Optical neural network-based distributed acoustic sensing system, and all-optical integration method

By using a distributed acoustic wave sensing system with an all-optical design and optical neural network for signal processing, the noise and complexity problems of traditional systems are solved, and real-time signal processing with high stability and low power consumption is achieved.

WO2026007164A1PCT designated stage Publication Date: 2026-01-08NANJING UNIV
View PDF 11 Cites 0 Cited by

Patent Information

Application Number
PCT/CN2024/105179
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2024-07-03
Filing Date
2024-07-12
Publication Date
2026-01-08

AI Technical Summary

Technical Problem

Traditional distributed acoustic wave sensing systems suffer from noise and distortion problems in signal processing, are highly complex, susceptible to electromagnetic interference, and consume a lot of energy, making it difficult to meet the requirements for real-time and accurate response.

Method used

The distributed acoustic wave sensing system, which adopts an all-optical design, uses an optical neural network for signal processing. It achieves signal weighting and activation function calculation through optical elements, eliminates the loss caused by photoelectric conversion, and uses a pure optical neural network for signal recognition and classification.

Benefits of technology

It improved system stability, simplified system architecture, enhanced parallel processing capabilities, reduced system power consumption, and increased data processing speed and capacity.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN2024105179_08012026_PF_FP_ABST
    Figure CN2024105179_08012026_PF_FP_ABST
Patent Text Reader

Abstract

The present invention relates to the technical fields of optical fiber sensing and artificial intelligence. Disclosed are an optical neural network-based distributed acoustic sensing (DAS) system, and an all-optical integration method. The DAS system comprises: an integrated DAS optical path portion, an external connection portion, and an integrated signal processing chip. The present invention combines an integrated optical delay line, a three-port-like detection structure, a pure-optical neural network module and the like to realize all-optical integration of the DAS system and signal processing in a pure-optical method, and has the advantages of reducing photoelectric conversion, enhancing a parallel processing capability, increasing processing speed, lowering energy consumption, improving system stability, and simplifying the system architecture, etc.
Need to check novelty before this filing date? Find Prior Art

Description

Distributed acoustic sensing system based on optical neural network and all-optical integration method TECHNICAL FIELD

[0001] The present application relates to the field of optical fiber sensing technology and artificial intelligence technology, and in particular to a distributed acoustic sensing system based on optical neural network and all-optical integration method. BACKGROUND

[0002] Distributed acoustic sensing (DAS) uses a narrow-linewidth laser to emit continuous light into a sensing optical fiber for transmission after pulse modulation. The backscattered Rayleigh scattering (RBS) signal generated by the sensing optical fiber is analyzed by a photodetector. The principles of phase change and interference effect are used to monitor and locate the weak events and changes in the optical fiber, thereby converting the optical fiber network into a continuous, high-resolution acoustic and vibration monitoring system. It has many advantages such as high sensitivity, high accuracy, fast response, wide coverage, real-time monitoring, etc. It is widely used in oil and gas exploration, pipeline and cable monitoring, perimeter security, etc. and has high application value and promotion prospect.

[0003] The traditional DAS sensing system uses the method of "photoelectric conversion + analog-digital conversion + digital signal processing" for signal processing. A series of electrical components such as photodetectors are introduced to assist in completing the demodulation of the signal. The signal is converted between light and electricity multiple times, and each conversion may introduce noise and distortion, thereby reducing the signal quality. At the same time, the electronic components have slow operation speed in the case of large amount of calculation or high accuracy, which is slightly insufficient in applications that require real-time and accurate response. In addition, due to the introduction of photoelectric conversion, the overall design of the system becomes more complex, and it is easily affected by electromagnetic interference, greatly increasing the maintenance cost and energy consumption, and reducing the system stability.

[0004] Optical neural network (ONN) is a technology that uses optical components instead of traditional electronic components to perform neural network operations. It uses optical phenomena such as interference, diffraction and nonlinear effects to perform weighting and activation function operations, thereby simulating the calculation and data processing in traditional neural networks. The signal is processed and transmitted in the form of light through optical components. Therefore, compared with the neural network operation performed by traditional electronic components, it has the advantages of high speed, low delay, low energy consumption, and parallel processing, and has great application prospect in the fields of artificial intelligence, image and signal processing, quantum computing, etc.

[0005] Therefore, the present application proposes a distributed acoustic sensing system based on optical neural network and all-optical integration method to solve the problems existing in the prior art, which is a problem urgently needed to be solved by those skilled in the art.

[0006] SUMMARY

[0007] Therefore, the application provides a long-distance wide-frequency-response distributed optical fiber acoustic wave sensing system and method based on photonic integration to solve the problems in the prior art.

[0008] To achieve the above-mentioned purpose, the application adopts the following technical solutions:

[0009] A distributed acoustic wave sensing system based on an optical neural network comprises a DAS optical path integrated part, an external part and a signal processing integrated chip, wherein,

[0010] The DAS optical path integrated part comprises a narrow linewidth laser, an intensity modulator and a first optical amplifier.

[0011] The external part comprises a circulator and a sensing optical fiber.

