Underground cable fault acquisition system and use method thereof

Abstract

The invention discloses an underground cable fault acquisition system and a use method thereof, and particularly relates to the technical field of cable fault detection, and the underground cable fault acquisition system comprises a multi-frequency pulse signal transmitting module, a reflected wave detection array module, a time-frequency joint analysis module, a three-dimensional positioning engine module, an augmented reality terminal and a fault identification module. The multi-frequency pulse signal transmitting module generates a specific excitation signal and injects the signal into a cable. The reflected wave detection array module collects a reflected signal. The time-frequency combined analysis module processes the signal to obtain a characteristic. The three-dimensional positioning engine module calculates a fault point coordinate. The augmented reality terminal maps the fault point coordinate and generates a maintenance path. And the fault identification module judges the fault category. Through cooperative work of multiple modules, accurate positioning, comprehensive analysis and efficient maintenance planning of underground cable faults are achieved, and the system has the advantages of being high in positioning accuracy, accurate in fault diagnosis and high in maintenance efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of cable fault detection, and more specifically, to an underground cable fault acquisition system and a method for using the same. Background Art

[0002] In modern society, the stability and reliability of power supply are crucial for economic development and people's daily lives. Underground cables, with advantages such as less land occupation, high safety, and little influence from the external environment, have become key facilities for urban power transmission. However, due to their being laid underground and the complex operating environment, they face many risk factors that can lead to faults.

[0003] From the perspective of the natural environment, moisture, acidity, alkalinity, and temperature changes in the soil will gradually erode the insulation layer of the cable. For example, in a humid soil environment, long-term penetration of moisture may cause the performance of the insulation material to decline, leading to insulation faults. At the same time, chemical substances in the soil will react with the cable outer sheath, reducing the mechanical strength and electrical performance of the cable.

[0004] Mechanical external forces can also pose a serious threat to underground cables. Urban construction and construction are frequent, and operations such as excavation and drilling may damage the cable if not careful. In addition, ground settlement or uplift will cause the cable to be stretched and squeezed, damaging its internal structure and thus triggering faults.

[0005] The aging problem of the cable itself cannot be ignored either. Long-term current transmission will cause the cable to generate heat, accelerating the aging process of the insulation material. Moreover, as the service life increases, the metal conductors inside the cable will also show corrosion and oxidation phenomena, affecting the power transmission efficiency.

[0006] Currently, there are many deficiencies in underground cable fault detection technologies. Traditional resistance measurement methods, insulation resistance test methods, etc. can only preliminarily judge whether there is a fault in the cable, and it is difficult to accurately determine the location of the fault point. Although the time domain reflectometry (TDR) can measure the fault distance, for long-distance cables or cases with multiple fault points, the measurement accuracy will be greatly affected, and error accumulation is likely to occur. And most of the existing technologies are difficult to accurately identify the fault type and cannot provide comprehensive information support for the maintenance work.

[0007] In the fault repair link, due to the lack of comprehensive consideration of factors such as the distribution of underground pipelines, geological conditions, and traffic conditions around the fault point, it is difficult for maintenance personnel to quickly plan a safe and efficient repair path, resulting in an extended repair time, which not only increases the repair cost but also has a greater negative impact on social production and residents' lives.

[0008] Therefore, an underground cable fault acquisition system and a method for using the same are proposed. Summary of the Invention

[0009] To overcome the above-mentioned defects of the prior art, the present invention provides an underground cable fault acquisition system and its usage method to solve the problems raised in the above-mentioned background art.

[0010] To achieve the above object, the present invention provides the following technical solution: An underground cable fault acquisition system, comprising:

[0011] A multi-frequency pulse signal transmitting module, configured to generate an excitation signal including a plurality of discrete frequency components;

[0012] A reflected wave detection array module, including a plurality of equally spaced sensor nodes arranged along the cable, and each of the sensors uses a dual-mode fiber optic sensor;

[0013] A time-frequency joint analysis module, performing a joint time-frequency transform and configured to process the reflected signal;

[0014] A three-dimensional positioning engine module, calculating the fault point based on a multi-sensor time difference positioning model;

[0015] An augmented reality terminal, mapping the fault point coordinates to a geographic information system and generating an optimal repair path.

