An automatic data collection system for synchronous line loss and electricity consumption

Abstract

The present invention relates to the technical field of power grid operation detection, and more specifically discloses a system for automatically collecting data on synchronous line loss electricity, comprising: collecting comprehensive line loss electricity data and environmental data; establishing an operation status function according to the comprehensive line loss electricity data; obtaining environmental impact factors according to the environmental data; constructing an intelligent sensor network deployed with a microgrid concept in power grid partitions, constructing software-defined grid technology, optimizing routing and efficient transmission of partition data, constructing an edge computing architecture, preliminarily analyzing and processing data, and summarizing and transmitting the data to a central server, constructing alarm logic in each partition, and automatically adjusting or suggesting local power grid configuration to optimize performance while issuing targeted alarms, constructing intelligent analysis functions in each partition, self-diagnosing faults, automatically executing repair plans, improving data processing speed and system response time, reducing dependence on central services, and improving maintenance efficiency and accuracy.

Description

Technical Field

[0001] The present invention relates to the technical field of power grid operation detection, and more specifically to an automatic acquisition system for synchronous line loss power data. Background Art

[0002] With the continuous development of power grid technology, power grid pipelines are constantly upgraded, especially the distribution network is gradually becoming intelligent, the line loss management level of power supply enterprises is gradually improving, and an integrated power and line loss management system is built to promote an efficient and energy-saving line loss management system. At the same time, the line loss system adopts a "one-level deployment, two-level application" construction model, horizontally covering the entire province for centralized deployment, and the bus runs through the city and county levels to achieve source collection of electric energy data, automatic generation of line losses, and real-time monitoring of indicators. Based on the "four-point" management model of "division, voltage, line, and substation", standardized, intelligent, and refined management of line losses is achieved. The ultimate goal is to achieve daily "four-point" line loss calculation and abnormal analysis and detection.

[0003] In the management of synchronous line losses in the distribution network, power companies can reduce line losses by improving the utilization efficiency of equipment. Power companies use big data technology to monitor distribution equipment in real time, fully grasp the operating status of equipment, promptly discover and repair equipment failures, and ensure the normal operation of equipment. At the same time, power companies can evaluate and optimize equipment usage based on big data analysis results, reduce equipment losses by improving equipment utilization efficiency, and thus reduce the synchronous line loss rate.

[0004] However, the existing automatic data collection system for line loss and electricity consumption in the same period cannot adopt multi-element monitoring, which leads to inaccurate data collection; in complex or vast power grid systems, data loss, errors and delays may occur during data transmission; centralized data processing is used, resulting in poor system timeliness and the risk of system overload. Summary of the invention

[0005] In order to overcome the above-mentioned defects of the prior art, the present invention provides a system for automatically collecting synchronous line loss electricity data to solve the problems existing in the above-mentioned background technology.

[0006] The present invention provides the following technical solution: a synchronous line loss power data automatic collection system, comprising:

[0007] S1, data acquisition module, collects comprehensive data of line loss and power and environmental data;

[0008] S2, data comprehensive processing module, establishes the operation status function according to the comprehensive data of line loss and electricity; obtains the environmental impact factor according to the environmental data;

[0009] S3, partition processing output module, builds a smart sensor network deployed with the concept of microgrid in the power grid partition, builds software-defined grid (SDN) technology, optimizes the routing and efficient transmission of partition data, builds an edge computing architecture, preliminarily analyzes and processes data, and summarizes and transmits the data to the central server. It builds alarm logic in each partition, and can automatically adjust or recommend local power grid configuration to optimize performance while issuing targeted alarms. It builds intelligent analysis functions in each partition, performs self-diagnosis of faults, automatically executes repair plans, and guides operators for those that cannot be automatically repaired.

[0010] Furthermore, the line loss power comprehensive data includes power parameter data and power archive data;

[0011] The electric energy parameter data include real-time line loss quantity, current, voltage, power factor, grid frequency, and changes in loss circuits before and after line loss control; the electric energy archive data include the physical parameters and configuration information of substations, distribution lines, distribution transformers, historical electricity quantity data, and abnormal record data.

[0012] The environmental data include meteorological environmental data and electromagnetic environmental data; the meteorological environmental data include temperature, humidity, environmental dust concentration and atmospheric pressure; the electromagnetic environmental data include the geomagnetic value and industrial frequency magnetic field strength of the area where the electric energy meter is located.