[0012] The signal processing integrated chip comprises a second optical amplifier, a first coupler, an optical delay line, a second coupler, a third coupler, a first polarization beam splitter, a second polarization beam splitter, a fourth coupler, a fifth coupler, a sixth coupler, an information conversion module, a space-time signal combination module, a pure optical neural network module and an optoelectronic conversion and control module.

[0013] Optionally, the narrow linewidth laser, the intensity modulator and the first optical amplifier in the DAS optical path integrated part are connected in sequence; the intensity modulator is connected with a first output end of the optoelectronic conversion and control module of the signal processing integrated chip.

[0014] An output end of the first optical amplifier is connected with a first port of the circulator of the external part.

[0015] The narrow linewidth laser is configured to output continuous narrow linewidth high-coherence laser to the intensity modulator.

[0016] The intensity modulator is modulated by the optoelectronic conversion and control module, and modulates the continuous probe light input by the narrow linewidth laser into pulsed probe light, which is output to the first optical amplifier.

[0017] The first optical amplifier is configured to amplify the power of the pulsed probe light modulated by the intensity modulator and output to the first port of the circulator.

[0018] Optionally, in the external part, a second port of the circulator is connected with the sensing optical fiber, and a third port of the circulator is connected with an input end of the second optical amplifier of the signal processing integrated chip.

[0019] The circulator is used for outputting the pulse probe light received by the first port of the circulator from the second port of the circulator to the sensing optical fiber, and outputting the Rayleigh backscattering light RBS signal received by the second port of the circulator from the third port of the circulator to the second optical amplifier of the signal processing integrated chip.

[0020] The sensing optical fiber is used for receiving the pulse probe light input by the second port of the circulator, and generating the Rayleigh backscattering light RBS signal.

[0021] The system, optionally, the first coupler, the optical delay line, the second coupler, the third coupler, the first polarization beam splitter, the second polarization beam splitter, the fourth coupler, the fifth coupler and the sixth coupler in the signal processing integrated chip form a temperature control and vibration isolation module.

[0022] The system, optionally, the second optical amplifier is used for amplifying the Rayleigh backscattering light RBS signal input by the third port of the circulator, and outputting the Rayleigh backscattering light RBS signal to the first coupler of the temperature control and vibration isolation module.

[0023] The system, optionally, the first coupler is used for splitting the optical signal input by the second optical amplifier into two paths, one path is output to the second coupler through the first output port of the first coupler, and the other path is output to the optical delay line through the second output port of the first coupler.

[0024] The optical delay line is used for transmitting the optical signal input by the second output port of the first coupler to the third coupler, eliminating multipath interference, realizing phase matching and time sequence synchronization.

[0025] The second coupler is used for splitting the optical signal input by the first output port of the first coupler into three paths, one path is output to the first polarization beam splitter through the first output port of the second coupler, one path is output to the fifth coupler through the second output port of the second coupler, and one path is output to the second polarization beam splitter through the third output port of the second coupler.

[0026] The third coupler is used for splitting the optical signal input by the optical delay line into three paths, one path is output to the fourth coupler through the first output port of the third coupler, one path is output to the fifth coupler through the second output port of the third coupler, and one path is output to the sixth coupler through the third output port of the third coupler.

[0027] The first polarization beam splitter is used for transmitting the optical signal input by the first output port of the second coupler to the fourth coupler, and generating a 2π / 3 phase shift at the same time.

[0028] The second polarization beam splitter is used for transmitting the optical signal input by the third output port of the second coupler to the sixth coupler, and generating a 4π / 3 phase shift at the same time.

[0029] A fourth coupler is configured to couple the optical signal input from the first polarization beam splitter and the optical signal input from the first output port of the third coupler, and output to the first input port of the information conversion module.

[0030] A fifth coupler is configured to couple the optical signal input from the second output port of the second coupler and the optical signal input from the second output port of the third coupler, and output to the second input port of the information conversion module.

[0031] A sixth coupler is configured to couple the optical signal input from the second polarization beam splitter and the optical signal input from the third output port of the third coupler, and output to the third input port of the information conversion module.

[0032] Optionally, the system further comprises an information conversion module configured to convert the optical intensity signals input from the fourth coupler, the fifth coupler and the sixth coupler into phase signals, and output to the space-time signal combination module.

[0033] The space-time signal combination module is controlled by the optoelectronic conversion and control module, and is configured to integrate the multiple groups of space-time two-dimensional phase signals input from the information conversion module to form an overall space-time two-dimensional phase signal, and then transmit the overall space-time two-dimensional phase signal to the pure optical neural network module.

[0034] The pure optical neural network module is configured to perform recognition and classification operations on the overall space-time two-dimensional phase signal input from the space-time signal combination module, and output to the optoelectronic conversion and control module.

[0035] The optoelectronic conversion and control module receives an external trigger signal Trigger, and is configured to set the detection light pulse parameters, control the working states of the space-time signal combination module and the first intensity modulator, and convert the time information, position information and event information input from the pure optical neural network module into electrical signals.

