Distribution network time-sharing segmentation line loss monitoring method and system

Abstract

The invention discloses a time-sharing segmented line loss monitoring method and system for a distribution network, and the method comprises the steps: building and training a line loss anomaly prediction model based on historical line loss data, and obtaining the line loss anomaly prediction data and line / transformer area information in a specific time period through the line loss anomaly prediction model; the cloud platform obtains real-time line loss data of the distribution network in real time, when the real-time line loss data exceeds a theoretical line loss threshold value, the line loss abnormity prediction data is called for comparison, and if the line loss abnormity prediction data exceeds the real-time line loss data, the corresponding line / transformer area is positioned as an abnormal line loss point; and acquiring real-time monitoring data of the corresponding line / transformer area based on the line / transformer area information corresponding to the abnormal line loss point, judging whether the line loss occurring at the current moment is abnormal line loss, and if so, issuing alarm information to a processing end of the corresponding line / transformer area. The system is used for implementing the method. According to the invention, line loss abnormity judgment and line loss abnormal point positioning can be rapidly realized.

Description

Technical Field

[0001] The present invention relates to the technical field of distribution network line loss monitoring, and in particular to a distribution network time-sharing and section-sharing line loss monitoring method and system. Background Art

[0002] In recent years, line loss management technology has been vigorously developed, and the phenomenon of continuous high losses has been basically eliminated. However, intermittent high losses and uneconomical equipment operation problems still exist, and "hidden" and "intermittent" abnormal line loss problems still need to be addressed.

[0003] With the rapid development of society, the current power distribution network often incorporates a large number of photovoltaic equipment or power-consuming equipment based on the original technology. The grid-connected equipment will cause negative losses in the lines / stations, which brings new challenges to line loss management. For example, in large cities, there are many distribution networks and complex lines. When line loss problems occur, the investigation cycle is long and difficult, which brings many challenges to line loss monitoring. The traditional medium-voltage line loss management methods can no longer meet the work requirements of lean line loss management. Summary of the invention

[0004] The purpose of the present invention is to provide a distribution network time-sharing and section-by-section line loss monitoring method, which uses a time-sharing and section-by-section method to monitor the line loss of the distribution network, and can quickly realize the judgment of line loss anomalies and the location of line loss anomaly points, thereby ensuring daily monitoring of line losses and rapid and accurate troubleshooting.

[0005] Based on this, in a first aspect, the present invention provides a distribution network time-division and section-division line loss monitoring method, comprising the following steps: Based on historical line loss data, a line loss anomaly prediction model is established and trained. Based on the line loss occurrence time, the line loss anomaly prediction data in a specific time period and the line / station area information where the line loss anomaly prediction occurs are obtained through the line loss anomaly prediction model, and the obtained line loss anomaly prediction data and line / station area information are sent to the cloud platform; The cloud platform acquires real-time line loss data of the distribution network in real time. When the real-time line loss data exceeds the theoretical line loss threshold, the abnormal line loss prediction data is retrieved for comparison. If the abnormal line loss prediction data exceeds the real-time line loss data, the line / station area corresponding to the abnormal line loss prediction data is located as an abnormal line loss point. Based on the line / substation information corresponding to the abnormal line loss point, obtain the real-time monitoring data of the corresponding line / substation to determine whether the line loss occurring at the current moment is abnormal line loss. If so, send alarm information to the processing end of the corresponding line / substation based on the line / substation information.

[0006] In this method, based on historical line loss data, a line loss prediction model is established and trained, specifically: Based on time, obtain the line loss data of the distribution network N days or N years ago, and use the line loss data as the first training set according to time, where N≥1; the line loss data at least includes the line loss value, power, resistance, reactance, susceptance and capacitance data of the line in the distribution network corresponding to the time of line loss occurrence; Based on time, abnormal line loss data in the same time period of the distribution network is obtained, and the abnormal line loss data is used as the second training set according to the occurrence time; the abnormal line loss data at least includes the line loss value, power, resistance, reactance, susceptance and capacitance data of the line in the distribution network corresponding to the abnormal line loss occurrence time; A prediction model is established, the first training set and the second training set are associated according to time to form an associated training set, and the associated training set is input into the prediction model to train it until the output line loss abnormality prediction data is within a preset error range, thereby obtaining a line loss abnormality prediction model.

