AI full-automatic diagnosis-based overhead-cable mixed line fault analysis method

Abstract

An overhead-cable mixed line fault analysis method based on AI full-automatic diagnosis comprises the following steps: installing a plurality of voltage acquisition points on a line to be detected, and using a voltage sensor to acquire voltage data of each voltage acquisition point in real time; the collected voltage data are processed and analyzed, feature information related to faults is extracted, and the feature information comprises voltage amplitude change, waveform distortion and phase deviation; according to the extracted feature information, the position of a line fault point is judged through an algorithm and AI model analysis; the position of the fault point is displayed on a monitoring interface in a visual mode, a fault report is generated, and the fault report comprises the position of the fault point, the fault type and the detection time for operation and maintenance personnel to refer to and process. Through application of the high-precision voltage sensor and a complex data processing algorithm, fault features can be captured and identified more accurately, so that accurate positioning of fault points is realized, misinformation and missing report are reduced, and operation and maintenance efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the field of line fault analysis, and in particular to an overhead-cable hybrid line fault analysis method based on AI fully automatic diagnosis. Background Art

[0002] In traditional power systems, line faults are common and have a significant impact. They may cause power outages, equipment damage, and even safety accidents. Quickly and accurately locating the fault point is crucial to reducing power outage time, reducing economic losses, and improving power supply reliability. Hybrid lines consisting of overhead lines and cable lines are widely present, but due to differences in their manufacturing process, installation location, and electrical parameters, fault locating of hybrid lines becomes complicated and difficult. However, traditional fault locating methods often rely on manual inspections and simple electrical tests. These methods are not only time-consuming and labor-intensive, but also easily affected by environmental factors, resulting in low positioning accuracy.

[0003] Therefore, we made improvements to this and proposed an overhead-cable hybrid line fault analysis method based on AI fully automatic diagnosis. Summary of the invention

[0004] The purpose of the present invention is to solve the problem that the existing fault location methods rely on manual inspection and simple electrical testing, which are not only time-consuming and labor-intensive, but also easily affected by environmental factors, resulting in low positioning accuracy.

[0005] In order to achieve the above-mentioned purpose of the invention, the present invention provides an overhead-cable hybrid line fault analysis method based on AI fully automatic diagnosis to improve the above-mentioned problems.

[0006] The present invention is specifically as follows:

[0007] An overhead-cable hybrid line fault analysis method based on AI fully automatic diagnosis includes the following steps:

[0008] S1. Voltage data collection: Install multiple voltage collection points on the line to be tested, and use voltage sensors to collect voltage data from each voltage collection point in real time;

[0009] S2. Fault feature extraction: Process and analyze the collected voltage data to extract feature information related to the fault, including voltage amplitude changes, waveform distortion, and phase shift;

[0010] S3. Fault location determination: Based on the extracted feature information, the location of the line fault point is determined through algorithm and AI model analysis;

[0011] S4. Fault report and indication: The location of the fault point is displayed in a visual form on the monitoring interface, and a fault report is generated. The fault report includes the location of the fault point, fault type and detection time for reference and processing by operation and maintenance personnel.

[0012] Furthermore, in step S3, the line is a uniform transmission line, and the voltage before and after the fault point changes significantly. The location of the fault point is estimated by using the sudden change of voltage data before and after the fault point.

[0013] Furthermore, the method of estimating the location of the fault point by using the mutation of the voltage data before and after the fault point specifically includes: assuming that the total length of the line is L, the total number of voltage collection points is N, and the collection points are evenly distributed, if the voltage difference between the i-th collection point and the i+1-th collection point exceeds a set voltage difference threshold, it is determined that the fault point is located between the two collection points.

[0014] Furthermore, let the distance from the i-th collection point to the starting point of the line be d i , then the distance d from the fault point to the starting point of the line fault The calculation formula is:

[0015] D fault =d i +ΔV i，i+1 / ΔV total ×(d i+1 -d i );

[0016] Where ΔV i，i+1 is the voltage difference between the i-th and i+1-th collection points; ΔV total is the maximum voltage change caused by a known fault.

[0017] Furthermore, the relative proportion of voltage change is used to calculate the location of the fault point. The voltage change is proportional to the distance from the fault point to the acquisition point. The formula is:

[0018]

[0019] represents the sum of the absolute values ​​of the voltage differences between all adjacent acquisition points, which serves as a normalization factor; It represents the ratio of the voltage change between the ith collection point and the i+1th collection point to the voltage change of the entire line;

[0020] In places close to the fault point, if the voltage change is not linear, a nonlinear weight is introduced, the formula is:

[0021]

[0022] Where α is a positive number used to prevent the denominator from being zero and adjust the degree of nonlinear effects.

[0023] Furthermore, in step S3, a model is selected to determine the fault location according to the fault type, line characteristics and data features, and a data set with known fault locations and corresponding voltage characteristics is used to train the model.