[0036] Optionally, the pure optical neural network is implemented in a lookup table manner, wherein the optical memory is composed of non-volatile waveguide phase shifters; the signal of the pure optical neural network module is an analog optical signal, which is a time-space two-dimensional graph representing phase information by intensity.

[0037] Optionally, the first optical amplifier and the second optical amplifier are semiconductor optical amplifiers.

[0038] An all-optical integration method of a distributed acoustic wave sensing system based on an optical neural network, applied to any one of the distributed acoustic wave sensing systems based on the optical neural network, comprises the following steps:

[0039] S1, the narrow linewidth laser transmits continuous narrow linewidth high coherence laser to the intensity modulator, the intensity modulator is modulated by photoelectric conversion and control module, the continuous probe light input by the narrow linewidth laser is modulated into pulsed probe light, and is output to the first port of the optical circulator; the first optical amplifier amplifies the power of the pulsed probe light modulated by the intensity modulator, and outputs to the first port of the optical circulator;

[0040] S2, the optical circulator outputs the pulsed probe light input to the first port from the second port to the sensing optical fiber, and outputs the backscattering Rayleigh scattering light RBS signal received by the second port from the third port to the second optical amplifier;

[0041] S3, the second optical amplifier amplifies the backscattering Rayleigh scattering light RBS signal input to the third port of the optical circulator, and outputs to the first coupler; the first coupler divides the light signal input by the optical amplifier into two paths, one path is output to the second coupler through the first output port of the first coupler, and the other path is output to the optical delay line through the second output port of the first coupler; the optical delay line transmits the light signal input by the second output port of the first coupler to the third coupler, while eliminating multipath interference, realizing phase matching and time synchronization;

[0042] S4, the second coupler divides the light signal input by the first output port of the first coupler into three paths, one path is output to the first polarization beam splitter through the first output port of the second coupler, one path is output to the fifth coupler through the second output port of the second coupler, and one path is output to the second polarization beam splitter through the third output port of the second coupler; the first polarization beam splitter transmits the light signal input by the first output port of the second coupler to the fourth coupler, while generating a 2π / 3 phase shift; the second polarization beam splitter transmits the light signal input by the third output port of the second coupler to the sixth coupler, while generating a 4π / 3 phase shift; the third coupler divides the light signal input by the optical delay line into three paths, one path is output to the fourth coupler through the first output port of the third coupler, one path is output to the fifth coupler through the second output port of the third coupler, and one path is output to the sixth coupler through the third output port of the third coupler;

[0043] S5, the fourth coupler couples the light signal input by the first polarization beam splitter and the light signal input by the first output port of the third coupler, and outputs to the information conversion module; the fifth coupler couples the light signal input by the second output port of the second coupler and the light signal input by the second output port of the third coupler, and outputs to the information conversion module; the sixth coupler couples the light signal input by the second polarization beam splitter and the light signal input by the third output port of the third coupler, and outputs to the information conversion module;

[0044] S6, the information conversion module converts the light intensity signals input by the fourth coupler, the fifth coupler and the sixth coupler into phase signals and outputs to the space-time signal combination module;

[0045] S7, the space-time signal combination module integrates the multiple groups of space-time two-dimensional phase signals input by the information conversion module under the control of the photoelectric conversion and control module to form an overall space-time two-dimensional phase signal, and then transmits to the pure optical neural network module;

[0046] S8, the pure optical neural network module performs identification and classification operations on the overall space-time two-dimensional phase signal input by the space-time signal combination module, and then outputs to the photoelectric conversion and control module;

[0047] S9, the photoelectric conversion and control module receives an external trigger signal Trigger for setting a detection light pulse parameter, controlling the working state of the space-time signal combination module and the first intensity modulator, and converting the time information, position information and event information input by the pure optical neural network module into an electrical signal output.

[0048] Through the above technical solution, compared with the prior art, the present application provides a distributed acoustic wave sensing system and all-optical integration method based on an optical neural network, which has the following beneficial effects:

[0049] 1) The all-optical circuit design is adopted to eliminate a series of losses caused by photoelectric conversion, improve system stability and simplify system architecture;

[0050] 2) The pure optical neural network (ONN) is used for signal identification and classification to enhance parallel processing capability, improve data processing speed and increase data processing capacity;

[0051] 3) The all-optical integration method is adopted to reduce the overall system size and weight, reduce system power consumption, and enhance system durability and reliability. BRIEF DESCRIPTION OF DRAWINGS

[0052] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiment or prior art description. Obviously, the drawings in the following description are only embodiments of the present application, and those skilled in the art can obtain other drawings according to the provided drawings without creative labor.

[0053] Fig. 1 is a structure block diagram of a distributed acoustic wave sensing system based on an optical neural network disclosed by the present application;

[0054] Fig. 2 is a flowchart of a possible method one of the information conversion module, the space-time signal combination module and the pure optical neural network module disclosed by the present application;

[0055] Fig. 3 is a structure diagram of a three-input lookup table in a second method that can be implemented by the information conversion module, the space-time signal combination module and the pure optical neural network module of the present application;

[0056] Fig. 4 is a cross-sectional structure diagram of a non-volatile waveguide phase shifter of an FEBTO crystal integrated in an optical memory in the ONN disclosed by the present application;

[0057] Fig. 5 is an all-optical integration method of a distributed acoustic wave sensing system based on an optical neural network disclosed by the present application. DETAILED DESCRIPTION

[0058] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present application.

