Cable joint partial discharge on-line monitoring probe, system and method

Abstract

The invention discloses a cable joint partial discharge on-line monitoring probe, system and method, and the system comprises a sensing optical fiber link which comprises a transmission optical fiber and a plurality of monitoring probes which are connected in series with the transmission optical fiber; the light emitting and modulating module generates and modulates narrow linewidth laser, generates a frequency division multiplexing chirped pulse light sequence and injects the frequency division multiplexing chirped pulse light sequence into a sensing optical fiber link; the polarization diversity receiving module is used for receiving the backward Rayleigh scattering light from the sensing optical fiber link, decomposing the backward Rayleigh scattering light into two paths of orthogonal linearly polarized light, respectively performing coherent detection on the two paths of orthogonal linearly polarized light and local oscillation light, and outputting two paths of electric signals containing orthogonal polarization components; and the data acquisition and processing module is used for carrying out matched filtering on the electric signal to generate a distance-response curve, setting a distance gate according to the physical position of the monitoring probe in the sensing optical fiber link, and extracting and demodulating the phase signal at each probe position. According to the invention, high-sensitivity, high-stability and distributed accurate monitoring of partial discharge signals of the cable intermediate joint can be realized.

Description

Technical Field

[0001] This invention belongs to the field of online monitoring technology for power equipment, and specifically relates to an online monitoring probe, system, and method for partial discharge of cable joints. Background Technology

[0002] Distribution cables are the lifeblood of urban power networks, and cable joints, as field-fabricated accessories in cable lines, are often high-risk sites for insulation failures due to their complex structure and demanding manufacturing processes. Statistics show that a large number of cable faults originate from insulation deterioration or breakdown at cable joints. Therefore, real-time and effective partial discharge (PD) monitoring of cable joints is of great significance for ensuring the safe operation of the power grid.

[0003] Because cables are typically laid in complex environments such as underground utility tunnels and cable trenches, they face challenges such as limited space, harsh environments, and a lack of power supply and communication facilities. Traditional monitoring methods mainly include offline oscillatory wave detection or online high-frequency current transformer (HFCT) and ultrasonic methods. However, offline detection cannot capture instantaneous faults, and traditional point-based online monitoring sensors are usually active devices that require on-site power extraction and are difficult to deploy on a large scale over long distances, resulting in high maintenance costs.

[0004] In recent years, distributed optical acoustic sensing (DAS) technology based on phase-sensitive optical time-domain reflectometry (Φ-OTDR) has been gradually applied to the condition monitoring of power cables due to its inherent safety, resistance to electromagnetic interference, lack of need for on-site power supply, and ability to achieve long-distance continuous monitoring.

[0005] For example, Chinese invention patent application CN115267457A discloses a multi-phase encoded φ-OTDR cable partial discharge monitoring system and its operating method. This scheme proposes a φ-OTDR system based on multi-phase encoding and time-frequency division multiplexing technology. It utilizes a narrow-linewidth laser as a light source, drives an electro-optic modulator through a time-frequency division multiplexed signal, and combines it with an acousto-optic modulator to generate multi-frequency, multi-phase encoded pulses. The system continuously and tightly winds sensing optical fibers around the entire length of the cable under test, and achieves distributed monitoring of partial discharge along the cable by demodulating the phase change of the backscattered Rayleigh light in the optical fiber.

[0006] While the aforementioned existing technologies have addressed the issue of long-distance monitoring to some extent, significant technical bottlenecks remain in practical applications for partial discharge monitoring of cable joints: The method of densely and continuously winding the sensing fiber around the entire cable body is difficult to implement, especially in narrow cable trenches. Furthermore, continuous winding makes it challenging to ensure a tight fit between the fiber and the joint surface, leading to uneven acoustic-optical coupling efficiency. Heterodyne coherent detection structures are affected by temperature changes, mechanical vibrations, or fiber bending, causing random birefringence changes in the signal light during transmission. This results in uncontrollable drift in the polarization state of the returned Rayleigh scattered light, affecting the long-term stability and alarm accuracy of the system. Continuous monitoring of the entire sensing fiber in this approach places a heavy data processing burden and results in unfocused targets, impacting the accuracy and efficiency of fault identification. Summary of the Invention

[0007] This invention provides an online monitoring probe, system, and method for partial discharge in cable joints, aiming to solve the problems of engineering deployment difficulties, low acoustic-optical coupling efficiency due to poor contact, unstable monitoring signals due to polarization fading, and difficulty in achieving both high signal-to-noise ratio and high spatial resolution in the application of existing distributed optical fiber sensing technology for cable joint monitoring.

[0008] To address the aforementioned technical problems, this invention proposes an online monitoring probe for partial discharge in cable joints, comprising: The strip probe body is formed by at least one tight-buffered optical fiber arranged in a continuous S-shape in a plane, and multiple reciprocating segments of the tight-buffered optical fiber are closely arranged in the width direction. The sealing layer is made of a flexible material that is resistant to high temperature and has high electrical insulation properties. It is used to seal the multi-turn tightly wrapped optical fiber as a whole, so that the strip probe body has a flat and bendable strip structure. The transmission port includes an optical fiber lead or connector extending from the sealing layer and connected to the tight-buffered optical fiber, for connecting the strip probe body in series to an external sensing optical path. The strip probe body is configured to surround and fit the outer surface of the cable intermediate joint, so that multiple tightly wrapped optical fibers form a sensing plane covering the outer periphery of the cable intermediate joint, which is used to sense the partial discharge signal generated at any position of the cable intermediate joint.

[0009] Preferably, the spacing of the multi-turn tight-buffered optical fibers around the outer periphery of the cable intermediate joint is configured such that the equivalent axial winding density after probe installation is not less than 50 turns / meter. The inner surface of the strip probe body is coated with an acoustic coupling layer, which is composed of silicone grease or polymer gel and is used to fill the microscopic gaps between the probe and the surface of the cable intermediate joint.

[0010] Preferably, the transmission port includes two leads that are respectively connected to both ends of the tight-buffered optical fiber, and the two leads extend from the same longitudinal end of the sealing layer.

[0011] A second aspect of the present invention also provides an online monitoring system for partial discharge of cable joints, comprising: A sensing fiber optic link includes a transmission fiber and a plurality of monitoring probes as described in the first aspect of the present invention, which are connected in series on the transmission fiber and are disposed on the outer surface of the cold shrink sleeve of the cable intermediate joint. The optical emission and modulation module is configured to generate and modulate narrow-linewidth laser light to generate a frequency-division multiplexed chirped pulse light sequence and inject it into the sensing fiber optic link. The polarization diversity receiving module is configured to receive back Rayleigh scattered light from the sensing fiber optic link, decompose it into two orthogonal linearly polarized lights, coherently detect them with the local oscillator light, and output two electrical signals containing orthogonal polarization components. The data acquisition and processing module is configured to use the frequency division multiplexing chirped coded signal as a reference to perform matched filtering on the two electrical signals to generate a distance-response curve, and to set a distance gate according to the physical location of the cable joint partial discharge online monitoring probe in the sensing fiber optic link to extract and demodulate the phase signal at each probe location.