[0013] The geomagnetic value and the industrial frequency magnetic field strength may affect the changes in the magnetic field in the area where the electric energy meter is located, which may cause changes in the sensitivity of the current sensor and thus affect the accuracy of current measurement; changes in the magnetic field may cause the generation of induced voltage in the electric energy meter; the induced voltage may be related to the frequency and strength of the external magnetic field, thereby introducing measurement errors; and strong magnetic fields may interfere with the electronic components in the electric energy meter, causing circuit instability and failure.

[0014] Furthermore, in the smart sensor network deployed by the microgrid concept, the microgrid is a small, self-made power supply network that can operate independently or connect to a larger main grid, using distributed voltage transformers to obtain voltage; using passive current transformers to obtain current; and the grid frequency is obtained in the following ways:

[0015] A digital control oscillation structure is arranged on a circuit for partitioning line loss; the digital control oscillation structure comprises an analog output terminal, a preamplifier filter terminal, an analog-to-digital conversion terminal, a digital signal processing terminal and a control circuit terminal; the digital signal processing terminal comprises an oscillation generating unit;

[0016] The analog output terminal outputs the analog input signal; the preamplifier filter terminal processes the analog input signal, eliminates noise and interference, and obtains an analog signal; the analog signal is converted into a digital signal through the analog-to-digital conversion terminal; the digital signal enters the digital signal processing terminal, and the oscillation generating unit generates a reference signal; the digital signal processing terminal is used to compare the reference signal with the digital signal, and the control circuit terminal adjusts the frequency and phase of the reference signal to make the reference signal match the digital signal; when the reference signal and the digital signal match, the output frequency of the oscillation generating unit is the grid frequency;

[0017] The temperature, humidity and atmospheric pressure are obtained by corresponding sensors installed in the working area of ​​the power grid;

[0018] The power factor is obtained by obtaining the phase difference between current and voltage; the power factor ;in, is the phase difference between current and voltage;

[0019] The method for obtaining the geomagnetic value includes:

[0020] Install the Hall sensor at the edge of the partition, and the detection surface of the Hall sensor should be parallel to the ground during installation;

[0021] The Hall sensor obtains the magnetic induction intensity in the horizontal x direction respectively , horizontal y direction magnetic induction intensity and the magnetic induction intensity perpendicular to the z direction ; Synthetic horizontal x-direction magnetic induction intensity and the horizontal y-direction magnetic induction intensity Get the horizontal magnetic induction intensity ; Synthesize the horizontal magnetic induction intensity and the vertical z-direction magnetic induction intensity to obtain the geomagnetic value ;

[0022] The method for obtaining the power frequency magnetic field strength includes:

[0023] Set at a distance of v meters from the energy meter The b magnetic sensors are arranged in a triangle or a rectangle;

[0024] Apply the fast Fourier transform algorithm to the data collected by the magnetic sensor to obtain the complex magnetic induction intensity of the power frequency component and its amplitude and phase;

[0025] Then the power frequency magnetic field strength at the electric energy meter is ;in, is the magnetic permeability; is the amplitude of the complex magnetic induction intensity of the power frequency component; is the phase of the complex magnetic induction intensity of the power frequency component; For energy meters to magnetic sensors The vector distance of represents an imaginary unit;

[0026] The establishment methods of the line loss power operation state function include:

[0027] Get voltage stability index based on voltage ;in, is the statistical cycle number of voltage; and is the voltage of two adjacent sampling points; is the average voltage within the statistical period; is the index of the sampling point voltage within the statistical period;

[0028] Get current stability index based on current ;in, is the first-order derivative of the original current curve; is the second-order derivative of the original current curve; the original current curve is obtained by dividing the current value in the current acquisition period by the time in the acquisition period. Perform curve fitting to obtain;

[0029] Obtain frequency stability indicators based on power grid frequency ,in, is the statistical cycle number of the power grid frequency; For the The grid frequency at the sub-sampling point; is the average grid frequency during the statistical period; is the index of the grid frequency of the sampling point within the statistical period;

[0030] Obtain electrical performance sub-functions based on voltage stability index, current stability index, frequency stability index, power factor, current harmonics and accumulated measurement values .

[0031] ;in, is the power factor; is the current harmonic; is the phase angle stability index; is the cumulative measurement error; and is the performance weight parameter;

[0032] ;in, It is the function of the phase angle changing with time, indicating the instantaneous phase difference between the input voltage and current. The changing function is about time. Function of

[0033] ;in, is the cumulative measurement value, It is the active electric energy measured by the metering chip.

