CN-121982210-A - Three-dimensional image detection method and device for underground cable based on terahertz data

CN121982210ACN 121982210 ACN121982210 ACN 121982210ACN-121982210-A

Abstract

The three-dimensional image detection method for the underground cable based on the terahertz data comprises the steps of obtaining the terahertz data of the underground cable, generating dielectric constant distribution of the underground cable through a multi-physical field coupling waveguide model according to the terahertz data, extracting an electric field time domain signal from the complex signal after the dielectric constant distribution is compressed into the complex signal by utilizing a quantum derivative compressed sensing algorithm, carrying out nonlinear phase compensation on the electric field time domain signal by adopting a preset Volterra series model to obtain a compensated electric field time domain signal, carrying out tensor decomposition on the compensated electric field time domain signal to obtain a feature matrix, carrying out Bayesian inversion on the feature matrix to obtain optimal cable parameters of the underground cable, generating an optimal three-dimensional parameter field according to the optimal cable parameters, and drawing a three-dimensional image of the underground cable according to the optimal three-dimensional parameter field. Thus, the problems of limited penetrating capacity and resolution of the existing metal detection, low-frequency electromagnetic wave and radar imaging in the underground complex environment are broken through.

Inventors

TANG QI
CHEN ZHIPING
HE ZILAN
HUANG JING

Assignees

广东电网有限责任公司佛山供电局

Dates

Publication Date: 20260505
Application Date: 20260127

Claims (10)

1. The three-dimensional image detection method for the underground cable based on the terahertz data is characterized by comprising the following steps of: Acquiring terahertz data of an underground cable, and generating dielectric constant distribution of the underground cable through a multi-physical field coupling waveguide model according to the terahertz data; After the dielectric constant distribution is compressed into complex signals by utilizing a quantum derived compressed sensing algorithm, an electric field time domain signal is extracted from the complex signals, and nonlinear phase compensation is carried out on the electric field time domain signal by adopting a preset Volterra series model, so that a compensated electric field time domain signal is obtained; Performing tensor decomposition on the compensated electric field time domain signal to obtain a feature matrix, and performing Bayesian inversion on the feature matrix to obtain optimal cable parameters of the underground cable; and generating an optimal three-dimensional parameter field according to the optimal cable parameter, so as to draw a three-dimensional image of the underground cable according to the optimal three-dimensional parameter field.
2. The terahertz data-based underground cable three-dimensional image detection method of claim 1, wherein the expression of the multi-physical field coupling waveguide model is: Wherein, the For the dielectric constant distribution to be described, Is a dielectric constant at a high frequency and is, As the contribution of the k-th order polarization, Is the damping coefficient [ (] ) Is used for correcting phase jump errors caused by soil humidity and temperature variation, For the plasma frequency, Is a terahertz wave frequency, and the frequency of the terahertz wave is the frequency of the terahertz wave, For the kth polarization resonance frequency, In imaginary units.
3. The method for three-dimensional image detection of an underground cable based on terahertz data according to claim 1, wherein the step of compressing the dielectric constant distribution into complex signals using a quantum-derived compressive sensing algorithm comprises: mapping the dielectric constant distribution into sparse signals through sparse transformation; and linearly compressing the sparse signal by adopting a quantum sparse basis matrix generated by quantum entanglement to obtain the complex signal.
4. The method for three-dimensional image detection of an underground cable based on terahertz data as set forth in claim 3, wherein the step of linearly compressing the sparse signal with a quantum sparse basis matrix generated from quantum entanglement to obtain the complex signal includes: The complex signal is generated using the following expression: Wherein, the In order to provide the complex signal in question, A sparse basis matrix generated for quantum entanglement, Is a physical sparse matrix for mapping the dielectric constant distribution to the sparse signal , In order to compress the matrix, As a vector of the noise it is, Is an allowable error threshold.
5. The method for three-dimensional image detection of an underground cable based on terahertz data according to claim 1, wherein the step of performing nonlinear phase compensation on the electric field time domain signal by using a preset Volterra series model to obtain a compensated electric field time domain signal comprises the steps of: Inputting the electric field time domain signal into the Volterra series model to estimate nonlinear phase errors in the electric field time domain signal based on a multi-order kernel function in the Volterra series model; and carrying out phase compensation on the electric field time domain signal according to the estimated nonlinear phase error so as to correct nonlinear distortion caused by medium characteristic change and obtain a compensated electric field time domain signal.
6. The method of claim 5, wherein the step of inputting the electric field time domain signal to the Volterra series model to estimate the nonlinear phase error in the electric field time domain signal based on a multi-order kernel function in the Volterra series model comprises: the nonlinear phase error is estimated using the following expression: Wherein, the For the Volterra series model, For the multi-order kernel function, At a point in time for the electric field time domain signal Is a value of (2).
7. The method for three-dimensional image detection of an underground cable based on terahertz data according to claim 1, wherein the step of performing bayesian inversion on the feature matrix to obtain optimal cable parameters of the underground cable comprises the steps of: Taking the weight coefficient of the feature matrix as observation data, and constructing a likelihood function based on the observation data; Acquiring prior distribution of the underground cable, combining the likelihood function and the prior distribution through Bayesian theorem, and calculating posterior distribution of the underground cable; And optimizing the posterior distribution by using a Markov chain Monte Carlo algorithm to solve the optimal cable parameters.
8. An underground cable three-dimensional image detection device based on terahertz data, which is characterized by comprising: The dielectric constant distribution generation module is used for acquiring terahertz data of the underground cable and generating dielectric constant distribution of the underground cable through a multi-physical field coupling waveguide model according to the terahertz data; The electric field time domain signal compensation module is used for extracting an electric field time domain signal from a complex signal after the dielectric constant distribution is compressed into the complex signal by utilizing a quantum derivative compressed sensing algorithm, and carrying out nonlinear phase compensation on the electric field time domain signal by adopting a preset Volterra series model to obtain a compensated electric field time domain signal; The optimal cable parameter determining module is used for performing tensor decomposition on the compensated electric field time domain signals to obtain a feature matrix, and performing Bayesian inversion on the feature matrix to obtain optimal cable parameters of the underground cable; and the three-dimensional image drawing module is used for generating an optimal three-dimensional parameter field according to the optimal cable parameter so as to draw a three-dimensional image of the underground cable according to the optimal three-dimensional parameter field.
9. A storage medium having stored therein computer readable instructions which, when executed by one or more processors, cause the one or more processors to perform the steps of the method for three-dimensional image detection of a subterranean cable based on terahertz data as set forth in any one of claims 1 to 7.
10. A computer device includes one or more processors and a memory; Stored in the memory are computer readable instructions which, when executed by the one or more processors, perform the steps of the terahertz data-based subsurface cable three-dimensional image detection method as recited in any one of claims 1 to 7.

