Method and system for detecting corrosion degree of wire steel core of overhead transmission line

Abstract

The invention provides a method and system for detecting the corrosion degree of a wire steel core of an overhead transmission line, and the method comprises the steps: carrying out the multi-angle X-ray transmission of a whole to-be-detected wire in an electrified state, and obtaining a projection sequence; inverting the projection sequence into a steel core three-dimensional density body, segmenting and identifying the steel core, the air and the corrosion area based on the steel core three-dimensional density body, and calculating the corrosion sectional area loss rate of the steel core; the chemical state of the Fe element is obtained in the corrosion area, satellite peaks of Fe-K alpha and iron oxide are collected respectively, and the concentration ratio of Fe < 0 > to Fe < 3 + > is calculated; according to the steel core corrosion sectional area loss rate and the concentration ratio of Fe < 0 > to Fe < 3 + >, the corresponding corrosion grade and residual strength of the wire to be detected are output, and the problems that in the prior art, the detection efficiency and reliability of the steel core corrosion degree of the wire of the overhead transmission line are not high are effectively solved. And the detection efficiency and reliability of the corrosion degree of the wire steel core of the overhead transmission line are effectively improved.

Description

Technical Field

[0001] This invention relates to the field of power transmission line operation and maintenance, and in particular to a method and system for detecting the degree of corrosion of the steel core of overhead power transmission line conductors. Background Technology

[0002] During the operation of power transmission lines, steel-cored aluminum stranded wires are subjected to the combined effects of mechanical tension, electrical loads, and complex environments over long periods. The internal steel core is highly susceptible to corrosion due to rainwater infiltration, salt spray erosion, electrochemical corrosion, or fretting wear. Once the steel core experiences cross-sectional loss or material deterioration, the mechanical strength of the entire conductor will rapidly decline, potentially leading to wire breakage, widespread power outages, and safety risks. However, due to the strong shielding effect of aluminum stranded wires on visible light, ultrasound, and conventional electromagnetic detection methods, current technologies struggle to quantitatively, non-destructively, and live-line detect the degree of steel core corrosion without stripping the outer aluminum strands.

[0003] Currently, the industry generally relies on methods such as manual tower climbing for spot checks, magnetic flux leakage or ultrasonic point testing, or single X-ray fluoroscopic imaging. While these methods can identify obvious geometric defects such as steel core fractures, they increase the difficulty and error rate in determining the chemical information of steel core corrosion through aluminum wires. This leads to difficulties in early corrosion identification, low reliability of identification, and delayed maintenance decisions. Consequently, the efficiency and reliability of detecting the corrosion degree of overhead transmission line conductors are not high.

[0004] To address this problem, the present invention provides a method and system for detecting the degree of corrosion of the steel core of overhead transmission line conductors, thereby solving the aforementioned problem. Summary of the Invention

[0005] To address the problems existing in the prior art, this invention innovatively proposes a method and system for detecting the degree of corrosion of the steel core of overhead transmission line conductors. This effectively solves the problem of low efficiency and reliability in detecting the degree of corrosion of the steel core of overhead transmission line conductors caused by the prior art, and effectively improves the efficiency and reliability of detecting the degree of corrosion of the steel core of overhead transmission line conductors.

[0006] The first aspect of this invention provides a method for detecting the degree of corrosion of the steel core of overhead transmission line conductors, comprising: Under energized conditions, the entire conductor under test is subjected to multi-angle X-ray transmission to obtain a projection sequence; The projection sequence is inverted into a three-dimensional density volume of the steel core. Based on the three-dimensional density volume of the steel core, the steel core, air and corrosion zone are segmented and identified, and the cross-sectional area loss rate of the steel core corrosion is calculated. The chemical state of Fe was obtained in the rusted area. Satellite peaks of Fe-Kα and iron oxide were collected, and Fe was calculated. 0 with Fe 3+ The concentration ratio between them; Based on the cross-sectional area loss rate of steel core corrosion, Fe 0 with Fe 3+ The concentration ratio between the two values ​​outputs the corrosion level and remaining strength of the conductor under test.

[0007] Optionally, the projection sequence is inverted into a three-dimensional density volume of the steel core. Based on the three-dimensional density volume of the steel core, the segmentation and identification of the steel core, air, and corrosion areas specifically include: The projection sequence is backprojected using Shepp-Logan filtering and then iterated using SART to generate a millimeter-scale three-dimensional density volume. The grayscale value of the three-dimensional density volume of the steel core of the conductor under test is obtained. Based on the grayscale value of the three-dimensional density volume of the steel core of the conductor under test and the preset first correspondence, the material attenuation coefficient corresponding to different grayscale value ranges of the three-dimensional density volume of the steel core of the conductor under test is obtained. The steel core, air and rust areas are segmented and identified based on the different material attenuation coefficients. The preset first correspondence is the correspondence between the material attenuation coefficient and the grayscale value range.

[0008] Furthermore, the specific calculation of the cross-sectional area loss rate due to corrosion of the steel core is as follows: △A=S1 / S2 Where △A is the cross-sectional area loss rate of steel core corrosion, S1 is the corrosion area of ​​the steel core of the conductor under test, and S2 is the original design cross-sectional area of ​​the steel core of the conductor under test.

[0009] Optionally, the chemical state of Fe is obtained in the rusted area, satellite peaks of Fe-Kα and iron oxide are collected, and Fe is calculated. 0 with Fe 3+ The concentration ratios between them specifically include: Based on the energy-intensity spectrum, sensitivity factors were established, where Fe... 0 The sensitivity factor is the net peak area of ​​Fe-Kα and Fe 0 Concentration ratio; Fe 3+ The sensitivity factor is the net peak area of ​​iron oxide and Fe. 3+ The ratio of concentrations; Net peak area and theoretical conditions based on laboratory standard sampling Fe 0 Concentration, to obtain normalized Fe 0 Sensitivity factor, Fe 3+ Sensitivity factor; Based on normalized Fe 0 Sensitivity factor, Fe 3+ Sensitivity factor, current sampling Fe 0 with Fe 3+ The net peak area was used to obtain Fe. 0 with Fe 3+ The concentration ratio between them.

[0010] Furthermore, based on the normalized Fe 0 Sensitivity factor, Fe 3+ The sensitivity factor and the current sampled net peak area are used to obtain Fe. 0 with Fe 3+ The specific concentration ratio between them is as follows: R = C1_Fe 0 / C1_Fe 3+ = (I1_Kα / S1_Kα I1_sat / S1_sat) / (I1_sat / S1_sat); Where R is Fe 0 with Fe 3+ The concentration ratio between them, C1_Fe 0 For Fe in the current sample 0 The concentration of C1_Fe 3+ For Fe in the current sample 3+ The concentration of Fe in the current sample, I1_Kα. 0 The net peak area, I1_sat is the Fe in the current sample. 3+ The net peak area, S1_Kα is the normalized Fe 0 The sensitivity factor, S1_sat is the normalized Fe 3+ The sensitivity factor.

[0011] Optionally, based on the cross-sectional area loss rate of steel core corrosion, Fe 0 with Fe 3+ Before outputting the corrosion level and residual strength of the conductor under test based on the concentration ratio, the following is also included: The cross-sectional area loss rate of steel core corrosion, Fe 0 with Fe 3+ The concentration ratios between them are spatiotemporally aligned.

