference such as strain and vibration. The system cost is relatively low, the system is simple, and the stability is good. Therefore, it has been widely used in the field of distributed temperature sensing. However, both spatial resolution and signal-to-noise ratio depend on pulse width. In order to improve spatial resolution, existing technical solutions often use optical pulses with shorter pulse widths, which sacrifices the signal-to-noise ratio, and correspondingly makes the temperature accuracy and effective temperature measurement length poor. Summary of the invention

[0006] The present invention provides a high spatial resolution distributed temperature sensing method and device based on a deconvolution algorithm to solve the irreconcilable contradiction between the spatial resolution and the signal-to-noise ratio of the current temperature sensor.

[0007] According to a first aspect of an embodiment of the present invention, a high spatial resolution distributed temperature sensing method based on a deconvolution algorithm is provided, comprising the following steps:

[0008] Step S100, dividing the temperature sensing optical fiber into a plurality of optical fiber segments, collecting scattered light intensity traces of each optical fiber segment on the temperature sensing optical fiber in real time, and demodulating the temperature traces of each optical fiber segment according to the scattered light intensity traces;

[0009] Step S200, for each optical fiber segment, taking into account the pulse broadening that occurs after the input optical signal is transmitted to each optical fiber segment, actual optical pulse data of the optical fiber segment is collected;

[0010] Step S300, taking into account the input optical signal of the temperature sensing optical fiber and the returned scattered light, and the power loss existing in the transmission process of the temperature sensing optical fiber, for each optical fiber segment, determining the inverse of the optimal signal-to-noise ratio corresponding to the optical fiber segment;

[0011] Step S400, performing edge blurring processing on the temperature traces demodulated corresponding to each optical fiber segment to offset the ringing effect of the demodulated temperature traces;

[0012] Step S500: for each optical fiber segment, using a deconvolution algorithm, according to the actual light pulse data corresponding to the optical fiber segment, the inverse of the optimal signal-to-noise ratio, and the temperature trace after edge blurring, determine the actual temperature trace of the optical fiber segment;

[0013] Step S600: splice together the actual temperature traces corresponding to the various optical fiber segments to obtain the actual temperature trace corresponding to the temperature sensing optical fiber.

[0014] In an optional implementation, step S300 includes:

[0015] Step S301, along the transmission direction of the input optical signal, starting from the end of each optical fiber segment on the temperature sensing optical fiber, respectively select sub-optical fiber segments of a set length toward the head end, and place each sub-optical fiber segment in a constant temperature box;

[0016] Step S302: for each optical fiber segment, a deconvolution algorithm is used to determine the actual temperature trace of each optical fiber segment according to the actual light pulse data corresponding to the optical fiber segment, the randomly selected inverse of the signal-to-noise ratio, and the temperature trace after edge blurring, thereby determining the actual temperature trace of the sub-optical fiber segment on each optical fiber segment;

[0017] Step S303, for each sub-fiber segment's actual temperature trace, obtain the root mean square error between the fiber temperature corresponding to the sub-fiber segment and the set temperature of the constant temperature box;

[0018] Step S304: for each fiber segment, iteratively adjust the randomly selected inverse signal-to-noise ratio, and repeat steps S302 to S303 until the root mean square error is minimized, and use the corresponding inverse signal-to-noise ratio at this time as the optimal inverse signal-to-noise ratio corresponding to the fiber segment.

[0019] In another optional implementation, the step S100 includes:

[0020] Step S101, using a modulator to modulate the DC laser provided by the laser into a pulse light signal;

[0021] Step S102, transmitting the pulse light signal to the temperature sensing optical fiber;

[0022] Step S103, converting the scattered light returned from the temperature sensing optical fiber into an electrical signal;

[0023] Step S104: The data acquisition card acquires the electrical signal to obtain the scattered light intensity trace of each optical fiber segment on the temperature sensing optical fiber.

[0024] In another optional implementation, the step S200 includes:

[0025] Step S201, placing optical fibers of different lengths in sequence between a modulator and a photoelectric converter in a real-time acquisition process of a light intensity trace, wherein for the i-th optical fiber segment on the temperature sensing optical fiber, the lengths of the first i optical fiber segments on the temperature sensing optical fiber are correspondingly added to obtain a corresponding optical fiber length, and the optical fiber of the length is placed between the modulator and the photoelectric converter, wherein i is an integer greater than 0;

[0026] Step S202: For each optical fiber of different length, the modulator is used to convert the DC laser provided by the laser into a pulsed light signal. The pulsed light signal is transmitted to the photoelectric converter through the optical fiber. The photoelectric converter converts the optical pulse signal into an electrical pulse signal, thereby collecting actual optical pulse data of the optical fiber segment corresponding to the length of the optical fiber.

[0027] In another optional implementation, the edge blurring process is performed on the demodulated temperature trace according to the following formula: I = α*T+(1-α)*(T**OTF)

[0028] Where ** is the convolution operation, * is the multiplication operation of the elements at the corresponding positions of the matrix, I is the temperature trace with blurred edges, T is the demodulated temperature trace, OTF is the optical transfer function corresponding to the actual light pulse data as the point spread function, α is the ambiguity, and α is expressed as:

[0029] Where real is the real part operation, is the inverse Fourier transform, abs is the modulo operation, is Fourier transform, N is a matrix whose dimension is the same as that of the demodulated temperature trace matrix, the first few elements in the N matrix are optical pulse data, and are scaled to ∑N=1; k is a matrix whose dimension is the same as that of the demodulated temperature trace matrix, and all elements are 1. When and only when the length of the optical pulse data matrix is ​​an odd number, a 1 is added to the end of the α matrix to ensure that its length is an even number. For each optical fiber segment, the dimension of the corresponding temperature trace matrix is ​​1*X, and X represents the number of temperature data points demodulated on the optical fiber segment.