[0059] In the present application, the relationship terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between the entities or operations, the term "comprise", "include" or any other variant thereof is intended to cover non-exclusive inclusion, so that the process, method, article or equipment including a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such process, method, article or equipment. Without more limitations, the element defined by the statement "comprising a" does not exclude the presence of another identical element in the process, method, article or equipment including the element.

[0060] Referring to Fig. 1, the present application discloses a distributed acoustic wave sensing system based on an optical neural network, comprising: a DAS optical path integration part, an external part and a signal processing integrated chip; wherein,

[0061] The DAS optical path integration part comprises: a narrow line width laser, an intensity modulator, a first optical amplifier;

[0062] The external part comprises: a circulator, a sensing optical fiber; the first port of the circulator is denoted by a in Fig. 1, the second port of the circulator is denoted by b in Fig. 1, and the third port of the circulator is denoted by c in Fig. 1;

[0063] The signal processing integrated chip comprises: a second optical amplifier, a first coupler, an optical delay line, a second coupler, a third coupler, a first polarization beam splitter, a second polarization beam splitter, a fourth coupler, a fifth coupler, a sixth coupler, an information conversion module, a space-time signal combination module, a pure optical neural network module ONN, and an optoelectronic conversion and control module.

[0064] The first polarization beam splitter is indicated by PS1 in FIG. 1, and the second polarization beam splitter is indicated by PS2 in FIG. 1.

[0065] Further, the narrow linewidth laser, the intensity modulator, and the first optical amplifier in the DAS optical path integrated part are sequentially connected; the intensity modulator is connected with a first output end of the optoelectronic conversion and control module of the signal processing integrated chip.

[0066] The output end of the first optical amplifier is connected with a first port of the circulator in the external part.

[0067] The narrow linewidth laser is used to output continuous narrow linewidth high-coherence laser to the intensity modulator.

[0068] The intensity modulator is modulated by the optoelectronic conversion and control module, and modulates the continuous probe light input by the narrow linewidth laser into pulsed probe light, which is output to the first optical amplifier.

[0069] The first optical amplifier is used to amplify the power of the pulsed probe light modulated by the intensity modulator, and output to the first port of the circulator.

[0070] Specifically, the narrow linewidth laser has a wavelength of 1550 nm and a linewidth of 100 kHz, and due to the limited coherence length, a self-correlation structure is used for phase demodulation. This design fully considers the core components with limited performance in current optoelectronic integration.

[0071] Further, in the external part, a second port of the circulator is connected with a sensing optical fiber, and a third port of the circulator is connected with an input end of the second optical amplifier of the signal processing integrated chip.

[0072] The circulator is used to output the pulsed probe light received by the first port of the circulator from the second port of the circulator to the sensing optical fiber, and output the Rayleigh backscattering light RBS signal received by the second port of the circulator from the third port of the circulator to the second optical amplifier of the signal processing integrated chip.

[0073] The sensing optical fiber is used to receive the pulsed probe light input by the second port of the circulator, and generate the Rayleigh backscattering light RBS signal.

[0074] Further, the first coupler, the optical delay line, the second coupler, the third coupler, the first polarization beam splitter, the second polarization beam splitter, the fourth coupler, the fifth coupler and the sixth coupler in the signal processing integrated chip form a temperature control and vibration isolation module. The temperature control and vibration isolation module greatly reduces the influence of external temperature and vibration on the stability of the system.

[0075] Further, the second optical amplifier is used for amplifying the Rayleigh backscattering light RBS signal input from the third port of the circulator and outputting to the first coupler of the temperature control and vibration isolation module.

[0076] Further, the first coupler is used for splitting the optical signal input from the second optical amplifier into two paths, one path being output to the second coupler through the first output port of the first coupler, and the other path being output to the optical delay line through the second output port of the first coupler.

[0077] The optical delay line is used for transmitting the optical signal input from the second output port of the first coupler to the third coupler, eliminating multipath interference, and realizing phase matching and timing synchronization.

[0078] The second coupler is used for splitting the optical signal input from the first output port of the first coupler into three paths, one path being output to the first polarization beam splitter through the first output port of the second coupler, one path being output to the fifth coupler through the second output port of the second coupler, and one path being output to the second polarization beam splitter through the third output port of the second coupler.

[0079] The third coupler is used for splitting the optical signal input from the optical delay line into three paths, one path being output to the fourth coupler through the first output port of the third coupler, one path being output to the fifth coupler through the second output port of the third coupler, and one path being output to the sixth coupler through the third output port of the third coupler.

[0080] The first polarization beam splitter is used for transmitting the optical signal input from the first output port of the second coupler to the fourth coupler, and generating a stable 2π / 3 phase shift.