[0012] Optionally, based on the cross-sectional area loss rate of steel core corrosion, Fe 0 with Fe 3+ The concentration ratio between the two values ​​outputs the corrosion level and residual strength of the conductor under test, specifically including: The cross-sectional area loss rate of steel core corrosion, Fe 0 with Fe 3+ The concentration ratios between them are normalized into two-dimensional feature vectors; If the two-dimensional feature vector contains the steel core corrosion cross-sectional area loss rate vector or Fe... 0 with Fe 3+If the concentration ratio vector between the two values ​​satisfies a first preset condition, then the corresponding level is dangerous corrosion, and the remaining strength is less than a first percentage threshold; wherein, the first preset condition is the steel core corrosion cross-sectional area loss rate vector or Fe in the two-dimensional feature vector. 0 with Fe 3+ The concentration ratio vector between them conforms to the first preset vector range.

[0013] Furthermore, based on the two-dimensional feature vector, the output of the corrosion level and remaining strength of the conductor under test specifically includes: If the two-dimensional feature vector contains the steel core corrosion cross-sectional area loss rate vector or Fe... 0 with Fe 3+ If the concentration ratio vector between the two values ​​satisfies the second preset condition, then the corresponding level is moderate to severe corrosion, and the remaining strength is greater than the first percentage threshold and less than the second percentage threshold; wherein, the second preset condition is that the steel core corrosion cross-sectional area loss rate vector in the two-dimensional feature vector is less than the first loss rate vector threshold and not less than the second loss rate vector threshold, or, Fe 0 with Fe 3+ The concentration ratio vector between them is greater than the first ratio vector threshold and not less than the second ratio vector threshold; If the two-dimensional feature vector contains the steel core corrosion cross-sectional area loss rate vector or Fe... 0 with Fe 3+ If the concentration ratio vector between the two values ​​satisfies the third preset condition, then the corresponding level is moderate to slight corrosion, and the remaining strength is greater than the second percentage threshold and less than the third percentage threshold. The third preset condition is that the steel core corrosion cross-sectional area loss rate vector in the two-dimensional feature vector is less than the second loss rate vector threshold and not less than the third loss rate vector threshold, and Fe... 0 with Fe 3+ The concentration ratio vector between them is greater than the second ratio vector threshold and not less than the third ratio vector threshold; If the two-dimensional feature vector contains the steel core corrosion cross-sectional area loss rate vector or Fe... 0 with Fe 3+ If the concentration ratio vector between the two values ​​satisfies the fourth preset condition, then the corresponding level is mild to severe corrosion, with the remaining strength greater than the third percentage threshold and less than the fourth percentage threshold. The fourth preset condition is that the steel core corrosion cross-sectional area loss rate vector in the two-dimensional feature vector is less than the third loss rate vector threshold and not less than the fourth loss rate vector threshold, and Fe... 0 with Fe 3+ The concentration ratio vector between them is greater than the fourth ratio vector threshold; If the two-dimensional feature vector contains the steel core corrosion cross-sectional area loss rate vector or Fe... 0 with Fe 3+If the concentration ratio vector between the two values ​​satisfies the fifth preset condition, then the corresponding level is mild to slight corrosion, and the remaining strength is greater than the fourth percentage threshold. The fifth preset condition is that the steel core corrosion cross-sectional area loss rate vector in the two-dimensional feature vector is less than the fourth loss rate vector threshold, and Fe... 0 with Fe 3+ The concentration ratio vector between them is greater than the third ratio vector threshold, wherein the first loss rate vector threshold, the second loss rate vector threshold, the third loss rate vector threshold, and the fourth loss rate vector threshold decrease in sequence, and the first ratio vector threshold, the second ratio vector threshold, the fourth ratio vector threshold, and the third ratio vector threshold increase in sequence.

[0014] Furthermore, the method for estimating residual strength is as follows: in: For the estimated residual strength, V[0] is the cross-sectional area loss rate vector of steel core corrosion in the two-dimensional feature vector, and V[1] is the Fe in the two-dimensional feature vector. 0 with Fe 3+ The vector of concentration ratios between them, σ_△A is the standard deviation of the cross-sectional area loss rate vector of steel core corrosion in the two-dimensional feature vector, μ_△A is the mean of the cross-sectional area loss rate vector of steel core corrosion in the two-dimensional feature vector, and σ_R is the concentration ratio vector of Fe in the two-dimensional feature vector. 0 with Fe 3+ The standard deviation of the concentration ratio vector between the two, μ_R is the Fe concentration in the two-dimensional feature vector. 0 with Fe 3+ The mean of the concentration ratio vector between them.

[0015] A second aspect of the present invention provides a system for detecting the degree of corrosion of the steel core of overhead transmission line conductors, comprising: The acquisition module performs multi-angle X-ray transmission on the entire conductor under test while it is energized to acquire a projection sequence. The segmentation and calculation module inverts the projection sequence into a three-dimensional density volume of the steel core. Based on the three-dimensional density volume of the steel core, it identifies the steel core, air and rusted areas through adaptive threshold segmentation and calculates the cross-sectional area loss rate of the steel core due to rust. The data acquisition and calculation module obtains the chemical state of Fe in the rusted area, collects satellite peaks of Fe-Kα and iron oxide, and calculates Fe. 0 with Fe 3+ The concentration ratio between them; The output module, based on the cross-sectional area loss rate of steel core corrosion and Fe... 0 with Fe 3+ The concentration ratio between the two values ​​outputs the corrosion level and remaining strength of the conductor under test.

[0016] The technical solution adopted in this invention has the following technical effects: 1. This invention involves performing multi-angle X-ray transmission on the entire conductor under test while it is charged to obtain a projection sequence; inverting the projection sequence into a three-dimensional density volume of the steel core; based on the three-dimensional density volume of the steel core, segmenting and identifying the steel core, air, and corrosion zones, and calculating the cross-sectional area loss rate of the steel core due to corrosion; obtaining the chemical state of Fe in the corrosion zone, collecting satellite peaks of Fe-Kα and iron oxide respectively, and calculating Fe... 0 with Fe 3+ The concentration ratio between them; based on the cross-sectional area loss rate of steel core corrosion, Fe 0 with Fe 3+ The concentration ratio between the two values ​​outputs the corrosion level and remaining strength of the conductor under test, effectively solving the problem of low detection efficiency and reliability of steel core corrosion degree of overhead transmission line conductors due to existing technology, and effectively improving the detection efficiency and reliability of steel core corrosion degree of overhead transmission line conductors.

[0017] 2. In the technical solution of this invention, the cross-sectional area loss rate of steel core corrosion and Fe are considered. 0 with Fe 3+ Before normalizing the concentration ratios between them to form a two-dimensional feature vector, it also includes: the cross-sectional area loss rate of steel core corrosion, Fe... 0 with Fe 3+ The concentration ratios between the values ​​were aligned in time and space to ensure the reliability of the detection of corrosion degree of steel core in overhead transmission line conductors.

[0018] 3. In the technical solution of this invention, the cross-sectional area loss rate of steel core corrosion and Fe are considered. 0 with Fe 3+ The concentration ratios between them are normalized into two-dimensional feature vectors; based on the steel core corrosion cross-sectional area loss rate vector or Fe in the two-dimensional feature vectors... 0 with Fe 3+ The numerical values ​​of the concentration ratio vector correspond to the output corrosion level and residual strength. Different on-site actions can be taken according to the corrosion level and residual strength to ensure the efficient and safe operation of the steel core of the overhead transmission line conductor.