[0030] In another optional implementation, the step S500 includes:

[0031] For each fiber segment, a deconvolution algorithm is used to determine the actual temperature trace of the fiber segment according to the actual light pulse data corresponding to the fiber segment, the inverse of the optimal signal-to-noise ratio, and the temperature trace after edge blurring according to the following formula:

[0032]

[0033] Where G(f) is the Fourier transform of the demodulated temperature trace, F(f) is the Fourier transform of the actual temperature trace, H(f) is the Fourier transform of the actual optical pulse data, and K is the inverse of the optimal signal-to-noise ratio.

[0034] In another optional implementation, the scattered light is Raman light, including Stokes light and anti-Stokes light, the light intensity trace includes a Stokes light intensity matrix and an anti-Stokes light intensity matrix, and the step S100 includes: according to the following formula, based on the Stokes light intensity matrix and the anti-Stokes light intensity matrix, demodulating the temperature trace of each optical fiber segment:

[0035]

[0036] Where h is Planck's constant, K is Boltzmann's constant, Δν is the Raman frequency shift, T ref is the temperature of the calibration fiber segment on the temperature sensing fiber when it is placed in a constant temperature box, that is, the temperature of the constant temperature box. At this time, the ratio of the Stokes light intensity to the anti-Stokes light intensity at a fiber point on the calibration fiber segment is the second intensity ratio. R is the first intensity ratio divided by the second intensity ratio. The first intensity ratio is the ratio of the Stokes light intensity to the anti-Stokes light intensity found from the corresponding positions in the Stokes light intensity matrix and the anti-Stokes light intensity matrix according to a fiber point on the temperature sensing fiber.

[0037] The present invention also provides a light intensity trace real-time acquisition device used in the above method, characterized in that it comprises a modulator, a wavelength division multiplexer, a photoelectric converter, a data acquisition card and a temperature sensing optical fiber, wherein the modulator is connected to the first end of the wavelength division multiplexer, the second end of the wavelength division multiplexer is connected to the temperature sensing optical fiber, the third end and the fourth end are respectively connected to the input end of the photoelectric converter, and the output end of the photoelectric converter is connected to the data acquisition card;

[0038] The modulator converts a direct current laser into a pulsed light signal and sends it to the wavelength division multiplexer; the wavelength division multiplexer forwards the pulsed light signal to the temperature sensing optical fiber; the temperature sensing optical fiber transmits a first scattered light and a second scattered light respectively containing a first wavelength and a second wavelength back to the wavelength division multiplexer; the wavelength division multiplexer transmits the first scattered light and the second scattered light to the photoelectric converter; the photoelectric converter converts the first scattered light and the second scattered light into a first electrical signal and a second electrical signal respectively; the data acquisition card acquires the first electrical signal and the second electrical signal to obtain a first light intensity trace and a second light intensity trace, and demodulates the temperature traces of each optical fiber segment on the temperature sensing optical fiber according to the first light intensity trace and the second light intensity trace.

[0039] In an optional implementation, it also includes an amplifier arranged between the modulator and the first end of the wavelength division multiplexer, the amplifier is used to amplify the pulse light signal, the photoelectric converter includes a first avalanche photodiode and a second avalanche photodiode, the third end of the wavelength division multiplexer is connected to the first input end of the data acquisition card through the first avalanche photodiode, and the fourth end is connected to the second input end of the data acquisition card through the second avalanche photodiode.

[0040] In another optional implementation, a narrow line width laser is selected as the DC laser, so that the pulse light signal formed by the modulation of the DC laser is not significantly broadened during the transmission process of the temperature sensing optical fiber, thereby ensuring the long-distance temperature sensing resolution; a high-gain erbium-doped fiber amplifier is selected to amplify the pulse light signal to reduce the temperature measurement time;

[0041] A high-isolation wavelength division multiplexer is selected to better filter out noise and ensure the signal-to-noise ratio of the demodulated temperature trace; a multimode optical fiber is selected as the temperature sensing optical fiber to allow a higher-power pulse optical signal to be input to the temperature sensing optical fiber, thereby ensuring the signal-to-noise ratio of the demodulated temperature trace;

[0042] The use of a high-bandwidth photoelectric converter and a data acquisition card can provide more information for the deconvolution algorithm, so that the actual temperature trace determined by the deconvolution algorithm can improve the signal-to-noise ratio of the temperature trace while improving the spatial resolution.

[0043] The beneficial effects of the present invention are:

[0044] 1. When the present invention uses a deconvolution algorithm to process the demodulated temperature trace, it involves actual optical pulse data, the inverse of the optimal signal-to-noise ratio and the demodulated temperature trace, wherein the collection of the actual optical pulse data takes into account the pulse broadening of the optical signal during optical fiber transmission, the inverse of the optimal signal-to-noise ratio takes into account the power loss of the optical signal and scattered light during optical fiber transmission, and the temperature trace takes into account the ringing effect introduced by the deconvolution algorithm. Therefore, when the deconvolution algorithm is used to determine the actual temperature trace based on the actual optical pulse data, the inverse of the optimal signal-to-noise ratio and the fuzzy processed temperature trace, the spatial resolution can be improved. At the same time, the reduction in the signal-to-noise ratio is reduced, that is, a smaller signal-to-noise ratio is sacrificed in exchange for a larger spatial resolution, thereby resolving the irreconcilable contradiction between the spatial resolution and the signal-to-noise ratio to a certain extent; in addition, the present invention divides the temperature sensing optical fiber into a plurality of optical fiber segments, and uses a deconvolution algorithm to process the temperature traces of each demodulated optical fiber segment respectively to obtain the actual temperature traces of each optical fiber segment, and then splices the actual temperature traces of each optical fiber segment together to obtain the actual temperature trace of the entire temperature sensing optical fiber, thereby even in long-distance sensing, the spatial resolution will not deteriorate, thereby realizing long-distance optical fiber temperature sensing;

