A Method for Locating Early Aging of Cables Based on Line Loss Difference Sequence Analysis

Abstract

This invention relates to the field of cable aging detection technology, specifically a method for locating early aging of cables based on line loss difference sequence analysis. The method includes using cable segments between adjacent utility poles as the smallest diagnostic unit and collecting electrical energy data for each segment; constructing a line loss difference sequence for the cable line to be diagnosed based on actual and theoretical line losses; calculating the arithmetic mean and standard deviation of the line loss difference sequence to construct a dynamic benchmark band for initial screening; for segments with suspicious markings, obtaining the line loss difference between the segments before and after the segment to determine local maxima, calculating the relative prominence to determine the cable health index, and determining the cable diagnosis result based on the cable health index. This method is highly sensitive to minor additional losses caused by water treeing and other aging processes, effectively adapts to changes in line load, overcomes measurement noise, and enables early identification of insulation defects.

Description

Technical Field

[0001] This invention relates to the field of cable aging detection technology, specifically a method for locating early aging of cables based on line loss difference sequence analysis. Background Technology

[0002] Distribution network cables, especially their joints and bodies, are prone to early insulation defects due to factors such as water treeing, aging, and joint oxidation during long-term operation. However, these defects are initially weak, and their gradual deterioration will eventually lead to insulation breakdown accidents.

[0003] Existing aging location methods typically employ offline dielectric loss testing, which requires power outages and limits operational efficiency. Online monitoring methods, on the other hand, are mostly limited to loss analysis of the entire line or large sections, making it difficult to accurately locate specific fault points on a scale of several meters, thus failing to meet the needs of precise operation and maintenance. Summary of the Invention

[0004] To address the aforementioned problems, the first aspect of this invention provides a method for locating early aging of cables based on line loss difference sequence analysis, comprising:

[0005] The cable segment between adjacent utility poles is taken as the smallest diagnostic unit. Based on the smallest diagnostic unit, the cable line to be diagnosed is divided into multiple continuous micro segments, and the power data of each micro segment is collected.

[0006] The actual line loss and theoretical line loss of each micro-segment are calculated based on the power data, and a line loss difference sequence of the cable line to be diagnosed is constructed based on the actual line loss and theoretical line loss.

[0007] The arithmetic mean and standard deviation of the line loss difference sequence are calculated respectively. A dynamic benchmark band is constructed based on the arithmetic mean and standard deviation. The line loss difference sequence is initially screened based on the dynamic benchmark band. Micro segments with line loss difference values ​​exceeding the dynamic benchmark band are marked as suspicious.

[0008] For a suspicious micro-segment, the difference in line loss between its two ends is used to determine the local maximum. If the difference in line loss of the suspicious micro-segment is greater than the difference in line loss between the two micro-segments, the relative prominence is calculated, and the cable health index is determined based on the relative prominence. The cable diagnosis result is then determined based on the cable health index.

[0009] The electrical energy data includes the inflow of active power P. in i. Outflowing active power P out i. The root mean square current Ii flowing through the micro segment.

[0010] The theoretical line loss calculation formula is as follows:

[0011]

[0012] Where Ii represents the root mean square current flowing through the micro-segment; R is the resistance per unit length at 20 °C; α represents the temperature coefficient of copper / aluminum; θ i Let represent the surface temperature of the micro-segment; Li represents the cable length of the i-th micro-segment.

[0013] The constructed dynamic reference band is [Low, High], where Low = μ - k * σ, High = μ + k * σ, k is the confidence coefficient; μ represents the arithmetic mean of the line loss difference sequence; and σ is the standard value of the line loss difference sequence.

[0014] The formula for relative protrusion is:

[0015]

[0016] In the formula, ΔP(i) represents the line loss difference of the i-th micro segment; ΔP(i-1) represents the line loss difference of the (i-1)-th micro segment; ΔP(i+1) represents the line loss difference of the (i+1)-th micro segment; and σ is the standard value of the line loss difference sequence.

[0017] The formula for the cable health index is:

[0018]

[0019] Where Tm represents the threshold value.

[0020] The initial screening of the line loss difference sequence based on the dynamic reference band specifically involves determining whether there is ΔP(i) > High or ΔP(i) < Low. If so, the line loss difference of the micro-segment is determined to exceed the dynamic reference band.

