Intelligent transformer area line anomaly detection method and system based on artificial intelligence

Abstract

The invention discloses an intelligent transformer area line anomaly detection method and system based on artificial intelligence, and relates to the technical field of transformer area line detection, and the method comprises the following steps: S1, obtaining the power supply amount of a transformer area, obtaining the power consumption of the transformer area, judging whether the line loss of the transformer area is normal or not through the power supply amount of the transformer area and the power consumption of the transformer area, and executing S2 if the line loss of the transformer area is abnormal; the line with serious line loss is found out from a plurality of branch lines through the leakage current sensor, different areas on the branch lines are detected through the leakage current sensor, secondary confirmation is carried out on the detection result through the characteristic that the temperature of the line loss position is high, and therefore the accuracy of the line loss positioning result can be improved.

Description

Technical Field

[0001] This invention relates to the field of transformer substation line detection technology, specifically to an intelligent transformer substation line anomaly detection method and system based on artificial intelligence. Background Technology

[0002] A distribution transformer area is a basic management unit in a power distribution system. It typically refers to the physical area supplied by a single distribution transformer, covering the low-voltage side (0.4kV) lines, equipment, and user groups. Its core functions include power distribution, load management, fault isolation, and power consumption monitoring. Smart distribution transformer areas integrate IoT, big data, and artificial intelligence technologies to achieve real-time monitoring, automated operation and maintenance, and energy efficiency optimization, thereby improving power supply reliability, reducing operation and maintenance costs, and supporting flexible control of diverse loads such as photovoltaics and energy storage under the new power system.

[0003] Transformer area line loss refers to the loss of electricity during transmission. Due to the numerous causes of line loss, such as aging lines causing leakage, electricity theft, long-term lack of calibration of user distribution meters, aging power supply equipment, and non-compliant power supply lines, it is difficult to trace the cause of line loss when it is severe. At the same time, due to the intricate network of lines, the location of faulty lines is also difficult to determine. Currently, most transformer area line loss still relies on power maintenance personnel to check each line one by one, which is not only time-consuming and labor-intensive but also puts a great deal of work pressure on power maintenance personnel. To address this, we propose a smart transformer area line anomaly detection method and system based on artificial intelligence. Summary of the Invention

[0004] (a) Technical problems to be solved To address the shortcomings of existing technologies, this invention provides a smart transformer substation line anomaly detection method and system based on artificial intelligence, in order to solve the aforementioned problems in the existing technologies.

[0005] (II) Technical Solution To achieve the above objectives, the present invention provides the following technical solution: a smart transformer substation line anomaly detection method based on artificial intelligence, comprising the following steps: S1: Obtain the power supply and power consumption of the transformer area. Determine whether the line loss of the transformer area is normal by comparing the power supply and power consumption. If the line loss of the transformer area is abnormal, execute S2. S2: Multiple leakage current sensors are installed on each branch line in the transformer area. The leakage current sensors are used to locate the branch line with line loss. The thermal infrared imaging scanner is used to scan the branch line with line loss to obtain a thermal image of the line. The multiple leakage current sensors on the branch line with line loss and the thermal image of the line are used to locate the low-temperature line loss line section and the high-temperature line loss line section. S3: Set the observation period, obtain the line loss change status within the observation period, and determine whether the line loss is caused by technical line loss or management line loss by the line loss change status, low temperature line loss line section and high temperature line loss line section within the observation period. S4: Obtain the thermal image of the line segment with line loss, determine the cause of technical line loss based on the thermal image, mark the location of line loss, and send the cause of technical line loss and the location of line loss to the management personnel; S5: Identify suspicious users by identifying high-temperature and low-temperature line loss sections, obtain historical verification records of the suspicious user's distribution meter, determine the cause of the line loss by using the historical verification marks of the suspicious user's distribution meter, mark the location of the suspicious user, and send the cause of the line loss and the location of the line loss to the management personnel.

[0006] Preferably, in S1, the line loss of the transformer area is judged to be normal by the power supply and power consumption of the transformer area. Specifically: S101: Obtain the power supply of the transformer area, obtain the power consumption of the transformer area, obtain the line loss power by subtracting the power supply of the transformer area from the power consumption of the transformer area, and obtain the line loss rate by dividing the line loss power by the power supply of the transformer area and multiplying it by 100%. S102: Set a preset threshold for safe line loss rate, determine whether the line loss rate is higher than the preset threshold for safe line loss rate. If the line loss rate is lower than or equal to the preset threshold for safe line loss rate, mark it as normal line loss and repeat step 101. If the line loss rate is higher than the preset threshold for safe line loss rate, mark it as abnormal line loss.

