Zone area line loss intelligent troubleshooting device and method adaptive to new energy tidal load

Abstract

The invention provides a transformer area line loss intelligent troubleshooting device and method adaptive to new energy tidal loads, and belongs to the technical field of power system line loss troubleshooting. The line loss intelligent troubleshooting device comprises a perception acquisition unit, an edge calculation unit, a visual display unit and a communication transmission assembly, the perception acquisition unit, the edge calculation unit and the visual display unit realize data interaction through the communication transmission assembly, and the whole device is a modularized plug-and-play architecture. The sensing acquisition units are respectively arranged at a total outlet of a low-voltage side of a transformer in a zone area, a head end of a branch line, a new energy grid-connected point, a tidal load key node and a user meter box inlet wire; and a data processing module, a topology identification module, a sub-scene line loss calculation module and an abnormity diagnosis module are arranged in the edge calculation unit. According to the invention, the topological structure of the transformer area can be accurately identified, the operation state of each line and equipment can be monitored in real time, and complex changes caused by new energy tidal loads can be effectively handled.

Description

Intelligent line loss detection device and method for transformer substations adapted to tidal loads of new energy sources Technical Field

[0001] This invention belongs to the field of power system line loss investigation technology, specifically relating to an intelligent line loss investigation device and method for transformer substations adapted to new energy tidal loads. Background Technology

[0002] Low-voltage line loss management is a core aspect for power supply companies to improve economic efficiency and promote the sustainable development of the power grid. With the large-scale integration of distributed new energy (photovoltaics and wind power) into low-voltage distribution transformer areas, the power grid in these areas has transformed from the traditional "one-way radial power supply" to a complex form of "two-way interaction between source and load." Coupled with the tidal load fluctuations of residential and commercial electricity consumption, the dynamic volatility of line loss rates in these areas has intensified, significantly increasing the difficulty of accurate diagnosis and management.

[0003] Existing transformer substation line loss detection technologies are insufficient to penetrate to micro-units such as branch lines and user meters for real-time dynamic sensing, which can easily lead to insufficient monitoring. At the same time, when considering the time sequence characteristics of tidal loads, the calculation of line loss rates under special operating conditions such as reverse power is prone to deviation, making it difficult to adapt to the dynamic changes of bidirectional power flow and tidal loads, and difficult to distinguish between losses related to renewable energy grid connection and losses from faults in traditional power grid equipment.

[0004] Therefore, in order to ensure that line losses can be quickly and accurately identified and located when anomalies occur, it is particularly important to improve the level of intelligent management of line losses in low-voltage distribution networks under the background of new power systems. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides an intelligent line loss detection device and method for transformer substations adapted to tidal loads from renewable energy sources.

[0006] To solve one or more of the above-mentioned technical problems, the technical solution adopted by this invention is: an intelligent line loss investigation device for transformer substations adapted to new energy tidal loads, comprising a sensing and acquisition unit, an edge computing unit, a visualization display unit, and a communication transmission component. The sensing and acquisition unit, edge computing unit, and visualization display unit all achieve data interaction through the communication transmission component, and the overall structure is a modular, plug-and-play architecture. The sensing and acquisition unit is respectively installed at the main outlet of the low-voltage side of the transformer substation, the beginning of branch lines, the new energy grid connection point, key nodes of tidal loads, and the inlet of the user meter box, for collecting data at each node. The system provides clock-level electrical quantity data including three-phase voltage, current, active and reactive power, forward and reverse active energy, and four-quadrant reactive energy. The edge computing unit includes a data processing module, a topology identification module, a scene-specific line loss calculation module, and an anomaly diagnosis module, which are used for multi-source data processing, topology analysis, line loss calculation, and anomaly detection, respectively. The visualization unit graphically displays the dynamic topology of the transformer area, real-time historical line loss curves, and the location and type of abnormal loss points. The communication transmission component enables data security and real-time interaction between the sensing and acquisition unit, the edge computing unit, and the visualization unit.

[0007] Furthermore, the sensing and acquisition unit includes a power acquisition module, a time synchronization module, and a data preprocessing component. The power acquisition module is used to achieve accurate acquisition of multiple electrical quantities, the time synchronization module is used to ensure that the time axis of the data at each acquisition point is consistent, and the data preprocessing component is used to perform preliminary filtering and format standardization on the raw data.