[0012] Preferably, the light emitting and modulation device includes: Narrow linewidth laser module, configured to output continuous coherent light, and utilizes feedforward phase noise compensation technology to compress linewidth; An acousto-optic modulator receives the continuous coherent light; An arbitrary waveform generator is configured to generate the frequency division multiplexing chirped coded signal to drive the acousto-optic modulator; The frequency division multiplexing chirped coded signal contains multiple linear frequency modulated pulse subbands that are orthogonal in the frequency domain and overlap in the time domain.

[0013] Preferably, the polarization diversity receiver includes: The first beam splitter, the second beam splitter, and the third beam splitter are used. The first beam splitter splits the received local oscillator light into two paths, which are then sent to the second beam splitter and the third beam splitter, respectively. The polarization beam splitter receives the signal light returned from the sensing fiber optic link and decomposes it into two mutually orthogonal polarization states, X and Y, which are then sent to the second and third beam splitters, respectively. The polarization state X interferes with one local oscillator beam in the second beam splitter, resulting in two beams with a phase difference of 180°. The polarization state Y interferes with one local oscillator light in the third beam splitter, resulting in two beams with a phase difference of 180°. Two photoelectric balanced detectors receive two beams from the second and third beam splitters, respectively, and output polarization states S and P after differential amplification.

[0014] Preferably, the signal processing device is specifically configured as follows: By using the frequency division multiplexing chirped coded signal at the transmitting end as a reference signal, cross-correlation operation is performed with the two received electrical signals to achieve pulse compression and obtain a distance-response curve with a high signal-to-noise ratio; On the distance-response curve, a time window is set as a distance gate corresponding to the physical location of each monitoring probe; Phase demodulation is performed only on data falling within the distance gate to separate probe responses from different frequency channels and spatial locations.

[0015] A third aspect of the present invention also provides an online monitoring method for partial discharge of cable joints, the method being based on the detection system described in the second aspect of the present invention, comprising the following steps: A frequency-division multiplexed chirped pulse light sequence is generated by a control optical emission and modulation device and enters the sensing optical fiber link; wherein, multiple monitoring probes in the sensing optical fiber link have been tightly wound and fixed to the outer surface of different intermediate joints of the cable to be monitored. Back-scattered Rayleigh light is collected using a polarization diversity receiver, and two baseband signals containing orthogonal polarization states are output through polarization beam splitting and dual-balanced detection. Using the frequency division multiplexing chirped coded signal as a reference template, the baseband signal is subjected to matched filtering and range gating to separate the vibration signal segments corresponding to each monitoring probe; The separated vibration signal segments are subjected to orthogonal demodulation and polarization synthesis to restore the ultrasonic waveform generated by partial discharge.

[0016] Preferably, the matched filtering and distance gate selection specifically include: The received baseband signal is cross-correlated with the original frequency division multiplexing chirped coded signal generated at the transmitter. The autocorrelation characteristics of the chirped pulse are used to achieve pulse compression, and the scattered signals that overlap due to frequency division multiplexing in optical fiber transmission are separated in the frequency domain and time domain. After obtaining the distance-response curve generated by cross-correlation calculation, an electronic distance gate with a predetermined time width is set on the distance-response curve according to the pre-calibrated physical position of each monitoring probe on the sensing fiber optic link. Data outside the electronic distance gate is filtered out, and only the data inside the gate is retained as the effective vibration signal segment of the corresponding probe.

[0017] Preferably, the method for orthogonal demodulation and polarization synthesis includes: Complex signals corresponding to two orthogonal polarization states are constructed based on the two baseband signals respectively; The instantaneous phase angle of each complex signal is calculated using the arctangent function, and a phase unwinding algorithm is used to recover the continuous absolute phase signal by detecting and compensating for phase jumps. Calculate the instantaneous power of the complex signals of orthogonal polarization states in real time, and perform any of the following operations to generate the final phase output: select the absolute phase signal corresponding to the polarization state with the larger instantaneous power as the output, or perform weighted synthesis of the two absolute phase signals according to the ratio of the instantaneous power of the two polarization states.

[0018] Compared with the prior art, the present invention has the following technical effects: 1. The monitoring probe proposed in this invention employs a serpentine (S-shaped) reciprocating arrangement of tightly-buffered optical fibers within a flexible substrate. Compared to traditional simple linear or low-density winding, this structure creates a high-density sensing plane on the cable joint surface. Regardless of the location of partial discharge within the joint, the resulting weak mechanical vibration signals can be effectively captured by multiple closely adjacent optical fiber segments on this sensing plane, completely eliminating monitoring blind spots. Combined with a dedicated acoustic coupling layer filling the microscopic gaps, this significantly reduces sound wave reflection loss at the contact interface, achieving high-sensitivity sensing of weak partial discharge signals.

[0019] 2. The monitoring probe proposed in this invention adopts a structural design where the optical signal transmission port is led out from the same side. In cable trenches or manholes with narrow spaces and complex environments, construction personnel only need to complete the winding and installation of the probe and the fusion splicing of the optical path from one side of the cable joint, without having to operate across both ends of the joint, greatly reducing construction difficulty and time costs. At the same time, the probe is connected to the trunk optical cable in a bypass series configuration, avoiding long-distance exposure and crossing of the transmission optical fiber on the joint surface, effectively reducing the risk of communication interruption due to external mechanical damage.

[0020] 3. The monitoring system proposed in this invention introduces a polarization diversity reception mechanism. Through the cooperation of a polarization beamsplitter and a dual-balanced detector, randomly polarized backscattered light is decomposed into two orthogonal linearly polarized components for separate detection. Regardless of how the polarization state of the optical fiber drifts due to environmental disturbances, the system can always obtain a stable signal output through selective or vector synthesis methods. This fundamentally eliminates the phenomenon of signal fluctuations or even loss caused by polarization mismatch, significantly improving the reliability and alarm accuracy of the online monitoring system.

[0021] 4. The monitoring system proposed in this invention employs frequency division multiplexing chirped pulse coding, combined with matched filtering at the receiver. On the one hand, chirped pulses allow for a significant increase in pulse width without sacrificing spatial resolution, thereby achieving a signal-to-noise ratio far exceeding that of traditional OTDRs, greatly enhancing the detection capability for weak partial discharge signals at the nanosecond level. On the other hand, frequency division multiplexing technology enables parallel processing of signals in different frequency bands, effectively separating temporally overlapping scattered signals and improving spatial positioning accuracy and anti-crosstalk capability when multiple connectors are monitored simultaneously.