[0007] The method further includes: Based on time, the weather type data and daily temperature data of the day corresponding to the line loss data of the first training set are obtained as correction data, and the correction data is associated with the line loss data according to the time reference to form the first training set.

[0008] In the method, the line loss abnormality prediction data in a specific time period and the line / station area information where the line loss abnormality prediction occurs are obtained through the line loss abnormality prediction model, and the obtained line loss abnormality prediction data and line / station area information are sent to the cloud platform, which also includes: According to the abnormal data type or severity, set the corresponding risk level and corresponding level threshold; Based on the acquired line loss abnormality prediction data, it is compared with the level threshold and risk classification is performed based on the comparison result to obtain risk level data, and the risk level data is associated with the line loss abnormality prediction data as associated data.

[0009] In this method, the cloud platform obtains real-time line loss data of the distribution network in real time. When the real-time line loss data exceeds the theoretical line loss threshold, the abnormal line loss prediction data is retrieved for comparison. If the abnormal line loss prediction data exceeds the real-time line loss data, the line / station area corresponding to the abnormal line loss prediction data is located as an abnormal line loss point. The method also includes: When there are multiple abnormal line loss prediction data exceeding the real-time line loss data, the abnormal line loss points corresponding to the abnormal line loss prediction data are sorted and defined as the first to Xth abnormal line loss points in sequence, where X≥1.

[0010] In this method, when the abnormal line loss prediction data is also associated with risk level data, when sorting the abnormal line loss points, the cloud platform retrieves the risk level data and sorts the abnormal line loss points according to the risk level; Based on the line / substation information corresponding to the abnormal line loss point, the real-time monitoring data of the corresponding line / substation is obtained in descending order according to the risk level, and it is determined whether the line loss occurring in the corresponding line / substation at the current moment is abnormal line loss. If so, an alarm message is sent to the processing end of the corresponding line / substation based on the line / substation information.

[0011] In this method, based on the line / substation information corresponding to the abnormal line loss point, real-time monitoring data of the corresponding line / substation is obtained to determine whether the line loss occurring at the current moment is abnormal line loss, specifically: Based on the line / station area information corresponding to the abnormal line loss point, the corresponding line / station area is decomposed into several branch line areas according to line branches or geographical areas; Acquire power information data and line loss rate at two end points of the line in each branch line area, wherein the power information data at least includes current, voltage and harmonics; Acquire historical power information data and historical line loss rate in the branch line area between the two end points, wherein the historical power information data at least includes historical current, historical voltage and historical harmonics; Based on the historical power information data and the historical line loss rate, determine whether the power information data and the line loss rate corresponding to the current moment have both undergone sudden changes. If both have undergone sudden changes, it is determined that abnormal line loss exists in the branch line area corresponding to the current moment.

[0012] The method further includes: When it is determined that there is abnormal line loss in the branch line area corresponding to the current moment, the branch line name information or geographical location coordinates of the corresponding branch line area are obtained, and the branch line name information or geographical location coordinates are associated with the alarm information as index information; When the processing end receives the alarm information, the abnormal line loss point is located based on the branch line name information or the geographical location coordinates.

[0013] The method further includes: When obtaining the power information data and line loss rate at both end points of the line in each branch line area, the information of the newly connected photovoltaic equipment or power equipment in the branch line area is simultaneously obtained; Based on the information of photovoltaic equipment or electric power equipment, obtain theoretical power information data and theoretical line loss rate of the corresponding photovoltaic equipment or electric power equipment when using electricity; When there is a newly connected photovoltaic device or electric power device in the corresponding branch line area, obtain the historical power information data and historical line loss rate of the corresponding branch line area when the new device is not connected, as well as the theoretical power information data and theoretical line loss rate when the corresponding photovoltaic device or electric power device is using electricity, recalculate the actual power information data and actual line loss rate in the corresponding branch line area, and use the actual power information data and actual line loss rate as the latest historical power information data and historical line loss rate data; Based on the latest historical power information data and historical line loss rate data, determine whether the power information data and line loss rate corresponding to the current moment have both undergone sudden changes. If both have undergone sudden changes, it is determined that abnormal line loss exists in the branch line area corresponding to the current moment.