[0081] The second polarization beam splitter is used for transmitting the optical signal input from the third output port of the second coupler to the sixth coupler, and generating a stable 4π / 3 phase shift.

[0082] The fourth coupler is used for coupling the optical signal input from the first polarization beam splitter and the optical signal input from the first output port of the third coupler, and outputting to the first input port of the information conversion module.

[0083] The fifth coupler is used for coupling the optical signal input from the second output port of the second coupler and the optical signal input from the second output port of the third coupler, and outputting to the second input port of the information conversion module.

[0084] A sixth coupler is configured to couple the optical signal input from the second polarization beam splitter and the optical signal input from the third output port of the third coupler, and output to a third input port of the information conversion module.

[0085] Specifically, the length of the optical delay line needs to be selected considering the spatial resolution, system bandwidth, delay line loss, system stability and other factors. Based on the above-mentioned one kind of distributed acoustic wave sensing system full optical integration method based on optical neural network, if the spatial resolution of the system is 10 m, the pulse width of the intensity modulator should be 50 ns, and if the maximum delay time required is 100 ns, the length of the optical delay line should be 20 m. At present, the methods of optical delay line include optical waveguide delay line, optical fiber external connection and integrated delay line. The average size of the optical waveguide delay line design is basically 50 cm 2 The size of the optical fiber external connection is generally several hundred cm 3 The integrated optical delay line has a size of only a few cm 2 , but its loss is too large to be practically applied.

[0086] An optimal solution can be adopted for the design of the optical delay line in the system, that is, a micro-ring resonator is made on a low-loss silicon nitride waveguide to realize the delay. When the wavelength of the optical signal meets the resonance condition of the ring optical waveguide, the optical signal will circulate in the ring optical waveguide, thereby realizing the effect of delay. In order to reduce the loss, a plurality of micro-ring resonators are connected in series. The total length of the delay line is 20 m, and the total loss is about 240 dB, including waveguide loss, coupling loss, bending loss and amplification loss. Then, a pump light source is added, and the delay line is divided into 20 parts, and an SOA is added at the connection position of each part. Each SOA can provide a gain of 12 dB, thereby offsetting the loss and achieving the balance between gain and attenuation.

[0087] Based on the above-mentioned one kind of distributed acoustic wave sensing system full optical integration method based on optical neural network, if the spatial resolution of the system is 10 m, the pulse width of the intensity modulator should be 50 ns, and if the maximum delay time required is 100 ns, the length of the optical delay line should be 20 m. An optimal solution can be adopted for the design of the optical delay line in the system, that is, a plurality of micro-ring resonators are connected in series on the premise of silicon-based integration. Each micro-ring resonator is an erbium-doped micro-ring resonator, and the design is optimized by adjusting the pump light power and the size of the ring, so that the optical path length of each micro-ring resonator is 1 cm, the gain is 2 dB, and the total attenuation of each micro-ring resonator is 2 dB. In addition, a plurality of pump light sources are distributed in the entire delay line to ensure that each micro-ring resonator receives appropriate pump power. Finally, 2000 micro-rings are cascaded to achieve the required delay distance and actual gain-attenuation balance.

[0088] Further, the information conversion module is used for converting the optical intensity signals input by the fourth coupler, the fifth coupler and the sixth coupler into phase signals and outputting the phase signals to the space-time signal combination module;

[0089] The space-time signal combination module is controlled by the photoelectric conversion and control module, is used for integrating the multiple groups of space-time two-dimensional phase signals input by the information conversion module to form an integral space-time two-dimensional phase signal, and then transmitting the integral space-time two-dimensional phase signal to the optical neural network module ONN;

[0090] The optical neural network module ONN is used for performing recognition and classification operations on the integral space-time two-dimensional phase signal input by the space-time signal combination module and outputting the integral space-time two-dimensional phase signal to the photoelectric conversion and control module.

[0091] The photoelectric conversion and control module receives an external trigger signal Trigger, is used for setting a detection light pulse parameter, controlling the working states of the space-time signal combination module and the first intensity modulator, and converting time information, position information and event information input by the optical neural network module ONN into an electrical signal and outputting the electrical signal.

[0092] Further, the optical neural network ONN is implemented in a lookup table mode, wherein an optical memory is formed by a non-volatile waveguide phase shifter; the signal of the optical neural network module ONN is an analog optical signal, which is a time-space two-dimensional graph and uses intensity to represent phase information. The optical neural network ONN performs signal recognition and classification, enhances parallel processing capability, improves data processing speed and increases data processing capacity.

[0093] Referring to FIG. 2, a possible implementation method one of the information conversion module, the space-time signal combination module and the optical neural network module is shown, which includes the following steps:

[0094] Step one: the information conversion module and the space-time signal combination module are integrally formed by an optical neural network, wherein the input information is three analog optical intensity signals representing phases output by a three-port structure, and the output information is an analog optical intensity signal representing a phase;

[0095] Step two: the optical neural network module is formed by multiple micro-ring resonators, the micro-ring resonators are connected by couplers to realize nonlinear processing and feature extraction of signals, and finally the processed signals are classified and recognized.