[0022] 5. The monitoring method proposed in this invention adopts a distributed architecture of transmission optical fiber + discrete probes. By setting an electronic distance gate in the signal processing flow, the system extracts valid data only for the physical location of each probe, automatically filtering out invalid background noise data on long-distance transmission optical fibers. This not only significantly reduces the data throughput load of the back-end signal processing unit, enabling the system to concentrate computing resources on the refined analysis and pattern recognition of complex partial discharge characteristics, but also effectively avoids false alarms due to environmental noise, achieving efficient and focused intelligent monitoring. Attached Figure Description

[0023] Figure 1 This is a schematic diagram of the monitoring probe described in an embodiment of the present invention; Figure 2 This is a schematic diagram of the monitoring system described in an embodiment of the present invention; Figure 3 This is a schematic diagram of the chirped pulse signal described in an embodiment of the present invention; Figure 4 This is a schematic diagram of the polarization diversity receiving device according to an embodiment of the present invention; Figure 5 This is a flowchart illustrating the monitoring method described in an embodiment of the present invention; Figure 6 This is a schematic diagram of the probe deployment method described in an embodiment of the present invention.

[0024] Reference numerals in the attached figures: 1. Sensing fiber optic link; 2. Optical circulator; 3. Fiber optic amplifier; 4. Acousto-optic modulator; 5. Coupler; 6. Narrow linewidth laser module; 7. Polarization diversity receiver; 8. Data acquisition and processing module; 9. Radio frequency amplifier. Detailed Implementation

[0025] To make the objectives, technical solutions, and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below in conjunction with specific embodiments of the present application and with reference to the accompanying drawings.

[0026] Before detailing the embodiments of this application, the monitoring principle involved in the embodiments of this application will first be explained: The transient acoustic pressure wave p(t) generated by partial discharge (PD) acts on the sensitive fiber segment close to the cable joint, causing axial strain ε(t) in the fiber. This strain simultaneously changes the fiber length L and the effective refractive index n of the fiber core, both of which together affect the phase of the light wave. The situation changes. The relationship can be described as follows:

[0027] in, Let be the wavelength of light in a vacuum. For a given optical fiber and wavelength, what is the phase change? With strain Proportional, that is Therefore, high-precision measurement of optical phase changes is crucial. This allows us to reverse the original waveform of the emitted sound wave.

[0028] Accordingly, the ultrasonic waves generated by partial discharge at the cable joint induce radial strain in the sensing optical fiber through structural coupling, thereby modulating the phase of the probe light in the fiber. The core function of the polarization diversity receiver is to stably convert the optical signal carrying this phase modulation information, but with randomly changing polarization state, into an electrical signal. Subsequent digital signal processing is responsible for accurately reconstructing the phase changes from these electrical signals and identifying the partial discharge characteristics.

[0029] Example 1 This embodiment is an online monitoring probe for partial discharge at cable joints. To achieve high sensitivity, omnidirectional capture, and precise positioning of discharge signals on the surface of cable joints, this embodiment uses a tightly coiled and sealed optical fiber to form a planar strip-shaped sensing probe. Figure 1 As shown, this design aims to solve the engineering problems of uneven signal coupling, monitoring blind spots, and inconvenient installation when traditional straight optical fibers are laid out on irregular surfaces at the joints.

[0030] Specifically, the monitoring probe described in this embodiment includes: The strip probe body is formed by at least one tight-buffered optical fiber arranged in a continuous S-shape in a plane, and multiple reciprocating segments of the tight-buffered optical fiber are closely arranged in the width direction. The sealing layer is made of a flexible material that is resistant to high temperature and has high electrical insulation properties. It is used to seal the multi-turn tightly wrapped optical fiber as a whole, so that the strip probe body has a flat and bendable strip structure. The transmission port includes an optical fiber lead or connector extending from the sealing layer and connected to the tight-buffered optical fiber, for connecting the strip probe body in series to an external sensing optical path. The strip probe body is configured to surround and fit the outer surface of the cable intermediate joint, so that multiple tightly wrapped optical fibers form a sensing plane covering the outer periphery of the cable intermediate joint, which is used to sense the ultrasonic signal accompanied by partial discharge generated at any position of the cable intermediate joint.

[0031] correspond Figure 1 The black lines represent the tight-buffered optical fiber wrapped in the sealing layer, while the red areas represent fiber management mechanisms, such as fiber clips or tape, used to organize the tight-buffered optical fiber into a planar strip structure. The optical fiber between the two fiber management mechanisms forms the sensing plane in this embodiment. The outer portions of the two fiber management mechanisms are the fiber folding areas, primarily designed to control the bending radius of the optical fiber within a preset range. This prevents mechanical damage and additional optical loss, ensuring the stability and reliability of optical signal transmission, extending the fiber's lifespan, and improving the overall safety and stability of the system. Figure 1 The upper and lower leads on the left side are connected to transmission ports, which connect the strip probe body (or monitoring probe) in series to the external sensing optical path.

[0032] The sealing layer encapsulates the strip probe body. Its flat, flexible strip structure allows it to fit snugly against the irregular, curved surfaces of cable joints, like a strap, enabling quick and stable installation via binding or adhesive. This ensures optimal contact and vibration transmission path between the sensing unit and the discharge signal source, maximizing signal coupling efficiency. Crucially, this strip structure essentially forms a high-density sensing plane. When partial discharge occurs at any point on the cable joint, the resulting ultrasonic signal will be sensed by at least several closely adjacent optical fibers on this plane. This not only effectively eliminates potential blind spots in linear deployments but also significantly improves the system's confidence in identifying weak discharge signals, its anti-interference capability, and the three-dimensional spatial positioning accuracy of the discharge source through spatial correlation and comparative analysis of multiple signals.

[0033] In this embodiment, the spacing of the multi-turn tight-buffered optical fibers around the cable intermediate joint is configured such that the equivalent axial winding density after probe installation is not less than 50 turns / meter.

[0034] The inner surface of the strip probe body is coated with an acoustic coupling layer, which is composed of silicone grease or polymer gel and is used to fill the microscopic gaps between the probe and the surface of the cable intermediate joint.

[0035] The transmission port includes two leads that are respectively connected to both ends of the tight-buffered optical fiber, and both leads extend from the same longitudinal end of the sealing layer. This design, with both leads extending from the same longitudinal end, simplifies the construction process for on-site personnel when using monitoring probes to wrap around cable joints and connect the monitoring probes to the transmission optical cable, making it particularly suitable for cable well environments with limited space.

[0036] Example 2 This embodiment is an online monitoring system for partial discharge of cable joints, such as... Figure 2 As shown, the blue lines represent optical paths and the black lines represent electrical circuits. This system includes: The sensing fiber optic link 1 includes a transmission fiber and multiple monitoring probes, as described in Embodiment 1, connected in series on the transmission fiber. The monitoring probes are disposed on the outer surface of the cold-shrink tubing of the cable joint. The cable joint consists of a conductor, a main insulation layer, a semi-conductive layer, a cold-shrink tubing, and an outer sheath from the inside out. If the sensing probe is directly installed inside the cold-shrink tubing (i.e., in contact with the semi-conductive layer), although the signal can be sensed to the greatest extent, it will directly damage the critical insulation structure of the joint, introducing unacceptable safety risks and long-term operational hazards. Conversely, if the probe is installed on the outermost outer sheath surface, the ultrasonic signal generated by the discharge needs to pass through multiple layers of media such as the cold-shrink tubing and the outer sheath, and the sound wave energy will be severely attenuated and distorted, resulting in an extremely weak effective signal that is difficult to detect.