[0019] 4. The technical solution of this invention can complete the entire inspection process without power interruption or aluminum layer stripping. Data can be collected by drones or robots in a single flight, completely eliminating the risks of high-altitude operations. Secondly, it deeply integrates geometric cross-sectional area loss with chemical valence state changes, allowing the degree of corrosion to leap from "empirical estimation" to "scientific quantification". It provides a replicable and scalable engineering paradigm for intelligent power grid inspection.

[0020] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit the invention. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 This is a flowchart illustrating the method of Embodiment 1 in the present invention; Figure 2 This is another flowchart illustrating the method of Embodiment 1 in the present invention; Figure 3 This is a schematic diagram illustrating the complete process of implementing the method in Embodiment 1 of the present invention; Figure 4 This is a schematic diagram of the system structure in Embodiment 2 of the present invention. Detailed Implementation

[0023] To clearly illustrate the technical features of this solution, the invention will be described in detail below through specific embodiments and in conjunction with the accompanying drawings. The following disclosure provides many different embodiments or examples for implementing different structures of the invention. To simplify the disclosure of the invention, components and arrangements of specific examples are described below. Furthermore, reference numerals and / or letters may be repeated in different examples. This repetition is for simplification and clarity and does not in itself indicate a relationship between the various embodiments and / or arrangements discussed. It should be noted that the components illustrated in the drawings are not necessarily drawn to scale. Descriptions of well-known components, processing techniques, and processes are omitted in this invention to avoid unnecessarily limiting the invention.

[0024] Example 1 like Figure 1 As shown, the present invention provides a method for detecting the degree of corrosion of the steel core of overhead transmission line conductors, comprising: S1, under energized conditions, perform multi-angle X-ray transmission on the entire conductor under test to obtain a projection sequence; S2, the projection sequence is inverted into a three-dimensional density volume of the steel core. Based on the three-dimensional density volume of the steel core, the steel core, air and rusted areas are segmented and identified, and the cross-sectional area loss rate of the steel core rusted is calculated. S3. Obtain the chemical state of Fe in the rusted area, collect satellite peaks for Fe-Kα and iron oxide respectively, and calculate Fe. 0 with Fe 3+ The concentration ratio between them; S4, based on the cross-sectional area loss rate of steel core corrosion, Fe 0 with Fe 3+ The concentration ratio between the two values ​​outputs the corrosion level and remaining strength of the conductor under test.

[0025] In step S1, the function is to perform multi-angle X-ray transmission along the entire conductor while it is energized to obtain a distortion-free projection sequence. Specifically, without interrupting power, a drone or a tower-mounted rail carrying a 160–320kV microfocus X-ray source (focus ≤ 0.4 mm, pulse frequency 1 kHz) and a 16-bit dynamic flat panel detector (amorphous silicon / CMOS flat panel, 0.2 mm pixel) flies at a constant speed along the conductor, acquiring over 90 high-definition projections in one lap. If a drone is chosen, the RTK positioning error must be ≤ 2 cm and the wind speed ≤ 10 m / s. The X-rays can penetrate a 40mm aluminum layer while maintaining a spatial resolution of 0.2mm; the detector synchronously records attitude, GPS, and timestamps with millisecond-level shutter speeds, providing distortion-free raw data for subsequent 3D reconstruction.

[0026] In step S2, the projection sequence is inverted into a three-dimensional density volume of the steel core. Based on the three-dimensional density volume of the steel core, the segmentation and identification of the steel core, air, and rust areas specifically include: S21, the projection sequence is back-projected using Shepp-Logan filtering and SART iteration to generate a millimeter-scale three-dimensional density volume; This step involves inverting the projection sequence into a steel-core 3D density volume and calculating the cross-sectional area loss rate ΔA. The filtering algorithm requires a Shepp-Logan core + GPU parallel FBP; the iterative algorithm requires the SART algorithm (5 iterations, CUDA acceleration, single frame <0.3 s). Compared to using a single algorithm, this combination offers advantages such as faster speed, lower aluminum layer artifacts, less sharp rust edges, and zero manual parameter tuning. It transforms 400 projections into millimeter-level artifact-free 3D volumes within 0.3 seconds, reducing traditional hour-level reconstruction to second-level timeframes, without the need for manual parameter tuning. The projection data is directly written to the edge GPU via a high-speed NVMe cache, completing Shepp-Logan filtering backprojection + 5 rounds of SART iterations within 0.3 seconds to generate a millimeter-level 3D density volume.

[0027] S22, obtain the grayscale value of the three-dimensional density of the steel core of the conductor under test, and obtain the material attenuation coefficient corresponding to different grayscale value ranges of the three-dimensional density of the steel core of the conductor under test according to the grayscale value of the three-dimensional density of the steel core of the conductor under test and the preset first correspondence relationship, and segment and identify the steel core, air and rust area according to the different material attenuation coefficients; wherein, the preset first correspondence relationship is the correspondence relationship between the material attenuation coefficient and the grayscale value range.

[0028] The material attenuation coefficient (μ) is the "rate of decrease" of X-ray intensity as it passes through matter, measured in cm. -1 The higher the value, the faster the ray attenuation and the higher the gray level of the reconstructed voxel. First, the μ-gray level correspondence table is determined using a standard step block (pre-defined first correspondence), as follows: Aluminum layer μ ≈ 0.25 cm -1 →Grayscale 400–1200; The steel core μ ≈ 0.96 cm -1 →Grayscale 1200–2200; The corrosion layer μ ≈ 0.75 cm -1 →Grayscale 900–1400; During reconstruction, adaptive thresholding is performed based on grayscale values ​​to back-calculate the μ value, thereby distinguishing between air, aluminum, rust, and steel core. Then, voxels with adjacent grayscale values ​​within the same threshold range are clustered into a three-dimensional "island." Clusters that are too small or isolated are considered artifacts, ensuring that the final rusted area and steel core are physically connected in three-dimensional space, rather than being scattered noise.

[0029] Specifically, the calculation of the cross-sectional area loss rate due to steel core corrosion is as follows: △A=S1 / S2 Where △A is the cross-sectional area loss rate of steel core corrosion, S1 is the corrosion area of ​​the steel core of the conductor under test, and S2 is the original design cross-sectional area of ​​the steel core of the conductor under test (based on the factory CAD model, find the design drawings of this type of conductor).

[0030] In step S3, the chemical state of Fe is obtained in the rusted area, satellite peaks of Fe-Kα and iron oxide are collected, and Fe is calculated. 0 with Fe 3+ The concentration ratios between them specifically include: S31, based on the energy-intensity spectrum, sensitivity factors were established, where Fe 0 The sensitivity factor is the net peak area of ​​Fe-Kα and Fe 0 Concentration ratio; Fe 3+ The sensitivity factor is the net peak area of ​​iron oxide and Fe. 3+ The ratio of concentrations; This step involves precisely acquiring the chemical state of Fe in the anomalous corrosion region and calculating the concentration ratio R. Within the anomalous corrosion region, a submillimeter-scale micro-beam XRF (Rh target, tube voltage 50 kV, tube current 200 µA, focal spot ≤100 μm) is accurately guided to the corrosion hotspot, acquiring Fe-Kα and iron oxide satellite peaks within 30 seconds. Based on the energy-intensity spectrum, a sensitivity factor is established using standard samples, followed by quantitative inversion of unknown samples. The entire process is automated after micro-beam XRF acquisition, requiring no manual threshold setting. Fe 0 The sensitivity factor is defined as: S_Kα = I_Kα 0 / C_Fe 0 Where S_Kα is Fe 0 The sensitivity factor, I_Kα0 The net peak area of ​​Fe-Kα, C_Fe 0 For Fe 0 concentration; Fe 3+ The sensitivity factor is defined as: S_sat = I_sat³ / C_Fe 3+ Where S_sat is Fe 3+ The sensitivity factor, I_sat³ is the net peak area of ​​iron oxide, C_Fe 3+ For Fe 3+ concentration; S32, net peak area obtained through laboratory standard sampling and theoretical conditions Fe 0 Concentration, to obtain normalized Fe 0 Sensitivity factor, Fe 3+ Sensitivity factor; The standard sample is: NIST SRM 1155 (pure Fe foil, Fe 0 (Fe content >99.9% of total iron content) and self-developed Fe2O3 thin film (XPS verified, Fe 3+ The content accounts for >98% of the total iron content.