[0024] Furthermore, the AI ​​model analysis fault judgment includes the following steps:

[0025] S11, input test data: input the voltage characteristic data to be detected into the trained model;

[0026] S12, model prediction: the model outputs the predicted fault location based on the input feature data;

[0027] S13, Result evaluation: Evaluate the accuracy of the model prediction and verify the performance of the model by comparing with the actual fault location;

[0028] S14, Feedback mechanism: Compare the model prediction results with the actual fault location, feed back the error information to the model, and optimize and adjust the model.

[0029] Furthermore, for fault conditions that the model cannot accurately determine, an exception handling mechanism is designed to remind operators to check.

[0030] Compared with the prior art, the present invention has the following beneficial effects:

[0031] 1. In order to solve the problem that the fault location methods in the prior art rely on manual inspections and simple electrical tests, which are not only time-consuming and labor-intensive, but also easily affected by environmental factors, resulting in low positioning accuracy, the present invention uses high-precision voltage sensors and complex data processing algorithms to more accurately capture and identify fault characteristics, thereby achieving accurate positioning of the fault point, which helps to reduce false alarms and missed alarms and improve operation and maintenance efficiency;

[0032] 2. The present invention collects and analyzes voltage data in real time, and can provide positioning results in the shortest time after a fault occurs, providing strong support for rapid power restoration, which is of great significance for ensuring safe and stable operation of the power grid;

[0033] 3. The present invention improves the reliability of fault location through multi-point voltage collection and comprehensive analysis. Even in the face of complex and changeable fault conditions, the accuracy of the positioning results can be ensured through multi-source data fusion and algorithm optimization. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 A flow chart of an overhead-cable hybrid line fault analysis method based on AI fully automatic diagnosis provided by the present invention;

[0035] Figure 2 Statistical chart of current, voltage and distance parameters provided by the present invention. DETAILED DESCRIPTION

[0036] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0037] Specifically, please refer to Figure 1 , an overhead-cable hybrid line fault analysis method based on AI fully automatic diagnosis, comprising the following steps:

[0038] S1. Voltage data collection: Install multiple voltage collection points on the line to be tested, and use voltage sensors to collect voltage data from each voltage collection point in real time;

[0039] S2. Fault feature extraction: Process and analyze the collected voltage data to extract feature information related to the fault, including voltage amplitude changes, waveform distortion, and phase shift;

[0040] S3. Fault location determination: Based on the extracted feature information, the location of the line fault point is determined through algorithm and AI model analysis;

[0041] S4. Fault report and indication: The location of the fault point is displayed in a visual form on the monitoring interface, and a fault report is generated. The fault report includes the location of the fault point, fault type and detection time for reference and processing by operation and maintenance personnel.

[0042] The present invention, through the application of high-precision voltage sensors and complex data processing algorithms, can more accurately capture and identify fault characteristics, thereby accurately locating the fault point, which helps to reduce false alarms and missed alarms and improve operation and maintenance efficiency.

[0043] In step S3, the line is a uniform transmission line, and the voltage before and after the fault point changes significantly. The location of the fault point is estimated by using the sudden change of voltage data before and after the fault point.

[0044] Determine the voltage mutation point in step S3: by comparing the voltage data of adjacent voltage collection points, find the point where the voltage changes significantly, calculate the absolute value of the voltage difference between adjacent points, and set a threshold value. Assuming that the total length of the line is L, the total number of voltage collection points is N, and the collection points are evenly distributed, if the voltage difference between the i-th collection point and the i+1-th collection point exceeds the set threshold, it is determined that the fault point is located between these two collection points.

[0045] Let the distance from the i-th collection point to the starting point of the line be d i , then the distance d from the fault point to the starting point of the line fault

[0046] The formula is: D fault =d i +ΔV i，i+1 / ΔV total ×(d i+1 -d i )

[0047] ΔV i，i+1 is the voltage difference between the i-th and i+1-th collection points; ΔV total is the maximum voltage change caused by a known fault,

[0048] The relative proportion of voltage change is used to calculate the fault location. The voltage change is proportional to the distance from the fault point to the acquisition point. The formula is:

[0049]

[0050] represents the sum of the absolute values ​​of the voltage differences between all adjacent acquisition points, which serves as a normalization factor; It represents the ratio of the voltage change between the i-th and i+1-th acquisition points to the voltage change of the entire line;

[0051] In places close to the fault point, if the voltage change is not linear, a nonlinear weight is introduced, the formula is:

[0052]

[0053] α is a positive number used to prevent the denominator from being zero and to adjust the degree of nonlinear effects.

[0054] In step S3, a model is selected to determine the fault location according to the fault type, line characteristics and data features, and a data set with known fault locations and corresponding voltage characteristics is used to train the model.

[0055] AI model analysis fault judgment includes the following steps:

[0056] S11, input test data: input the voltage characteristic data to be detected into the trained model;

[0057] S12, model prediction: the model outputs the predicted fault location based on the input feature data;

[0058] S13, Result evaluation: Evaluate the accuracy of the model prediction and verify the performance of the model by comparing with the actual fault location;

[0059] S14, Feedback mechanism: Compare the model prediction results with the actual fault location, feed back the error information to the model, and optimize and adjust the model.