[0021] Another aspect of the present invention provides a cable early aging location system based on line loss difference sequence analysis, comprising:

[0022] The micro-segment acquisition module is used to take the cable segment between adjacent utility poles as the smallest diagnostic unit, divide the cable line to be diagnosed into multiple continuous micro-segments according to the smallest diagnostic unit, and collect the power data of each micro-segment.

[0023] The line loss calculation module is used to calculate the actual line loss and theoretical line loss of each micro-segment based on the power data, and to construct a line loss difference sequence of the cable line to be diagnosed based on the actual line loss and theoretical line loss.

[0024] The screening and marking module is used to calculate the arithmetic mean and standard deviation of the line loss difference sequence, construct a dynamic benchmark band based on the arithmetic mean and standard deviation, perform initial screening of the line loss difference sequence based on the dynamic benchmark band, and mark the micro segments whose line loss difference exceeds the dynamic benchmark band as suspicious.

[0025] The diagnostic module is used to determine the local maximum value of the line loss difference between the two ends of a suspicious marker micro-segment. If the line loss difference of the suspicious marker micro-segment is greater than the line loss difference between the two ends of the micro-segment, the relative protrusion is calculated, and the cable health index is determined based on the relative protrusion. The cable diagnosis result is determined based on the cable health index.

[0026] Beneficial effects: This invention is a method for locating early aging of cables based on line loss difference sequence analysis. It can achieve meter-level accurate location of cable aging faults using only existing automated system data, without power outages or additional investment. By dividing the cable line into micro-segments and calculating the actual and theoretical line losses of each micro-segment to construct a line loss difference sequence, and confirming trend changes through dynamic benchmark screening, local maxima judgment, and relative prominence, it can be highly sensitive to minor additional losses caused by water tree aging, effectively adapt to changes in line load, overcome measurement noise, achieve early identification of insulation defects, improve operation and maintenance detection efficiency, and avoid insulation breakdown accidents. Detailed Implementation

[0027] Exemplary embodiments of this disclosure will now be described in more detail.

[0028] Example 1

[0029] This embodiment provides a method for locating early aging of cables based on line loss difference sequence analysis. The specific implementation steps are as follows:

[0030] Step S1: Take the cable segment between adjacent utility poles as the smallest diagnostic unit, divide the cable line to be diagnosed into multiple continuous micro segments according to the smallest diagnostic unit, and collect the power data of each micro segment respectively.

[0031] The cable line to be diagnosed is divided into n consecutive micro-segments, denoted as S1, S2, ..., Sn, with adjacent utility poles as the boundaries and the cable segment between adjacent utility poles as the smallest diagnostic unit.

[0032] The power distribution automation system collects power data for each micro-segment of the cable, including the inflow of active power P. in i. Outflowing active power P out i. The root mean square current Ii flowing through the micro segment.

[0033] Step S2: Calculate the actual line loss and theoretical line loss of each micro-segment based on the power data, and construct a line loss difference sequence for the cable line to be diagnosed based on the actual line loss and theoretical line loss.

[0034] Based on the inflow and outflow active power collected from each micro-segment, the actual line loss is calculated using the power difference method, and the formula is as follows:

[0035] Pa(i) = P in i - P out i.

[0036] Meanwhile, the theoretical line loss is calculated based on the root mean square current of each micro-segment, using the following formula:

[0037]

[0038] Where Ii represents the root mean square current flowing through the micro-segment; R is the resistance per unit length at 20 °C; α represents the temperature coefficient of copper / aluminum (0.00393 for copper, 0.00403 for aluminum); θ i Let represent the surface temperature of the micro-segment; Li represents the cable length of the i-th micro-segment.

[0039] The difference between the calculated actual line loss and the theoretical line loss is used to obtain the line loss difference value ΔP(i) = Pa(i) - Pt(i) for each micro segment. The line loss difference values ​​of all micro segments are arranged in physical spatial order to obtain the line loss difference value sequence D of the entire cable line to be diagnosed, D=[ΔP(1), ΔP(2), ..., ΔP(n)].