[0037] This location offers the following key advantages: Without damaging the main insulation, the probe does not come into contact with the internal semiconductive layer and the main insulation, thus fully maintaining the original insulation integrity and electrical safety performance of the cable joint; Signal attenuation is minimized at this location, which is closest to the discharge point (usually located at the main insulation or interface) and only requires passing through a single layer of cold shrink tubing (whose acoustic impedance is relatively matched to the insulation material), thus preserving the strength and characteristics of the ultrasonic signal to the greatest extent.

[0038] To achieve optimal coupling, during installation, the flexible strip probe is tightly wrapped and fixed along the surface of the cold-shrink tubing, and a special acoustic emission coupling agent (such as silicone grease or polymer gel) is evenly applied to its contact interface. This coupling agent effectively fills the microscopic gaps between the fiber optic probe and the tubing surface, significantly reducing the reflection loss of sound waves at the interface and ensuring that vibration energy is efficiently and without distortion transmitted to the sensing fiber.

[0039] Material properties enhance sensing; cold-shrink tubing is typically made of silicone rubber, which has a relatively low Young's modulus (usually in the range of 1-10 MPa). This means that under the action of partial discharge acoustic vibration signals, the same stress will produce stronger deformation on the surface of the tubing, thereby providing a more significant mechanical vibration signal for the fiber optic probe attached to it, further improving the sensitivity and signal-to-noise ratio of the sensing system.

[0040] The optical emission and modulation module is configured to generate and modulate a narrow-linewidth laser to produce a frequency-division multiplexed chirped pulse light sequence injected into the sensing fiber optic link. Specifically, it includes a narrow-linewidth laser module 6 (NLL), a coupler (5), an acousto-optic modulator 4 (AOM), an arbitrary waveform generator (AWG), and an RF amplifier 9 positioned between the arbitrary waveform generator and the acousto-optic modulator 4. The arbitrary waveform generator (AWG) is used to generate a parasitic frequency-division multiplexed linear frequency-modulated (chirped) RF pulse sequence as a driving signal. This sequence contains multiple chirped pulse subbands that are orthogonal in the frequency domain and overlappable in the time domain. The RF amplifier 9 amplifies the highly precise and complex signal generated by the front-end AWG, providing sufficient energy to drive the AOM. After this signal is loaded into the AOM, it modulates the input continuous light into a frequency-division multiplexed chirped light pulse sequence at high speed and with a high extinction ratio. Subsequently, a matched filter is used to separate the time-domain fiber response of each pulse.

[0041] The polarization diversity receiver module 7 (PDR) is configured to receive backscattered Rayleigh light from the sensing fiber optic link and decompose it into two orthogonal linearly polarized lights, which are then coherently detected with the local oscillator light to output two electrical signals containing orthogonal polarization components.

[0042] The data acquisition and processing module 8 is configured to use the frequency division multiplexing chirped coded signal as a reference to perform matched filtering on the two electrical signals to generate a distance-response curve, and to set a distance gate according to the physical location of the cable joint partial discharge online monitoring probe in the sensing fiber optic link to extract and demodulate the phase signal at each probe location. Specifically, it includes a data acquisition DAQ and a host computer. The data acquisition DAQ samples the analog waveform output by the PDR, converts it into a digital sequence, and sends the digital sequence to the host computer (such as a PC) for a series of digital signal processing steps.

[0043] Specifically, the light emitting and modulation device includes: Narrow-linewidth laser module 6 is configured to output continuous coherent light and utilizes feedforward phase noise compensation technology to compress the linewidth. In this embodiment, after system startup, narrow-linewidth laser module 6 generates a beam of continuous coherent light with a center wavelength of 1550 nm. To obtain a high-coherence light source at a lower cost, this embodiment employs feedforward laser linewidth compression technology. This continuous light first enters an external phase modulator, where a controller drives the modulator to perform feedforward compensation based on a prediction of the laser's inherent phase noise. This actively compresses the laser output linewidth from the MHz level to the kHz level, significantly improving the coherence length of the light source and laying the foundation for subsequent high-precision phase demodulation.

[0044] An arbitrary waveform generator (AWG) is configured to generate the frequency division multiplexing chirped coded signal to drive the acousto-optic modulator 4.

[0045] Acousto-optic modulator 4 and coupler 5 receive the continuous coherent light; the narrow-linewidth continuous light, after linewidth compression, enters the acousto-optic modulator AOM by 90% after passing through coupler 5. Arbitrary waveform generator (AWG) generates a set of frequency-division multiplexed linear frequency-modulated (chirped) radio frequency pulse sequences as drive signals, as shown in the following diagram. Figure 3 As shown, this sequence contains multiple chirped pulse subbands that are orthogonal in the frequency domain and overlappable in the time domain. After this signal is loaded into the AOM, the input continuous light is modulated into a frequency-division multiplexed chirped optical pulse sequence with high speed and high extinction ratio. Subsequently, a matched filter is used to separate the time-domain fiber response of each pulse.

[0046] In this embodiment, since the power of the modulated optical pulse is low, it needs to be amplified by an erbium-doped fiber amplifier 3 to obtain sufficient probe light power for injection into the sensing fiber link 1. The amplified optical pulse enters the system through the first port, Port 1, of an optical circulator 2. The optical circulator 2 is a three-port non-reciprocal device, and light can only be transmitted along the directions of Port 1→Port 2 and Port 2→Port 3.

[0047] An optical pulse is output from Port 2 of optical circulator 2 and injected into sensing fiber optic link 1. As the pulse propagates forward in the fiber, it generates weak backscattered Rayleigh light at each location. When the pulse reaches the fiber optic sensing probe specifically positioned at the cable connector, the ultrasonic waves generated by partial discharge at the connector cause strain in the fiber at the probe, thereby modulating the phase of the optical pulse propagating at that point. All backscattered light (both modulated and unmodulated) returns along the original path.

[0048] The backscattered Rayleigh light carrying vibration information returns to Port 2 of the circulator. According to the circulator's characteristics, this light will be output from Port 3, and together with 10% of the intrinsic light output from coupler 5, it will enter the polarization diversity receiver 7 (PDR) for photoelectric conversion. Figure 4 As shown, the polarization diversity receiving device 7 includes: The system comprises a first beam splitter BS1, a second beam splitter BS2, and a third beam splitter BS3. The first beam splitter BS1 splits the received local oscillator light (LO) into two paths, which are then fed into the second beam splitter BS2 and the third beam splitter BS3, respectively. In this embodiment, 10% of the continuous light output from the narrow linewidth laser module 6 is split off by the coupler 5 and sent as local oscillator light to the first beam splitter BS1. The local oscillator light is then split into two paths in a 50:50 ratio and sent directly to the second beam splitter BS2 and the third beam splitter BS3.