[0014] At the same time, the second aspect of the present invention provides a distribution network time-sharing and section-sharing line loss monitoring system, comprising: The line loss anomaly prediction module is used to establish and train a line loss anomaly prediction model based on historical line loss data. Based on the line loss occurrence time, the line loss anomaly prediction data in a specific time period and the line / station area information where the line loss anomaly prediction occurs are obtained through the line loss anomaly prediction model, and the obtained line loss anomaly prediction data and line / station area information are sent to the cloud platform; The cloud platform is used to obtain the real-time line loss data of the distribution network in real time. When the real-time line loss data exceeds the theoretical line loss threshold, the abnormal line loss prediction data is retrieved for comparison. If the abnormal line loss prediction data exceeds the real-time line loss data, the line / station area corresponding to the abnormal line loss prediction data is located as the abnormal line loss point; And, a judgment module is used to obtain real-time monitoring data of the corresponding line / substation based on the line / substation information corresponding to the abnormal line loss point, and determine whether the line loss occurring at the current moment is abnormal line loss. If so, an alarm message is sent to the processing end of the corresponding line / substation based on the line / substation information.

[0015] Compared with the prior art, the beneficial effects of the present invention are as follows: the present invention obtains the line loss abnormality prediction data for the corresponding specific time period and the line / substation information where the line loss abnormality prediction occurs based on the line loss abnormality prediction model, and uses this as auxiliary information to combine with the real-time line loss data for comparative judgment, thereby increasing the rationality and correctness of the abnormal line loss judgment, and the basic information of the line loss object or object area can also be directly determined through the line / substation information, so that when it is subsequently determined to be an abnormal line loss, the line loss point can be quickly found, so that the corresponding alarm information can be quickly sent to the processing end of the area, so that the subsequent line loss abnormality processing matters can be responded to quickly, and precise line loss monitoring, management, early warning and processing can be realized. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The drawings described herein are used to provide a further understanding of the embodiments of the present invention, constitute a part of this application, and do not constitute a limitation of the embodiments of the present invention. In the drawings: Figure 1 It is a flow chart of the time-sharing and section-sharing line loss monitoring method of the distribution network. DETAILED DESCRIPTION

[0017] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with embodiments and drawings. The exemplary embodiments of the present invention and their description are only used to explain the present invention and are not intended to limit the present invention.

[0018] like Figure 1 The first embodiment of the present invention provides a distribution network time-division and section-division line loss monitoring method, comprising the following steps: S1. Based on historical line loss data, a line loss anomaly prediction model is established and trained. Based on the line loss occurrence time, the line loss anomaly prediction data in a specific time period and the line / station area information where the line loss anomaly prediction occurs are obtained through the line loss anomaly prediction model, and the obtained line loss anomaly prediction data and line / station area information are sent to the cloud platform; In this step, the specific time period obtained is a certain period of time delayed from the current moment, so that the line loss abnormality at the current moment, that is, a certain time period in the future of the current moment, can be obtained through the line loss abnormality prediction model, thereby achieving a certain early warning effect.

[0019] For example, based on the time when line loss occurs, at 0:00 in the morning, the line loss anomaly prediction model can predict the line loss anomaly prediction data within 24 hours of the same day, and use this as the line loss anomaly data for the day. In actual applications, as time goes by, the line loss anomaly prediction model will continue to predict and modify the above data, thereby making the prediction data more accurate.