[0045] 2. When the deconvolution algorithm is used to process the temperature trace, the temperature trace and the actual light pulse data used are in the form of Fourier transform, that is, the deconvolution algorithm is set before the temperature demodulation, which can improve the operation speed. In addition, the expressions of the deconvolution algorithm and the edge blurring algorithm of the present invention are simple, so the actual temperature trace obtained by the present invention is real-time;

[0046] 3. The present invention uses Raman light, Stokes light or anti-Stokes light as scattered light, which can improve the sensitivity of temperature sensing;

[0047] 4. In the light intensity trace real-time acquisition device of the present invention, after receiving the pulse light signal, the temperature sensing optical fiber generates two paths of scattered light of different wavelengths. The two paths of scattered light of different wavelengths can provide more information for the determination of the temperature trace, thereby improving the accuracy of the determination of the actual temperature trace; the present invention uses an avalanche photodiode as a photoelectric converter, which can convert the scattered light into an electrical signal and amplify the electrical signal. As a photoelectric converter, it has a simple structure and can realize more functions;

[0048] 5. In the temperature trace real-time adopting device of the present invention, the component components are selected to further improve the spatial resolution while ensuring the signal-to-noise ratio. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 It is a flow chart of an embodiment of a high spatial resolution distributed temperature sensing method based on a deconvolution algorithm of the present invention;

[0050] Figure 2 It is a schematic structural diagram of an embodiment of a light intensity trace real-time acquisition device of the present invention;

[0051] Figure 3 It is a schematic diagram of the structure of an embodiment of the actual optical pulse data acquisition device of the present invention;

[0052] Figure 4 It is a schematic structural diagram of an embodiment of a device for determining the inverse of the optimal signal-to-noise ratio of the present invention. DETAILED DESCRIPTION

[0053] In order to enable those skilled in the art to better understand the technical solutions in the embodiments of the present invention and to make the above-mentioned purposes, features and advantages of the embodiments of the present invention more obvious and understandable, the technical solutions in the embodiments of the present invention are further described in detail below in conjunction with the accompanying drawings.

[0054] In the description of the present invention, unless otherwise specified and limited, it should be noted that the term "connection" should be understood in a broad sense. For example, it can be a mechanical connection or an electrical connection, or it can be the internal connection between two elements. It can be a direct connection or an indirect connection through an intermediate medium. For ordinary technicians in this field, the specific meanings of the above terms can be understood according to specific circumstances.

[0055] See also Figure 1 , is a schematic diagram of an embodiment of a high spatial resolution distributed temperature sensing method based on a deconvolution algorithm of the present invention. The high spatial resolution distributed temperature sensing method based on a deconvolution algorithm may include the following steps:

[0056] Step S100, dividing the temperature sensing optical fiber into a plurality of optical fiber segments, collecting scattered light intensity traces of each optical fiber segment on the temperature sensing optical fiber in real time, and demodulating the temperature traces of each optical fiber segment according to the scattered light intensity traces;

[0057] Step S200, for each optical fiber segment, taking into account the pulse broadening that occurs after the input optical signal is transmitted to each optical fiber segment, actual optical pulse data of the optical fiber segment is collected;

[0058] Step S300, taking into account the input optical signal of the temperature sensing optical fiber and the returned scattered light, and the power loss existing in the transmission process of the temperature sensing optical fiber, for each optical fiber segment, determining the inverse of the optimal signal-to-noise ratio corresponding to the optical fiber segment;

[0059] Step S400, performing edge blurring processing on the temperature traces demodulated corresponding to each optical fiber segment to offset the ringing effect of the demodulated temperature traces;

[0060] Step S500: for each optical fiber segment, using a deconvolution algorithm, according to the actual light pulse data corresponding to the optical fiber segment, the inverse of the optimal signal-to-noise ratio, and the temperature trace after edge blurring, determine the actual temperature trace of the optical fiber segment;

[0061] Step S600: splice together the actual temperature traces corresponding to the various optical fiber segments to obtain the actual temperature trace corresponding to the temperature sensing optical fiber.

[0062] When the present invention uses the deconvolution algorithm to process the demodulated temperature trace, it is necessary to involve the demodulated temperature trace, the pre-saved light pulse data transmitted to the temperature sensing optical fiber, and the estimated inverse of the signal-to-noise ratio. However, since the optical signal is affected by factors such as mode-related dispersion, wavelength-related dispersion, and polarization-related dispersion during the optical fiber transmission process, pulse width broadening will inevitably occur. Therefore, directly using the input light pulse of the temperature sensing optical fiber as the light pulse data is not suitable for the entire long-distance optical fiber. For this reason, the present invention considers the pulse broadening that occurs after the optical signal is transmitted to each optical fiber segment for each optical fiber segment, and collects the actual light pulse data of each optical fiber segment separately, which can improve the accuracy of the light pulse data involved in the deconvolution algorithm. When the deconvolution algorithm is processed based on the actual light pulse data, the accuracy of the deconvolution algorithm processing can be improved.