[0031] Energy window (keV): Fe-Kα: 6.404 ± 0.015 (Fe 0 ); Fe-Kβ: 7.058±0.020 (all Fe); Fe-oxide satellite peak: 6.270±0.020 (Fe 3+ feature); Laboratory calibration was performed over a 30-day period, with 10 samples collected from each of the two standard samples, and an integration time of 30 seconds. The net peak areas were obtained as follows: Pure Fe: I_Kα 0 Fe2O3: I_sat; S_Kα= I_Kα 0 / C_Fe 0 ; S_sat = I_sat / C_Fe 3+ ; Where C_Fe 0 =1g / g (theoretical or ideal conditions), C_Fe 3+ = 0.7 g / g (mass fraction of Fe in Fe2O3, theoretical or ideal).

[0032] Finally, the average value was taken and normalized: S_Kα = 1.000 (normalized), S_sat = 0.78 ± 0.02.

[0033] S33, based on normalized Fe 0 Sensitivity factor, Fe 3+ Sensitivity factor, current sampling Fe 0 with Fe 3+ The net peak area was used to obtain Fe. 0 with Fe 3+ The concentration ratio between them.

[0034] Among them, based on the normalized Fe 0 Sensitivity factor, Fe 3+ The sensitivity factor and the current sampled net peak area are used to obtain Fe. 0 with Fe 3+ The specific concentration ratio between them is as follows: R = C1_Fe 0 / C1_Fe 3+ = (I1_Kα / S1_Kα I1_sat / S1_sat) / (I1_sat / S1_sat); Where R is Fe 0 with Fe 3+ The concentration ratio between them, C1_Fe 0 For Fe in the current sample 0 The concentration of C1_Fe 3+ For Fe in the current sample 3+ The concentration of Fe in the current sample, I1_Kα. 0 The net peak area, I1_sat is the Fe in the current sample. 3+ The net peak area, S1_Kα is the normalized Fe 0 The sensitivity factor, S1_sat is the normalized Fe 3+ The sensitivity factor.

[0035] Unknown sample current sampling measurement (on-site 30 s), after acquiring the spectrum, background is automatically subtracted (Shirley method), and the net peak area is obtained: I1_Kα (Fe in the current sample) 0 (net peak area), I_sat (current Fe in the sample) 3+ (net peak area); Concentration inversion using a linear superposition model: I1_Kα= S1_Kα·(C1_Fe 0 + C1_Fe 3+ ); I1_sat = S1_sat·C1_Fe 3+ ; Solving for: C1_Fe3+ = I1_sat / S1_sat; C1_Fe 0 = (I1_Kα / S1_Kα) C1_Fe 3+ ; Fe 0 / Fe 3+ Ratio R calculation: R = C1_Fe 0 / C1_Fe 3+ = (I1_Kα / S1_Kα I1_sat / S1_sat) / (I1_sat / S1_sat); Substituting the normalized sensitivity factors, S1_Kα = 1, S1_sat = 0.78: R = (I1_Kα 0.78·I1_sat) / (0.78·I1_sat); The above formula is embedded in the AI ​​chip ROM as a 32-bit floating point, and the R value is automatically output every 30 seconds without manual intervention.

[0036] like Figure 2 As shown, the present invention also provides a method for detecting the degree of corrosion of the steel core of overhead transmission line conductors, which includes the following steps before step S4: S40, the cross-sectional area loss rate of steel core corrosion, Fe 0 with Fe 3+ The concentration ratios between them are spatiotemporally aligned.

[0037] Specifically, in step S40, the R value obtained in this step is automatically matched with △A using timestamps and three-dimensional coordinates (Fe). 0 with Fe 3+ The "spatiotemporal alignment" of the concentration ratio R and the steel core corrosion cross-sectional area loss rate ΔA is achieved through "IMU unified clock + three-dimensional voxel coordinates + timestamp hash," eliminating the need for manual point selection. The specific process is as follows: S401, unified clock source: UAV / rail-mounted IMU + GNSS receiver (a device capable of simultaneously receiving signals from two or more global navigation satellite systems), PPS pulse (pulses per second) is triggered once per second; The IMU (Inertial Measurement Unit), X-ray flat panel, XRF probe (X-ray fluorescence spectroscopy), and edge computing unit are all connected to PTP (Precision Time Protocol, network clock synchronization protocol), with a clock error of <1 µs. All data packets have a 64-bit Unix timestamp (ns level) written to their headers.

[0038] S402, Scan-Reconstruction Coordinate System Establishment: While scanning, the exterior orientation elements (X, Y, Z, ω, φ, κ, i.e., three-dimensional coordinates, acceleration, angular velocity, etc.) of each frame of projection are recorded; after reconstruction, the origin of the voxel coordinate system is defined as the GNSS position of the first projection, and the voxel size is 0.1 mm; thus, the 4×4 homogeneous transformation matrix T of "voxel coordinates → world coordinates" is obtained.

[0039] S403, XRF Triggered Coordinate Writing: Before acquiring data at each point using the micro-beam XRF, the current IMU position (x_imu, y_imu, z_imu) and attitude are read; then converted to world coordinates using the hand-eye calibration matrix E (pre-calibrated, error <0.2 mm). [X_xrf, Y_xrf, Z_xrf, 1]ᵀ= T·E·[0, 0, 0, 1]ᵀ; This coordinate, along with the current ns timestamp, is written to the XRF spectrum file header.

[0040] S404, Spatiotemporal hash matching (completed in 1 ms on edge chip): An Octree is established for the reconstructed voxel mesh, with a depth of 10 layers and leaf nodes of 0.1 mm; all voxels are searched within a sphere with a radius of 0.5 mm centered on XRF world coordinates; the mean ΔA value within the sphere is taken as the geometric corrosion value at that point; secondary verification is performed using timestamp hashing. |t_xrf–t_voxel| ≤ 200 ms; (200 ms corresponds to a drone flight distance of 0.16 m, covering positioning drift).

[0041] S405, output matching results: Successful match: generate (ΔA, R, coordinates, timestamp) quadruple; Failed match: XRF point marked "orphan", local cached for the next round of testing.

[0042] Laboratory: Lead point calibration block 0.5 mm, average deviation between XRF aiming center and voxel center 0.18 mm; Field: 10 live-line operations, spatiotemporal matching success rate 100%, ΔA-R corresponding correlation coefficient R²=0.97.

[0043] By using a "unified clock + voxel Octree + time hash", we ensure that the "geometric photographs" and "chemical fingerprints" of the rust correspond one-to-one in both three-dimensional space and time.

