CN104568120B - Composite principle optical fiber sensing system and sensing method - Google Patents

Abstract

The invention discloses a composite principle optical fiber sensing system and sensing method. A laser light source generates continuous light, which is divided into local oscillator light through a coupler, two beams of detection light, one beam of detection light passes through an acousto-optic One end of the optical fiber is injected, and the other detection light is converted into an optical pulse by another acousto-optic modulator, and injected from the other end of the sensing optical fiber; in front of the photodetector, the local oscillator light, forward detection light, and backscattered light interfere with each other. The beat frequency signal is generated, and the photodetector converts the beat frequency signal into an electrical signal. The electrical signal is filtered by a band-pass filter circuit to filter out the difference frequency component, and then the beat frequency is reduced by a mixer. The signal is collected and amplified by the acquisition card. In the host computer, the AM demodulation of different frequencies is realized, and the forward detection signal and the backscattering signal are respectively obtained. Use the improved moving average algorithm and peak-average ratio algorithm to obtain the vibration position for the backscattered signal; analyze and identify the forward detection signal to determine the type of vibration and reduce the false alarm rate.

Description

A composite principle optical fiber sensing system and sensing method

technical field

The invention relates to the field of optical fiber sensing, in particular to an optical fiber sensing system and sensing method based on the principles of a Mach-Zehnder interferometer and a phase-type optical time domain reflectometer.

Background technique

Due to its high sensitivity, immunity to electromagnetic interference, wide detection range, and low cost, distributed optical fiber sensing systems are widely used in long-distance oil and gas pipeline monitoring, perimeter security, and building structure health monitoring. research hotspots.

The Mach-Zehnder/Sagnac interferometer distributed optical fiber sensing system uses the phase difference change caused by external disturbances in the two sensing optical paths to be detected, and is positioned by the correlation time delay estimation method, which can sense vibration well. However, due to the difficulty of determining the time delay in the correlation time delay estimation method itself, the positioning accuracy of this method is not high, and it is difficult to accurately judge the vibration point.

The distributed optical fiber sensing system based on optical time domain reflectometer (OTDR) technology uses the Rayleigh scattering phenomenon that occurs when light waves are transmitted in the optical fiber, and detects the intensity of Rayleigh scattered light in the back to obtain the loss change of the optical fiber and precisely locate the optical fiber. point of failure. Since this technique measures the intensity of Rayleigh scattered light, its measurement sensitivity is relatively low and it can only respond to changes in static loss.

The Φ-OTDR (Phase Sensitive OTDR) technology based on coherent Rayleigh scattering uses a long-term coherent light source to detect the coherent result of the light pulse return light. Its interference method can effectively achieve dynamic response, and can simultaneously achieve high positioning accuracy and high sensitivity detection , especially for the detection of weak disturbance signals. However, because the frequency of its emitted pulse is limited by the length of the optical fiber, its frequency response is very low, which makes it impossible to effectively identify vibration events, and the false alarm rate is high.

Contents of the invention

The invention provides a composite principle optical fiber sensing system and sensing method. The invention successfully combines Mach-Zehnder interferometer and phase-sensitive OTDR technology, so that it has high positioning accuracy and high frequency response at the same time, and can well respond to Vibration events are identified to effectively reduce the false alarm rate. See the description below for details:

A composite principle optical fiber sensing system, consisting of laser light source, FPGA, EDFA and its filter, acousto-optic modulator, optical fiber circulator, coupler, photodetector, band-pass filter circuit, mixer, host computer and transmission Sensing optical fiber composition,

The continuous light generated by the laser light source is divided into three parts through the coupler, one part is the local oscillator light, and the other two parts are the detection light. One beam of detection light is injected from one end of the sensing fiber after passing through the acousto-optic modulator, and the other detection light An optical modulator converts light pulses, which are injected from the other end of the sensing fiber;

In front of the photodetector, the local oscillator light, forward probe light, and backscattered light interfere with each other to generate a beat frequency signal. The photodetector converts the beat frequency signal into an electrical signal, and the electrical signal passes through a bandpass filter circuit to filter out the difference frequency. The beat frequency is reduced through the mixer, and the signal is collected and amplified by the acquisition card. In the host computer, the amplitude modulation and demodulation of different frequencies are realized, and the forward detection signal and backscattering signal are respectively obtained.