[0007] Preferably, in S2, the leakage current sensor is used to lock the branch line with line loss, specifically as follows: S201: Obtain the leakage current on each branch line and mark it as the starting leakage current, and set the safe leakage current preset threshold; S202: Determine whether the starting leakage current of each branch line is higher than the preset threshold for safe leakage current. If the starting leakage current of the branch line is higher than the preset threshold for safe leakage current, mark the branch line as a line loss branch line. If the starting leakage current of the branch line is lower than or equal to the preset threshold for safe leakage current, mark the branch line as a normal branch line.

[0008] Preferably, in S2, the low-temperature line loss section and the high-temperature line loss section are identified by multiple leakage current sensors and thermal images of the line loss branch line. Specifically: S203: Obtain the leakage current values ​​of multiple leakage current sensors on the branch line respectively, and calculate the leakage current difference by subtracting the leakage current values ​​of two adjacent leakage current sensors in each group. S204: Obtain the starting leakage current on the branch line, set the fault-tolerant leakage current, obtain the minimum starting leakage current by subtracting the starting leakage current from the fault-tolerant leakage current, and obtain the leakage current judgment range by taking the minimum starting leakage current as the minimum value and the starting leakage current as the maximum value. S205: Determine whether multiple leakage current differences are within the leakage current judgment range. If the leakage current difference is within the leakage current judgment range, mark the line between the corresponding two leakage current sensors as a suspected line loss line segment. S206: Obtain the thermal image of the line section with suspected line loss, obtain the thermal image of the line section with non-suspected line loss on the line loss branch line, and obtain the temperature fluctuation value by subtracting the line temperature of the line section with suspected line loss from the line temperature of the line section with non-suspected line loss on the line loss branch line. S207: Set a preset threshold for temperature fluctuation value, determine whether the temperature fluctuation value is greater than the preset threshold for temperature fluctuation value. If the temperature fluctuation value is greater than the preset threshold for temperature fluctuation value, mark the suspected line loss line segment as a high temperature line loss line segment. If the temperature fluctuation value is less than or equal to the preset threshold for temperature fluctuation value, mark the suspected line loss line segment as a low temperature line loss line segment.

[0009] Preferably, in S3, the change status of line loss within the observation period is obtained, specifically as follows: S301: Set the observation period, measure the leakage current of the line loss branch once at 9:00, 15:00 and 23:00 every day, and sum the three leakage currents every day and take the average to obtain the daily average leakage current. S302: Arrange all daily average leakage currents in chronological order of date, subtract the daily average leakage currents of all two adjacent days to obtain the leakage current difference, obtain the lowest leakage current difference and the highest leakage current difference, take the absolute value of the lowest leakage current difference and the highest leakage current difference respectively, and sum them to obtain the leakage current fluctuation value. S303: Set a preset threshold for leakage current fluctuation value, determine whether the leakage current fluctuation value is greater than the preset threshold for leakage current fluctuation value. If the leakage current fluctuation value is less than or equal to the preset threshold for leakage current fluctuation value, mark the line loss change state as stable. If the leakage current fluctuation value is greater than the preset threshold for leakage current fluctuation value, mark the line loss change state as jumping.

[0010] Preferably, in S3, the cause of line loss is determined as either technical or managerial loss by observing the changes in line loss within the observation period, and identifying line segments with low-temperature and high-temperature losses. Specifically: S304: Determine whether the line loss section is a low-temperature line loss section or a high-temperature line loss section. If the line loss section is a low-temperature line loss section, mark the cause of the line loss as technical line loss. If the line loss section is a high-temperature line loss section, execute S305. S305: Obtain the line loss change status within the observation period. If the line loss change status within the observation period is a jump, mark the line loss cause as management line loss. If the line loss change status within the observation period is stable, mark the line loss cause as technical line loss.

[0011] Preferably, in S4, the cause of technical line loss is determined and the location of line loss is marked using the line thermal image, specifically as follows: S401: Obtain a thermal image of the line loss section, mark ten target points with equal spacing on the cable in the thermal image, obtain the temperature of each of the ten target points, sum the temperatures of the ten target points and take the average value to obtain the mean temperature. S402: Obtain the highest temperature among the ten target points, calculate the temperature difference by subtracting the highest temperature from the average temperature, set a preset threshold for the temperature difference, and determine whether the temperature difference is greater than the preset threshold. If the temperature difference is greater than or equal to the preset threshold, mark the cause of the technical line loss as cable aging and leakage and mark the point as the location of the line loss. If the temperature difference is less than the preset threshold, execute S403. S403: Based on the cable width of ten target points, select the minimum value among the cable widths of the ten target points and mark it as the minimum cable width. Set a preset threshold for cable width and determine whether the minimum cable width is greater than the preset threshold. If the minimum cable width is less than or equal to the preset threshold, mark the cause of technical line loss as the paint being burned through due to the cable being too thin and mark that point as the location of line loss.