[0020] S2. The cloud platform obtains real-time line loss data of the distribution network in real time. When the real-time line loss data exceeds the theoretical line loss threshold, the abnormal line loss prediction data is retrieved for comparison. If the abnormal line loss prediction data exceeds the real-time line loss data, the line / station area corresponding to the abnormal line loss prediction data is located as the abnormal line loss point; In this step, the theoretical line loss threshold can be preset according to the actual situation. For example, when a large number of grid-connected equipment are newly introduced on the same day, when there are large-scale activities, or when there is concentrated power consumption, the value can be adjusted according to the actual power consumption on the same day to ensure that there will be no misjudgment. When the real-time line loss data exceeds the theoretical line loss threshold, it means that there may be problems with the line loss in the distribution network at this time, so the abnormal line loss prediction data is used for comparison. If the abnormal line loss prediction data exceeds the real-time line loss data, it means that there is abnormal line loss in both prediction and real-time. In this case, it can be concluded that there must be abnormal line loss in the distribution network at this time. Under this circumstance, the situation can be directly identified as an abnormal line loss situation, eliminating the auxiliary troubleshooting process in the process and saving a lot of computing time.

[0021] S3. Based on the line / substation information corresponding to the abnormal line loss point, obtain the real-time monitoring data of the corresponding line / substation, and determine whether the line loss occurring at the current moment is abnormal line loss. If so, send an alarm message to the processing end of the corresponding line / substation based on the line / substation information.

[0022] In this step, real-time monitoring data of the corresponding line / substation is obtained, which can be obtained through the acquisition equipment on the corresponding line, such as ADC data acquisition module, current and voltage sensor, line loss tester, load detector, etc. According to the required data, the corresponding equipment and instrument can be selected for acquisition. Since the corresponding power monitoring equipment and system applications are relatively mature in the prior art, this embodiment will no longer specifically describe the acquisition equipment or acquisition method related content.

[0023] In the specific implementation process, in step S1, a line loss prediction model is established and trained based on historical line loss data, specifically: Based on time, obtain the line loss data of the distribution network N days or N years ago, and use the line loss data as the first training set according to time, where N≥1; the line loss data at least includes the line loss value, power, resistance, reactance, susceptance and capacitance data of the line in the distribution network corresponding to the time of line loss occurrence; Based on time, abnormal line loss data in the same time period of the distribution network is obtained, and the abnormal line loss data is used as the second training set according to the occurrence time; the abnormal line loss data at least includes the line loss value, power, resistance, reactance, susceptance and capacitance data of the line in the distribution network corresponding to the abnormal line loss occurrence time; A prediction model is established, the first training set and the second training set are associated according to time to form an associated training set, and the associated training set is input into the prediction model to train it until the output line loss abnormality prediction data is within a preset error range, thereby obtaining a line loss abnormality prediction model.

[0024] In practical applications, external factors may cause changes in the physical state of the distribution network line, such as changes in line temperature. Therefore, during actual line loss monitoring, external factors may also affect changes in the line loss of the distribution network line. Therefore, the line loss prediction model also needs to comprehensively consider changes in external factors to improve its prediction accuracy. Therefore, step S1 may further include: Based on time, the weather type data and daily temperature data of the day corresponding to the line loss data of the first training set are obtained as correction data, and the correction data is associated with the line loss data according to the time reference to form the first training set.

[0025] As a specific implementation, the prediction model can adopt the LSSVM model or the Lasso model. The specific model algorithm and training process belong to the prior art, and this embodiment will not be further described. The corresponding content can be obtained by referring to the prior art such as the patent number 202311671428.4, which is named the public content of the line loss detection method and system based on big data. In the specific implementation process, in the step S1, the line loss abnormality prediction data in a specific time period and the line / station area information where the line loss abnormality prediction occurs are obtained through the line loss abnormality prediction model, and the acquired line loss abnormality prediction data and line / station area information are sent to the cloud platform, which also includes: According to the abnormal data type or severity, set the corresponding risk level and corresponding level threshold; Based on the acquired line loss abnormality prediction data, it is compared with the level threshold and risk classification is performed based on the comparison result to obtain risk level data, and the risk level data is associated with the line loss abnormality prediction data as associated data.