A sensing method for a composite principle optical fiber sensing system, said method comprising the following steps:

Use the improved moving average algorithm and average peak ratio algorithm to obtain the vibration position for the backscattered signal;

Analyze and identify the positive detection signal, determine the type of vibration, and reduce the false alarm rate.

The steps of obtaining the vibration position using the improved moving average algorithm and the peak-to-average ratio algorithm for the backscattered signal are specifically:

Select K traces, the number of averages M and the interval parameter n, average the 1st to the Mth, and average the nth to the M-n+1th...to get T=int((K-M)/ n)+1 average curves;

Extract the time-domain signal; calculate the peak-to-peak ratio of each time-domain signal separately, and obtain the vibration position through the peak-to-peak ratio.

The steps of calculating the peak-to-peak ratio of each time-domain signal are specifically:

The peak-to-peak ratio is calculated as C _j =max ₁₀₀ (|t _j |)/average(|t _j |), where max ₁₀₀ (|t _j |) means taking the average of the maximum 100 values of |t _j |; average( |t _j |) is to calculate the average value of |t _j |, which is used to describe the light intensity and noise level; the peak-to-average ratio C _j is used as a comprehensive evaluation of the K traces on the optical fiber.

The beneficial effect of the technical solution provided by the present invention is that the system has the same positioning accuracy as the phase-sensitive OTDR, and at the same time can obtain high-frequency details of vibration, and then perform accurate vibration type identification, effectively reducing the false alarm rate of the system. When used, the system successfully realizes the multiplexing of light source, sensing optical fiber, detector, signal conditioning and other parts, making the system structure relatively simple and low cost. The invention overcomes the problem that the existing system cannot have high positioning accuracy and high frequency resolution at the same time, and the invention successfully reuses parts such as light source, sensing optical fiber, detector, signal conditioning circuit, etc., effectively simplifying the system structure. At the same time, the system has the characteristics of distributed, small interference from electromagnetic and other external sources, which are unique to distributed optical fiber monitoring and detection systems, and is easy to install. It can well meet various vibration detection and monitoring applications, especially long-distance pipeline monitoring and monitoring. Perimeter security, etc.

Description of drawings

Fig. 1 is a structural schematic diagram of a composite principle optical fiber sensing system;

Figure 2(a) is a schematic diagram of the time-domain waveform of the vibration signal;

Figure 2(b) is a schematic diagram of the wavelet packet decomposition results of the vibration signal.

In the accompanying drawings, the list of parts represented by each label is as follows:

1. Laser light source; 2. 1:99 coupler;

3. The first 1:1 coupler; 4. 80MHz acousto-optic modulator;

5. 110MHz acousto-optic modulator; 6. The first EDFA and its filter;

7. The second EDFA and its filter; 8. The circulator;

9. Sensing optical fiber; 10. The second 1:1 coupler;

11. The third EDFA and its filter; 12. Photodetector;

13. Band-pass filter circuit; 14. Signal generator;

15. Mixer machine; 16. Low-pass filter circuit;

17. Signal acquisition and conditioning module; 18. Host computer;

19. FPGAs.

detailed description

In order to make the purpose, technical solution and advantages of the present invention clearer, the implementation manners of the present invention will be further described in detail below.

A composite principle optical fiber sensing system, see Figure 1, the composite principle optical fiber sensing system consists of: laser light source 1, 1:99 coupler 2, first 1:1 coupler 3, 80MHz acousto-optic modulator 4, 110MHz Acousto-optic modulator 5, first EDFA (erbium-doped fiber amplifier) and its filter 6, second EDFA and its filter 7, circulator 8, sensing fiber 9, second 1:1 coupler 10, third EDFA and its filter 11, photodetector 12, band-pass filter circuit 13, signal generator 14, mixer 15, low-pass filter circuit 16, signal acquisition and conditioning module 17, host computer 18 and FPGA19.