[0008] Furthermore, the data processing module of the edge computing unit has a built-in time synchronization submodule, a data interpolation submodule, and a filtering submodule. The time synchronization submodule is used to achieve accurate alignment of multi-source data from the sensing and acquisition unit. The data interpolation submodule is used to intelligently complete missing data caused by intermittent power output from new energy sources. The filtering submodule is used to eliminate electrical quantity data interference caused by tidal load fluctuations, and at the same time, it is used to ensure the consistency of forward and reverse power statistical logic under bidirectional power flow.

[0009] Furthermore, the topology identification module of the edge computing unit includes a characteristic current injection submodule and a new energy branch concealment and avoidance submodule. The characteristic current injection submodule is used to realize automatic topology identification, and the new energy branch concealment and avoidance submodule is used to dynamically adjust the identification strategy and correct the topology results during peak power generation of distributed power sources.

[0010] Furthermore, the edge computing unit's scenario-based line loss calculation module incorporates a midday photovoltaic peak sub-model module and a nighttime electricity consumption peak sub-model module. The midday photovoltaic peak sub-model module is used to calculate the losses of new energy reverse transmission lines and additional losses due to limited absorption. The nighttime electricity consumption peak sub-model module is used to exclude the impact of new energy output, focus on traditional load peaks, and calculate the conventional losses of equipment and lines.

[0011] Furthermore, the anomaly diagnosis module of the edge computing unit includes an indicator system construction submodule, a dynamic weight adjustment submodule, and a loss type stripping submodule. The indicator system construction submodule builds a multi-dimensional diagnostic indicator system including equipment health, line parameters, real-time renewable energy consumption status, and typical tidal load patterns. The dynamic weight adjustment submodule dynamically adjusts the weights of each indicator according to typical electricity consumption periods. The loss type stripping submodule is used to distinguish between qualitative renewable energy-related characteristic losses and traditional abnormal losses of the power grid.

[0012] A method for intelligent line loss investigation in transformer substations adapted to renewable energy tidal loads includes: S1, Deployment and debugging of sensing and acquisition units: Deploying sensing and acquisition units at key nodes in the transformer substation area, completing link debugging between communication transmission components and each unit, setting minute-level data freeze and acquisition frequency for the sensing and acquisition units, and starting the time synchronization module to achieve time calibration at all points; S2, Multi-source data acquisition and modular processing: Acquiring electrical quantity data from each node through the power acquisition module, integrating transformer substation meter data, renewable energy monitoring data, and tidal load time series data to form a multi-source heterogeneous dataset, completing data processing through the data processing module in the edge computing unit, and uploading standardized data to the edge computing unit through the communication transmission component; S3, Dynamic topology analysis by the topology identification module: The edge computing unit starts the topology identification module, injects non-invasive characteristic current, collects feedback signals, and combines them with renewable energy data. The topology structure is identified; if a peak in new energy power generation occurs, the topology result is corrected through the new energy branch concealment and avoidance sub-module; S4, the scenario-specific line loss calculation module performs precise calculations: based on the operating status, the edge computing unit automatically matches the midday photovoltaic peak or nighttime electricity consumption peak sub-model module with the scenario-specific line loss calculation module, retrieves the topology result to calculate the line loss value and the actual line loss value, and analyzes the deviation; S5, the anomaly diagnosis module performs multi-dimensional source tracing: the edge computing unit starts the anomaly diagnosis module, retrieves diagnostic indicators, adapts indicator weights, correlates and analyzes line loss deviations, identifies loss types, locates anomalies and determines their types; S6, the visualization display unit outputs results and provides operation and maintenance guidance: the edge computing unit sends the anomaly diagnosis results, line loss calculation data, and topology structure data to the visualization display unit through the communication transmission component for data display, clearly identifies anomaly information, and generates targeted on-site operation and maintenance handling guidance.

[0013] Furthermore, in step S2, the tidal load time series data includes load change data of morning peak, midday commercial load peak, evening peak, and nighttime trough. After being processed by the data processing module, it forms a standardized power grid area and time series dataset. The dataset achieves bidirectional synchronization between the edge computing unit and the sensing and acquisition unit through the communication transmission component, supporting real-time data completion by the sensing and acquisition unit.