[0063] Since the input optical signal of the temperature sensing optical fiber and the returned scattered light will be lost accordingly during the optical fiber transmission process, the signal power will decrease. Therefore, the signal-to-noise ratio pre-estimated by the existing method is not applicable to the entire long-distance optical fiber. For this reason, the present invention determines the inverse of the optimal signal-to-noise ratio corresponding to each optical fiber segment in consideration of the input optical signal of the temperature sensing optical fiber and the returned scattered light, and the power loss existing in the temperature sensing optical fiber transmission process, which can improve the accuracy of the inverse of the signal-to-noise ratio involved in the deconvolution algorithm. In addition, when the deconvolution algorithm is directly used to process the temperature trace, a non-negligible ringing effect will be introduced, showing that abnormally large noise appears near the leftmost optimal temperature trace and the temperature mutation zone. For this reason, the present invention performs edge fuzzy processing on the temperature trace corresponding to each optical fiber segment, thereby offsetting the ringing effect of the demodulated temperature trajectory.

[0064] It can be seen from the above embodiments that when the present invention uses a deconvolution algorithm to process the demodulated temperature trace, it involves actual optical pulse data, the inverse of the optimal signal-to-noise ratio and the demodulated temperature trace, wherein the collection of the actual optical pulse data takes into account the pulse broadening of the optical signal during the optical fiber transmission process, the inverse of the optimal signal-to-noise ratio takes into account the power loss of the optical signal and the scattered light during the optical fiber transmission process, and the temperature trace takes into account the ringing effect introduced by the deconvolution algorithm. Therefore, when the deconvolution algorithm is used to determine the actual temperature trace based on the actual optical pulse data, the inverse of the optimal signal-to-noise ratio and the fuzzy processed temperature trace, the spatial resolution can be improved. While increasing the spatial resolution, the reduction in signal-to-noise ratio is reduced, that is, a smaller signal-to-noise ratio is sacrificed in exchange for a larger spatial resolution, thereby resolving the irreconcilable contradiction between spatial resolution and signal-to-noise ratio to a certain extent; in addition, the present invention divides the temperature sensing optical fiber into multiple optical fiber segments, and uses a deconvolution algorithm to process the temperature traces of each demodulated optical fiber segment respectively to obtain the actual temperature trace of each optical fiber segment, and then splices the actual temperature traces of each optical fiber segment together to obtain the actual temperature trace of the entire temperature sensing optical fiber, thereby even in long-distance sensing, the spatial resolution will not deteriorate, thereby realizing long-distance optical fiber temperature sensing.

[0065] See also Figure 2, is a schematic diagram of the structure of an embodiment of a real-time acquisition device for light intensity traces, the device may include a modulator (which may be an acousto-optic modulator), a wavelength division multiplexer, a photoelectric converter, a data acquisition card and a temperature sensing optical fiber, wherein the modulator is connected to the first end of the wavelength division multiplexer, the second end of the wavelength division multiplexer is connected to the temperature sensing optical fiber, the third end and the fourth end are respectively connected to the input end of the photoelectric converter, and the output end of the photoelectric converter is connected to the data acquisition card; the modulator converts a DC laser into a pulsed optical signal and sends it to the wavelength division multiplexer; the wavelength division multiplexer forwards the pulsed optical signal to the temperature sensing optical fiber; A temperature sensing optical fiber; the temperature sensing optical fiber transmits the first scattered light and the second scattered light respectively containing the first wavelength and the second wavelength back to the wavelength division multiplexer; the wavelength division multiplexer transmits the first scattered light and the second scattered light to the photoelectric converter; the photoelectric converter converts the first scattered light and the second scattered light into the first electrical signal and the second electrical signal respectively; the data acquisition card collects the first electrical signal and the second electrical signal to obtain the first light intensity trace and the second light intensity trace, and demodulates the temperature traces of each optical fiber segment on the temperature sensing optical fiber according to the first light intensity trace and the second light intensity trace. In addition, the light intensity trace real-time acquisition device can also include an amplifier arranged between the modulator and the first end of the wavelength division multiplexer, the amplifier is used to amplify the pulse light signal, the photoelectric converter includes a first avalanche photodiode and a second avalanche photodiode, the third end of the wavelength division multiplexer is connected to the first input end of the data acquisition card through the first avalanche photodiode, and the fourth end is connected to the second input end of the data acquisition card through the second avalanche photodiode.

[0066] Among them, the first light intensity trace and the second light intensity trace can be a Stokes light intensity matrix and an anti-Stokes light intensity matrix respectively. The controller (such as a computer) can be connected to the laser, amplifier and data acquisition card used to generate the DC laser respectively, and can also be connected to the modulator through a signal generator and a modulation drive circuit in sequence. Therefore, the controller can control the laser to generate the corresponding DC laser, control the amplification factor of the amplifier, control the acquisition of the data acquisition card, and control the signal generator to generate the corresponding pulse signal. The signal generator uses the modulation drive circuit to enable the modulator to modulate the DC laser based on the pulse signal to generate a pulse light signal. For example, the laser can generate a DC laser with a wavelength of 1550nm and a power of 20mW; the acousto-optic modulator modulates the DC laser into a pulse laser with a repetition frequency of 10Khz and a pulse width of 14ns; the amplifier increases the peak power of the pulsed light signal to 10W; the wavelength division multiplexer will output the input pulsed light signal from its 1550 end (i.e., the first end) to the temperature sensing optical fiber through its com end (i.e., the second end). After the temperature sensing optical fiber receives the pulsed light signal with a wavelength of 1550nm, it generates a pulsed light signal that is transmitted back to the wavelength division multiplexer. The first scattered light and the second scattered light have wavelengths of 1450nm and 1650nm respectively; the first scattered light and the second scattered light are transmitted to the first avalanche photodiode and the second avalanche photodiode respectively by the 1450 end (i.e., the third end) and the 1650 end (i.e., the fourth end) of the wavelength division multiplexer, the first avalanche photodiode converts the first scattered light into a first electrical signal and amplifies it, the second avalanche photodiode converts the second scattered light into a second electrical signal and amplifies it, and the data acquisition card collects the first electrical signal and the second electrical signal. The data acquisition card can transmit the collected first electrical signal and the second electrical signal to a computer for processing according to the method of the present invention, so as to obtain the actual temperature traces corresponding to each optical fiber segment on the temperature sensing optical fiber.