[0016] Preferably, the expression of the composite excitation signal with discrete frequency components generated by the multi-frequency pulse signal generator is:

[0017]

[0018] Wherein, A i ∈ [0.5, 2.0] is an adjustable amplitude coefficient, f i ∈ [1 kHz, 10 MHz] is a frequency component, is a random phase;

[0019] And the frequency components satisfy:

[0020] f i = f min + (i - 1)·Δf + (-1) i ·δf

[0021] δf = 0.15Δf

[0022] Wherein, Spectral leakage is reduced by alternating frequency offsets.

[0023] Preferably, the formula for the time-frequency joint analysis module to process the reflected signal is:

[0024]

[0025] Wherein, is the Morlet wavelet basis function, a is the scale parameter, and b is the translation parameter;

[0026] The time-frequency joint analysis module uses Hilbert-Huang transform to obtain the Intrinsic Mode Function (IMF) through Empirical Mode Decomposition:

[0027]

[0028] and calculate the instantaneous frequency where H is the Hilbert transform. Preferably, the three-dimensional positioning engine module sets the coordinates of the fault point in the multi-sensor time difference positioning model as (x, y, z), and its calculation formula is:

[0029] and its calculation formula is:

[0030]

[0031] And the three-dimensional positioning engine module uses the Levenbery-Marquardt optimization algorithm, and the iteration formula is:

[0032] θ k+1 = θ k -(J T J + λT) -1 J T r

[0033] where θ = [x, y, z] T is the coordinate vector, J is the Jacobian matrix, and the element λ is taken as λ k+1 = λ k .

[0034] Preferably, the specific formula for the integrated path planning of the augmented reality terminal is:

[0035]

[0036] where d i is the road section length, s i ∈[0, 1] is the excavation difficulty coefficient, r i is the safety distance from underground pipelines, and the weights satisfy w d + w s + w r = 1, r0 = 1m.

[0037] Preferably, it further includes a fault identification module based on the Convolutional Neural Network (CNN) model:

[0038] CNN(X) = Softmax(FC(ReLU(Pool(Conv2D(X)))))

[0039] Input the time-frequency matrix The output categories include:

[0040] Short - circuit fault (output neuron 1);

[0041] Open - circuit fault (output neuron 2);

[0042] Insulation degradation (output neuron 3).

[0043] An underground cable fault location method based on the underground cable fault acquisition system described in the above item, comprising the following steps:

[0044] S1. Inject the transmitted composite pulse signal S(t) into the cable through the multi - frequency pulse signal transmitting module;

[0045] S2. Then collect the reflected signal through the reflected wave detection array module where h m (t) is the channel response of the m - th sensor;

[0046] S3. Use the three - dimensional positioning engine module to calculate the cross - correlation peak time delay of each sensor signal:

[0047]

[0048] Then construct an over - determined system of equations:

[0049]

[0050] S4. Solve the fault coordinates through the three - dimensional positioning engine module:

[0051]

[0052] where, and c = 2.0 is the Huber loss threshold;

[0053] S5. Finally, render the three - dimensional coordinates of the fault point and the repair path through the augmented reality terminal, and at the same time output the fault category using the fault identification module.

[0054] The technical effects and advantages of the present invention:

[0055] 1. The multi - frequency pulse signal transmitting module generates an excitation signal containing multiple discrete frequency components through a unique frequency setting formula, which can effectively reduce spectral leakage, improve signal quality, and provide a good signal basis for subsequent accurate positioning. The three - dimensional positioning engine module is based on the multi - sensor time - difference positioning model and combines the Levenbery - Marquardt optimization algorithm to solve the fault coordinates, and can accurately calculate the three - dimensional coordinates of the fault point. Compared with traditional positioning methods, the positioning accuracy is higher.

[0056] 2. The time-frequency joint analysis module processes the reflected signal using the Morlet wavelet basis function and Hilbert-Huang transform, which can obtain rich time-frequency features. By performing empirical mode decomposition to obtain the intrinsic mode functions and calculating the instantaneous frequency, it helps to deeply analyze the characteristics of the fault signal, thereby more accurately judging the fault type. The fault identification module analyzes the time-frequency matrix based on the convolutional neural network model and can quickly and accurately identify common fault categories such as short-circuit faults, open-circuit faults, and insulation deterioration, providing strong support for fault diagnosis.