[0014] Furthermore, in the modular closed-loop process of steps S1-S6, the responses and calculations of each unit and module are all achieved through data interaction by the communication transmission component.

[0015] Compared with the prior art, the beneficial effects of the present invention are as follows: Combined with the corresponding annotations and document content in the background art, the beneficial effects are as follows: 1. By deploying sensing and acquisition units at various key nodes in the transformer area, the present invention can achieve accurate acquisition of multiple electrical quantities, and integrate transformer area meter data, new energy monitoring data and tidal load time series data to form a multi-source heterogeneous dataset, which can enable line loss investigation to penetrate to micro-units such as branch lines and user meters, thereby improving monitoring strength and real-time dynamic sensing capabilities.

[0016] 2. The edge computing unit's built-in data processing module, topology identification module, scenario-specific line loss calculation module, and anomaly diagnosis module have been optimized for multi-source data processing, topology analysis, line loss calculation, and anomaly detection, respectively. In particular, the time synchronization submodule, data interpolation submodule, and filtering submodule in the data processing module can effectively solve the data loss problem caused by intermittent output of new energy sources, greatly reduce the interference of tidal load fluctuations on electrical quantity data, and ensure the consistency of forward and reverse power statistical logic under bidirectional power flow.

[0017] 3. The topology identification module automatically identifies the topology structure by injecting non-invasive characteristic current and combining it with new energy data. It also dynamically adjusts the identification strategy and corrects the topology results during peak periods of new energy power generation, thereby improving the accuracy and adaptability of topology identification.

[0018] 4. The scenario-specific line loss calculation module automatically matches the midday photovoltaic peak or nighttime electricity consumption peak sub-model module according to the operating status, achieving accurate line loss calculation and effectively distinguishing between losses related to renewable energy grid connection and losses from traditional power grid equipment faults. The anomaly diagnosis module constructs a multi-dimensional diagnostic indicator system, dynamically adjusts indicator weights, correlates and analyzes line loss deviations, identifies loss types, locates anomalies and determines their types, achieving accurate identification and location, and facilitating targeted on-site handling guidance for operation and maintenance personnel. Attached Figure Description

[0019] The present invention will now be described in further detail with reference to the accompanying drawings.

[0020] Figure 1: Schematic diagram of the structure of the tool half of Embodiment 1 of the present invention; wherein: 1, sensing and acquisition unit; 11, power acquisition module; 12, time synchronization module; 13, data preprocessing component; 2, edge computing unit; 21, data processing module; 22, topology recognition module; 23, scene-specific line loss calculation module; 24, anomaly diagnosis module; 3, visualization display unit; 4, communication transmission component. Detailed Implementation

[0021] To better understand the present invention, the content of the invention is further clearly illustrated below with reference to embodiments and accompanying drawings. However, the scope of protection of the present invention is not limited to the embodiments described below. Numerous specific details are set forth in the following description to provide a more thorough understanding of the invention. However, it will be apparent to those skilled in the art that the present invention can be practiced without one or more of these details.

[0022] Referring to Figure 1, the purpose of this embodiment is to provide an intelligent line loss investigation device for transformer areas that is adapted to the tidal load of new energy. It includes a sensing and acquisition unit 1, an edge computing unit 2, a visualization display unit 3, and a communication transmission component 4. The sensing and acquisition unit 1, the edge computing unit 2, and the visualization display unit 3 all realize data interaction through the communication transmission component 4. The whole is a modular plug-and-play architecture with built-in functional modules.

[0023] The sensing and acquisition unit 1 is the front-end data acquisition unit of the device. It adopts a miniaturized and modular networking design and includes an energy acquisition module 11, a time synchronization module 12, and a data preprocessing component 13. It supports flexible deployment at multiple points at the main outlet of the low-voltage side of the transformer in the distribution area, the beginning of branch lines, the grid connection point of new energy sources, key nodes of tidal loads, and the inlet of user meter boxes, realizing full-path coverage monitoring of the power flow in the distribution area from "main trunk to branch to user to power source". Among them, the energy acquisition module 11 is responsible for accurately collecting minute-level three-phase voltage, current, active and reactive power, forward and reverse active power, and four-quadrant reactive power data at each node; the time synchronization module 12 adopts high-precision clock synchronization technology to ensure the consistency of the time axis of the data at all acquisition points; the data preprocessing component 13 completes the preliminary filtering, format conversion, and standardization of the raw data, laying the foundation for subsequent data processing.