[0067] After receiving the pulse light signal, the temperature sensing optical fiber of the present invention generates two paths of scattered light of different wavelengths. The two paths of scattered light of different wavelengths can provide more information for the determination of the temperature trace, thereby improving the accuracy of the determination of the actual temperature trace; the present invention uses an avalanche photodiode as a photoelectric converter, which can convert the scattered light into an electrical signal while amplifying the electrical signal. As a photoelectric converter, it has a simple structure and can realize more functions. It should be noted that the repetition frequency of the pulse light signal modulated by the present invention is determined by the amplification factor of the amplifier and the balance of noise, while the pulse width of the pulse light signal is determined by the balance of the two factors of spatial resolution and signal-to-noise ratio. That is, when the deconvolution algorithm of the present invention is applied to process the demodulated temperature trace, by adjusting the pulse width of the optical signal input to the temperature sensing optical fiber, the reduction in the signal-to-noise ratio can be minimized while improving the spatial resolution, that is, sacrificing the minimum signal-to-noise ratio in exchange for a larger spatial resolution. In this embodiment, the pulse width of the optical pulse signal is 14ns, which is the result of repeated optimization and adjustment.

[0068] In addition, since the dispersion effect of narrow-linewidth lasers is weaker, compared with other types of lasers, the pulse broadening that occurs during the temperature sensing optical fiber transmission process is not obvious. Therefore, the present invention selects narrow-linewidth lasers as the DC lasers, so that the pulse light signal formed by the modulation of the DC laser is not obviously broadened during the temperature sensing optical fiber transmission process, thereby ensuring the spatial sensing resolution of long-distance temperature. Since a high-gain erbium-doped fiber amplifier can amplify a pulsed light signal with a higher repetition frequency, the present invention uses a high-gain erbium-doped fiber amplifier to amplify the pulsed light signal, which can reduce the temperature measurement time; since the scattered light returned by the temperature sensing optical fiber contains Rayleigh scattered light with a higher power, and the Rayleigh scattered light is equivalent to a non-zero mean low-frequency noise, the present invention can use a high-isolation wavelength division multiplexer to better filter out the noise and ensure the signal-to-noise ratio of the demodulated temperature trace; since the stimulated Raman scattering threshold of the multimode optical fiber is relatively high, the present invention can use a multimode optical fiber as the temperature sensing optical fiber to allow a pulsed light signal with a higher power to be input to the temperature sensing optical fiber, thereby ensuring the signal-to-noise ratio of the demodulated temperature trace; the present invention can also use a high-bandwidth photoelectric converter and a data acquisition card to provide more information for the deconvolution algorithm, so that the actual temperature trace determined according to the deconvolution algorithm can improve the spatial resolution while ensuring the signal-to-noise ratio of the temperature trace.

[0069] Wherein, in step S100, the temperature trajectory of each optical fiber segment on the temperature sensing optical fiber is collected in real time according to the following steps:

[0070] Step S101, using a modulator to modulate the DC laser provided by the laser into a pulse light signal;

[0071] Step S102, transmitting the pulse light signal to the temperature sensing optical fiber;

[0072] Step S103, converting the scattered light returned from the temperature sensing optical fiber into an electrical signal;

[0073] Step S104: The data acquisition card acquires the electrical signal to obtain the scattered light intensity trace of each optical fiber segment on the temperature sensing optical fiber.

[0074] See also Figure 3 , is a schematic diagram of the structure of an embodiment of an actual optical pulse data acquisition device. The actual optical pulse data acquisition device may include a modulator, optical fibers of different lengths, a photoelectric converter, and a data acquisition card, wherein the modulator is connected to the data acquisition card in sequence through optical fibers of corresponding lengths and photoelectric converters, and the modulator, photoelectric converter, and data acquisition card may use corresponding devices in the light intensity trace real-time acquisition device, and the pulse light signal output by the modulator, the sampling rate of the data acquisition card, etc. may be equal to the relevant parameters in the light intensity trace real-time acquisition process. The step S200 may include:

[0075] Step S201, placing optical fibers of different lengths in sequence between a modulator and a photoelectric converter in a real-time acquisition process of a light intensity trace, wherein for the i-th optical fiber segment on the temperature sensing optical fiber, the lengths of the first i optical fiber segments on the temperature sensing optical fiber are correspondingly added to obtain a corresponding optical fiber length, and the optical fiber of the length is placed between the modulator and the photoelectric converter, wherein i is an integer greater than 0;

[0076] Step S202: For each optical fiber of different length, the modulator is used to convert the DC laser provided by the laser into a pulsed light signal. The pulsed light signal is transmitted to the photoelectric converter through the optical fiber. The photoelectric converter converts the optical pulse signal into an electrical pulse signal, thereby collecting actual optical pulse data of the optical fiber segment corresponding to the length of the optical fiber.

[0077] The photoelectric converter may use an avalanche photodiode. To avoid damage to the avalanche photodiode, the output end of the modulator may be connected to an attenuator or a coupler. The attenuator attenuates the pulse light signal output by the modulator and then transmits it to the avalanche photodiode through an optical fiber of a corresponding length. The coupler may transmit a portion of the pulse light signal output by the modulator to the avalanche photodiode through an optical fiber of a corresponding length. Since the avalanche photodiode will output a noise voltage of about 0.002V when there is no signal input, the data acquisition card may use the pulse light signal with an amplitude higher than the threshold as optical pulse data.