[0057] 3. The augmented reality terminal maps the fault point coordinates to the geographic information system and generates the optimal repair path according to the path planning formula that includes factors such as the road section length, excavation difficulty coefficient, and safety distance from underground pipelines. This not only facilitates the maintenance personnel to quickly reach the fault point but also comprehensively considers the construction difficulty and safety, effectively improving the repair efficiency and reducing the repair time and cost. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 It is a schematic flow chart of the method for fault location using the underground cable fault acquisition system of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0059] Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0060] An underground cable fault acquisition system provided by the present invention includes:

[0061] A multi-frequency pulse signal transmitting module for generating an excitation signal containing multiple discrete frequency components;

[0062] A reflected wave detection array module including a plurality of equally spaced sensor nodes arranged along the cable, and each of the sensors uses a dual-mode fiber optic sensor;

[0063] A time-frequency joint analysis module for performing joint time-frequency transformation to process the reflected signal;

[0064] A three-dimensional positioning engine module for calculating the fault point based on the multi-sensor time difference positioning model;

[0065] An augmented reality terminal for mapping the fault point coordinates to the geographic information system and generating the optimal repair path.

[0066] During specific implementation, assemble the multi-frequency pulse signal transmitting module, the reflected wave detection array module, the time-frequency joint analysis module, the three-dimensional positioning engine module, the augmented reality terminal, and the fault identification module to ensure correct line connection and smooth communication between the modules;

[0067] Then, parameter settings are carried out. For the multi-frequency pulse signal transmission module, according to the formula

[0068]

[0069] set the adjustable amplitude coefficient A i ∈[0.5, 2.0], the frequency component f i ∈[1 kHz, 10 MHz] and the random phase And through the formula f i = f min +(i - 1)·Δf + (-1) i ·δf, δf = 0.15Δf, calculate and set the discrete frequency components to generate a composite excitation signal to reduce spectral leakage by alternating frequency offset.

[0070] For the time-frequency joint analysis module, determine the Morlet wavelet basis function scale parameter a, translation parameter b, and process the reflected signal according to the formula Set the relevant parameters of the Hilbert-Huang transform to obtain the intrinsic mode function (IMF) by empirical mode decomposition, and calculate the instantaneous frequency according to the formula.

[0071] Then, set the initial coordinate estimate of the fault point through the three-dimensional positioning engine module, and determine the parameter λ in the Levenbery-Marquardt optimization algorithm.

[0072] Then, according to the actual requirements, set the path planning formula through the augmented reality terminal

[0073]

[0074] where, d i is the section length, s i ∈[0, 1] is the excavation difficulty coefficient, r i is the safety distance from the underground pipeline, and the weights satisfy w d + w s + w r = 1, r0 = 1 m.

[0075] Finally, the fault identification module uses a large amount of time-frequency matrix data of known fault types to train the convolutional neural network (CNN) model, and adjusts the model parameters to enable it to accurately identify fault types such as short-circuit faults, open-circuit faults, and insulation deterioration.

[0076] As Figure 1 shown, when using the set underground cable fault acquisition system for fault location, first, start the multi-frequency pulse signal transmission module, according to the formula

[0077]

[0078] Generate a composite pulse signal, and through the formula f i = f min +(i - 1)·Δf + (-1) i ·δf, δf = 0.15Δf, Calculate and set the discrete frequency components to generate a composite excitation signal and inject it into the underground cable;

[0079] The reflection wave detection array module laid along the cable starts to work. It includes multiple equally spaced sensor nodes, and each sensor uses a dual-mode fiber optic sensor. The sensors collect the reflection signals where h m (t) is the channel response of the m-th sensor, and n m (t) is the noise.

[0080] Use the three-dimensional positioning engine module to calculate the cross-correlation peak time delay of each sensor signal. The formula is According to the position information (x m , y m , z m ) of each sensor and the signal propagation speed v, construct an overdetermined system of equations (x, y, z) are the coordinates of the fault point.

[0081] Solve the fault coordinates through the three-dimensional positioning engine module using the Levenbery - Marquardt optimization algorithm. First, set the initial estimated value θ k = [x k , y k , z k T , and perform iterative calculations according to the iterative formula θ k+1 = θ k - (J T J + λT) -1 J T r, where J is the Jacobian matrix, and the element λ is adjusted according to λ k+1 = λ k until the convergence condition is met to obtain the fault point coordinates (x, y, z).