[0044] In step S4, based on the cross-sectional area loss rate of steel core corrosion and Fe... 0 with Fe 3+ The concentration ratio between the two values ​​outputs the corrosion level and residual strength of the conductor under test, specifically including: The cross-sectional area loss rate of steel core corrosion, Fe 0 with Fe 3+ The concentration ratios between them are normalized into two-dimensional feature vectors; In this step, the R value obtained is matched with △A to form a (△A, R) tuple. The (△A, R) tuple is then merged into a unified two-dimensional feature vector to eliminate single-dimensional misjudgments. That is, after z-score normalization, △A and R form a two-dimensional feature vector V=[△A_norm,R_norm].

[0045] The generation of the two-dimensional feature vector V = [ΔA_norm, R_norm] is completed in one step within the edge AI chip, with zero manual intervention and zero floating-point parameter tuning throughout the entire process. Specifically, it includes: (1) Raw data entry point: ΔA: Cross-sectional area loss rate of steel core, range 0–20%, resolution 0.1%.

[0046] R:Fe 0 / Fe 3+ Ratio, range 0–15, resolution 0.01.

[0047] (2) Missing value and outlier interception If ΔA > 20% or R < 0, it is marked as exceeding the limit or data abnormality, directly judged as red (dangerous corrosion), and not entered into the two-dimensional feature vector calculation.

[0048] If ΔA or R is NaN / Null, fill it with the value from the previous frame (time sliding window 1 s).

[0049] (3) Z-score normalization (curing coefficient) Training set: 100,000 field data points (0–20%, ΔA; 0–15%, R).

[0050] Statistics (already burned into ROM): μ_ΔA = 10%, σ_ΔA = 5%; μ_R = 7.5, σ_R = 3.75; Where σ_△A is the standard deviation of the steel core corrosion cross-sectional area loss rate vector in the two-dimensional feature vector, μ_△A is the mean of the steel core corrosion cross-sectional area loss rate vector in the two-dimensional feature vector, and σ_R is the Fe value in the two-dimensional feature vector. 0 with Fe 3+ The standard deviation of the concentration ratio vector between the two, μ_R is the Fe concentration in the two-dimensional feature vector. 0 with Fe 3+ The mean of the vector of concentration ratios between them; Normalization formula (32-bit floating point): ΔA_norm = (ΔA–μ_ΔA) / σ_ΔA; R_norm = (R–μ_R) / σ_R; Result range: ΔA_norm ∈ [-2, 2]; R_norm ∈ [-2, 2]; (4) Assembly of two-dimensional feature vector V, memory layout (Little-Endian, 8 bytes): Byte 0–3: ΔA_norm (IEEE-754 float); Byte 4–7: R_norm (IEEE-754 float); Furthermore, s can be defined as the fusion score, which compresses the normalized △A and R into a continuous number for positioning in a 6-level grading table.

[0051] Fusion score: s = W·V = 0.6·ΔA_norm + 0.4·R_norm; W represents the weight vector, which was trained using 100,000 field data points (w_△A=0.6, indicating the weight of 60% for the geometric loss of the steel core corrosion cross-sectional area; w_R = 0.4, indicating the weight of Fe). 0 with Fe 3+ The concentration ratio between them accounts for 40% of the weight. These two values ​​of 0.6 and 0.4 were calculated using least squares regression from 100,000 field breaking force samples. (The chip is directly solidified, and no further parameter adjustments are needed on-site.)

[0052] The corrosion level and remaining strength percentage are output based on the two-dimensional feature vector V. The criterion table for the V vector fused into the AI ​​chip is shown in Table 1. Table 1: V Vector Criterion Table

[0053] In this context, the color depth of green A is greater than that of green B. For example, green A is dark green and green B is light green; the color depth of yellow A is less than that of yellow B. For example, yellow A is light yellow and yellow B is dark yellow; and the color depth of red A is less than that of red B. For example, red A is light red and red B is dark red.

[0054] Specifically, such as Figure 3 As shown, if the cross-sectional area loss rate vector of steel core corrosion or Fe in the two-dimensional feature vector... 0 with Fe 3+If the concentration ratio vector between the two values ​​satisfies the first preset condition, then the corresponding level is dangerous corrosion, and the remaining strength is less than the first percentage threshold (60%). The first preset condition is that the steel core corrosion cross-sectional area loss rate vector in the two-dimensional feature vector is not less than the first loss rate vector threshold (1.5), or, Fe... 0 with Fe 3+ The concentration ratio vector between them is not greater than the first ratio vector threshold (-0.5), that is, 1.5≤△A_norm or R_norm≤-0.5.

[0055] Preferably, the dangerous corrosion level can be further divided into severe dangerous corrosion (corresponding to red B) and mild dangerous corrosion (corresponding to red A). When the cross-sectional area loss rate vector of steel core corrosion in the two-dimensional feature vector or Fe 0 with Fe 3+ When the concentration ratio vector between the two values ​​satisfies: 1.5 ≤ ΔA_norm < 2.0 or R_norm ≤ -0.5, the corresponding corrosion level is light dangerous corrosion (corresponding to red A). At this time, the remaining strength is 50%-60%, and the on-site action is to immediately arrange for line replacement; when the steel core corrosion cross-sectional area loss rate vector or Fe in the two-dimensional feature vector... 0 with Fe 3+ When the concentration ratio vector between them satisfies the condition that △A_norm≥2.0 and R_norm≤-1.0, the corresponding corrosion level is severe dangerous corrosion (corresponding to red B). At this time, the remaining strength is less than 50%, and the on-site action is to immediately stop operation and carry out emergency reinforcement.

[0056] If the two-dimensional feature vector contains the steel core corrosion cross-sectional area loss rate vector or Fe... 0 with Fe 3+ If the concentration ratio vector between the two values ​​satisfies the second preset condition, then the corresponding level is moderate to severe corrosion, with the remaining strength greater than the first percentage threshold (60%) and less than the second percentage threshold (70%). The second preset condition is that the steel core corrosion cross-sectional area loss rate vector in the two-dimensional feature vector is less than the first loss rate vector threshold (1.5) and not less than the second loss rate vector threshold (1.0), or, Fe... 0 with Fe 3+ The concentration ratio vector between them is greater than the first ratio vector threshold (-0.5) and not less than the second ratio vector threshold (0), that is, 1 ≤ ΔA_norm < 1.5, or -0.5. <R_norm≤-0; If the two-dimensional feature vector contains the steel core corrosion cross-sectional area loss rate vector or Fe... 0 with Fe 3+If the concentration ratio vector between them satisfies the third preset condition, the corresponding grade is moderate and slightly rusted, and the remaining strength is greater than the second percentage threshold (70%) and less than the third percentage threshold (80%); where the third preset condition is that the loss rate vector of the cross-sectional area of the steel core rust in the two-dimensional feature vector is less than the second loss rate vector threshold (1.0) and not less than the third loss rate vector threshold (0), and, Fe 0 and Fe 3+ the concentration ratio vector between them is greater than the second ratio vector threshold (0) and not less than the third ratio vector threshold (1.0), that is, 0 ≤ △A_norm < 1.0, and, -0 < R_norm ≤ 1.0; If the loss rate vector of the cross-sectional area of the steel core rust in the two-dimensional feature vector or Fe 0 and Fe 3+ the concentration ratio vector between them satisfies the fourth preset condition, the corresponding grade is mild and severely rusted, and the remaining strength is greater than the third percentage threshold (80%) and less than the fourth percentage threshold (90%); where the fourth preset condition is that the loss rate vector of the cross-sectional area of the steel core rust in the two-dimensional feature vector is less than the third loss rate vector threshold (0) and not less than the fourth loss rate vector threshold (-1.0), and, Fe 0 and Fe 3+ the concentration ratio vector between them is greater than the fourth ratio vector threshold (0.5), that is, -1.0 ≤ △A_norm < 0, and, 0.5 < R_norm; If the loss rate vector of the cross-sectional area of the steel core rust in the two-dimensional feature vector or Fe 0 and Fe 3+ the concentration ratio vector between them satisfies the fifth preset condition, the corresponding grade is mild and slightly rusted, and the remaining strength is greater than the fourth percentage threshold (90%); where the fifth preset condition is that the loss rate vector of the cross-sectional area of the steel core rust in the two-dimensional feature vector is less than the fourth loss rate vector threshold (-1.0), and, Fe 0 and Fe 3+ the concentration ratio vector between them is greater than the third ratio vector threshold (1.0), that is, △A_norm < -1.0, and, 1.0 < R_norm; where the first loss rate vector threshold (1.5), the second loss rate vector threshold (1.0), the third loss rate vector threshold (0), the fourth loss rate vector threshold (-1.0) decrease in turn, and the first ratio vector threshold (-0.5), the second ratio vector threshold (0), the fourth ratio vector threshold (0.5), the third ratio vector threshold (1.0) increase in turn.