[0096] Referring to FIG. 3, a possible implementation method two of the information conversion module, the space-time signal combination module and the optical neural network module is shown, which includes the following steps:

[0097] Step one: the information conversion module, the space-time signal combination module and the pure optical neural network module are realized by using a look-up-table (LUT), and the optical storage is an optical on-chip storage made of a non-volatile material. The LUT is a three-input LUT, the three analog optical signals input into the LUT represent phase information by intensity, and an analog optical intensity signal representing the optical phase is output after the LUT. The cross-sectional structure of the non-volatile waveguide phase shifter integrated with the FEBTO crystal is shown in FIG. 4, and the phase shifter can be used as a basic component of the photonic storage.

[0098] Step two: the analog optical intensity signal representing the optical phase is input into the pure optical neural network module for final classification and recognition.

[0099] Further, the first optical amplifier and the second optical amplifier are semiconductor optical amplifiers (SOAs).

[0100] Corresponding to the system described in FIG. 1, the application also discloses a full-optical integration method of a distributed acoustic wave sensing system based on an optical neural network, which is applied to any one of the distributed acoustic wave sensing systems based on the optical neural network, and the specific steps are shown in FIG. 5:

[0101] S1, the narrow linewidth laser transmits continuous narrow linewidth high coherence laser to the intensity modulator, the intensity modulator is modulated by the photoelectric conversion and control module, the continuous probe light input by the narrow linewidth laser is modulated into pulsed probe light, and output to the first optical amplifier, the first optical amplifier amplifies the power of the pulsed probe light modulated by the intensity modulator, and outputs to the first port of the circulator;

[0102] S2, the circulator outputs the pulsed probe light input into the first port from the second port to the sensing optical fiber, and outputs the backscattering Rayleigh scattering (RBS) signal received by the second port from the third port to the second optical amplifier;

[0103] S3, the second optical amplifier amplifies the backscattering Rayleigh scattering (RBS) signal input into the third port of the circulator, and outputs to the first coupler, the first coupler is a 1*2 optical coupler (the splitting ratio is 50:50), which divides the optical signal input by the optical amplifier into two paths, one path is output from the first output port of the first coupler to the second coupler, and the other path is output from the second output port of the first coupler to the optical delay line, the length of the optical delay line is 20m, which transmits the optical signal input by the second output port of the first coupler to the third coupler, and eliminates the multipath interference, realizes phase matching and time sequence synchronization;

[0104] S4, the second coupler adopts a 1*3 optical coupler (splitting ratio is 33:33:33), which divides the optical signal input from the first output port of the first coupler into three paths, one path is output to the first polarization beam splitter through the first output port of the second coupler, one path is output to the fifth coupler through the second output port of the second coupler, and one path is output to the second polarization beam splitter through the third output port of the second coupler; the first polarization beam splitter transmits the optical signal input from the first output port of the second coupler to the fourth coupler, while generating a 2π / 3 phase shift; the second polarization beam splitter transmits the optical signal input from the third output port of the second coupler to the sixth coupler, while generating a 4π / 3 phase shift; the third coupler adopts a 1*3 optical coupler (splitting ratio is 33:33:33), which divides the optical signal input from the optical delay line into three paths, one path is output to the fourth coupler through the first output port of the third coupler, one path is output to the fifth coupler through the second output port of the third coupler, and one path is output to the sixth coupler through the third output port of the third coupler;

[0105] S5, the fourth coupler couples the optical signal input from the first polarization beam splitter and the optical signal input from the first output port of the third coupler, and outputs to the information conversion module; the fifth coupler couples the optical signal input from the second output port of the second coupler and the optical signal input from the second output port of the third coupler, and outputs to the information conversion module; the sixth coupler couples the optical signal input from the second polarization beam splitter and the optical signal input from the third output port of the third coupler, and outputs to the information conversion module;

[0106] S6, the information conversion module converts the optical intensity signals input from the fourth coupler, the fifth coupler and the sixth coupler into phase signals, and outputs to the space-time signal combination module;

[0107] S7, the space-time signal combination module is controlled by the photoelectric conversion and control module, integrates the multiple groups of space-time two-dimensional phase signals input from the information conversion module, forms an overall space-time two-dimensional phase signal, and then transmits to the pure optical neural network module ONN;

[0108] S8, the pure optical neural network module ONN performs identification, classification and other operations on the overall space-time two-dimensional phase signal input from the space-time signal combination module, and then outputs to the photoelectric conversion and control module;

[0109] S9, the photoelectric conversion and control module receives an external trigger signal Trigger, which is used to set the detection light pulse parameters, control the working state of the space-time signal combination module and the first intensity modulator, and convert the time, position and event type information input from the pure optical neural network module ONN into an electrical signal output.