[0049] The polarization beam splitter PS receives the signal light returned from the sensing fiber (in this embodiment, sensing fiber link 1 via Port 2-Port 3 of optical circulator 2) and decomposes it into two mutually orthogonal polarization states, X and Y (i.e., ...). Figure 4 The S and P components marked in the diagram are fed into the second beam splitter BS2 and the third beam splitter BS3, respectively.

[0050] The polarization state X (S component) interferes with one local oscillator light in the second beam splitter BS2, outputting two beams with a phase difference of 180°; the polarization state Y (P component) interferes with one local oscillator light in the third beam splitter, outputting two beams with a phase difference of 180°.

[0051] Two photoelectric balanced detectors, BPD1 and BPD2, are used. BPD1 receives the two beams from the second beam splitter BS2 and is responsible for detecting the interference signal in the S-polarization channel, outputting an analog voltage signal (corresponding to...). 、 Information). BPD2 receives two beams from the third beam splitter BS3, is responsible for detecting the interference signal in the P-polarization channel, and outputs an analog voltage signal (corresponding to...). 、 (Information). That is, each BPD converts the phase change of the optical signal into two differential electrical signals (I and Q paths) through coherent beat frequency. Since environmental disturbances can cause the polarization state of the returned signal light to rotate randomly, the output of a single BPD will attenuate as a result. However, the PS decomposes the input light of arbitrary polarization state onto two fixed bases, X and Y, to ensure that at least one channel can always receive a stronger signal.

[0052] The working principle of polarization diversity receiver 7 is as follows: The PDR coherently mixes the returned signal light carrying phase information with the local oscillator light. Its key outputs are the in-phase (I) and quadrature (Q) electrical signals corresponding to the P and S orthogonal polarization states, respectively.

[0053] For each polarization path (taking the P-path as an example), the output of the BPD essentially reflects the interference result between the signal light and the local oscillator light. Ideally, these two orthogonal components can be expressed as:

[0054]

[0055] in, For signal amplitude, This refers to the phase change caused by the partial discharge that ultimately needs to be demodulated. This includes common terms such as laser phase noise and slowly changing phase in the environment. The expression for the S-path is similar, but may differ due to changes in polarization state.

[0056] The polarization diversity structure of PDR ensures that no matter how randomly the polarization state of the returned light rotates, its optical field can always be decomposed into two fixed orthogonal bases, P and S. Thus, at least one path (P or S) can maintain high coherence efficiency and output stable I and Q signals, fundamentally avoiding the polarization fading problem in traditional single-path coherent detection.

[0057] The signal processing device is specifically configured as follows: By using the frequency division multiplexing chirped coded signal at the transmitting end as a reference signal, cross-correlation operation is performed with the two received electrical signals to achieve pulse compression and obtain a distance-response curve with a high signal-to-noise ratio; On the distance-response curve, a time window is set as a distance gate corresponding to the physical location of each monitoring probe; Phase demodulation is performed only on data falling within the distance gate to separate probe responses from different frequency channels and spatial locations.

[0058] Specifically, the subsequent PC-side acquisition card processes the two electrical signals output from the two BPDs. By taking the square root of the sum of the squares of the two orthogonal signals, a stable voltage signal V(t) independent of the polarization state of the input signal is synthesized. This process fundamentally eliminates the polarization fading problem in traditional single-channel coherent detection.

[0059] The data acquisition card synchronously converts the two analog signals output by the PDR into digital sequences. However, directly demodulating these data is inefficient and inaccurate because: the system transmits coded long pulses (such as chirped pulses), and the backscattered signals from different locations on the optical fiber (such as multiple connector probes) overlap in the time domain; the acoustic signal generated by partial discharge is extremely weak, and direct demodulation is easily overwhelmed by noise. Therefore, matched filtering (pulse compression) is required first. This step is crucial for improving system performance. The process typically involves synthesizing the energy of the two polarization signals to obtain a stable amplitude signal V(t) independent of the polarization state.

[0060] This V(t) signal contains amplitude information of all scattering points along the fiber, but loses phase information. It is mainly used for preliminary event detection and distance positioning.

[0061] The acquired digital signal V[n] is cross-correlated with the original coded sequence (reference signal) generated by the arbitrary waveform generator (AWG) at the transmitting end. This operation is equivalent to an optimal filter; when the received signal contains components that match the reference signal, a sharp correlation peak will appear in the output.

[0062] This process achieves two core functions: By compressing the energy of a long coded pulse into an extremely narrow time window, a sharp peak with a high signal-to-noise ratio (SNR) is generated, which greatly enhances the ability to detect weak partial discharge signals. The time delay of the correlation peak precisely corresponds to the round-trip propagation time of the optical pulse in the optical fiber. By calculating this delay, the distance information z of the vibration event can be obtained, and a high-resolution distance-response curve R(z) can be output. In the scheme using frequency division multiplexing (FDM) chirped pulses, matched filtering can also use coding templates of different frequencies to separate the responses of different frequency channels that overlap in the time domain, realizing multi-channel parallel detection.

[0063] After obtaining the high signal-to-noise ratio distance-response curve R(z), the physical position of each sensitizing probe on the optical fiber is precisely calibrated beforehand. , ,..., At the corresponding distance on the R(z) curve, a distance gate is set to extract the original four digital signal segments corresponding to the time window within each gate: , , , These signal segments are the demodulated signals corresponding to the i-th probe position, which have undergone pulse compression and distance gating, resulting in a significantly improved signal-to-noise ratio.

[0064] High-precision phase demodulation is performed on the signal segment extracted by each probe. This process is based on the principle of quadrature demodulation, which recovers the phase information from the in-phase (I) and quadrature (Q) components.

[0065] First, construct a complex signal for each polarization path:

[0066]

[0067] Where j is the imaginary unit.

[0068] For each complex signal, the phase angle is calculated to obtain the instantaneous phase of each polarization:

[0069]

[0070] here This is the arctangent function in the four quadrants, with an output range of (- , The result is the wrapped phase.

[0071] Due to the actual phase change It may be far more than 2 , and Phase jumps may occur. A phase unwinding algorithm is needed to detect and compensate for these jumps to recover the continuous absolute phase.

[0072]

[0073] Due to environmental disturbances, and The signal-to-noise ratio of the two signals fluctuates over time. A stable final phase signal is synthesized using the polarization selection method, and the power of the two signals at each moment is calculated. , Choose the phase of the path with higher power as the output at that moment. .

[0074] The final result The signal directly and with high fidelity reflects the phase change in the optical fiber caused by the partial discharge acoustic wave, and its relationship with strain... This is directly proportional to the data, providing the core data foundation for subsequent partial discharge feature recognition.

[0075] To achieve targeted monitoring of cable joints, this embodiment employs a quasi-distributed enhanced-sensitivity probe design. The sensing fiber consists of a continuous single-mode fiber connected in series with multiple discrete, strip-shaped enhanced-sensitivity fiber probes for partial discharge acoustic waves, with each probe corresponding to one cable joint to be monitored.