[0078] See also Figure 4, is a schematic diagram of the structure of an embodiment of a device for determining the inverse of the optimal signal-to-noise ratio. Figure 2 The difference between the light intensity trace real-time acquisition device shown is that, along the transmission direction of the input optical signal, starting from the end of each optical fiber segment on the temperature sensing optical fiber, sub-optical fiber segments of a set length (for example, 0.5m) are selected toward the head end, and each sub-optical fiber segment is placed in a constant temperature box. The step S300 may include: step S301, along the transmission direction of the input optical signal, starting from the end of each optical fiber segment on the temperature sensing optical fiber, sub-optical fiber segments of a set length are selected toward the head end, and each sub-optical fiber segment is placed in a constant temperature box (for example, a constant temperature water tank).

[0079] Step S302: for each optical fiber segment, a deconvolution algorithm is used to determine the actual temperature trace of each optical fiber segment according to the actual optical pulse data corresponding to the optical fiber segment, the randomly selected inverse of the signal-to-noise ratio, and the temperature trace after edge blurring, thereby determining the actual temperature trace of the sub-optical fiber segment on each optical fiber segment. In this step, the temperature trace corresponding to the optical fiber segment can be collected in the same way as in the above step S100, and the actual optical pulse data corresponding to the optical fiber segment can be collected in the same way as in the above step S200.

[0080] Step S303: for each sub-fiber segment's actual temperature trace, obtain the root mean square error between the fiber temperature corresponding to the sub-fiber segment and the set temperature of the constant temperature box.

[0081] In this step, the root mean square error is calculated according to the following formula:

[0082]

[0083] Where RMSE stands for root mean square error, X i represents the ith temperature data point on the actual temperature trace corresponding to the sub-fiber segment, X0 represents the set temperature of the constant temperature box, and n represents the number of temperature data points on the actual temperature trace corresponding to the sub-fiber segment.

[0084] Step S304: for each fiber segment, iteratively adjust the randomly selected inverse signal-to-noise ratio, and repeat steps S302 to S303 until the root mean square error is minimized, and use the corresponding inverse signal-to-noise ratio at this time as the optimal inverse signal-to-noise ratio corresponding to the fiber segment.

[0085] In the present invention, whether in step S400, step S500, or step S302, the edge-blurred temperature trace is subjected to edge blurring processing according to the following formula: I = α*T+(1-α)*(T**OTF)

[0086] Where ** is the convolution operation, * is the multiplication operation of the elements at the corresponding positions of the matrix, I is the temperature trace with blurred edges, T is the demodulated temperature trace, OTF is the optical transfer function corresponding to the actual light pulse data as the point spread function, α is the ambiguity, and α is expressed as:

[0087] Where real is the real part operation, is the inverse Fourier transform, abs is the modulo operation, is Fourier transform, N is a matrix whose dimension is the same as that of the demodulated temperature trace matrix, the first few elements in the N matrix are optical pulse data, and are scaled to ∑N=1; k is a matrix whose dimension is the same as that of the demodulated temperature trace matrix, and all elements are 1. When and only when the length of the optical pulse data matrix is ​​an odd number, a 1 is added to the end of the α matrix to ensure that its length is an even number. For each optical fiber segment, the dimension of the corresponding temperature trace matrix is ​​1*X, and X represents the number of temperature data points demodulated on the optical fiber segment.

[0088] In addition, the step S500 includes:

[0089] For each fiber segment, a deconvolution algorithm is used to determine the actual temperature trace of the fiber segment according to the actual light pulse data corresponding to the fiber segment, the inverse of the optimal signal-to-noise ratio, and the temperature trace after edge blurring according to the following formula:

[0090]

[0091] Where G(f) is the Fourier transform of the demodulated temperature trace, F(f) is the Fourier transform of the actual temperature trace, H(f) is the Fourier transform of the actual optical pulse data, and K is the inverse of the optimal signal-to-noise ratio.

[0092] When the present invention uses a deconvolution algorithm to process the temperature trace, the temperature trace and actual optical pulse data used are in the form of Fourier transform, that is, the deconvolution algorithm is set before the temperature demodulation, which can improve the operation speed, and the expressions of the deconvolution algorithm and the edge blurring algorithm of the present invention are simple, so the actual temperature trace obtained by the present invention is real-time. It should be noted that: the above-mentioned scattered light can be Raman light. Since this light is very sensitive to temperature, the present invention uses Raman light, Stokes light or anti-Stokes light as scattered light, which can improve the temperature sensing sensitivity. The above-mentioned deconvolution algorithm can be a Wiener deconvolution algorithm. Including Stokes light and anti-Stokes light, the light intensity trace includes a Stokes light intensity matrix and an anti-Stokes light intensity matrix, and the step S100 includes: according to the following formula, according to the Stokes light intensity matrix and the anti-Stokes light intensity matrix, the temperature trace of each optical fiber segment is demodulated:

[0093]

[0094] Where h is Planck's constant, K is Boltzmann's constant, Δν is the Raman frequency shift, T ref is the temperature of the calibration fiber section on the temperature sensing fiber when it is placed in a constant temperature box, that is, the temperature of the constant temperature box. At this time, the ratio of the Stokes light intensity to the anti-Stokes light intensity at a fiber point on the calibration fiber section is the second light intensity ratio. R is the first light intensity ratio divided by the second light intensity ratio. The first light intensity ratio is the ratio of the Stokes light intensity to the anti-Stokes light intensity found from the corresponding position in the Stokes light intensity matrix and the anti-Stokes light intensity matrix at a fiber point on the temperature sensing fiber. For example, a 20m calibration fiber section on the temperature sensing fiber is placed in a constant temperature box, and the temperature of the constant temperature box is T. ref At this time, the real-time acquisition device of the light intensity trace of the present invention can still obtain the Stokes light intensity matrix and the anti-Stokes light intensity matrix. Each fiber point on the temperature sensing fiber corresponds to the position in the two matrices respectively. For the matrix with a dimension of 1*X, when determining the second light intensity ratio, the Stokes light intensity and the anti-Stokes light intensity are first found from the corresponding positions of the two matrices respectively, and then the two intensities are divided; similarly, when the light intensity trace is officially collected to determine the actual temperature of each fiber point on the temperature sensing fiber, for each fiber point on the position sensing fiber, in order to determine the first light intensity ratio, the Stokes light intensity and the anti-Stokes light intensity can also be found from the corresponding positions of the two matrices respectively, and then the two intensities are divided.