Description

Three-dimensional image detection method and device for underground cable based on terahertz data Technical Field The application relates to the technical field of underground cable detection, in particular to an underground cable three-dimensional image detection method and device based on terahertz data. Background Underground cables are widely laid underground in roads, residential areas and industrial parks as an important infrastructure of urban power and communication networks, and the safe and stable operation of the underground cables is directly related to the normal operation of urban functions. Along with the acceleration of the urban process, underground cable lines are increasingly dense, and a direct-buried laying mode is adopted, so that the periodic detection and fault positioning of the cables become important links for guaranteeing the power supply safety. The existing underground cable detection mainly depends on technologies such as metal detection, audio signal detection, low-frequency electromagnetic wave detection and the like, and the methods have a certain effect on metal cable positioning and are mature in application. Meanwhile, some detection means assist in judging the buried state of the cable by means of radar detection and infrared imaging. However, due to the complex underground environment and various mediums, the methods still have defects in terms of detection precision, penetration depth and resolution, and particularly when the nonmetallic material is wrapped or nonmetallic sheath cable is faced, the penetration capability and imaging effect of the prior art are limited, and positioning deviation or detection blind areas are easy to cause. Disclosure of Invention The present application aims to solve at least one of the above technical drawbacks, especially the technical drawbacks of the prior art that the penetration capability and imaging effect are limited, and the positioning deviation or the detection blind area is easily caused. In a first aspect, the application provides a three-dimensional image detection method of an underground cable based on terahertz data, which comprises the following steps: Acquiring terahertz data of an underground cable, and generating dielectric constant distribution of the underground cable through a multi-physical field coupling waveguide model according to the terahertz data; After the dielectric constant distribution is compressed into complex signals by utilizing a quantum derivative compressed sensing algorithm, an electric field time domain signal is extracted from the complex signals, and nonlinear phase compensation is carried out on the electric field time domain signal by adopting a preset Volterra series model, so that a compensated electric field time domain signal is obtained; performing tensor decomposition on the compensated electric field time domain signal to obtain a feature matrix, and performing Bayesian inversion on the feature matrix to obtain optimal cable parameters of the underground cable; And generating an optimal three-dimensional parameter field according to the optimal cable parameters so as to draw a three-dimensional image of the underground cable according to the optimal three-dimensional parameter field. In one embodiment, the expression for the multiple physical field-coupled waveguide model is: Wherein, the In order to provide a dielectric constant distribution,Is a dielectric constant at a high frequency and is,As the contribution of the k-th order polarization,Is the damping coefficient [ (]) Is used for correcting phase jump errors caused by soil humidity and temperature variation,For the plasma frequency,Is a terahertz wave frequency, and the frequency of the terahertz wave is the frequency of the terahertz wave,For the kth polarization resonance frequency,In imaginary units. In one embodiment, the step of compressing the dielectric constant distribution into a complex signal using a quantum derived compressive sensing algorithm comprises: Mapping the dielectric constant distribution into sparse signals through sparse conversion; and linearly compressing the sparse signal by adopting a quantum sparse basis matrix generated by quantum entanglement to obtain a complex signal. In one embodiment, the step of linearly compressing the sparse signal with a quantum sparse basis matrix generated from quantum entanglement to obtain a complex signal includes: The complex signal is generated using the following expression: Wherein, the In the case of a complex signal, the signal is,A sparse basis matrix generated for quantum entanglement,Is a physical sparse matrix for mapping dielectric constant distribution into sparse signals,In order to compress the matrix,As a vector of the noise it is,Is an allowable error threshold. In one embodiment, the step of performing nonlinear phase compensation on the electric field time domain signal by using a preset Volterra series model to obtain a compensated electric f