[0076] This embodiment also incorporates Frequency Division Multiplexing (FDM) technology. In the chirped pulse sequence transmitted by the system, different pulses or different time periods within the same pulse can be modulated to different center frequencies. When these differently frequency-coded pulses propagate through the optical fiber and return, their backscattered signals overlap in the time domain. By matched filtering and demodulation using different frequency coding templates, the responses from different frequency channels can be separated in the frequency domain. This means that the system can simultaneously and independently process signals corresponding to different fiber segments or different events, achieving multi-channel parallel detection and demultiplexing.

[0077] Furthermore, after obtaining a high signal-to-noise ratio range-response curve through matched filtering, the system sets an electronic range gate at the corresponding peak on the curve based on the pre-calibrated physical position of each enhanced probe on the optical fiber. For systems employing frequency division multiplexing (FDM), this process is divided into two layers: First, using matched filtering and corresponding frequency templates, the mixed received signal is separated into multiple independent frequency channel data streams, each corresponding to a transmitted FDM code. Then, a range gate is set on each separated frequency channel data stream (i.e., its own curve) according to the probe position. The original signal segment corresponding to the time window within each gate is extracted. This extracts an independent time-domain vibration signal corresponding to the i-th probe position, processed by pulse compression and FDM / TDM multiplexing. This method achieves distributed measurement, precisely concentrating data processing and analysis resources on preset key monitoring points, and effectively avoiding crosstalk between signals from different locations.

[0078] Example 3 This embodiment describes an online monitoring method for partial discharge in cable joints. Figure 5 As shown, the method is based on the detection system described in Embodiment 2 and includes the following steps: Step 1: The optical emission and modulation device generates a frequency-division multiplexed chirped pulse light sequence, which enters the sensing fiber optic link; wherein, multiple monitoring probes in the sensing fiber optic link are tightly wound and fixed to the outer surface of different intermediate joints of the cable to be monitored, as shown in the following arrangement. Figure 6 As shown in the figure, the orange part is the cable core. The cross-section is divided into the main insulation layer and the cold shrink tubing in the radial direction. The monitoring probe is tightly wrapped and fixed to the outer surface of the cold shrink tubing.

[0079] The optical emission and modulation module controls the generation of a frequency-division multiplexed chirped pulse light sequence into the sensing fiber optic link. The optical emission and modulation module is configured to generate and modulate a narrow-linewidth laser to generate a frequency-division multiplexed chirped pulse light sequence for injection into the sensing fiber optic link. Specifically, it includes a narrow-linewidth laser module 6, a coupler (5), an acousto-optic modulator 4, an arbitrary waveform generator (AWG), and an RF amplifier 9 positioned between the arbitrary waveform generator and the acousto-optic modulator 4. The arbitrary waveform generator (AWG) is used to generate a parasitic frequency-division multiplexed linear frequency-modulated (chirped) RF pulse sequence as a driving signal. This sequence contains multiple chirped pulse subbands that are orthogonal in the frequency domain and overlappable in the time domain. The RF amplifier 9 amplifies the highly precise and complex signal generated by the front-end AWG, providing sufficient energy to drive the AOM (Agent Optimizer). After this signal is loaded into the AOM, it modulates the input continuous light into a frequency-division multiplexed chirped light pulse sequence at high speed and with a high extinction ratio. A matched filter is then used to separate the time-domain fiber response of each pulse.

[0080] In this embodiment, after system startup, the narrow-linewidth laser module 6 generates a beam of continuous coherent light with a center wavelength of 1550nm. After passing through coupler 5, 90% of the continuous light enters the acousto-optic modulator (AOM). The arbitrary waveform generator (AWG) generates a set of frequency-division multiplexed linear frequency-modulated (chirped) radio frequency pulse sequences as driving signals. The radio frequency pulse sequence is as follows: Figure 3 As shown, this sequence contains multiple chirped pulse subbands that are orthogonal in the frequency domain and overlappable in the time domain. After this signal is loaded into the AOM, the input continuous light is modulated into a frequency-division multiplexed chirped optical pulse sequence with high speed and high extinction ratio. Subsequently, a matched filter is used to separate the time-domain fiber response of each pulse.

[0081] Step 2: Use polarization diversity receiver 7 to collect backscattered Rayleigh light, and output two baseband signals containing orthogonal polarization states through polarization beam splitting and dual-balanced detection.

[0082] Specifically, the polarization diversity receiver 7 includes: The system comprises a first beam splitter BS1, a second beam splitter BS2, and a third beam splitter BS3. The first beam splitter BS1 splits the received local oscillator light (LO) into two paths, which are then fed into the second beam splitter BS2 and the third beam splitter BS3, respectively. In this embodiment, 10% of the continuous light output from the narrow linewidth laser module 6 is split off by the coupler 5 and sent as local oscillator light to the first beam splitter BS1. The local oscillator light is then split into two paths in a 50:50 ratio and sent directly to the second beam splitter BS2 and the third beam splitter BS3.

[0083] The polarization beam splitter PS receives the signal light returned from the sensing fiber (in this embodiment, sensing fiber link 1 via Port 2-Port 3 of optical circulator 2) and decomposes it into two mutually orthogonal polarization states, X and Y (i.e., ...). Figure 4The S and P components marked in the diagram are fed into the second beam splitter BS2 and the third beam splitter BS3, respectively.

[0084] The polarization state X (S component) interferes with one local oscillator light in the second beam splitter BS2, outputting two beams with a phase difference of 180°; the polarization state Y (P component) interferes with one local oscillator light in the third beam splitter, outputting two beams with a phase difference of 180°.

[0085] Two photoelectric balanced detectors, BPD1 and BPD2, are used. BPD1 receives the two beams from the second beam splitter BS2 and is responsible for detecting the interference signal in the S-polarization channel, outputting an analog voltage signal (corresponding to...). 、 Information). BPD2 receives two beams from the third beam splitter BS3, is responsible for detecting the interference signal in the P-polarization channel, and outputs an analog voltage signal (corresponding to...). 、 (Information). Each BPD converts the phase change of the optical signal into two differential electrical signals (I and Q paths) via coherent beat frequency. Because environmental disturbances cause the polarization state of the returned signal light to rotate randomly, the output of a single BPD will attenuate. However, the PS decomposes the input light of arbitrary polarization state onto two fixed bases, X and Y, ensuring that at least one channel always receives a strong signal. The two analog signals are sent as baseband signals to the data acquisition DAQ.

[0086] Step 3: Using the frequency division multiplexing chirped coded signal as a reference template, perform matched filtering and distance gating on the baseband signal to separate the vibration signal segments corresponding to each monitoring probe.

[0087] The matched filtering and distance gating described in this step specifically include steps S31 to S33: S31: Perform cross-correlation calculations between the received baseband signal and the original frequency division multiplexing chirped coded signal generated at the transmitter, utilize the autocorrelation characteristics of the chirped pulse to achieve pulse compression, and separate the scattered signals that overlap in optical fiber transmission due to frequency division multiplexing in the frequency and time domains.