[0057] Furthermore, the specific value of the remaining rust strength corresponding to the wire under test can also be output according to the two-dimensional feature vector. The estimation method of the remaining strength is specifically as follows: Where: For the estimated residual strength, V[0] is the cross-sectional area loss rate vector of steel core corrosion in the two-dimensional feature vector, and V[1] is the Fe in the two-dimensional feature vector. 0 with Fe 3+ The vector of concentration ratios between them, σ_△A is the standard deviation of the cross-sectional area loss rate vector of steel core corrosion in the two-dimensional feature vector, μ_△A is the mean of the cross-sectional area loss rate vector of steel core corrosion in the two-dimensional feature vector, and σ_R is the concentration ratio vector of Fe in the two-dimensional feature vector. 0 with Fe 3+ The standard deviation of the concentration ratio vector between the two, μ_R is the Fe concentration in the two-dimensional feature vector. 0 with Fe 3+ The mean of the concentration ratio vector between them.

[0058] The AI ​​chip can directly input V without further inverse normalization, and the error is ≤3%.

[0059] The judgment result is transmitted back within 5 minutes via 4G / 5G or Bluetooth 5.0. The fields include device ID, GPS, △A, R, level, remaining strength and on-site photos; local 48h cache ensures traceability even if wireless interruption occurs, realizing closed-loop management of "scan and judge immediately, judge and leave immediately".

[0060] This invention's technical solution enables full-process inspection without power outages or aluminum layer stripping. Data can be collected in a single flight by a drone or robot, completely eliminating the risks of high-altitude operations. Secondly, it deeply integrates geometric cross-sectional area loss with chemical valence state changes, achieving an ΔA error of ≤3% and an R error of ≤5%, transforming the degree of corrosion from "empirical estimation" to "scientific quantification." Thirdly, the edge AI chip can complete scanning, modeling, and grading within 5 minutes, and output suggestions for line replacement or recoating, truly achieving "instant judgment after scanning." Finally, the platform adopts a modular design, allowing seamless adaptation of a single hardware set to various composite conductors such as LJG and OPGW, reducing maintenance costs by more than 60% and providing a replicable and scalable engineering paradigm for intelligent power grid inspection.

[0061] To more clearly illustrate the detection method of this solution, the following uses the live detection of a UAV as an example to explain the specific implementation method.

[0062] The following equipment is required: Drone: Hexacopter DJI Matrice 300 RTK, RTK base station error ≤1 cm; X-ray chamber: Customized 200kV microfocus, 2.8kg, focus 0.3mm, 6S battery, 22.2V, 10Ah, power consumption 120W; Detector: Varex 4336DX wireless, Wi-Fi 6, 300 Mbps, 25W power consumption; Flight controller: Pixhawk 6X + CUAV v5+, with open MAVLink protocol; Edge computing chip (AI chip): Jetson Xavier NX, 2×USB 3.0, 1×HDMI, power consumption 15W; Communications and Data: 5G CPE: Huawei 5G CPE Pro 2, uplink 100 Mbps; Software requirements: 3D reconstruction: ASTRA Toolbox 2.1 (CUDA accelerated), with specific parameters including Shepp-Logan filter kernel, 5 SART iterations, and 6GB GPU memory.

[0063] △AR Normalized Model: The model is based on PyTorch 1.12, MobileNet-V3-Small, with 1.2 M parameters after pruning, and INT8 quantization. The training set consists of 100,000 field data points (△A: 0-20%, R: 0-15).

[0064] Calibration cycle: Geometric calibration is performed monthly using a NIST SRM 1155+ dual circular plate, and energy calibration is performed weekly using pure Fe sheets and Fe2O3 thin films.

[0065] The drone's flight path can be manually controlled by the drone pilot or planned in advance using a cloud platform. The scanning speed should ideally be 0.8 m / s. During flight, pay attention to the ambient wind speed; if the wind speed is ≥10 m / s, the drone should immediately return to base.

[0066] The scanning angle should meet the following requirements: First, the cone angle formed by the X-ray source-object-detector should be ≥15° to ensure the stereo resolution of the tomographic reconstruction. Second, the angular interval between adjacent projections should be ≤2°, and the total number of projections should be ≥90 to meet the Nyquist sampling of the FBP / SART algorithm. Third, the overlap of adjacent scanning bands of the UAV along the traverse direction should be ≥70° to ensure no gaps. Fourth, the angle between the traverse axis and the X-ray centerline should be controlled within 0°±3° to avoid elliptical distortion. Fifth, when encountering a tension angle tower, an additional compensation angle of 5°-8° (an additional deflection angle between the X-ray cone beam centerline and the traverse axis) should be added on the outside to ensure that the X-ray can still penetrate the steel core perpendicularly.

[0067] After scanning is complete, the system performs intelligent analysis via a judgment and analysis module. The analysis results are then transmitted to the maintenance personnel's terminal in JSON format via a data transmission module. Maintenance personnel can then make further maintenance and repair decisions for the transmission lines based on the scan results.

[0068] This invention involves performing multi-angle X-ray transmission on the entire conductor under test while it is charged, obtaining a projection sequence; inverting the projection sequence into a three-dimensional density volume of the steel core; based on the three-dimensional density volume of the steel core, segmenting and identifying the steel core, air, and corrosion zones, and calculating the cross-sectional area loss rate of the steel core due to corrosion; obtaining the chemical state of Fe in the corrosion zone, collecting satellite peaks of Fe-Kα and iron oxide respectively, and calculating Fe... 0 with Fe 3+ The concentration ratio between them; based on the cross-sectional area loss rate of steel core corrosion, Fe 0 with Fe 3+ The concentration ratio between the two values ​​outputs the corrosion level and remaining strength of the conductor under test, effectively solving the problem of low detection efficiency and reliability of steel core corrosion degree of overhead transmission line conductors due to existing technology, and effectively improving the detection efficiency and reliability of steel core corrosion degree of overhead transmission line conductors.

[0069] The technical solution of this invention includes the cross-sectional area loss rate of steel core corrosion and Fe. 0 with Fe 3+ Before normalizing the concentration ratios between them to form a two-dimensional feature vector, it also includes: the cross-sectional area loss rate of steel core corrosion, Fe... 0 with Fe 3+ The concentration ratios between the values ​​were aligned in time and space to ensure the reliability of the detection of corrosion degree of steel core in overhead transmission line conductors.