[0110] The various embodiments described in this specification are described with reference to a particular sequence or order, but the order of the steps can be modified so that particular sequences or orders make no significant contribution to the progress of the art. Moreover, certain features and subcombinations are of utility and can be employed without reference to other features and subcombinations. For purposes of clarity, not every embodiment or feature in this specification is described or shown. Embodiments that provide real benefits can include any embodiment or combination of features described in this specification— even if the range of benefits realized is not the full range of benefits. Those of ordinary skill can understand that information and signals can be represented using any of a variety of technologies and techniques. For the purposes of this description, the terms "information" and "signals" can be regarded as synonymous. Those of ordinary skill can appreciate that the signals can be analog or digital, and the like.

[0111] The foregoing description of the disclosed embodiments enables one of ordinary skill in the art to make or utilize the application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein can be applied to other embodiments without the use of the inventive faculty. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. An optical neural network based distributed acoustic sensing system, characterized in that, The system comprises a DAS optical path integration part, an external part, and a signal processing integrated chip, wherein The DAS optical path integration part comprises a narrow linewidth laser, an intensity modulator, and a first optical amplifier; The external part comprises a circulator and a sensing optical fiber; The signal processing integrated chip comprises a second optical amplifier, a first coupler, an optical delay line, a second coupler, a third coupler, a first polarization beam splitter, a second polarization beam splitter, a fourth coupler, a fifth coupler, a sixth coupler, an information conversion module, a space-time signal combination module, a pure optical neural network module, and an optoelectronic conversion and control module.

2. The distributed acoustic sensing system based on an optical neural network according to claim 1, wherein The narrow linewidth laser, the intensity modulator, and the first optical amplifier in the DAS optical path integration part are connected in sequence; the intensity modulator is connected to a first output end of the optoelectronic conversion and control module of the signal processing integrated chip; An output end of the first optical amplifier is connected to a first port of the circulator of the external part; The narrow linewidth laser is configured to output continuous narrow linewidth high-coherence laser to the intensity modulator; The intensity modulator is modulated by the optoelectronic conversion and control module, and is configured to modulate the continuous probe light input by the narrow linewidth laser into pulsed probe light and output the pulsed probe light to the first optical amplifier; The first optical amplifier is configured to amplify the power of the pulsed probe light modulated by the intensity modulator and output the pulsed probe light to the first port of the circulator.

3. The distributed acoustic sensing system based on an optical neural network according to claim 2, wherein In the external part, a second port of the circulator is connected to the sensing optical fiber, and a third port of the circulator is connected to an input end of the second optical amplifier of the signal processing integrated chip; The circulator is configured to output the pulsed probe light received by the first port of the circulator from the second port of the circulator to the sensing optical fiber, and output the Rayleigh backscattering (RBS) signal received by the second port of the circulator from the third port of the circulator to the second optical amplifier of the signal processing integrated chip; The sensing optical fiber is configured to receive the pulsed probe light input by the second port of the circulator and generate the RBS signal.

4. The distributed acoustic sensing system based on an optical neural network according to claim 3, wherein The first coupler, the optical delay line, the second coupler, the third coupler, the first polarization beam splitter, the second polarization beam splitter, the fourth coupler, the fifth coupler, and the sixth coupler in the signal processing integrated chip constitute a temperature-controlled vibration isolation module.

5. The distributed acoustic sensing system based on an optical neural network according to claim 4, wherein The second optical amplifier is configured to amplify the RBS signal input by the third port of the circulator and output the RBS signal to the first coupler of the temperature-controlled vibration isolation module.

6. The distributed acoustic sensing system based on an optical neural network according to claim 5, wherein ​ a first coupler, configured to split the optical signal input from the second optical amplifier into two paths, one path output from a first output port of the first coupler to a second coupler, and the other path output from a second output port of the first coupler to an optical delay line; the optical delay line, configured to transmit the optical signal input from the second output port of the first coupler to a third coupler, while eliminating multipath interference, achieving phase matching and timing synchronization; a second coupler, configured to split the optical signal input from the first output port of the first coupler into three paths, one path output from a first output port of the second coupler to a first polarization beam splitter, one path output from a second output port of the second coupler to a fifth coupler, and one path output from a third output port of the second coupler to a second polarization beam splitter; a third coupler, configured to split the optical signal input from the optical delay line into three paths, one path output from a first output port of the third coupler to a fourth coupler, one path output from a second output port of the third coupler to the fifth coupler, and one path output from a third output port of the third coupler to a sixth coupler; the first polarization beam splitter, configured to transmit the optical signal input from the first output port of the second coupler to the fourth coupler, while generating a 2π / 3 phase shift; the second polarization beam splitter, configured to transmit the optical signal input from the third output port of the second coupler to the sixth coupler, while generating a 4π / 3 phase shift; the fourth coupler, configured to couple the optical signal input from the first polarization beam splitter and the optical signal input from the first output port of the third coupler, and output to a first input port of an information conversion module; the fifth coupler, configured to couple the optical signal input from the second output port of the second coupler and the optical signal input from the second output port of the third coupler, and output to a second input port of the information conversion module; the sixth coupler, configured to couple the optical signal input from the second polarization beam splitter and the optical signal input from the third output port of the third coupler, and output to a third input port of the information conversion module.