[0012] Preferably, in S5, suspicious users are identified by identifying high-temperature and low-temperature line loss sections, historical verification records of the suspicious user's distribution meter are obtained, and the cause of the line loss is determined by the historical verification marks of the suspicious user's distribution meter, and the location of the suspicious user is marked. Specifically: S501: Obtain user information registered at the location of high-temperature line loss lines and low-temperature line loss lines and mark them as suspicious users; obtain the historical verification records of distribution meters of suspicious users. S502: Set the meter verification cycle. Determine whether the meter of a suspected user has exceeded the verification cycle by checking the historical verification records of the meter. If the meter of a suspected user has exceeded the verification cycle, mark the management line loss as overdue meter verification and mark the location of the suspected user. If the meter of a suspected user has not exceeded the verification cycle, mark the management line loss as electricity theft and mark the location of the suspected user.

[0013] An AI-based intelligent transformer substation line anomaly detection system includes the following modules: The transformer area line loss detection module obtains the power supply and power consumption of the transformer area, and determines whether the line loss of the transformer area is normal by using the power supply and power consumption of the transformer area. If the line loss of the transformer area is abnormal, S2 is executed. The line loss location module installs multiple leakage current sensors on each branch line in the transformer area. The leakage current sensors are used to locate the branch line with line loss. A thermal infrared imaging scanner is used to scan the branch line with line loss to obtain a thermal image of the line. The multiple leakage current sensors on the branch line with line loss and the thermal image of the line are used to locate the low-temperature line loss line section and the high-temperature line loss line section. The line loss cause analysis module sets the observation period, obtains the line loss change status within the observation period, and determines whether the line loss is technical line loss or management line loss based on the line loss change status within the observation period, the line loss section with low temperature loss, and the line loss section with high temperature loss. The technical line loss cause analysis and marking module acquires the line thermal image of the line segment with line loss, determines the cause of technical line loss based on the line thermal image, marks the location of line loss, and sends the cause of technical line loss and the location of line loss to the management personnel; The line loss cause analysis and marking module identifies suspicious users by identifying high-temperature and low-temperature line loss sections, obtains historical verification records of the suspicious user's distribution meter, determines the cause of the line loss by marking the historical verification records of the suspicious user's distribution meter, marks the location of the suspicious user, and sends the cause of the line loss and the location of the line loss to the management personnel.

[0014] (III) Beneficial Effects This invention provides a method and system for detecting line anomalies in intelligent transformer substations based on artificial intelligence, which has the following beneficial effects: (1) In this scheme, the leakage current sensor is used to find the line with serious line loss among many branch lines. Then, the leakage current sensor is used to detect different areas on the branch line. Since the leakage current sensor is distributed on the line and is easily damaged by wind and rain, the detection results are inaccurate. Therefore, after the branch line is detected by the leakage current sensor, the branch line is scanned by the thermal infrared sensor. The high temperature at the line loss point is used to confirm the detection results for the second time, which helps to improve the accuracy of the line loss location results.

[0015] (2) In this scheme, the cause of line loss is determined by observing the change of line loss within the observation period, the line segment with low temperature loss, and the line segment with high temperature loss. Since the cause of line loss is due to two reasons, namely, user electricity theft and user distribution meter not being calibrated for a long time, and the cause of line loss is due to two reasons, namely, cable aging and leakage and cable thickness non-compliant heating and burning through the paint, the leakage change of line loss is usually manifested as large leakage current fluctuation, while the leakage change of line loss is usually manifested as small and stable leakage current fluctuation. Therefore, the cause of line loss is analyzed by the change of line loss. At the same time, the line loss of management line loss is usually manifested as high temperature of the line, and the line loss of technical line loss is also usually manifested as high temperature of the line, but the temperature is far less than that of the line with management line loss. Therefore, the cause of line loss can be determined by temperature change, which facilitates comprehensive analysis of the cause of line loss from multiple angles and directions, and avoids the inaccuracy of the analysis result caused by the simplification of the analysis process. Attached Figure Description

[0016] Figure 1 This is a flowchart of an intelligent transformer substation line anomaly detection method based on artificial intelligence according to the present invention. Figure 2 This is a schematic diagram of the module structure of an intelligent transformer substation line anomaly detection system based on artificial intelligence according to the present invention. Figure 3 This is a schematic diagram of the logical structure of an intelligent transformer substation line anomaly detection method based on artificial intelligence according to the present invention. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] Please see Figures 1-3 This invention provides a smart transformer substation line anomaly detection method based on artificial intelligence, comprising the following steps: S1: Obtain the power supply and power consumption of the transformer area. Determine whether the line loss of the transformer area is normal by comparing the power supply and power consumption. If the line loss of the transformer area is abnormal, execute S2. S2: Multiple leakage current sensors are installed on each branch line in the transformer area. The leakage current sensors are used to locate the branch line with line loss. The thermal infrared imaging scanner is used to scan the branch line with line loss to obtain a thermal image of the line. The multiple leakage current sensors on the branch line with line loss and the thermal image of the line are used to locate the low-temperature line loss line section and the high-temperature line loss line section. S3: Set the observation period, obtain the line loss change status within the observation period, and determine whether the line loss is caused by technical line loss or management line loss by the line loss change status, low temperature line loss line section and high temperature line loss line section within the observation period. S4: Obtain the thermal image of the line segment with line loss, determine the cause of technical line loss based on the thermal image, mark the location of line loss, and send the cause of technical line loss and the location of line loss to the management personnel; S5: Identify suspicious users by identifying high-temperature and low-temperature line loss sections, obtain historical verification records of the suspicious user's distribution meter, determine the cause of the line loss by using the historical verification marks of the suspicious user's distribution meter, mark the location of the suspicious user, and send the cause of the line loss and the location of the line loss to the management personnel.