[0095] In one example, the temperature sensor can be divided into multiple fiber segments, and the length of each fiber segment is 1 km. The spatial resolution of existing temperature sensors can only reach the meter level, preferably about 1m, and can only distinguish the temperature time above 1m. The temperature time below 1m cannot be accurately measured, which causes the distortion of the temperature trace. The spatial resolution will be further deteriorated during long-distance sensing, and the real-time performance of existing temperature sensors is not good. The present invention can achieve a spatial resolution of 0.5m, and the spatial resolution will not continue to deteriorate due to the increase in sensing length. The method of the present invention is simple and can obtain the actual temperature trace in real time.

[0096] Those skilled in the art will readily appreciate other embodiments of the present invention after considering the specification and practicing the invention disclosed herein. This application is intended to cover any variations, uses or adaptations of the present invention that follow the general principles of the present invention and include common knowledge or customary techniques in the art that are not disclosed by the present invention. The specification and examples are to be considered exemplary only, and the true scope and spirit of the present invention are indicated by the following claims.

[0097] It will be appreciated that the invention is not limited to the precise construction that has been described above and shown in the drawings and that various modifications and changes may be made without departing from its scope. The scope of the invention is governed solely by the appended claims.

Claims

1. A high spatial resolution distributed temperature sensing method based on a deconvolution algorithm, characterized in that: The following steps are involved: Step S100, dividing the temperature sensing optical fiber into a plurality of optical fiber segments, collecting scattered light intensity traces of each optical fiber segment on the temperature sensing optical fiber in real time, and demodulating the temperature traces of each optical fiber segment according to the scattered light intensity traces; Step S200, for each optical fiber segment, taking into account the pulse broadening that occurs after the input optical signal is transmitted to each optical fiber segment, actual optical pulse data of the optical fiber segment is collected; Step S300, taking into account the input optical signal of the temperature sensing optical fiber and the returned scattered light, and the power loss existing in the transmission process of the temperature sensing optical fiber, for each optical fiber segment, determining the inverse of the optimal signal-to-noise ratio corresponding to the optical fiber segment; Step S400, performing edge blurring processing on the temperature traces demodulated corresponding to each optical fiber segment to offset the ringing effect of the demodulated temperature traces; Step S500: for each optical fiber segment, using a deconvolution algorithm, according to the actual light pulse data corresponding to the optical fiber segment, the inverse of the optimal signal-to-noise ratio, and the temperature trace after edge blurring, determine the actual temperature trace of the optical fiber segment; Step S600, splicing the actual temperature traces corresponding to the various optical fiber segments together to obtain the actual temperature trace corresponding to the temperature sensing optical fiber; The step S300 includes: Step S301, along the transmission direction of the input optical signal, starting from the end of each optical fiber segment on the temperature sensing optical fiber, respectively select sub-optical fiber segments of a set length toward the head end, and place each sub-optical fiber segment in a constant temperature box; Step S302: for each optical fiber segment, a deconvolution algorithm is used to determine the actual temperature trace of each optical fiber segment according to the actual light pulse data corresponding to the optical fiber segment, the randomly selected inverse of the signal-to-noise ratio, and the temperature trace after edge blurring, thereby determining the actual temperature trace of the sub-optical fiber segment on each optical fiber segment; Step S303, for each sub-fiber segment's actual temperature trace, obtain the root mean square error between the fiber temperature corresponding to the sub-fiber segment and the set temperature of the constant temperature box; Step S304: for each fiber segment, iteratively adjust the randomly selected inverse signal-to-noise ratio, and repeat steps S302 to S303 until the root mean square error is minimized, and use the corresponding inverse signal-to-noise ratio at this time as the optimal inverse signal-to-noise ratio corresponding to the fiber segment.

2. The high spatial resolution distributed temperature sensing method based on the deconvolution algorithm according to claim 1 is characterized in that: The step S100 includes: Step S101, using a modulator to modulate the DC laser provided by the laser into a pulsed light signal; Step S102, transmitting the pulse light signal to the temperature sensing optical fiber; Step S103, converting the scattered light returned from the temperature sensing optical fiber into an electrical signal; Step S104: The data acquisition card acquires the electrical signal to obtain the scattered light intensity trace of each optical fiber segment on the temperature sensing optical fiber.

3. The high spatial resolution distributed temperature sensing method based on the deconvolution algorithm according to claim 2 is characterized in that: The step S200 includes: Step S201, placing optical fibers of different lengths in sequence between a modulator and a photoelectric converter in a real-time acquisition process of a light intensity trace, wherein for the i-th optical fiber segment on the temperature sensing optical fiber, the lengths of the first i optical fiber segments on the temperature sensing optical fiber are correspondingly added to obtain a corresponding optical fiber length, and the optical fiber of the length is placed between the modulator and the photoelectric converter, wherein i is an integer greater than 0; Step S202: For each optical fiber of different length, the modulator is used to convert the DC laser provided by the laser into a pulsed light signal. The pulsed light signal is transmitted to the photoelectric converter through the optical fiber. The photoelectric converter converts the pulsed light signal into an electrical pulse signal, thereby collecting actual light pulse data of the optical fiber segment corresponding to the length of the optical fiber.