[0088] Specifically, the data acquisition card synchronously converts the two analog signals output by the PDR into digital sequences. However, directly demodulating these data is inefficient and inaccurate because: the system transmits coded long pulses (such as chirped pulses), and the backscattered signals from different locations on the optical fiber (such as multiple connector probes) overlap in the time domain; the acoustic signal generated by partial discharge is extremely weak, and direct demodulation is easily overwhelmed by noise. Therefore, matched filtering (pulse compression) is required first. This step is crucial for improving system performance. The process typically involves synthesizing the energy of the two polarization signals to obtain a stable amplitude signal V(t) independent of the polarization state.

[0089] This V(t) signal contains amplitude information of all scattering points along the fiber, but loses phase information. It is mainly used for preliminary event detection and distance positioning.

[0090] The acquired digital signal V[n] is cross-correlated with the original coded sequence (reference signal) generated by the arbitrary waveform generator at the transmitting end. This operation is equivalent to an optimal filter; when the received signal contains a component that matches the reference signal, a sharp correlation peak will appear in the output.

[0091] This process achieves two core functions: By compressing the energy of a long coded pulse into an extremely narrow time window, a sharp peak with a high signal-to-noise ratio (SNR) is generated, which greatly enhances the ability to detect weak partial discharge signals. The time delay of the correlation peak precisely corresponds to the round-trip propagation time of the optical pulse in the optical fiber. By calculating this delay, the distance information z of the vibration event can be obtained, and a high-resolution distance-response curve R(z) can be output. In the scheme using frequency division multiplexing (FDM) chirped pulses, matched filtering can also use coding templates of different frequencies to separate the responses of different frequency channels that overlap in the time domain, realizing multi-channel parallel detection.

[0092] S32: Obtain the distance-response curve generated after cross-correlation calculation, and set an electronic distance gate with a predetermined time width on the distance-response curve according to the pre-calibrated physical position of each monitoring probe on the sensing fiber optic link.

[0093] Specifically, after obtaining the high signal-to-noise ratio distance-response curve R(z), the physical position of each sensitizing probe on the optical fiber is precisely calibrated beforehand. , ,..., At the corresponding distance on the R(z) curve, a distance gate is set to extract the original four digital signal segments corresponding to the time window within each gate: , , , These signal segments are the demodulated signals corresponding to the i-th probe position, which have undergone pulse compression and distance gating, resulting in a significantly improved signal-to-noise ratio.

[0094] S33: Filter out data other than the electronic distance gate, and retain only the data inside the gate as the effective vibration signal segment of the corresponding probe.

[0095] Step 4: Perform orthogonal demodulation and polarization synthesis on the separated vibration signal segments to restore the ultrasonic waveform generated by partial discharge.

[0096] The method for orthogonal demodulation and polarization combining described in this step includes steps S41 to S43: S41: Construct complex signals corresponding to two orthogonal polarization states based on the two baseband signals respectively, and recover the phase information from the in-phase (I) and quadrature (Q) components:

[0097]

[0098] Where j is the imaginary unit.

[0099] S42: Calculate the instantaneous phase angle of each complex signal using the arctangent function. The instantaneous phase angle is calculated as follows:

[0100]

[0101] here This is the arctangent function in the four quadrants, with an output range of (- , The result is the wrapped phase.

[0102] Due to the actual phase change It may be far more than 2 , and Phase jumps may occur. A phase unwinding algorithm is needed to detect and compensate for these jumps to recover the continuous absolute phase.

[0103]

[0104] S43: Calculate the instantaneous power of the complex signals of orthogonal polarization states in real time, and perform any of the following operations to generate the final phase output: select the absolute phase signal corresponding to the polarization state with the larger instantaneous power as the output, or perform weighted synthesis of the two absolute phase signals according to the ratio of the instantaneous power of the two polarization states.

[0105] Due to environmental disturbances, and The signal-to-noise ratio of the two signals fluctuates over time. A stable final phase signal is synthesized using the polarization selection method, and the power of the two signals at each moment is calculated. , Choose the phase of the path with higher power as the output at that moment. .

[0106] The final result The signal directly and with high fidelity reflects the phase change in the optical fiber caused by the partial discharge acoustic wave, and its relationship with strain... This is directly proportional to the data provided for subsequent partial discharge feature recognition.

[0107] The spatial positioning principle of this embodiment is optical time domain reflection (OTDR) and pulse compression.

[0108] The positioning method is based on the principle of optical time-domain reflectometry (OTDR). By injecting coded light pulses into the sensing fiber, any point along the fiber with a non-uniform refractive index (such as the sensitizing probe, fusion splice, or fiber end) will generate backscattered Rayleigh light as the pulse propagates. By measuring the time difference Δt between pulse emission and reception of the scattered light at a certain point, the distance Z between the event point and the transmitting end can be calculated based on the speed of light v in the fiber (approximately 2 × 10⁸ m / s).

[0109] However, traditional OTDRs use simple light pulses, and their spatial resolution is limited by the pulse width (the narrower the pulse width, the higher the resolution, but the lower the energy, which limits the detection distance and signal-to-noise ratio).

[0110] Therefore, this embodiment employs frequency division multiplexing (FDM) chirped pulse coding technology. AWG generates a set of linear frequency-modulated (chirped) pulse sequences that are orthogonal in the frequency domain and partially overlapper in the time domain. At the receiver, matched filtering (cross-correlation operation) is performed on the acquired backscattered signal V(t) and the transmitted original coded sequence. This process achieves pulse compression: concentrating the energy of the long coded pulse into an extremely narrow time window, generating a sharp correlation peak. This brings two major advantages: without shortening the physical pulse width or sacrificing transmit energy, the effective detection pulse width is compressed through signal processing, achieving a spatial resolution far exceeding that of traditional pulse OTDRs; the matched filter is the optimal filter that maximizes the output signal-to-noise ratio against a white noise background, significantly enhancing the detection capability of weak partial discharge signals.

[0111] The output of the matched filter is a clear distance-response curve R(z), and each peak on the curve corresponds to a strong reflection or scattering point on the optical fiber (i.e., the position of the enhanced probe), thus achieving high-precision spatial positioning.

[0112] Furthermore, to achieve targeted monitoring of cable joints, this embodiment employs a quasi-distributed enhanced-sensitivity probe design. The sensing fiber consists of a continuous single-mode fiber connected in series with multiple discrete, strip-shaped enhanced-sensitivity fiber probes for partial discharge acoustic waves, with each probe corresponding to a cable joint to be monitored. By combining frequency division multiplexing (FDM) technology, different pulses or different time periods within the same pulse in the chirped pulse sequence emitted by the AWG can be modulated to different center frequencies. When these differently frequency-coded pulses propagate in the fiber and return, their backscattered signals overlap in the time domain. Through matched filtering and demodulation using different frequency coding templates, the responses from different frequency channels can be separated in the frequency domain. This means that the system can simultaneously and independently process signals corresponding to different fiber segments or different events, achieving multi-channel parallel detection and demultiplexing.