[0026] In application, the hazards caused by different abnormal line losses are of different sizes and degrees, and the hazards caused by abnormal line losses are fast and the degree of harm is also different. For this reason, in step S1 of this embodiment, the corresponding risk classification is obtained by comparing the abnormal line loss prediction data with the corresponding level threshold, so that when the abnormal line loss is subsequently processed, it can be processed according to the risk level, so as to deal with big problems early and avoid serious accidents.

[0027] In a specific implementation process, the cloud platform obtains real-time line loss data of the distribution network in real time. When the real-time line loss data exceeds the theoretical line loss threshold, the abnormal line loss prediction data is retrieved for comparison. If the abnormal line loss prediction data exceeds the real-time line loss data, the line / station area corresponding to the abnormal line loss prediction data is located as an abnormal line loss point. The following also includes: When there are multiple abnormal line loss prediction data exceeding the real-time line loss data, the abnormal line loss points corresponding to the abnormal line loss prediction data are sorted and defined as the first to Xth abnormal line loss points in sequence, where X≥1.

[0028] During application, multiple abnormal line loss points may appear in the distribution network at the same time, which may be in the same line / substation or in different lines / substations. Therefore, when such a situation occurs, each abnormal line loss point may be correct or valid, so that multiple abnormal line loss points can be sorted for subsequent investigation one by one.

[0029] It should be noted that when the line loss abnormality prediction data is also associated with risk level data, the abnormal line loss points can be sorted according to their respective risk levels in the risk level data, so that the line loss abnormal points with high risk levels can be placed in front for subsequent priority processing.

[0030] Furthermore, in a specific application: when the abnormal line loss prediction data is also associated with risk level data, when sorting the abnormal line loss points, the cloud platform retrieves the risk level data and sorts the abnormal line loss points according to the risk level; Based on the line / substation information corresponding to the abnormal line loss point, the real-time monitoring data of the corresponding line / substation is obtained in descending order according to the risk level, and it is determined whether the line loss occurring in the corresponding line / substation at the current moment is abnormal line loss. If so, an alarm message is sent to the processing end of the corresponding line / substation based on the line / substation information.

[0031] As a specific implementation method, based on the line / substation information corresponding to the abnormal line loss point, real-time monitoring data of the corresponding line / substation is obtained to determine whether the line loss occurring at the current moment is abnormal line loss, specifically: Based on the line / station area information corresponding to the abnormal line loss point, the corresponding line / station area is decomposed into several branch line areas according to line branches or geographical areas; Acquire power information data and line loss rate at two end points of the line in each branch line area, wherein the power information data at least includes current, voltage and harmonics; Acquire historical power information data and historical line loss rate in the branch line area between the two end points, wherein the historical power information data at least includes historical current, historical voltage and historical harmonics; Based on the historical power information data and the historical line loss rate, determine whether the power information data and the line loss rate corresponding to the current moment have both undergone sudden changes. If both have undergone sudden changes, it is determined that abnormal line loss exists in the branch line area corresponding to the current moment.

[0032] When applied, in order to realize rapid judgment, identification and positioning of abnormal line loss, this embodiment splits the corresponding line / substation area to obtain several branch line areas, and judges the power information and line loss rate and historical data of each branch line area, so as to quickly determine the abnormal line loss point, and can also quickly locate and identify the specific branch line area of ​​the abnormal line loss point, so as to facilitate the subsequent rapid processing of abnormal line loss problems and save a lot of time in finding and locating abnormal points.

[0033] Furthermore, in order to better locate and locate the abnormal line loss point, this step also includes: When it is determined that there is abnormal line loss in the branch line area corresponding to the current moment, the branch line name information or geographical location coordinates of the corresponding branch line area are obtained, and the branch line name information or geographical location coordinates are associated with the alarm information as index information; When the processing end receives the alarm information, the abnormal line loss point is located based on the branch line name information or the geographical location coordinates.