[0024] Edge computing unit 2 is the core computing unit of the device, deployed at the local operation and maintenance terminal of the transformer area. Edge computing unit 2 has four major functional modules: data processing module 21, topology identification module 22, scene-specific line loss calculation module 23, and anomaly diagnosis module 24. Each module has sub-modules, which can complete multi-source data deep processing, dynamic topology identification of the transformer area, accurate calculation of scene-specific line loss, and multi-dimensional diagnosis and tracing of line loss anomalies. It is the core computing hub of the device.

[0025] The visualization unit 3 is the result output and human-machine interaction unit of the device. It adopts an industrial-grade touch display architecture, receives the calculation results of the edge computing unit 2, and displays information such as the dynamic topology map of the transformer area, real-time / historical line loss curves, abnormal loss point locations, and abnormality types through a graphical and visual interface. It provides operation and maintenance personnel with intuitive and clear troubleshooting guidance and supports manual interaction and operation command issuance by operation and maintenance personnel.

[0026] The communication transmission component 4 is the core component for cross-unit data interaction of the device. It adopts the 4G secure communication standard and has a built-in encrypted transmission module to realize secure, real-time and stable interaction of massive high-frequency data between the sensing and acquisition unit 1, the edge computing unit 2 and the visualization display unit 3, ensuring the collaborative operation of each unit and module, and supporting bidirectional data synchronization and remote transmission of instructions.

[0027] Referring to Figure 1, the data processing module 21 internally includes a high-precision time synchronization submodule, a new energy power output data interpolation submodule, and a tidal load filtering submodule. Its core function is to improve the quality of multi-source heterogeneous data and solve the data processing challenges under bidirectional power flow. Specifically: the high-precision time synchronization submodule achieves accurate time alignment of data from multiple points in the sensing and acquisition unit 1, with a synchronization error ≤1μs; the data interpolation submodule intelligently completes missing data caused by intermittent power output based on the time sequence characteristics of new energy power output, with a completion accuracy ≥98%; the tidal load filtering submodule adopts an adaptive filtering algorithm to eliminate electrical quantity data interference caused by tidal load fluctuations, while correcting the logical deviation of forward and reverse power statistics under bidirectional power flow to ensure data consistency.

[0028] The topology identification module 22 internally includes a non-intrusive characteristic current injection submodule and a new energy branch concealment avoidance submodule. Its core function is to achieve automatic and accurate identification of the dynamic topology of the transformer area. Specifically, the characteristic current injection submodule injects a low-amplitude, high-recognition characteristic current signal into the transformer area lines, collects and analyzes the line feedback signal, and realizes automatic identification of the transformer area topology. The new energy branch concealment avoidance submodule monitors the power output data of new energy sources in real time. When it detects that the new energy source is in its peak power generation period, it dynamically adjusts the topology identification strategy, corrects the topology results, and completely solves the problem of inaccurate line loss allocation and tracing caused by "new energy branch concealment".

[0029] The scenario-specific line loss calculation module 23 includes a midday photovoltaic peak sub-model module and a nighttime electricity peak sub-model module. Designed for the typical characteristics of high renewable energy penetration (over 50% in some rural areas) and tidal loads in Nanyang, its core function is to achieve scenario-specific and accurate line loss calculation. Specifically, the midday photovoltaic peak sub-model module accurately calculates line losses during renewable energy backfeeding and additional losses due to limited absorption, targeting high-penetration photovoltaic areas with high reverse power and high output scenarios. The nighttime electricity peak sub-model module excludes the impact of renewable energy output, focusing on traditional residential / commercial peak loads and calculating conventional losses caused by equipment overload, aging, faults, or abnormal power consumption.