[0040] Step S3: Calculate the arithmetic mean and standard deviation of the line loss difference sequence respectively, construct a dynamic benchmark band based on the arithmetic mean and standard deviation, perform preliminary screening of the line loss difference sequence based on the dynamic benchmark band, and mark the micro segments with line loss difference values ​​exceeding the dynamic benchmark band as suspicious.

[0041] Calculate the arithmetic mean μ and standard deviation σ of the line loss difference series, respectively, using the following formulas:

[0042]

[0043]

[0044] A dynamic reference band is constructed, which is [Low, High], where Low = μ - k * σ, High = μ + k * σ, and k is the confidence coefficient, k=2.

[0045] This embodiment does not use an absolute threshold to filter the line loss interpolation sequence. Instead, it compares the line loss difference of each micro-segment with the overall background level and fluctuation range of the entire line. The arithmetic mean represents the average background loss of the cable line under the current operating conditions, and the standard deviation characterizes the random fluctuation amplitude of the normal micro-segment.

[0046] The line loss difference sequence is initially screened based on the dynamic reference band. Line loss differences that exceed the dynamic reference band are selected, i.e., ΔP(i) > High or ΔP(i) < Low. The micro-segments corresponding to the selected line loss differences are marked as suspicious.

[0047] Step S4: For the micro-segment of the suspicious mark, obtain the line loss difference between the two ends of the line segment before and after it and make a local maximum judgment. If the line loss difference of the micro-segment of the suspicious mark is greater than the line loss difference between the two micro-segments before and after it, calculate the relative protrusion and determine the cable health index based on the relative protrusion. Determine the cable diagnosis result based on the cable health index.

[0048] For the suspicious segment ΔP(i) selected in step S3, the line loss difference values ​​ΔP(i-1) and ΔP(i+1) of the two segments before and after the segment in the line loss difference value sequence are obtained for local maxima judgment. If ΔP(i) > ΔP(i-1) and ΔP(i) > ΔP(i+1), then the relative prominence is calculated based on the line loss difference values ​​of the two segments before and after the segment. The formula is as follows:

[0049]

[0050] Based on relative prominence, a true, isolated early fault point has a spatially localized impact. Therefore, it should appear as a sharp peak rather than a flat plateau on the line loss difference sequence D. The local maximum condition ensures that the peak of the sequence is identified, excluding interference from continuous, slightly aging sections. Relative prominence S(i) is the core quantitative indicator, and its numerator [ΔP(i)-(ΔP(i-1) + ΔP(i+1)) / 2] measures the absolute height of the peak relative to its left and right neighboring background. Dividing by σ is for normalization, making this height comparable to the random fluctuation amplitude of the entire line.

[0051] Based on the calculated relative bulge, the cable health index H(i) is calculated using the following formula:

[0052]

[0053] Where Tm represents the threshold value, Tm=3.

[0054] The cable diagnosis result is determined based on the cable health index. Different ranges of the cable health index are set, and each range corresponds to a different degree of aging of the cable line. For example, when S(i)=3, H(i)=0, which corresponds to extremely significant aging; when S(i)≤0, H(i)=1, which is completely healthy.

[0055] By obtaining cable diagnostic results, maintenance personnel can maintain aging cable lines and locate cable aging based on suspicious marked micro-segments.

[0056] In addition, a cable early aging location system based on line loss difference sequence analysis is also provided, including:

[0057] The micro-segment acquisition module is used to take the cable segment between adjacent utility poles as the smallest diagnostic unit, divide the cable line to be diagnosed into multiple continuous micro-segments according to the smallest diagnostic unit, and collect the power data of each micro-segment.

[0058] The line loss calculation module is used to calculate the actual line loss and theoretical line loss of each micro-segment based on the power data, and to construct a line loss difference sequence of the cable line to be diagnosed based on the actual line loss and theoretical line loss.

[0059] The screening and marking module is used to calculate the arithmetic mean and standard deviation of the line loss difference sequence, construct a dynamic benchmark band based on the arithmetic mean and standard deviation, perform initial screening of the line loss difference sequence based on the dynamic benchmark band, and mark the micro segments whose line loss difference exceeds the dynamic benchmark band as suspicious.