[0019] In this embodiment, the solution uses the power supply and power consumption of the transformer area to determine whether the line loss of the transformer area is normal, which facilitates the subsequent capture of the line loss location and analysis of the cause of the line loss. In this solution, a leakage current sensor is used to identify the branch lines with severe line loss among many branch lines. Then, the leakage current sensor is used to detect different areas on the branch lines. Since the leakage current sensor is distributed on the line and is easily damaged by wind and rain over a long period of time, the detection results are inaccurate. Therefore, after detecting the branch lines with the leakage current sensor, a thermal infrared sensor is used to scan the branch lines. The high temperature at the line loss point is used to confirm the detection results a second time, which helps to improve the accuracy of the line loss location results. This solution determines whether line loss is technical or managerial by observing the changes in line loss over a period of time, and by identifying low-temperature and high-temperature line loss sections. Managerial line loss can be caused by user electricity theft or long-term non-calibration of user distribution meters, while technical line loss can be caused by cable aging and leakage or overheating and burning through the enamel of cables of non-compliant thickness. The leakage current of managerial line loss is usually characterized by large fluctuations, while the leakage current of technical line loss is usually characterized by small and stable fluctuations. Therefore, the cause of line loss can be analyzed by observing the changes in line loss. At the same time, both managerial and technical line loss are usually characterized by high line temperatures, but the temperature is much lower than that of managerial line loss. Therefore, the cause of line loss can be determined by observing temperature changes. This allows for a comprehensive analysis of the cause of line loss from multiple angles and perspectives, avoiding inaccurate results caused by a single analysis process. This solution uses thermal imaging of the line to determine whether the technical line loss is caused by cable aging and leakage or by overheating and burning through the coating due to non-compliant cable thickness. This allows maintenance personnel to carry out targeted repairs when they arrive at the site. This solution identifies suspicious users by identifying high-temperature and low-temperature line loss sections. By retrieving the historical calibration records of the distribution meters of these suspicious users, it is determined whether the high line loss is caused by the distribution meters not being calibrated for a long time. Often, such long-term uncalibrated distribution meters are due to aging and poorly maintained lines that have not been replaced in a timely manner. Therefore, the leakage of current in the surrounding lines of these distribution meters is relatively serious. If there is no historical calibration record for the distribution meters, it indicates that there may be users stealing electricity. Therefore, the location of the suspicious users is sent to the management personnel to arrange for power personnel to conduct targeted on-site investigations. It is worth mentioning that the value of the preset threshold in this scheme can be obtained through weight analysis, which will not be elaborated on here.

[0020] In S1, the line loss of a transformer area is determined by the power supply and power consumption of that area. Specifically: S101: Obtain the power supply of the transformer area, obtain the power consumption of the transformer area, obtain the line loss power by subtracting the power supply of the transformer area from the power consumption of the transformer area, and obtain the line loss rate by dividing the line loss power by the power supply of the transformer area and multiplying it by 100%. S102: Set a preset threshold for safe line loss rate, determine whether the line loss rate is higher than the preset threshold for safe line loss rate. If the line loss rate is lower than or equal to the preset threshold for safe line loss rate, mark it as normal line loss and repeat step 101. If the line loss rate is higher than the preset threshold for safe line loss rate, mark it as abnormal line loss.

[0021] In S2, the leakage current sensor is used to pinpoint the branch line with the line loss, specifically as follows: S201: Obtain the leakage current on each branch line and mark it as the starting leakage current, and set the safe leakage current preset threshold; S202: Determine whether the starting leakage current of each branch line is higher than the preset threshold for safe leakage current. If the starting leakage current of the branch line is higher than the preset threshold for safe leakage current, mark the branch line as a line loss branch line. If the starting leakage current of the branch line is lower than or equal to the preset threshold for safe leakage current, mark the branch line as a normal branch line.

[0022] In this embodiment, a leakage current sensor is used to identify the branch line with severe line loss among many branch lines. Then, the leakage current sensor is used to detect different areas on the branch line to lock the suspected leakage area of ​​the branch line. This makes it easier to narrow down the investigation scope step by step and reduce the difficulty and workload of power workers in the investigation.