4. The high spatial resolution distributed temperature sensing method based on the deconvolution algorithm according to claim 1, characterized in that: The edge blurring process of the demodulated temperature trace is performed according to the following formula: I = α*T+(1-α)*(T**OTF) Where ** is the convolution operation, * is the multiplication operation of the elements at the corresponding positions of the matrix, I is the temperature trace with blurred edges, T is the demodulated temperature trace, OTF is the optical transfer function corresponding to the actual light pulse data as the point spread function, α is the ambiguity, and α is expressed as: Where real is the real part operation, is the inverse Fourier transform, abs is the modulo operation, is Fourier transform, N is a matrix whose dimension is the same as that of the demodulated temperature trace matrix, the first few elements in the N matrix are optical pulse data, and are scaled to ∑H=1; k is a matrix whose dimension is the same as that of the demodulated temperature trace matrix, and all elements are 1. When and only when the length of the optical pulse data matrix is ​​an odd number, a 1 is added to the end of the α matrix to ensure that its length is an even number. For each optical fiber segment, the dimension of the corresponding temperature trace matrix is ​​1*X, and X represents the number of temperature data points demodulated on the optical fiber segment.

5. The high spatial resolution distributed temperature sensing method based on the deconvolution algorithm according to claim 1, characterized in that: The step S500 includes: For each fiber segment, a deconvolution algorithm is used to determine the actual temperature trace of the fiber segment according to the actual light pulse data corresponding to the fiber segment, the inverse of the optimal signal-to-noise ratio, and the temperature trace after edge blurring according to the following formula: Where G(f) is the Fourier transform of the demodulated temperature trace, F(f) is the Fourier transform of the actual temperature trace, H(f) is the Fourier transform of the actual optical pulse data, and K is the inverse of the optimal signal-to-noise ratio.

6. The high spatial resolution distributed temperature sensing method based on the deconvolution algorithm according to claim 1 or 2, characterized in that: The scattered light is Raman light, including Stokes light and anti-Stokes light, the light intensity trace includes a Stokes light intensity matrix and an anti-Stokes light intensity matrix, and the step S100 includes: according to the following formula, based on the Stokes light intensity matrix and the anti-Stokes light intensity matrix, demodulating the temperature trace of each optical fiber segment: Where h is Planck's constant, K is Boltzmann's constant, Δν is the Raman frequency shift, T ref is the temperature of the calibration fiber segment on the temperature sensing fiber when it is placed in a constant temperature box, that is, the temperature of the constant temperature box. At this time, the ratio of the Stokes light intensity to the anti-Stokes light intensity at a fiber point on the calibration fiber segment is the second intensity ratio. R is the first intensity ratio divided by the second intensity ratio. The first intensity ratio is the ratio of the Stokes light intensity to the anti-Stokes light intensity found from the corresponding positions in the Stokes light intensity matrix and the anti-Stokes light intensity matrix according to a fiber point on the temperature sensing fiber.

7. A light intensity trace real-time acquisition device used in the method according to any one of claims 1 to 6, characterized in that: It includes a modulator, a wavelength division multiplexer, a photoelectric converter, a data acquisition card and a temperature sensing optical fiber, wherein the modulator is connected to the first end of the wavelength division multiplexer, the second end of the wavelength division multiplexer is connected to the temperature sensing optical fiber, the third end and the fourth end are respectively connected to the input end of the photoelectric converter, and the output end of the photoelectric converter is connected to the data acquisition card; The modulator converts the direct current laser into a pulse light signal and sends it to the wavelength division multiplexer; the wavelength division multiplexer forwards the pulse light signal to the temperature sensing optical fiber; the temperature sensing optical fiber transmits the first scattered light and the second scattered light having the first wavelength and the second wavelength respectively back to the wavelength division multiplexer; the wavelength division multiplexer transmits the first scattered light and the second scattered light to the photoelectric converter; The photoelectric converter converts the first scattered light and the second scattered light into a first electrical signal and a second electrical signal respectively; the data acquisition card acquires the first electrical signal and the second electrical signal to obtain a first light intensity trace and a second light intensity trace, and demodulates the temperature traces of each optical fiber segment on the temperature sensing optical fiber according to the first light intensity trace and the second light intensity trace.

8. The device according to claim 7, characterized in that It also includes an amplifier arranged between the modulator and the first end of the wavelength division multiplexer, the amplifier is used to amplify the pulse light signal, the photoelectric converter includes a first avalanche photodiode and a second avalanche photodiode, the third end of the wavelength division multiplexer is connected to the first input end of the data acquisition card through the first avalanche photodiode, and the fourth end is connected to the second input end of the data acquisition card through the second avalanche photodiode.

9. The device according to claim 8, characterized in that A narrow line width laser is selected as the DC laser, so that the pulse light signal formed by the modulation of the DC laser is not significantly broadened during the transmission process of the temperature sensing optical fiber, thereby ensuring the long-distance temperature sensing resolution; a high-gain erbium-doped fiber amplifier is selected to amplify the pulse light signal to reduce the temperature measurement time; A high-isolation wavelength division multiplexer is selected to better filter out noise and ensure the signal-to-noise ratio of the demodulated temperature trace; a multimode optical fiber is selected as the temperature sensing optical fiber to allow a higher-power pulse optical signal to be input to the temperature sensing optical fiber, thereby ensuring the signal-to-noise ratio of the demodulated temperature trace; The use of a high-bandwidth photoelectric converter and a data acquisition card can provide more information for the deconvolution algorithm, so that the actual temperature trace determined by the deconvolution algorithm can improve the signal-to-noise ratio of the temperature trace while improving the spatial resolution.

Images