[0082] The augmented reality terminal maps the fault point coordinates (x, y, z) calculated by the three-dimensional positioning engine module to the geographic information system, renders the three-dimensional coordinates of the fault point on the terminal interface, and according to the formula

[0083] ​

[0084] Generate the optimal maintenance path. Meanwhile, the fault identification module uses the time-frequency matrix as the input based on the convolutional neural network (CNN) model CNN(X) = Softmax(FC(ReLU(Pool(Conv2D(X))))), and then outputs the fault categories, including short circuit fault, open circuit fault, and insulation deterioration. As the input, and then outputs the fault categories, including: short circuit fault, open circuit fault, and insulation deterioration.

[0085] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An underground cable fault acquisition system, characterized in that, Including: A multi-frequency pulse signal transmitting module for generating an excitation signal containing multiple discrete frequency components; A reflected wave detection array module including a plurality of equally spaced sensor nodes laid along the cable, and each of the sensors uses a dual-mode fiber optic sensor; A time-frequency joint analysis module that performs joint time-frequency transformation for processing reflected signals; A three-dimensional positioning engine module that calculates the fault point based on a multi-sensor time difference positioning model; An augmented reality terminal that maps the fault point coordinates to a geographic information system and generates an optimal repair path.

2. The underground cable fault acquisition system according to claim 1, wherein The expression for the composite excitation signal with discrete frequency components generated by the multi-frequency pulse signal generator is: Among them, A i ∈[0.5, 2.0] is an adjustable amplitude coefficient, f i ∈[1 kHz, 10 MHz] is a frequency component, is a random phase; And the frequency components satisfy: f i = f min +(i - 1)·Δf + (-1) i ·δ f δf = 0.15Δf Among them, Reduce spectral leakage by alternating frequency offset.

3. The underground cable fault acquisition system according to claim 1, characterized in that, The formula for the time-frequency joint analysis module to process the reflected signal is: Among them, is the Morlet wavelet basis function, a is the scale parameter, and b is the translation parameter; The time-frequency joint analysis module uses Hilbert-Huang transform and obtains intrinsic mode functions (IMFs) through empirical mode decomposition: And calculate the instantaneous frequency where H i is the Hilbert transform.

4. The underground cable fault acquisition system according to claim 1, wherein In the multi-sensor time difference positioning model, the three-dimensional positioning engine module sets the fault point coordinates as (x, y, z), and its calculation formula is: And the three-dimensional positioning engine module uses the Levenbery-Marquardt optimization algorithm, and the iterative formula is: θ k+1 = θ k -(J T J + λT) -1 J T r where θ = [x, y, z] T is the coordinate vector, J is the Jacobian matrix, and the element λ is according to λ k+1 = λ k .

5. The underground cable fault acquisition system according to claim 1, characterized in that The specific formula for the augmented reality terminal to integrate path planning is: Among them, d i is the length of the road section, s i ∈[0,1] is the excavation difficulty coefficient, r i is the safety distance from the underground pipeline, and the weights satisfy w d +w s +w r = 1, r0 = 1m.

6. The underground cable fault acquisition system according to claim 1, characterized in that, It further includes a fault identification module based on a convolutional neural network (CNN) model: CNN(X) = Softmax(FC(ReLU(Pool(Conv2D(X))))) Input time-frequency matrix The output categories include: Short circuit fault (output neuron 1); Open circuit fault (output neuron 2); Insulation deterioration (output neuron 3).

7. A method for locating underground cable faults of the underground cable fault acquisition system according to any one of claims 1-6, characterized in that, Including the following steps: S1. Inject the transmitted composite pulse signal S(t) into the cable through the multi-frequency pulse signal transmitting module; S2. Then, collect the reflected signals through the reflected wave detection array module where h m (t) is the channel response of the m-th sensor S3. Use the three-dimensional positioning engine module to calculate the cross-correlation peak time delay of each sensor signal: Then construct an overdetermined system of equations: S4. Solve the fault coordinates through the three-dimensional positioning engine module: Among them, and c = 2.0 is the Huber loss threshold; S5. Finally, render the three-dimensional coordinates of the fault point and the repair path through the augmented reality terminal, and at the same time use the fault identification module to output the fault category.

Images