[0034] In this step, the processing end may be a network end or a device end of a power maintenance organization in the corresponding branch line area.

[0035] Furthermore, if there are newly connected devices, such as photovoltaic devices, power-consuming devices, etc., temporarily in the corresponding branch line area, then in the above steps, the historical power information data and historical line loss rate data are used to determine whether the power information data and line loss rate corresponding to the current moment have all changed suddenly, which will inevitably lead to misjudgment. Therefore, in order to avoid the above problems, this step also includes: When obtaining the power information data and line loss rate at both end points of the line in each branch line area, the information of the newly connected photovoltaic equipment or power equipment in the branch line area is simultaneously obtained; Based on the information of photovoltaic equipment or electric power equipment, obtain theoretical power information data and theoretical line loss rate of the corresponding photovoltaic equipment or electric power equipment when using electricity; When there is a newly connected photovoltaic device or electric power device in the corresponding branch line area, obtain the historical power information data and historical line loss rate of the corresponding branch line area when the new device is not connected, as well as the theoretical power information data and theoretical line loss rate when the corresponding photovoltaic device or electric power device is using electricity, recalculate the actual power information data and actual line loss rate in the corresponding branch line area, and use the actual power information data and actual line loss rate as the latest historical power information data and historical line loss rate data; Based on the latest historical power information data and historical line loss rate data, determine whether the power information data and line loss rate corresponding to the current moment have both undergone sudden changes. If both have undergone sudden changes, it is determined that abnormal line loss exists in the branch line area corresponding to the current moment.

[0036] In summary, this embodiment compares and judges the line loss abnormality prediction data in a certain time period and the line / substation information where the line loss abnormality prediction occurs in combination with the real-time line loss data in a time period manner, and at the same time, segments the detection area according to the line or substation, so as to directly determine the basic information of the line loss object or object area. When it is subsequently determined to be an abnormal line loss, the line loss point can be quickly found, thereby realizing line loss monitoring of all lines or substations based on the idea of ​​dividing the detection time into time periods and lines or substations, and realizing precise line loss monitoring, management, early warning and processing.

[0037] The second embodiment of the present invention is also based on the above distribution network time-division and section line loss monitoring method, and provides a distribution network time-division and section line loss monitoring system, including: The line loss anomaly prediction module is used to establish and train a line loss anomaly prediction model based on historical line loss data. Based on the line loss occurrence time, the line loss anomaly prediction data in a specific time period and the line / station area information where the line loss anomaly prediction occurs are obtained through the line loss anomaly prediction model, and the obtained line loss anomaly prediction data and line / station area information are sent to the cloud platform; The cloud platform is used to obtain the real-time line loss data of the distribution network in real time. When the real-time line loss data exceeds the theoretical line loss threshold, the abnormal line loss prediction data is retrieved for comparison. If the abnormal line loss prediction data exceeds the real-time line loss data, the line / station area corresponding to the abnormal line loss prediction data is located as the abnormal line loss point; And, a judgment module is used to obtain real-time monitoring data of the corresponding line / substation based on the line / substation information corresponding to the abnormal line loss point, and determine whether the line loss occurring at the current moment is abnormal line loss. If so, an alarm message is sent to the processing end of the corresponding line / substation based on the line / substation information.

[0038] In summary, the distribution network time-sharing and section-by-section line loss monitoring method and system of the present embodiment can obtain the line loss abnormality prediction data for the corresponding specific time period and the line / substation information where the line loss abnormality prediction occurs based on the line loss abnormality prediction model, and use this as auxiliary information to combine with the real-time line loss data for comparative judgment, thereby increasing the rationality and correctness of the abnormal line loss judgment, and the basic information of the line loss object or object area can also be directly determined through the line / substation information, so that when it is subsequently determined to be an abnormal line loss, the line loss point can be quickly found, the corresponding alarm information can be quickly sent to the processing end of the area, and the subsequent line loss abnormality processing matters can be quickly responded to, so as to achieve Accurate line loss monitoring, management, early warning and processing. At the same time, when determining whether the line loss occurring at the current moment is abnormal, the corresponding line / substation area is decomposed, and the abnormal line loss points are accurately found and located in different areas using several branch line areas. Combined with the changes in grid line loss caused by the newly connected power equipment, the abnormal line loss points are comprehensively judged, thereby improving the accuracy of abnormal line loss judgment, and accurately finding the location of abnormal line loss points, which provides convenience and a lot of spare time for subsequent processing of abnormal line loss problems, and realizes accurate line loss abnormality detection, abnormal line loss judgment, and abnormal line loss location, etc., and has a good application prospect.