[0070] The technical solution of this invention includes the cross-sectional area loss rate of steel core corrosion and Fe. 0 with Fe 3+ The concentration ratios between them are normalized into two-dimensional feature vectors; based on the steel core corrosion cross-sectional area loss rate vector or Fe in the two-dimensional feature vectors... 0 with Fe 3+ The numerical values ​​of the concentration ratio vector correspond to the output corrosion level and residual strength. Different on-site actions can be taken according to the corrosion level and residual strength to ensure the efficient and safe operation of the steel core of the overhead transmission line conductor.

[0071] The technical solution of this invention can complete the entire inspection process without power interruption or aluminum layer stripping. Data can be collected by drones or robots in a single flight, completely eliminating the risks of high-altitude operations. Secondly, it deeply integrates geometric cross-sectional area loss with chemical valence state changes, allowing the degree of corrosion to leap from "empirical estimation" to "scientific quantification". It provides a replicable and scalable engineering paradigm for intelligent power grid inspection.

[0072] Example 2 like Figure 4 As shown, the present invention also provides a system for detecting the degree of corrosion of the steel core of overhead transmission line conductors, comprising: The acquisition module 101, under energized conditions, performs multi-angle X-ray transmission on the entire conductor under test to acquire a projection sequence; The segmentation and calculation module 102 inverts the projection sequence into a three-dimensional density volume of the steel core. Based on the three-dimensional density volume of the steel core, it identifies the steel core, air and rust zone through adaptive threshold segmentation and calculates the cross-sectional area loss rate of the steel core corrosion. The data acquisition and calculation module 103 obtains the chemical state of Fe in the rusted area, collects satellite peaks of Fe-Kα and iron oxide respectively, and calculates Fe. 0 with Fe 3+ The concentration ratio between them; Output module 104, based on the cross-sectional area loss rate of steel core corrosion, Fe 0 with Fe 3+ The concentration ratio between the two values ​​outputs the corrosion level and remaining strength of the conductor under test.

[0073] It should be noted that the implementation process of the acquisition module 101, the segmentation and calculation module 102, the acquisition and calculation module 103, and the output module 104 in this embodiment is the same as the method steps of the embodiment, and will not be described again in this embodiment.

[0074] This invention involves performing multi-angle X-ray transmission on the entire conductor under test while it is charged, obtaining a projection sequence; inverting the projection sequence into a three-dimensional density volume of the steel core; based on the three-dimensional density volume of the steel core, segmenting and identifying the steel core, air, and corrosion zones, and calculating the cross-sectional area loss rate of the steel core due to corrosion; obtaining the chemical state of Fe in the corrosion zone, collecting satellite peaks of Fe-Kα and iron oxide respectively, and calculating Fe... 0 with Fe 3+ The concentration ratio between them; based on the cross-sectional area loss rate of steel core corrosion, Fe 0 with Fe 3+ The concentration ratio between the two values ​​outputs the corrosion level and remaining strength of the conductor under test, effectively solving the problem of low detection efficiency and reliability of steel core corrosion degree of overhead transmission line conductors due to existing technology, and effectively improving the detection efficiency and reliability of steel core corrosion degree of overhead transmission line conductors.

[0075] The technical solution of this invention includes the cross-sectional area loss rate of steel core corrosion and Fe. 0 with Fe 3+ Before normalizing the concentration ratios between them to form a two-dimensional feature vector, it also includes: the cross-sectional area loss rate of steel core corrosion, Fe... 0 with Fe 3+ The concentration ratios between the values ​​were aligned in time and space to ensure the reliability of the detection of corrosion degree of steel core in overhead transmission line conductors.

[0076] The technical solution of this invention includes the cross-sectional area loss rate of steel core corrosion and Fe. 0 with Fe3+ The concentration ratios between them are normalized into two-dimensional feature vectors; based on the steel core corrosion cross-sectional area loss rate vector or Fe in the two-dimensional feature vectors... 0 with Fe 3+ The numerical values ​​of the concentration ratio vector correspond to the output corrosion level and residual strength. Different on-site actions can be taken according to the corrosion level and residual strength to ensure the efficient and safe operation of the steel core of the overhead transmission line conductor.

[0077] The technical solution of this invention can complete the entire inspection process without power interruption or aluminum layer stripping. Data can be collected by drones or robots in a single flight, completely eliminating the risks of high-altitude operations. Secondly, it deeply integrates geometric cross-sectional area loss with chemical valence state changes, allowing the degree of corrosion to leap from "empirical estimation" to "scientific quantification". It provides a replicable and scalable engineering paradigm for intelligent power grid inspection.

[0078] While the specific embodiments of the present invention have been described above in conjunction with the accompanying drawings, this is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art without creative effort based on the technical solutions of the present invention are still within the scope of protection of the present invention.

Claims

1. A method for detecting the degree of corrosion of the steel core of overhead transmission line conductors, characterized in that, include: Under energized conditions, the entire conductor under test is subjected to multi-angle X-ray transmission to obtain a projection sequence; The projection sequence is inverted into a three-dimensional density volume of the steel core. Based on the three-dimensional density volume of the steel core, the steel core, air and corrosion zone are segmented and identified, and the cross-sectional area loss rate of the steel core corrosion is calculated. The chemical state of Fe was obtained in the rusted area. Satellite peaks of Fe-Kα and iron oxide were collected, and Fe was calculated. 0 with Fe 3+ The concentration ratio between them; Based on the cross-sectional area loss rate of steel core corrosion, Fe 0 with Fe 3+ The concentration ratio between the two values ​​outputs the corrosion level and remaining strength of the conductor under test.

2. The method for detecting the degree of corrosion of the steel core of an overhead transmission line conductor according to claim 1, characterized in that, The projection sequence is inverted into a three-dimensional density volume of the steel core. Based on the three-dimensional density volume of the steel core, the steel core, air, and rust areas are segmented and identified, specifically including: The projection sequence is backprojected using Shepp-Logan filtering and then iterated using SART to generate a millimeter-scale three-dimensional density volume. The grayscale value of the three-dimensional density volume of the steel core of the conductor under test is obtained. Based on the grayscale value of the three-dimensional density volume of the steel core of the conductor under test and the preset first correspondence, the material attenuation coefficient corresponding to different grayscale value ranges of the three-dimensional density volume of the steel core of the conductor under test is obtained. The steel core, air and rust areas are segmented and identified based on the different material attenuation coefficients. The preset first correspondence is the correspondence between the material attenuation coefficient and the grayscale value range.

3. The method for detecting the degree of corrosion of the steel core of an overhead transmission line conductor according to claim 2, characterized in that, The specific calculation of the cross-sectional area loss rate due to corrosion of the steel core is as follows: △A=S1 / S2 Where △A is the cross-sectional area loss rate of steel core corrosion, S1 is the corrosion area of ​​the steel core of the conductor under test, and S2 is the original design cross-sectional area of ​​the steel core of the conductor under test.

4. The method for detecting the degree of corrosion of the steel core of an overhead transmission line conductor according to claim 1, characterized in that, The chemical state of Fe was obtained in the rusted area. Satellite peaks of Fe-Kα and iron oxide were collected, and Fe was calculated. 0 with Fe 3+ The concentration ratios between them specifically include: Based on the energy-intensity spectrum, sensitivity factors were established, where Fe... 0 The sensitivity factor is the net peak area of ​​Fe-Kα and Fe 0 Concentration ratio; Fe 3+ The sensitivity factor is the net peak area of ​​iron oxide and Fe. 3+ The ratio of concentrations; Net peak area and theoretical conditions based on laboratory standard sampling Fe 0 Concentration, to obtain normalized Fe 0 Sensitivity factor, Fe 3+ Sensitivity factor; Based on normalized Fe 0 Sensitivity factor, Fe 3+ Sensitivity factor, current sampling Fe 0 with Fe 3+ The net peak area was used to obtain Fe. 0 with Fe 3+ The concentration ratio between them.