[0030] The anomaly diagnosis module 24 internally includes an indicator system construction submodule, a dynamic weight adjustment submodule, and a loss type identification submodule. Its core function is to achieve multi-dimensional diagnosis of line loss anomalies, loss type identification, and accurate source tracing. Specifically: the indicator system construction submodule builds a standardized diagnostic indicator system encompassing four dimensions: equipment health, line parameters, real-time renewable energy absorption status, and typical tidal load patterns; the dynamic weight adjustment submodule dynamically adjusts the weights of each indicator according to typical electricity consumption periods, ensuring the all-weather adaptability of the diagnostic system. Typical electricity consumption periods include peak photovoltaic output, midday commercial load peak, and nighttime residential electricity consumption peak; the loss type identification submodule uses algorithms such as correlation analysis and time series comparison to accurately distinguish and characterize renewable energy-related characteristic losses (photovoltaic fluctuation statistical deviation, reverse power metering deviation, and additional losses due to absorption limitations) from traditional grid anomaly losses (equipment failure, suspected electricity theft, and line overload).

[0031] This invention also provides an intelligent method for investigating line losses in transformer substations that adapts to tidal loads from new energy sources. Specifically, it includes the following steps: S1. Deployment and modular debugging of the sensing and acquisition unit 1: The sensing and acquisition unit 1 is modularly networked according to the power flow path of the transformer substation, and deployed at key nodes along the entire path, such as the main outlet on the low-voltage side of the transformer, the beginning of branch lines, and the new energy grid connection point; the communication transmission component 4 is debugged with the sensing and acquisition unit 1, the edge computing unit 2, and the visualization display unit 3 to ensure smooth data interaction; the 5-minute data freeze and acquisition frequency of the sensing and acquisition unit 1 are set, and the time synchronization module 12 of the sensing and acquisition unit 1 is activated to achieve accurate time calibration at all acquisition points. The entire deployment process is plug-and-play, requires no power outage, and does not affect the normal power supply to the transformer substation.

[0032] S2. Multi-source data acquisition and modular processing: The raw electrical quantity data of each node is acquired through the power acquisition module 11 of the sensing acquisition unit 1. At the same time, the original meter data of the transformer area, the distributed new energy monitoring data, and the tidal load time series data are integrated to form a multi-source heterogeneous dataset. The dataset is transmitted to the data processing module 21 of the edge computing unit 2. The high-precision time synchronization submodule, the new energy output data interpolation submodule, and the tidal load filtering submodule are used to complete data cleaning, time alignment, missing data interpolation, and fluctuation interference filtering, so as to solve the logical consistency problem of forward and reverse power statistics under bidirectional power flow. The dataset standardized by the data processing module 21 is synchronized bidirectionally between the sensing acquisition unit 1 and the edge computing unit 2 by the communication transmission component 4.

[0033] S3. Topology Identification Module 22 Dynamic Topology Analysis: Edge computing unit 2 starts topology identification module 22, injects characteristic current signals into the transformer area lines through non-intrusive characteristic current injection submodule, collects line feedback signals and combines them with the real-time power output data of new energy sources uploaded by sensing and acquisition unit 1 to automatically identify the real-time topology structure of the transformer area; if the power output of new energy sources reaches more than 70% of the rated power (peak power generation period), the new energy branch concealment and avoidance submodule is immediately started to correct the initial topology identification results and output accurate dynamic topology structure of the transformer area, providing basic data for line loss calculation.

[0034] S4. Precise Calculation by Scenario-Specific Line Loss Module 23: The edge computing unit 2 analyzes the current operating status of the transformer area (new energy output ratio, time period, load type) based on the real-time data uploaded by the sensing and acquisition unit 1. The scenario-specific line loss calculation module 23 automatically matches the corresponding sub-model module: when the proportion of new energy output in the transformer area to the total load is ≥30% and it is during the midday period, the midday photovoltaic peak sub-model module is triggered; when the proportion of new energy output in the transformer area to the total load is <10% and it is during the nighttime period, the nighttime electricity consumption peak sub-model module is triggered. Each sub-model module retrieves the line parameters and node connection relationships output by the topology identification module 22, calculates the theoretical line loss value and the actual line loss value of the transformer area respectively, analyzes the line loss deviation data, and transmits the data to the anomaly diagnosis module 24.