[0039] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A distribution network time-division and section-division line loss monitoring method, characterized in that: The following steps are involved: Based on historical line loss data, a line loss anomaly prediction model is established and trained. Based on the line loss occurrence time, the line loss anomaly prediction data in a specific time period and the line / station area information where the line loss anomaly prediction occurs are obtained through the line loss anomaly prediction model, and the obtained line loss anomaly prediction data and line / station area information are sent to the cloud platform; The cloud platform acquires real-time line loss data of the distribution network in real time. When the real-time line loss data exceeds the theoretical line loss threshold, the abnormal line loss prediction data is retrieved for comparison. If the abnormal line loss prediction data exceeds the real-time line loss data, the line / station area corresponding to the abnormal line loss prediction data is located as an abnormal line loss point. Based on the line / substation information corresponding to the abnormal line loss point, obtain the real-time monitoring data of the corresponding line / substation to determine whether the line loss occurring at the current moment is abnormal line loss. If so, send alarm information to the processing end of the corresponding line / substation based on the line / substation information.

2. The method according to claim 1, characterized in that Based on the historical line loss data, a line loss prediction model is established and trained, specifically: Based on time, obtain the line loss data of the distribution network N days or N years ago, and use the line loss data as the first training set according to time, where N≥1; the line loss data at least includes the line loss value, power, resistance, reactance, susceptance and capacitance data of the line in the distribution network corresponding to the time of line loss occurrence; Based on time, abnormal line loss data in the same time period of the distribution network is obtained, and the abnormal line loss data is used as the second training set according to the occurrence time; the abnormal line loss data at least includes the line loss value, power, resistance, reactance, susceptance and capacitance data of the line in the distribution network corresponding to the abnormal line loss occurrence time; A prediction model is established, the first training set and the second training set are associated according to time to form an associated training set, and the associated training set is input into the prediction model to train it until the output line loss abnormality prediction data is within a preset error range, thereby obtaining a line loss abnormality prediction model.

3. The method according to claim 2, characterized in that Also includes: Based on time, the weather type data and daily temperature data of the day corresponding to the line loss data of the first training set are obtained as correction data, and the correction data is associated with the line loss data according to the time reference to form the first training set.

4. The method according to claim 1, characterized in that The method further includes obtaining line loss anomaly prediction data and line / station area information of a specific time period through a line loss anomaly prediction model, and sending the obtained line loss anomaly prediction data and line / station area information to a cloud platform. According to the abnormal data type or severity, set the corresponding risk level and corresponding level threshold; Based on the acquired line loss abnormality prediction data, it is compared with the level threshold and risk classification is performed based on the comparison result to obtain risk level data, and the risk level data is associated with the line loss abnormality prediction data as associated data.

5. The method according to claim 4, characterized in that The cloud platform acquires real-time line loss data of the distribution network in real time. When the real-time line loss data exceeds the theoretical line loss threshold, the abnormal line loss prediction data is retrieved for comparison. If the abnormal line loss prediction data exceeds the real-time line loss data, the line / station area corresponding to the abnormal line loss prediction data is located as an abnormal line loss point. The cloud platform also includes: When there are multiple abnormal line loss prediction data exceeding the real-time line loss data, the abnormal line loss points corresponding to the abnormal line loss prediction data are sorted and defined as the first to Xth abnormal line loss points in sequence, where X≥1.