5. The method for detecting the degree of corrosion of the steel core of an overhead transmission line conductor according to claim 4, characterized in that, Based on normalized Fe 0 Sensitivity factor, Fe 3+ The sensitivity factor and the current sampled net peak area are used to obtain Fe. 0 with Fe 3+ The specific concentration ratios between them are as follows: R = C1_Fe 0 / C1_Fe 3+ = (I1_Kα / S1_Kα I1_sat / S1_sat) / (I1_sat / S1_sat); Where R is Fe 0 with Fe 3+ The concentration ratio between them, C1_Fe 0 For Fe in the current sample 0 The concentration of C1_Fe 3+ For Fe in the current sample 3+ The concentration of Fe in the current sample, I1_Kα. 0 The net peak area, I1_sat is the Fe in the current sample. 3+ The net peak area, S1_Kα is the normalized Fe 0 The sensitivity factor, S1_sat is the normalized Fe 3+ The sensitivity factor.

6. A method for detecting the degree of corrosion of the steel core of an overhead transmission line conductor according to any one of claims 1-5, characterized in that, Based on the cross-sectional area loss rate of steel core corrosion, Fe 0 with Fe 3+ The concentration ratio between the two values ​​outputs the corrosion level and residual strength of the conductor under test, and also includes: The cross-sectional area loss rate of steel core corrosion, Fe 0 with Fe 3+ The concentration ratios between them are spatiotemporally aligned.

7. The method for detecting the degree of corrosion of the steel core of an overhead transmission line conductor according to claim 1, characterized in that, Based on the cross-sectional area loss rate of steel core corrosion, Fe 0 with Fe 3+ The concentration ratio between the two values ​​outputs the corrosion level and residual strength of the conductor under test, specifically including: The cross-sectional area loss rate of steel core corrosion, Fe 0 with Fe 3+ The concentration ratios between them are normalized into two-dimensional feature vectors; If the two-dimensional feature vector contains the steel core corrosion cross-sectional area loss rate vector or Fe... 0 with Fe 3+ If the concentration ratio vector between the two values ​​satisfies a first preset condition, then the corresponding level is dangerous corrosion, and the remaining strength is less than a first percentage threshold; wherein, the first preset condition is that the steel core corrosion cross-sectional area loss rate vector in the two-dimensional feature vector is not less than a first loss rate vector threshold, or, Fe 0 with Fe 3 + The concentration ratio vector between them is not greater than the first ratio vector threshold.

8. The method for detecting the degree of corrosion of the steel core of an overhead transmission line conductor according to claim 7, characterized in that, Based on the two-dimensional feature vector, the output of the corrosion level and remaining strength of the conductor under test also includes: If the two-dimensional feature vector contains the steel core corrosion cross-sectional area loss rate vector or Fe... 0 with Fe 3+ If the concentration ratio vector between the two values ​​satisfies the second preset condition, then the corresponding level is moderate to severe corrosion, and the remaining strength is greater than the first percentage threshold and less than the second percentage threshold; wherein, the second preset condition is that the steel core corrosion cross-sectional area loss rate vector in the two-dimensional feature vector is less than the first loss rate vector threshold and not less than the second loss rate vector threshold, or, Fe 0 with Fe 3+ The concentration ratio vector between them is greater than the first ratio vector threshold and not less than the second ratio vector threshold; If the two-dimensional feature vector contains the steel core corrosion cross-sectional area loss rate vector or Fe... 0 with Fe 3+ If the concentration ratio vector between the two values ​​satisfies the third preset condition, then the corresponding level is moderate to slight corrosion, and the remaining strength is greater than the second percentage threshold and less than the third percentage threshold. The third preset condition is that the steel core corrosion cross-sectional area loss rate vector in the two-dimensional feature vector is less than the second loss rate vector threshold and not less than the third loss rate vector threshold, and Fe... 0 with Fe 3+ The concentration ratio vector between them is greater than the second ratio vector threshold and not less than the third ratio vector threshold; If the two-dimensional feature vector contains the steel core corrosion cross-sectional area loss rate vector or Fe... 0 with Fe 3+ If the concentration ratio vector between the two values ​​satisfies the fourth preset condition, then the corresponding level is mild to severe corrosion, with the remaining strength greater than the third percentage threshold and less than the fourth percentage threshold. The fourth preset condition is that the steel core corrosion cross-sectional area loss rate vector in the two-dimensional feature vector is less than the third loss rate vector threshold and not less than the fourth loss rate vector threshold, and Fe... 0 with Fe 3+ The concentration ratio vector between them is greater than the fourth ratio vector threshold; If the two-dimensional feature vector contains the steel core corrosion cross-sectional area loss rate vector or Fe... 0 with Fe 3+ If the concentration ratio vector between the two values ​​satisfies the fifth preset condition, then the corresponding level is mild to slight corrosion, and the remaining strength is greater than the fourth percentage threshold. The fifth preset condition is that the steel core corrosion cross-sectional area loss rate vector in the two-dimensional feature vector is less than the fourth loss rate vector threshold, and Fe... 0 with Fe 3+ The concentration ratio vector between them is greater than the third ratio vector threshold, wherein the first loss rate vector threshold, the second loss rate vector threshold, the third loss rate vector threshold, and the fourth loss rate vector threshold decrease in sequence, and the first ratio vector threshold, the second ratio vector threshold, the fourth ratio vector threshold, and the third ratio vector threshold increase in sequence.

9. The method for detecting the degree of corrosion of the steel core of an overhead transmission line conductor according to claim 8, characterized in that, The specific method for estimating residual strength is as follows: in: For the estimated residual strength, V[0] is the cross-sectional area loss rate vector of steel core corrosion in the two-dimensional feature vector, and V[1] is the Fe in the two-dimensional feature vector. 0 with Fe 3+ The vector of concentration ratios between them, σ_△A is the standard deviation of the cross-sectional area loss rate vector of steel core corrosion in the two-dimensional feature vector, μ_△A is the mean of the cross-sectional area loss rate vector of steel core corrosion in the two-dimensional feature vector, and σ_R is the concentration ratio vector of Fe in the two-dimensional feature vector. 0 with Fe 3+ The standard deviation of the concentration ratio vector between the two, μ_R is the Fe concentration in the two-dimensional feature vector. 0 with Fe 3+ The mean of the concentration ratio vector between them.

10. A system for detecting the degree of corrosion of steel core in overhead transmission line conductors, characterized in that, include: The acquisition module performs multi-angle X-ray transmission on the entire conductor under test while it is energized to acquire a projection sequence. The segmentation and calculation module inverts the projection sequence into a three-dimensional density volume of the steel core. Based on the three-dimensional density volume of the steel core, it identifies the steel core, air and rusted areas through adaptive threshold segmentation and calculates the cross-sectional area loss rate of the steel core due to rust. The data acquisition and calculation module obtains the chemical state of Fe in the rusted area, collects satellite peaks of Fe-Kα and iron oxide, and calculates Fe. 0 with Fe 3+ The concentration ratio between them; The output module, based on the cross-sectional area loss rate of steel core corrosion and Fe... 0 with Fe 3+ The concentration ratio between the two values ​​outputs the corrosion level and remaining strength of the conductor under test.

Images