[0035] S5. Anomaly Diagnosis Module 24: Multi-dimensional Source Tracing and Qualitative Analysis: Edge Computing Unit 2 initiates Anomaly Diagnosis Module 24. Through the indicator system construction sub-module, it retrieves the multi-dimensional diagnostic indicator system. The dynamic weight adjustment sub-module dynamically adjusts the weights of each indicator according to the current time period (such as midday photovoltaic peak and nighttime electricity consumption peak). It performs correlation analysis between the line loss deviation data and the adjusted indicator system, integrates the correlation analysis of "new energy output - line loss rate" and the time series comparison results of "tidal load curve - loss change trend", and the loss type stripping sub-module accurately separates the new energy-related characteristic losses from the traditional abnormal losses of the power grid, locates the specific section, equipment or user of the line loss anomaly, and clearly determines the anomaly type (photovoltaic fluctuation statistical deviation, reverse power metering deviation, equipment failure, suspected electricity theft, etc.).

[0036] S6. Visualization Unit 3 Output and Maintenance Guidance: Edge computing unit 2 securely transmits all information, including anomaly diagnosis results, line loss calculation data, and dynamic topology of the transformer area, to visualization unit 3 via communication transmission component 4. Visualization unit 3 displays various data intuitively in a graphical and visual interface, clearly identifying the specific location, anomaly type, and loss percentage of abnormal loss points. Simultaneously, based on the anomaly diagnosis results, it generates standardized and targeted on-site maintenance and handling guidelines to guide maintenance personnel in accurately carrying out troubleshooting and handling work, achieving one-click tracing and precise handling.

[0037] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Any other modifications or equivalent substitutions made by those skilled in the art to the technical solutions of the present invention, as long as they do not depart from the spirit and scope of the technical solutions of the present invention, should be covered within the scope of the claims of the present invention.

Claims

1. A smart line loss detection device for transformer substations adapted to tidal loads of new energy sources, characterized in that, The system comprises a sensing and acquisition unit, an edge computing unit, a visualization unit, and a communication transmission component. The sensing and acquisition unit, edge computing unit, and visualization unit all interact with each other via the communication transmission component, forming a modular, plug-and-play architecture. The sensing and acquisition unit is located at the main outlet of the low-voltage side of the transformer in the distribution area, the beginning of branch lines, the grid connection point of new energy sources, key nodes of tidal loads, and the inlet of user meter boxes. It collects minute-level electrical quantity data for each node, including three-phase voltage, current, active and reactive power, forward and reverse active energy, and four-quadrant reactive energy. The edge computing unit includes a data processing module, a topology identification module, a scene-specific line loss calculation module, and an anomaly diagnosis module, which are used for multi-source data processing, topology analysis, line loss calculation, and anomaly detection, respectively. The visualization unit graphically displays the dynamic topology of the distribution area, real-time historical line loss curves, and the location and type information of abnormal loss points. The communication transmission component ensures data security and real-time interaction between the sensing and acquisition unit, the edge computing unit, and the visualization unit.

2. The intelligent line loss detection device for transformer substations adapted to new energy tidal loads as described in claim 1, characterized in that, The sensing and acquisition unit includes a power acquisition module, a time synchronization module, and a data preprocessing component. The power acquisition module is used to achieve accurate acquisition of multiple electrical quantities. The time synchronization module is used to ensure that the time axis of the data at each acquisition point is consistent. The data preprocessing component is used to perform preliminary filtering and format standardization on the raw data.

3. The intelligent line loss detection device for transformer substations adapted to new energy tidal loads according to claim 1, characterized in that, The data processing module of the edge computing unit has a built-in time synchronization submodule, a data interpolation submodule, and a filtering submodule. The time synchronization submodule is used to achieve accurate alignment of multi-source data from the sensing and acquisition unit. The data interpolation submodule is used to intelligently complete missing data caused by intermittent power output from new energy sources. The filtering submodule is used to eliminate electrical quantity data interference caused by tidal load fluctuations and to ensure consistency of forward and reverse power statistical logic under bidirectional power flow.