The continuous light with narrow line width is generated by the laser light source 1, and the probe light and local oscillator light are generated through the 1:99 coupler 2. The local oscillator light reaches the second 1:1 coupler 10 directly. The probe light is divided into two beams of probe light via the first 1:1 coupler 3 . The first probe light is shifted by 100MHz through the 110MHz acousto-optic modulator 5 , amplified by the second EDFA and the filter 7 , injected into the sensing fiber 9 , and reaches the second 1:1 coupler 10 through the circulator 8 . Another beam of probe light passes through the 80MHz acousto-optic modulator 4, obtains a frequency shift of 80MHz and modulates it into a narrow pulse with a pulse width of hundreds of nanoseconds (the signal form of which is controlled by FPGA19), is amplified by the first EDFA and its filter 6, and passes through The circulator 8 is injected into the sensing fiber 9 , and the backscattered light at various positions in the sensing fiber 9 returns to the second 1:1 coupler 10 via the circulator 8 . At the second 1:1 coupler 10, the local oscillator light, the forward detection light, and the backscattered detection light interfere with each other to form a beat frequency, and the interference result is amplified by the third EDFA and its filter 11 and then detected by the photodetector 12 Converted to photocurrent, the photocurrent passes through the band-pass filter circuit 13 to filter out shot noise, the signal generator 14 sends out a 90MHz sine wave, and the photocurrent is mixed in the mixer 15, and then the high frequency in the current is eliminated through the low-pass filter circuit 16 The components are collected by the signal acquisition and conditioning module 17, and sent to the host computer 18 to complete AM demodulation, positioning and alarm signal classification, and obtain the vibration conditions along the distributed sensors.

The system lays a sensing optical fiber 9 within the range where vibration needs to be detected, and uses the sensing optical fiber 9 as a sensor and signal transmission medium to realize vibration detection and monitoring. The laser light source 1 is an ultra-narrow linewidth light source, the coherence length is greater than the sensing distance, and it outputs continuous light, which is divided into three parts through the coupler, one part is used as a local oscillator, and directly reaches the photodetector 12, and the other two parts are used as the first detection light and a second probe light. The first detection light is modulated by the 110MHz acousto-optic modulator 5, and the light wave obtains _a frequency shift f1 (110MHz can be selected), and is still a continuous light wave, which is amplified by the EDFA, and after filtering out the spontaneous amplitude of the EDFA, it is injected into the sensing fiber 9, Pass through the sensing fiber 9, go through the circulator to 8 and reach the photodetector 12. At this time, the light wave carries rich frequency information along the sensing fiber 9. The second probe light is modulated by the 80MHz acousto-optic modulator 4, and the light wave obtains another frequency shift f2 ₍ 80MHz can be selected), and is modulated into a narrow pulse at the same time, and is amplified by another EDFA, and after the spontaneous amplitude of the EDFA is filtered, it is passed through The circulator 8 is injected from the other end of the sensing fiber 9, and the backscattered light passes through the circulator 8 and then reaches the photodetector 12. At this time, the light wave carries precise vibration location information. At the second 1:1 coupler 10 in front of the photodetector 12, the backscattered light emitted by the local oscillator light, the first probe light, and the second probe light interferes, forming frequencies f ₁ , f ₂ , f ₁ - The beat frequency signal of f ₂ is detected by a high-speed photodetector and converted into a corresponding electrical signal. By adjusting the magnification of the EDFA, the intensity of the three beams of light reaching the photodetector 12 can be made similar. The electrical signal of the photodetector 12 first passes through the band-pass filter circuit 13 to remove high-frequency noise and low-frequency interference of the system, and then the band-pass filter result and the preset sine wave are mixed by the mixer 15 (its frequency is f ₂ < f ₀ <f ₁ , can be set to 90MHz), after mixing, the beat frequency will drop to tens of megahertz, namely f ₁ -f ₀ ,f ₀ -f ₂ ,f ₀ -(f ₁ -f ₂ ) , through the low-pass filter circuit 16, the interference beat f ₀ -(f ₁ -f ₂ ) is removed, and then the conditioning signal is amplified and collected by the acquisition card at a rate of 100 MHz. In the host computer, the amplitude modulation demodulation is performed on the beat signal with the frequency of |f _i -f ₀ |(i=1,2), and the backscatter signal S ₂ and the forward detection signal S ₁ will be demodulated respectively, According to the repetition frequency of the modulation signal of the 80MHz acousto-optic modulator 4, S ₂ is truncated into a plurality of scattering traces in time sequence r={r ₁ ,r ₂ ,r ₃ ,…,ri ,…, _{r k }}