7. The distributed acoustic sensing system based on optical neural network according to claim 6, wherein the information conversion module, configured to convert the optical intensity signals input from the fourth coupler, the fifth coupler and the sixth coupler into phase signals, and output to a space-time signal combination module; the space-time signal combination module, controlled by the optoelectronic conversion and control module, configured to integrate the multiple groups of two-dimensional phase signals input from the information conversion module, to form an overall two-dimensional phase signal, and then transmit to a pure optical neural network module; the pure optical neural network module, configured to perform identification and classification operations on the overall two-dimensional phase signal input from the space-time signal combination module, and output to the optoelectronic conversion and control module; the optoelectronic conversion and control module, receiving an external trigger signal Trigger, configured to set the detection light pulse parameters, control the working states of the space-time signal combination module and the first intensity modulator, and convert the time information, position information and event information input from the pure optical neural network module into electrical signals.

8. The distributed acoustic sensing system based on optical neural network according to claim 1, wherein The pure optical neural network is implemented by using a lookup table, and an optical memory is composed of non-volatile waveguide phase shifters; the signal of the pure optical neural network module is an analog optical signal, which is a time-space two-dimensional graph using intensity to represent phase information.

9. The distributed acoustic sensing system based on optical neural network according to claim 1, wherein, The first optical amplifier and the second optical amplifier are semiconductor optical amplifiers.

10. An all-optical integration method of a distributed acoustic sensing system based on an optical neural network, characterized in that, The distributed acoustic sensing system based on optical neural network is applied to any one of claims 1-9, comprising the following steps: S1, the narrow linewidth laser transmits continuous narrow linewidth high coherence laser to the intensity modulator, the intensity modulator is modulated by the photoelectric conversion and control module, the continuous probe light input by the narrow linewidth laser is modulated into pulsed probe light, and output to the first port of the circulator; the first optical amplifier amplifies the power of the pulsed probe light modulated by the intensity modulator, and outputs to the first port of the circulator; S2, the circulator outputs the pulsed probe light input to the first port from the second port to the sensing optical fiber, and outputs the backscattering Rayleigh scattering light RBS signal received by the second port from the third port to the second optical amplifier; S3, the second optical amplifier amplifies the backscattering Rayleigh scattering light RBS signal input by the third port of the circulator, and outputs to the first coupler; the first coupler divides the optical signal input by the optical amplifier into two paths, one path is output to the second coupler through the first output port of the first coupler, and the other path is output to the optical delay line through the second output port of the first coupler; the optical delay line transmits the optical signal input by the second output port of the first coupler to the third coupler, while eliminating multipath interference, realizing phase matching and time synchronization; S4, the second coupler divides the optical signal input by the first output port of the first coupler into three paths, one path is output to the first polarization beam splitter through the first output port of the second coupler, one path is output to the fifth coupler through the second output port of the second coupler, and one path is output to the second polarization beam splitter through the third output port of the second coupler; the first polarization beam splitter transmits the optical signal input by the first output port of the second coupler to the fourth coupler, while generating a 2π / 3 phase shift; the second polarization beam splitter transmits the optical signal input by the third output port of the second coupler to the sixth coupler, while generating a 4π / 3 phase shift; the third coupler divides the optical signal input by the optical delay line into three paths, one path is output to the fourth coupler through the first output port of the third coupler, one path is output to the fifth coupler through the second output port of the third coupler, and one path is output to the sixth coupler through the third output port of the third coupler; ​ S5, the fourth coupler couples the optical signal input by the first polarization beam splitter and the optical signal input by the first output port of the third coupler, and outputs to the information conversion module; the fifth coupler couples the optical signal input by the second output port of the second coupler and the optical signal input by the second output port of the third coupler, and outputs to the information conversion module; the sixth coupler couples the optical signal input by the second polarization beam splitter and the optical signal input by the third output port of the third coupler, and outputs to the information conversion module; S6, the information conversion module converts the optical intensity signals input by the fourth coupler, the fifth coupler and the sixth coupler into phase signals, and outputs to the space-time signal combination module; S7, the space-time signal combination module controls the multiple groups of space-time two-dimensional phase signals input by the information conversion module to integrate into an overall space-time two-dimensional phase signal under the control of the photoelectric conversion and control module, and then transmits to the pure optical neural network module; S8, the pure optical neural network module performs identification and classification operations on the overall space-time two-dimensional phase signal input by the space-time signal combination module, and then outputs to the photoelectric conversion and control module; S9, the photoelectric conversion and control module receives an external trigger signal Trigger, which is used to set the detection light pulse parameters, control the working state of the space-time signal combination module and the first intensity modulator, and convert the time information, position information and event information input by the pure optical neural network module into electrical signals for output.

Citation Information

Patent Citations

  • Distributed acoustic wave pipeline online monitoring system

    CN110185938A

  • Photon neural network chip

    CN110247728A

  • Fiber-optic distributed acoustic wave sensing system

    CN110440900A

  • Intelligent optical fiber distributed acoustic wave sensing system based on AI chip and method

    CN110487391A

  • Photonic neural network training method based on forward propagation

    CN110956259A