4. The intelligent line loss detection device for transformer substations adapted to new energy tidal loads according to claim 3, characterized in that, The topology identification module of the edge computing unit includes a characteristic current injection submodule and a new energy branch concealment and avoidance submodule. The characteristic current injection submodule is used to realize automatic topology identification, and the new energy branch concealment and avoidance submodule is used to dynamically adjust the identification strategy and correct the topology results during peak periods of distributed power generation.

5. The intelligent line loss detection device for transformer substations adapted to new energy tidal loads according to claim 4, characterized in that, The edge computing unit's scenario-based line loss calculation module includes a midday photovoltaic peak sub-model module and a nighttime electricity consumption peak sub-model module. The midday photovoltaic peak sub-model module is used to calculate the losses of new energy reverse transmission lines and the additional losses due to limited absorption. The nighttime electricity consumption peak sub-model module is used to exclude the impact of new energy output, focus on traditional load peaks, and calculate the conventional losses of equipment and lines.

6. The intelligent line loss detection device for transformer substations adapted to new energy tidal loads according to claim 3, characterized in that, The anomaly diagnosis module of the edge computing unit includes an indicator system construction submodule, a dynamic weight adjustment submodule, and a loss type stripping submodule. The indicator system construction submodule builds a multi-dimensional diagnostic indicator system that includes equipment health, line parameters, real-time renewable energy consumption status, and typical tidal load patterns. The dynamic weight adjustment submodule dynamically adjusts the weights of each indicator according to typical electricity consumption periods. The loss type stripping submodule is used to distinguish between qualitative renewable energy-related characteristic losses and traditional abnormal losses of the power grid.

7. A method for intelligently investigating line losses in transformer substations adapted to tidal loads from renewable energy sources, characterized in that, Includes the following steps: S1. Deployment and debugging of sensing and acquisition units: Deploy sensing and acquisition units at key nodes in the distribution area, complete the link debugging between communication transmission components and each unit, set the minute-level data freeze and acquisition frequency of the sensing and acquisition units, and start the time synchronization module to achieve time calibration at all points; S2. Multi-source data acquisition and modular processing: Obtain electrical quantity data of each node through the power acquisition module, integrate the distribution area meter data, new energy monitoring data, and tidal load time series data to form a multi-source heterogeneous dataset, complete the data processing through the data processing module in the edge computing unit, and upload the standardized data to the edge computing unit through the communication transmission component; S3. Dynamic Topology Analysis via Topology Identification Module: The edge computing unit activates the topology identification module, injects non-intrusive characteristic current, collects feedback signals, and identifies the topology structure in conjunction with new energy data. If a new energy power generation peak occurs, the topology result is corrected through the new energy branch concealment and avoidance sub-module. S4. Precise Calculation of Line Loss by Scenario: Based on the operating status, the edge computing unit automatically matches the midday photovoltaic peak or nighttime electricity consumption peak sub-model module with the scenario-specific line loss calculation module, retrieves the topology results, calculates the line loss value and the actual line loss value, and analyzes the deviation. S5. Multi-Dimensional Source Tracing via Anomaly Diagnosis Module: The edge computing unit activates the anomaly diagnosis module, retrieves diagnostic indicators, adapts indicator weights, correlates and analyzes line loss deviations, identifies loss types, locates anomalies, and determines their types. S6. Output and Maintenance Guidance from Visualization Unit: The edge computing unit sends the anomaly diagnosis results, line loss calculation data, and topology data to the visualization unit via the communication transmission component for data display, clearly identifies anomaly information, and generates targeted on-site maintenance handling guidance.

8. The intelligent method for detecting line losses in transformer substations adapted to tidal loads of new energy sources, as described in claim 7, is characterized in that... In step S2, the tidal load time series data includes load change data of morning peak, midday commercial load peak, evening peak, and nighttime trough. After being processed by the data processing module, it forms a standardized power grid area and time series dataset. The dataset achieves bidirectional synchronization between the edge computing unit and the sensing and acquisition unit through the communication transmission component, supporting real-time data completion by the sensing and acquisition unit.

9. The intelligent method for investigating line losses in transformer substations adapted to tidal loads of new energy sources, as described in claim 7, is characterized in that... In the modular closed-loop process of steps S1-S6, the responses and operations of each unit and module are all achieved through data interaction by the communication transmission component.

Images