In order to realize the location of the vibration event, the improved moving average algorithm is used to reduce the signal noise, and the peak-to-average ratio method is used to eliminate the influence of the attenuation along the optical fiber and quickly and accurately locate the location of the vibration. The optical fiber sensing method specifically includes the following processes:

101: Select K traces r={r ₁ ,r ₂ ,r ₃ ,…,r _i ,…,r _k }, select the average times M (recommended M>50), interval parameter n (recommended n=5) , for K traces: average from the 1st to the Mth, and average from the nth to the M-n+1th...to get T=int((KM)/n)+1 average curves ,which is:

Among them, R _i is the curve obtained from the i-th moving average, and int() is the rounding down operation.

102: Extract the time-domain signal t _j ={ _{xij}, where x ij is} the jth value of R _i . Here j represents the airspace, corresponding to the location information;

103: Calculate the peak-to-average ratio of each time-domain signal respectively.

Among them, the peak-to-average ratio is calculated as C _j =max ₁₀₀ (|t _j |)/average(|t _j |), where max ₁₀₀ (|t _j |) means taking the average value of the maximum 100 values of |t _j |, In order to eliminate the impact of occasional burrs, average(|t _j |) is the average value of |t _j |, which is used to describe the light intensity and noise level. The peak-to-average ratio C _j is used as a comprehensive evaluation of the K traces on the fiber, see Figure 2(a). The larger the value, the stronger the vibration. Through the formula A=(j c)/(2n _c f _s ) can determine the vibration position. Where c is the speed of light, n _c is the refractive index of the optical fiber, and f _s is the sampling rate of the acquisition card.

In order to reduce the false alarm rate of the system, after locating the vibration, it is necessary to analyze and identify the vibration details to determine the type of vibration. When the peak-to-average ratio C _j shows that vibration occurs, the forward detection signal S ₁ within the corresponding time period of the K traces is extracted. Since the forward detection signal S1 is _a continuous light detection, its frequency resolution is fs/2=50MHz. There are many ways to classify the signal after the vibration details are obtained (for example: empirical mode decomposition method, multi-scale chaos method, wavelet packet energy method, etc.), and here only the wavelet packet energy method is used as an example. Specifically include the following processes:

1) Since the vibration frequency along the pipeline is usually below 20KHz, the signal frequency down _- samples the signal S1 to reduce the sampling rate to 40KHz. Afterwards, 4 layers of wavelet packet decomposition are performed to obtain 16 wavelet packet components, and the first 8 components are selected, as shown in Figure 2(b), as the input feature vector of the BP neural network;

2) Before the application, the event signal collected in advance is used as the training and verification set to train and verify the BP network;

3) The discrimination of common events can be realized by using the trained neural network.

For example: manual excavation, mechanical excavation, etc. for pipeline applications.

In the embodiments of the present invention, unless otherwise specified, the models of the devices are not limited, as long as they can complete the above functions.

Those skilled in the art can understand that the accompanying drawing is only a schematic diagram of a preferred embodiment, and the serial numbers of the above-mentioned embodiments of the present invention are for description only, and do not represent the advantages and disadvantages of the embodiments.

The above descriptions are only preferred embodiments of the present invention, and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included in the protection of the present invention. within range.