[0023] In S2, multiple leakage current sensors and thermal images of the line loss branch lines are used to identify low-temperature and high-temperature line loss sections. Specifically: S203: Obtain the leakage current values ​​of multiple leakage current sensors on the branch line respectively, and calculate the leakage current difference by subtracting the leakage current values ​​of two adjacent leakage current sensors in each group. S204: Obtain the starting leakage current on the branch line, set the fault-tolerant leakage current, obtain the minimum starting leakage current by subtracting the starting leakage current from the fault-tolerant leakage current, and obtain the leakage current judgment range by taking the minimum starting leakage current as the minimum value and the starting leakage current as the maximum value. S205: Determine whether multiple leakage current differences are within the leakage current judgment range. If the leakage current difference is within the leakage current judgment range, mark the line between the corresponding two leakage current sensors as a suspected line loss line segment. S206: Obtain the thermal image of the line section with suspected line loss, obtain the thermal image of the line section with non-suspected line loss on the line loss branch line, and obtain the temperature fluctuation value by subtracting the line temperature of the line section with suspected line loss from the line temperature of the line section with non-suspected line loss on the line loss branch line. S207: Set a preset threshold for temperature fluctuation value, determine whether the temperature fluctuation value is greater than the preset threshold for temperature fluctuation value. If the temperature fluctuation value is greater than the preset threshold for temperature fluctuation value, mark the suspected line loss line segment as a high temperature line loss line segment. If the temperature fluctuation value is less than or equal to the preset threshold for temperature fluctuation value, mark the suspected line loss line segment as a low temperature line loss line segment.

[0024] In this embodiment, since the leakage current fluctuates greatly at both ends of the cable, the leakage point can be determined between two leakage current sensors by monitoring and comparing the values ​​of multiple leakage current sensors. This segment is the suspected line loss section. Since the leakage current sensors distributed on the line are easily damaged by wind and rain over a long period of time, resulting in inaccurate detection results, after detecting the branch line with the leakage current sensor, a thermal infrared sensor is used to scan the branch line. The high temperature at the line loss point is used to confirm the detection results a second time, which helps to improve the accuracy of the line loss location result.

[0025] In S3, the change status of line loss within the observation period is obtained, specifically as follows: S301: Set the observation period, measure the leakage current of the line loss branch once at 9:00, 15:00 and 23:00 every day, and sum the three leakage currents every day and take the average to obtain the daily average leakage current. S302: Arrange all daily average leakage currents in chronological order of date, subtract the daily average leakage currents of all two adjacent days to obtain the leakage current difference, obtain the lowest leakage current difference and the highest leakage current difference, take the absolute value of the lowest leakage current difference and the highest leakage current difference respectively, and sum them to obtain the leakage current fluctuation value. S303: Set a preset threshold for leakage current fluctuation value, determine whether the leakage current fluctuation value is greater than the preset threshold for leakage current fluctuation value. If the leakage current fluctuation value is less than or equal to the preset threshold for leakage current fluctuation value, mark the line loss change state as stable. If the leakage current fluctuation value is greater than the preset threshold for leakage current fluctuation value, mark the line loss change state as jumping.

[0026] In this embodiment, since the leakage line discharges at different degrees, the leakage current sensor detects different leakage current data under different discharge levels. By analyzing the leakage current fluctuations, the line loss change state can be divided into jump and stable, which facilitates the subsequent analysis of the cause of line loss.

[0027] In S3, the cause of line loss is determined by observing the changes in line loss within the observation period, as well as the line loss in low-temperature and high-temperature sections. Specifically: S304: Determine whether the line loss section is a low-temperature line loss section or a high-temperature line loss section. If the line loss section is a low-temperature line loss section, mark the cause of the line loss as technical line loss. If the line loss section is a high-temperature line loss section, execute S305. S305: Obtain the line loss change status within the observation period. If the line loss change status within the observation period is a jump, mark the line loss cause as management line loss. If the line loss change status within the observation period is stable, mark the line loss cause as technical line loss.

[0028] In this embodiment, the leakage current change state of managed line loss is usually characterized by large fluctuations in leakage current, while the leakage current change state of technical line loss is usually characterized by small and stable fluctuations in leakage current. Therefore, the cause of line loss can be analyzed by the change state of line loss. At the same time, managed line loss is usually characterized by high line temperature, and technical line loss is also usually characterized by high line temperature, but the temperature is much lower than that of managed line loss. Therefore, the cause of line loss can be determined by temperature change, which facilitates a comprehensive analysis of the cause of line loss from multiple angles and perspectives, and avoids inaccurate analysis results caused by a single analysis process.