[0060] The diagnostic module is used to determine the local maximum value of the line loss difference between the two ends of a suspicious marker micro-segment. If the line loss difference of the suspicious marker micro-segment is greater than the line loss difference between the two ends of the micro-segment, the relative protrusion is calculated, and the cable health index is determined based on the relative protrusion. The cable diagnosis result is determined based on the cable health index.

Claims

1. A method for locating early aging of cables based on line loss difference sequence analysis, characterized in that, include: The cable segment between adjacent utility poles is taken as the smallest diagnostic unit. Based on the smallest diagnostic unit, the cable line to be diagnosed is divided into multiple continuous micro segments, and the power data of each micro segment is collected. The actual line loss and theoretical line loss of each micro-segment are calculated based on the power data, and a line loss difference sequence of the cable line to be diagnosed is constructed based on the actual line loss and theoretical line loss. The arithmetic mean and standard deviation of the line loss difference sequence are calculated respectively. A dynamic baseline band is constructed based on the arithmetic mean and standard deviation. The line loss difference sequence is initially screened based on the dynamic baseline band, and micro-segments with line loss differences exceeding the dynamic baseline band are marked as suspicious. The constructed dynamic baseline band is [Low, High], where Low = μ - k * σ, High = μ + k * σ, k is the confidence coefficient, μ represents the arithmetic mean of the line loss difference sequence, and σ is the standard value of the line loss difference sequence. For a suspicious micro-segment, the difference in line loss between its two ends is used to determine the local maximum. If the difference in line loss of the suspicious micro-segment is greater than the difference in line loss between the two micro-segments, the relative protrusion is calculated, and the cable health index is determined based on the relative protrusion. The cable diagnosis result is then determined based on the cable health index. The formula for relative protrusion is: In the formula, ΔP(i) represents the line loss difference of the i-th micro segment; ΔP(i-1) represents the line loss difference of the (i-1)-th micro segment; ΔP(i+1) represents the line loss difference of the (i+1)-th micro segment; σ is the standard value of the line loss difference sequence; The formula for the cable health index is: Where Tm represents the threshold value.

2. The method for locating early aging of cables according to claim 1, characterized in that, The electrical energy data includes the inflow of active power P. in i. Outflowing active power P out i. The root mean square current Ii flowing through the micro segment.

3. The method for locating early aging of cables according to claim 1, characterized in that, The theoretical line loss calculation formula is as follows: Where Ii represents the root mean square current flowing through the micro-segment; R is the resistance per unit length at 20 °C; α represents the temperature coefficient of copper / aluminum; θ i Let represent the surface temperature of the micro-segment; Li represents the cable length of the i-th micro-segment.

4. The method for locating early aging of cables according to claim 1, characterized in that, The initial screening of the line loss difference sequence based on the dynamic reference band specifically involves determining whether there is ΔP(i) > High or ΔP(i) < Low. If so, the line loss difference of the micro-segment is determined to exceed the dynamic reference band.

5. A cable early aging location system based on line loss difference sequence analysis, implementing the cable early aging location method based on line loss difference sequence analysis as described in claim 1, characterized in that, include: The micro-segment acquisition module is used to take the cable segment between adjacent utility poles as the smallest diagnostic unit, divide the cable line to be diagnosed into multiple continuous micro-segments according to the smallest diagnostic unit, and collect the power data of each micro-segment. The line loss calculation module is used to calculate the actual line loss and theoretical line loss of each micro-segment based on the power data, and to construct a line loss difference sequence of the cable line to be diagnosed based on the actual line loss and theoretical line loss. The screening and marking module is used to calculate the arithmetic mean and standard deviation of the line loss difference sequence, construct a dynamic benchmark band based on the arithmetic mean and standard deviation, perform initial screening of the line loss difference sequence based on the dynamic benchmark band, and mark the micro segments whose line loss difference exceeds the dynamic benchmark band as suspicious. The diagnostic module is used to determine the local maximum value of the line loss difference between the two ends of a suspicious marker micro-segment. If the line loss difference of the suspicious marker micro-segment is greater than the line loss difference between the two ends of the micro-segment, the relative protrusion is calculated, and the cable health index is determined based on the relative protrusion. The cable diagnosis result is determined based on the cable health index.

Images