Claims (4)

1. a kind of composite principle optical fiber sensing system, including：LASER Light Source, FPGA, EDFA and its wave filter, acousto-optic modulator, Optical fiber circulator, the 1st：1 coupler, the 2nd 1：1 coupler, photodetector, bandwidth-limited circuit, frequency mixer, host computer And sensor fibre, it is characterised in that

LASER Light Source is super-narrow line width light source, and coherence length is more than distance sensing, exports continuous light, and pass through the 1st：1 coupling Device is divided into 3 parts, and a part be used as local oscillator light, directly reaches photodetector, and another two parts detect light and the as first Two detection light, the first detection light are modulated by 110MHz acousto-optic modulators, and light wave obtains a frequency displacement f₁, it is still continuous light wave, Amplify by EDFA, and after filtering out the spontaneous amplitudes of EDFA, inject sensor fibre, by sensor fibre, arrived by optical fiber circulator Up to photodetector, now light wave carries frequency information abundant along sensor fibre；

Second detection light is modulated via 80MHz acousto-optic modulators, and light wave obtains another frequency displacement f₂, while burst pulse is modulated to, Amplify by another EDFA, and after filtering out the spontaneous amplitudes of EDFA, injected via optical fiber circulator from the other end of sensor fibre, Its back-scattering light is by reaching photodetector after optical fiber circulator, now light wave carries accurate vibration location information；

The 2nd 1 before photodetector：At 1 coupler, the backscattering of local oscillator light, the first detection light and the second detection light Light transmitting interference, forming frequency f₁,f₂,f₁-f₂Beat signal, and detected with photodetector, be converted into corresponding telecommunications Number；By the multiplication factor for adjusting EDFA, so as to get the 3 beam light light intensity up to photodetector are close；

The electric signal of photodetector first passes through bandwidth-limited circuit, removes the high-frequency noise and low-frequency disturbance of system, passes through Frequency mixer is by bandpass filtering result and default sinusoidal wave mixed frequency, after mixing, claps wave frequency rate and will drop to tens of megahertzs, i.e. f₁- f₀,f₀-f₂,f₀-(f₁-f₂), by low-pass filter circuit, f is clapped into interference₀-(f₁-f₂) remove, amplify conditioned signal afterwards, pass through Gathered by capture card with 100MHz speed；

In host computer, frequency is existed | f_i-f₀|, i=1,2 bat signal carries out demodulation, will demodulate backscattering respectively Signal S₂And positive detectable signal S₁, according to the repetition rate of the modulated signal of 80MHz acousto-optic modulators, by S₂Temporally Order is blocked as a plurality of scattering trace r.

A kind of 2. method for sensing of composite principle optical fiber sensing system for described in claim 1, it is characterised in that the side Method comprises the following steps：

Vibration position is obtained using modified rolling average algorithm and equal peak ratio algorithm to backscatter signals；

Analysis identification is carried out to positive detectable signal, vibration species is judged, reduces rate of false alarm.

3. method for sensing according to claim 2, it is characterised in that described that backscatter signals are moved using modified Average algorithm and equal peak ratio algorithm obtain the step of vibration position and are specially：

K bars trace, average time M and spacing parameter n are chosen, average, nth bar is carried out to the 1st article to the M articles to the M-n+1 articles It is averaged, there are+1 averaged curve of T=int ((K-M)/n)；

Extract time-domain signal；The equal peak ratio of each time-domain signal is calculated respectively, by equal peak than obtaining vibration position.

4. method for sensing according to claim 2, it is characterised in that the equal peak for calculating each time-domain signal than step It is rapid to be specially：

Equal peak ratio is calculated as C_j=max₁₀₀(|t_j|)/average(|t_j|), wherein max₁₀₀(|t_j|) represent to take | t_j| maximum 100 The average value of individual value；average(|t_j|) be to calculate | t_j| average value, to describe light intensity and noise level；Equal peak is compared into C_j As to the overall merit on optical fiber in K bars trace time；t_jTo extract time-domain signal.