6. The method according to claim 5, characterized in that When the abnormal line loss prediction data is also associated with risk level data, when sorting the abnormal line loss points, the cloud platform retrieves the risk level data and sorts the abnormal line loss points according to the risk level; Based on the line / substation information corresponding to the abnormal line loss point, the real-time monitoring data of the corresponding line / substation is obtained in descending order according to the risk level, and it is determined whether the line loss occurring in the corresponding line / substation at the current moment is abnormal line loss. If so, an alarm message is sent to the processing end of the corresponding line / substation based on the line / substation information.

7. The method according to claim 6, characterized in that The method of obtaining the real-time monitoring data of the corresponding line / substation based on the line / substation information corresponding to the abnormal line loss point and determining whether the line loss occurring at the current moment is abnormal line loss is specifically as follows: Based on the line / station area information corresponding to the abnormal line loss point, the corresponding line / station area is decomposed into several branch line areas according to line branches or geographical areas; Acquire power information data and line loss rate at two end points of the line in each branch line area, wherein the power information data at least includes current, voltage and harmonics; Acquire historical power information data and historical line loss rate in the branch line area between the two end points, wherein the historical power information data at least includes historical current, historical voltage and historical harmonics; Based on the historical power information data and the historical line loss rate, determine whether the power information data and the line loss rate corresponding to the current moment have both undergone sudden changes. If both have undergone sudden changes, it is determined that abnormal line loss exists in the branch line area corresponding to the current moment.

8. The method according to claim 7, characterized in that Also includes: When it is determined that there is abnormal line loss in the branch line area corresponding to the current moment, the branch line name information or geographical location coordinates of the corresponding branch line area are obtained, and the branch line name information or geographical location coordinates are associated with the alarm information as index information; When the processing end receives the alarm information, the abnormal line loss point is located based on the branch line name information or the geographical location coordinates.

9. The method according to claim 7, characterized in that: Also includes: When obtaining the power information data and line loss rate at both end points of the line in each branch line area, the information of the newly connected photovoltaic equipment or power equipment in the branch line area is simultaneously obtained; Based on the information of photovoltaic equipment or electric power equipment, obtain theoretical power information data and theoretical line loss rate of the corresponding photovoltaic equipment or electric power equipment when using electricity; When there is a newly connected photovoltaic device or electric power device in the corresponding branch line area, obtain the historical power information data and historical line loss rate of the corresponding branch line area when the new device is not connected, as well as the theoretical power information data and theoretical line loss rate when the corresponding photovoltaic device or electric power device is using electricity, recalculate the actual power information data and actual line loss rate in the corresponding branch line area, and use the actual power information data and actual line loss rate as the latest historical power information data and historical line loss rate data; Based on the latest historical power information data and historical line loss rate data, determine whether the power information data and line loss rate corresponding to the current moment have both undergone sudden changes. If both have undergone sudden changes, it is determined that abnormal line loss exists in the branch line area corresponding to the current moment.

10. A distribution network time-sharing and section-sharing line loss monitoring system, characterized in that: include: The line loss anomaly prediction module is used to establish and train a line loss anomaly prediction model based on historical line loss data. Based on the line loss occurrence time, the line loss anomaly prediction data in a specific time period and the line / station area information where the line loss anomaly prediction occurs are obtained through the line loss anomaly prediction model, and the obtained line loss anomaly prediction data and line / station area information are sent to the cloud platform; The cloud platform is used to obtain the real-time line loss data of the distribution network in real time. When the real-time line loss data exceeds the theoretical line loss threshold, the abnormal line loss prediction data is retrieved for comparison. If the abnormal line loss prediction data exceeds the real-time line loss data, the line / station area corresponding to the abnormal line loss prediction data is located as the abnormal line loss point; And, a judgment module is used to obtain real-time monitoring data of the corresponding line / substation based on the line / substation information corresponding to the abnormal line loss point, and determine whether the line loss occurring at the current moment is abnormal line loss. If so, an alarm message is sent to the processing end of the corresponding line / substation based on the line / substation information.

Images