[0029] In S4, the cause of technical line loss is determined and the location of line loss is marked using line thermal images. Specifically: S401: Obtain a thermal image of the line loss section, mark ten target points with equal spacing on the cable in the thermal image, obtain the temperature of each of the ten target points, sum the temperatures of the ten target points and take the average value to obtain the mean temperature. S402: Obtain the highest temperature among the ten target points, calculate the temperature difference by subtracting the highest temperature from the average temperature, set a preset threshold for the temperature difference, and determine whether the temperature difference is greater than the preset threshold. If the temperature difference is greater than or equal to the preset threshold, mark the cause of the technical line loss as cable aging and leakage and mark the point as the location of the line loss. If the temperature difference is less than the preset threshold, execute S403. S403: Based on the cable width of ten target points, select the minimum value among the cable widths of the ten target points and mark it as the minimum cable width. Set a preset threshold for cable width and determine whether the minimum cable width is greater than the preset threshold. If the minimum cable width is less than or equal to the preset threshold, mark the cause of technical line loss as the paint being burned through due to the cable being too thin and mark that point as the location of line loss.

[0030] In this embodiment, the temperature of the damaged line segment is analyzed by taking a thermal image of the line when determining the location of the line loss. The temperature condition includes the uniformity of temperature distribution and the temperature level at different points in the cable. If the cable diameter is too small, leakage is usually manifested as excessively high temperature and uniform temperature distribution throughout the cable. If the cable is aged, leakage is usually manifested as high temperature and uneven temperature distribution at the damaged point. This characteristic can be used to analyze the specific cause of the technical line loss, so that maintenance personnel can carry out targeted maintenance work when they arrive at the site.

[0031] In S5, suspicious users are identified by identifying high-temperature and low-temperature line loss sections. Historical verification records of the suspicious users' distribution meters are obtained. The causes of line loss are determined by analyzing these historical verification records, and the locations of the suspicious users are marked. Specifically: S501: Obtain user information registered at the location of high-temperature line loss lines and low-temperature line loss lines and mark them as suspicious users; obtain the historical verification records of distribution meters of suspicious users. S502: Set the meter verification cycle. Determine whether the meter of a suspected user has exceeded the verification cycle by checking the historical verification records of the meter. If the meter of a suspected user has exceeded the verification cycle, mark the management line loss as overdue meter verification and mark the location of the suspected user. If the meter of a suspected user has not exceeded the verification cycle, mark the management line loss as electricity theft and mark the location of the suspected user.

[0032] In this embodiment, suspicious users are identified by identifying high-temperature and low-temperature line loss sections. By retrieving the historical verification records of the distribution meters of suspicious users, it is determined whether the high line loss is caused by the distribution meters not being verified for a long time. Often, such distribution meters that have not been verified for a long time are mostly due to the fact that the lines are old and poorly maintained and not replaced in a timely manner. Therefore, the leakage of the lines around such distribution meters is relatively serious. If there is no historical verification record of the distribution meters, it indicates that there may be users stealing electricity. Therefore, the location of the suspicious users is sent to the management personnel to arrange for power personnel to conduct targeted on-site investigations.

[0033] Please see Figures 1-3 This invention provides an intelligent transformer substation line anomaly detection system based on artificial intelligence, comprising the following modules: The transformer area line loss detection module obtains the power supply and power consumption of the transformer area, and determines whether the line loss of the transformer area is normal by using the power supply and power consumption of the transformer area. If the line loss of the transformer area is abnormal, S2 is executed. The line loss location module installs multiple leakage current sensors on each branch line in the transformer area. The leakage current sensors are used to locate the branch line with line loss. A thermal infrared imaging scanner is used to scan the branch line with line loss to obtain a thermal image of the line. The multiple leakage current sensors on the branch line with line loss and the thermal image of the line are used to locate the low-temperature line loss line section and the high-temperature line loss line section. The line loss cause analysis module sets the observation period, obtains the line loss change status within the observation period, and determines whether the line loss is technical line loss or management line loss based on the line loss change status within the observation period, the line loss section with low temperature loss, and the line loss section with high temperature loss. The technical line loss cause analysis and marking module acquires the line thermal image of the line segment with line loss, determines the cause of technical line loss based on the line thermal image, marks the location of line loss, and sends the cause of technical line loss and the location of line loss to the management personnel; The line loss cause analysis and marking module identifies suspicious users by identifying high-temperature and low-temperature line loss sections, obtains historical verification records of the suspicious user's distribution meter, determines the cause of the line loss by marking the historical verification records of the suspicious user's distribution meter, marks the location of the suspicious user, and sends the cause of the line loss and the location of the line loss to the management personnel.

[0034] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented in software, the above embodiments can be implemented, in whole or in part, as a computer program product. Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution.