[0060] For fault conditions that the model cannot accurately judge, an exception handling mechanism is designed to remind operators to check.

[0061] Figure 2 The present invention provides a model training data result of an overhead-cable hybrid line fault analysis method based on AI fully automatic diagnosis. The fault point analysis conclusion is obtained through model training under different preset current and voltage conditions. The results show that the fault point prediction accuracy obtained through multi-source data fusion and algorithm optimization is relatively high.

[0062] Obviously, the embodiments described above are only some embodiments of the present invention, rather than all embodiments. The preferred embodiments of the present invention are given in the accompanying drawings, but they do not limit the patent scope of the present invention. The present invention can be implemented in many different forms. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosure of the present invention more thorough and comprehensive. Although the present invention has been described in detail with reference to the aforementioned embodiments, for those skilled in the art, it is still possible to modify the technical solutions recorded in the aforementioned specific embodiments, or to perform equivalent replacements for some of the technical features therein. Any equivalent structure made using the contents of the specification and drawings of the present invention, directly or indirectly used in other related technical fields, is also within the scope of patent protection of the present invention.

Claims

1. A method for fault analysis of overhead-cable hybrid lines based on AI fully automatic diagnosis, characterized in that: The following steps are involved: S1. Voltage data collection: Install multiple voltage collection points on the line to be tested, and use voltage sensors to collect voltage data from each voltage collection point in real time; S2. Fault feature extraction: Process and analyze the collected voltage data to extract feature information related to the fault, including voltage amplitude changes, waveform distortion, and phase shift; S3. Fault location determination: Based on the extracted feature information, the location of the line fault point is determined through algorithm and AI model analysis; S4. Fault report and indication: The location of the fault point is displayed in a visual form on the monitoring interface, and a fault report is generated. The fault report includes the location of the fault point, fault type and detection time for reference and processing by operation and maintenance personnel.

2. According to claim 1, a method for analyzing overhead-cable hybrid line faults based on AI automatic diagnosis is characterized in that: In step S3, the line is a uniform transmission line, and the voltage before and after the fault point changes significantly. The location of the fault point is estimated by using the sudden change of voltage data before and after the fault point.

3. The method for analyzing overhead-cable hybrid line faults based on AI automatic diagnosis according to claim 2 is characterized in that: The method of estimating the location of the fault point by utilizing the mutation of voltage data before and after the fault point specifically includes: assuming that the total length of the line is L, the total number of voltage collection points is N, and the collection points are evenly distributed, if the voltage difference between the i-th collection point and the i+1-th collection point exceeds a set voltage difference threshold, it is determined that the fault point is located between the two collection points.

4. The method for analyzing overhead-cable hybrid line faults based on AI automatic diagnosis according to claim 3 is characterized in that: Let the distance from the i-th collection point to the starting point of the line be d i , then the distance d from the fault point to the starting point of the line fault The calculation formula is: D fault =d i +ΔV i，i+1 / ΔV total ×(d i+1 -d i ); Where ΔV i，i+1 is the voltage difference between the i-th and i+1-th collection points; ΔV total is the maximum voltage change caused by a known fault.

5. The method for analyzing overhead-cable hybrid line faults based on AI automatic diagnosis according to claim 4 is characterized in that: The relative proportion of voltage change is used to calculate the location of the fault point. The voltage change is proportional to the distance from the fault point to the acquisition point. The formula is: represents the sum of the absolute values ​​of the voltage differences between all adjacent acquisition points, which serves as a normalization factor; It represents the ratio of the voltage change between the ith collection point and the i+1th collection point to the voltage change of the entire line; In places close to the fault point, if the voltage change is not linear, a nonlinear weight is introduced, the formula is: Where α is a positive number used to prevent the denominator from being zero and adjust the degree of nonlinear effects.

6. The method for analyzing overhead-cable hybrid line faults based on AI automatic diagnosis according to claim 1 is characterized in that: In step S3, a model is selected to determine the fault location according to the fault type, line characteristics and data features, and a data set with known fault locations and corresponding voltage characteristics is used to train the model.

7. The method for analyzing overhead-cable hybrid line faults based on AI automatic diagnosis according to claim 6 is characterized in that: The AI ​​model analysis fault judgment includes the following steps: S11, input test data: input the voltage characteristic data to be detected into the trained model; S12, model prediction: the model outputs the predicted fault location based on the input feature data; S13, Result evaluation: Evaluate the accuracy of the model prediction and verify the performance of the model by comparing with the actual fault location; S14, Feedback mechanism: Compare the model prediction results with the actual fault location, feed back the error information to the model, and optimize and adjust the model.

8. The method for analyzing overhead-cable hybrid line faults based on AI automatic diagnosis according to claim 7 is characterized in that: For fault conditions that the model cannot accurately judge, an exception handling mechanism is designed to remind operators to check.

Images