[0113] The above description is only a preferred embodiment of the present invention. It should be noted that those skilled in the art can make several modifications and improvements without departing from the inventive concept of the present invention, and these all fall within the protection scope of the present invention.

Claims

1. An online monitoring probe for partial discharge of cable joints, characterized in that, include: The strip probe body is formed by at least one tight-buffered optical fiber arranged in a continuous S-shape in a plane, and multiple reciprocating segments of the tight-buffered optical fiber are closely arranged in the width direction. The sealing layer is made of a flexible material that is resistant to high temperature and has high electrical insulation properties. It is used to seal the multi-turn tightly wrapped optical fiber as a whole, so that the strip probe body has a flat and bendable strip structure. The transmission port includes an optical fiber lead or connector extending from the sealing layer and connected to the tight-buffered optical fiber, for connecting the strip probe body in series to an external sensing optical path. The strip probe body is configured to surround and fit the outer surface of the cable intermediate joint, so that multiple tightly wrapped optical fibers form a sensing plane covering the outer periphery of the cable intermediate joint, which is used to sense the partial discharge signal generated at any position of the cable intermediate joint.

2. The probe according to claim 1, characterized in that, The spacing of the multi-turn tight-buffered optical fibers around the outer periphery of the cable intermediate joint is configured such that the equivalent axial winding density after probe installation is not less than 50 turns / meter. The inner surface of the strip probe body is coated with an acoustic coupling layer, which is composed of silicone grease or polymer gel and is used to fill the microscopic gaps between the probe and the surface of the cable intermediate joint.

3. The probe according to claim 1, characterized in that, The transmission port includes two leads that are respectively connected to both ends of the tight-buffered optical fiber, and both leads extend from the same longitudinal end of the sealing layer.

4. A partial discharge online monitoring system for cable joints, characterized in that, include: A sensing fiber optic link includes a transmission fiber and a plurality of monitoring probes as described in claim 1, which are connected in series on the transmission fiber and are disposed on the outer surface of the cold shrink sleeve of the cable intermediate joint. The optical emission and modulation module is configured to generate and modulate narrow-linewidth laser light to generate a frequency-division multiplexed chirped pulse light sequence and inject it into the sensing fiber optic link. The polarization diversity receiving module is configured to receive back Rayleigh scattered light from the sensing fiber optic link, decompose it into two orthogonal linearly polarized lights, coherently detect them with the local oscillator light, and output two electrical signals containing orthogonal polarization components. The data acquisition and processing module is configured to use the frequency division multiplexing chirped coded signal as a reference to perform matched filtering on the two electrical signals to generate a distance-response curve, and to set a distance gate according to the physical location of the cable joint partial discharge online monitoring probe in the sensing fiber optic link to extract and demodulate the phase signal at each probe location.

5. The system according to claim 4, characterized in that, The optical emission and modulation device includes: Narrow linewidth laser module, configured to output continuous coherent light, and utilizes feedforward phase noise compensation technology to compress linewidth; An acousto-optic modulator receives the continuous coherent light; An arbitrary waveform generator is configured to generate the frequency division multiplexing chirped coded signal to drive the acousto-optic modulator; The frequency division multiplexing chirped coded signal contains multiple linear frequency modulated pulse subbands that are orthogonal in the frequency domain and overlap in the time domain.

6. The system according to claim 4, characterized in that, The polarization diversity receiver includes: The first beam splitter, the second beam splitter, and the third beam splitter are used. The first beam splitter splits the received local oscillator light into two paths, which are then sent to the second beam splitter and the third beam splitter, respectively. The polarization beam splitter receives the signal light returned from the sensing fiber optic link and decomposes it into two mutually orthogonal polarization states, X and Y, which are then sent to the second and third beam splitters, respectively. The polarization state X interferes with one local oscillator beam in the second beam splitter, resulting in two beams with a phase difference of 180°. The polarization state Y interferes with one local oscillator light in the third beam splitter, resulting in two beams with a phase difference of 180°. Two photoelectric balanced detectors receive two beams from the second and third beam splitters, respectively, and output polarization states S and P after differential amplification.

7. The system according to claim 4, characterized in that, The signal processing device is specifically configured as follows: By using the frequency division multiplexing chirped coded signal at the transmitting end as a reference signal, cross-correlation operation is performed with the two received electrical signals to achieve pulse compression and obtain a distance-response curve with a high signal-to-noise ratio; On the distance-response curve, a time window is set as a distance gate corresponding to the physical location of each monitoring probe; Phase demodulation is performed only on data falling within the distance gate to separate probe responses from different frequency channels and spatial locations.

8. A method for online monitoring of partial discharge in cable joints, characterized in that, The method is based on the detection system as described in claim 4 and includes the following steps: A frequency-division multiplexed chirped pulse light sequence is generated by a control optical emission and modulation device and enters the sensing optical fiber link; wherein, multiple monitoring probes in the sensing optical fiber link have been tightly wound and fixed to the outer surface of different intermediate joints of the cable to be monitored. Back-scattered Rayleigh light is collected using a polarization diversity receiver, and two baseband signals containing orthogonal polarization states are output through polarization beam splitting and dual-balanced detection. Using the frequency division multiplexing chirped coded signal as a reference template, the baseband signal is subjected to matched filtering and range gating to separate the vibration signal segments corresponding to each monitoring probe; The separated vibration signal segments are subjected to orthogonal demodulation and polarization synthesis to restore the ultrasonic waveform generated by partial discharge.

9. The method according to claim 8, characterized in that, The matched filtering and distance gate selection specifically include: The received baseband signal is cross-correlated with the original frequency division multiplexing chirped coded signal generated at the transmitter. The autocorrelation characteristics of the chirped pulse are used to achieve pulse compression, and the scattered signals that overlap due to frequency division multiplexing in optical fiber transmission are separated in the frequency domain and time domain. After obtaining the distance-response curve generated by cross-correlation calculation, an electronic distance gate with a predetermined time width is set on the distance-response curve according to the pre-calibrated physical position of each monitoring probe on the sensing fiber optic link. Data outside the electronic distance gate is filtered out, and only the data inside the gate is retained as the effective vibration signal segment of the corresponding probe.

10. The method according to claim 8, characterized in that, The method of orthogonal demodulation and polarization synthesis includes: Complex signals corresponding to two orthogonal polarization states are constructed based on the two baseband signals respectively; The instantaneous phase angle of each complex signal is calculated using the arctangent function, and a phase unwinding algorithm is used to recover the continuous absolute phase signal by detecting and compensating for phase jumps. Calculate the instantaneous power of the complex signals of orthogonal polarization states in real time, and perform any of the following operations to generate the final phase output: select the absolute phase signal corresponding to the polarization state with the larger instantaneous power as the output, or perform weighted synthesis of the two absolute phase signals according to the ratio of the instantaneous power of the two polarization states.

Images