[0035] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0036] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A method for detecting line anomalies in intelligent transformer substations based on artificial intelligence, characterized in that, Includes the following steps: S1: Obtain the power supply and power consumption of the transformer area. Determine whether the line loss of the transformer area is normal by comparing the power supply and power consumption. If the line loss of the transformer area is abnormal, execute S2. S2: Multiple leakage current sensors are installed on each branch line in the transformer area. The leakage current sensors are used to locate the branch line with line loss. The thermal infrared imaging scanner is used to scan the branch line with line loss to obtain a thermal image of the line. The multiple leakage current sensors on the branch line with line loss and the thermal image of the line are used to locate the low-temperature line loss line section and the high-temperature line loss line section. S3: Set the observation period, obtain the line loss change status within the observation period, and determine whether the line loss is caused by technical line loss or management line loss by the line loss change status, low temperature line loss line section and high temperature line loss line section within the observation period. S4: Obtain the thermal image of the line segment with line loss, determine the cause of technical line loss based on the thermal image, mark the location of line loss, and send the cause of technical line loss and the location of line loss to the management personnel; S5: Identify suspicious users by identifying high-temperature and low-temperature line loss sections, obtain historical verification records of the suspicious user's distribution meter, determine the cause of the line loss by using the historical verification marks of the suspicious user's distribution meter, mark the location of the suspicious user, and send the cause of the line loss and the location of the line loss to the management personnel.

2. The method for detecting line anomalies in intelligent transformer substations based on artificial intelligence according to claim 1, characterized in that: In S1, the line loss of a transformer area is determined by the power supply and power consumption of that area. Specifically: S101: Obtain the power supply of the transformer area, obtain the power consumption of the transformer area, obtain the line loss power by subtracting the power supply of the transformer area from the power consumption of the transformer area, and obtain the line loss rate by dividing the line loss power by the power supply of the transformer area and multiplying it by 100%. S102: Set a preset threshold for safe line loss rate, determine whether the line loss rate is higher than the preset threshold for safe line loss rate. If the line loss rate is lower than or equal to the preset threshold for safe line loss rate, mark it as normal line loss and repeat step 101. If the line loss rate is higher than the preset threshold for safe line loss rate, mark it as abnormal line loss.

3. The method for detecting line anomalies in intelligent transformer substations based on artificial intelligence according to claim 1, characterized in that: In S2, the leakage current sensor is used to pinpoint the branch line with the line loss, specifically as follows: S201: Obtain the leakage current on each branch line and mark it as the starting leakage current, and set the safe leakage current preset threshold; S202: Determine whether the starting leakage current of each branch line is higher than the preset threshold for safe leakage current. If the starting leakage current of the branch line is higher than the preset threshold for safe leakage current, mark the branch line as a line loss branch line. If the starting leakage current of the branch line is lower than or equal to the preset threshold for safe leakage current, mark the branch line as a normal branch line.

4. The method for detecting line anomalies in intelligent transformer substations based on artificial intelligence according to claim 2, characterized in that: In S2, multiple leakage current sensors and thermal images of the line loss branch lines are used to identify low-temperature and high-temperature line loss sections. Specifically: S203: Obtain the leakage current values ​​of multiple leakage current sensors on the branch line respectively, and calculate the leakage current difference by subtracting the leakage current values ​​of two adjacent leakage current sensors in each group. S204: Obtain the starting leakage current on the branch line, set the fault-tolerant leakage current, obtain the minimum starting leakage current by subtracting the starting leakage current from the fault-tolerant leakage current, and obtain the leakage current judgment range by taking the minimum starting leakage current as the minimum value and the starting leakage current as the maximum value. S205: Determine whether multiple leakage current differences are within the leakage current judgment range. If the leakage current difference is within the leakage current judgment range, mark the line between the corresponding two leakage current sensors as a suspected line loss line segment. S206: Obtain the thermal image of the line section with suspected line loss, obtain the thermal image of the line section with non-suspected line loss on the line loss branch line, and obtain the temperature fluctuation value by subtracting the line temperature of the line section with suspected line loss from the line temperature of the line section with non-suspected line loss on the line loss branch line. S207: Set a preset threshold for temperature fluctuation value, determine whether the temperature fluctuation value is greater than the preset threshold for temperature fluctuation value. If the temperature fluctuation value is greater than the preset threshold for temperature fluctuation value, mark the suspected line loss line segment as a high temperature line loss line segment. If the temperature fluctuation value is less than or equal to the preset threshold for temperature fluctuation value, mark the suspected line loss line segment as a low temperature line loss line segment.

5. The method for detecting line anomalies in intelligent transformer substations based on artificial intelligence according to claim 1, characterized in that: In S3, the change status of line loss within the observation period is obtained, specifically as follows: S301: Set the observation period, measure the leakage current of the line loss branch once at 9:00, 15:00 and 23:00 every day, and sum the three leakage currents every day and take the average to obtain the daily average leakage current. S302: Arrange all daily average leakage currents in chronological order of date, subtract the daily average leakage currents of all two adjacent days to obtain the leakage current difference, obtain the lowest leakage current difference and the highest leakage current difference, take the absolute value of the lowest leakage current difference and the highest leakage current difference respectively, and sum them to obtain the leakage current fluctuation value. S303: Set a preset threshold for leakage current fluctuation value, determine whether the leakage current fluctuation value is greater than the preset threshold for leakage current fluctuation value. If the leakage current fluctuation value is less than or equal to the preset threshold for leakage current fluctuation value, mark the line loss change state as stable. If the leakage current fluctuation value is greater than the preset threshold for leakage current fluctuation value, mark the line loss change state as jumping.

6. The method for detecting line anomalies in intelligent transformer substations based on artificial intelligence according to claim 5, characterized in that: In S3, the cause of line loss is determined by observing the changes in line loss within the observation period, as well as the line loss in low-temperature and high-temperature sections. Specifically: S304: Determine whether the line loss section is a low-temperature line loss section or a high-temperature line loss section. If the line loss section is a low-temperature line loss section, mark the cause of the line loss as technical line loss. If the line loss section is a high-temperature line loss section, execute S305. S305: Obtain the line loss change status within the observation period. If the line loss change status within the observation period is a jump, mark the line loss cause as management line loss. If the line loss change status within the observation period is stable, mark the line loss cause as technical line loss.

7. The method for detecting line anomalies in intelligent transformer substations based on artificial intelligence according to claim 1, characterized in that: In S4, the cause of technical line loss is determined and the location of line loss is marked using line thermal images. Specifically: S401: Obtain a thermal image of the line loss section, mark ten target points with equal spacing on the cable in the thermal image, obtain the temperature of each of the ten target points, sum the temperatures of the ten target points and take the average value to obtain the mean temperature. S402: Obtain the highest temperature among the ten target points, calculate the temperature difference by subtracting the highest temperature from the average temperature, set a preset threshold for the temperature difference, and determine whether the temperature difference is greater than the preset threshold. If the temperature difference is greater than or equal to the preset threshold, mark the cause of the technical line loss as cable aging and leakage and mark the point as the location of the line loss. If the temperature difference is less than the preset threshold, execute S403. S403: Based on the cable width of ten target points, select the minimum value among the cable widths of the ten target points and mark it as the minimum cable width. Set a preset threshold for cable width and determine whether the minimum cable width is greater than the preset threshold. If the minimum cable width is less than or equal to the preset threshold, mark the cause of technical line loss as the paint being burned through due to the cable being too thin and mark that point as the location of line loss.

8. The method for detecting line anomalies in intelligent transformer substations based on artificial intelligence according to claim 1, characterized in that: In S5, suspicious users are identified by identifying high-temperature and low-temperature line loss sections. Historical verification records of the suspicious users' distribution meters are obtained. The causes of line loss are determined by analyzing these historical verification records, and the locations of the suspicious users are marked. Specifically: S501: Obtain user information registered at the location of high-temperature line loss lines and low-temperature line loss lines and mark them as suspicious users; obtain the historical verification records of distribution meters of suspicious users. S502: Set the meter verification cycle. Determine whether the meter of a suspected user has exceeded the verification cycle by checking the historical verification records of the meter. If the meter of a suspected user has exceeded the verification cycle, mark the management line loss as overdue meter verification and mark the location of the suspected user. If the meter of a suspected user has not exceeded the verification cycle, mark the management line loss as electricity theft and mark the location of the suspected user.

9. An AI-based intelligent transformer substation line anomaly detection system, applied to the AI-based intelligent transformer substation line anomaly detection method described in any one of claims 1-8, characterized in that, Includes the following modules: The transformer area line loss detection module obtains the power supply and power consumption of the transformer area, and determines whether the line loss of the transformer area is normal by using the power supply and power consumption of the transformer area. If the line loss of the transformer area is abnormal, S2 is executed. The line loss location module installs multiple leakage current sensors on each branch line in the transformer area. The leakage current sensors are used to locate the branch line with line loss. A thermal infrared imaging scanner is used to scan the branch line with line loss to obtain a thermal image of the line. The multiple leakage current sensors on the branch line with line loss and the thermal image of the line are used to locate the low-temperature line loss line section and the high-temperature line loss line section. The line loss cause analysis module sets the observation period, obtains the line loss change status within the observation period, and determines whether the line loss is technical line loss or management line loss based on the line loss change status within the observation period, the line loss section with low temperature loss, and the line loss section with high temperature loss. The technical line loss cause analysis and marking module acquires the line thermal image of the line segment with line loss, determines the cause of technical line loss based on the line thermal image, marks the location of line loss, and sends the cause of technical line loss and the location of line loss to the management personnel; The line loss cause analysis and marking module identifies suspicious users by identifying high-temperature and low-temperature line loss sections, obtains historical verification records of the suspicious user's distribution meter, determines the cause of the line loss by marking the historical verification records of the suspicious user's distribution meter, marks the location of the suspicious user, and sends the cause of the line loss and the location of the line loss to the management personnel.

Images