[0034] Further, the edge computing architecture is set up at the local processing nodes of the power grid partition to perform preliminary analysis and process data;

[0035] The preliminary analysis steps include: data collection, preprocessing, anomaly detection and data cleaning;

[0036] The data detected by the intelligent sensor network is preprocessed, including filtering, denoising and normalization, to eliminate random errors and systematic errors in the data, improve the availability and consistency of the data, and ensure the accuracy and consistency of subsequent analysis through preprocessing;

[0037] The filtering process step comprises:

[0038] The collected current, voltage, power and other data are preliminarily cleaned to remove obvious outliers and noise for normalization;

[0039] According to the characteristics of different partitions and noise types, select appropriate filtering algorithms, including but not limited to sliding average filtering, median filtering and low-pass filtering;

[0040] Set filtering parameters and conduct filtering tests using simulated data;

[0041] The data of each partition is filtered separately, and the power data of each monitoring point in the partition is fused;

[0042] Use the filtered data to perform outlier detection and identify data points that deviate significantly from the normal range;

[0043] Integrate the filtered data of each partition, and compare and analyze the data before and after to form a complete power grid power analysis model;

[0044] It should be noted that the collected data needs to be preliminarily cleaned to remove obvious outliers and noise, and normalized to meet the requirements of subsequent filtering algorithms. Sliding average filtering is suitable for high-frequency oscillation systems and can effectively suppress periodic interference; while median filtering is suitable for eliminating impulse noise caused by accidental factors, and determining the parameters required for the filtering algorithm, such as window size, threshold, etc. These parameters directly affect the filtering effect and need to be optimized according to actual conditions. Through the simulation data, the filtering test is carried out to observe the filtering effect and adjust the parameters to optimize the performance. In each partition, the power data of different monitoring points in each partition are fused to form a unified data representation of the partition. For the detected outliers, they are corrected or eliminated to ensure the quality of the data set. At the same time, the abnormal information is fed back to the operation and inspection department for further investigation and processing. Through intuitive observation of the data comparison before and after filtering, quantitative evaluation is carried out in combination with specific indicators (such as signal-to-noise ratio, mean square error, etc.), the filtering effect is fully understood, the data of the entire line is integrated, the integrated data is analyzed, the cause of the abnormal line loss is found, and an alarm is issued. Targeted loss reduction measures are taken according to the cause to continuously optimize the operation efficiency of the power grid;

[0045] Furthermore, the environmental impact factor is obtained by:

[0046] Map temperature, humidity, atmospheric pressure, geomagnetic value and power frequency magnetic field strength to a numerical range of 0-1 to obtain corresponding normalized parameters;

[0047] Environmental impact factors ;

[0048] in, ; , , and is the meteorological parameter; is the temperature normalization parameter; Normalized parameters for humidity; is the normalized parameter for atmospheric pressure;

[0049] ; In the formula, and is the electromagnetic parameter; Normalization parameters for geomagnetic values; It is the normalized parameter of power frequency magnetic field strength.

[0050] It should be noted that the performance weight parameters, temperature influence parameters, usage rate influence parameters, state weight parameters, electrical performance sub-function weight parameters, own state sub-function weight parameters, meteorological parameters, electromagnetic parameters and data adjustment parameters in the above formulas are all obtained through training and solving with a large amount of measured data.

[0051] Furthermore, the method for constructing an edge computing framework includes:

[0052] Based on the smart sensor network deployed based on the concept of building a microgrid, select appropriate edge computing technologies and devices, such as edge nodes, gateway devices, and servers;

[0053] Install edge node devices (such as smart gateways or dedicated edge computing servers) at key points of the power grid to form partitions and number them. These devices should have the ability to collect, process and communicate data.

[0054] Choose a suitable operating system, such as Linux or Windows, based on the selected hardware platform. The operating system should support multi-threading and multi-tasking to fully utilize the computing resources of edge nodes, so that the edge data in the partition can be processed separately with better effect and faster response speed;

[0055] Configure appropriate transmission protocols (such as MQTT, CoAP, etc.) for the meters and sensors in the partition to ensure that they can send data at the preset frequency and format. These transmission protocols should support lightweight and low-power data transmission, conform to the partition edge computing environment, and have better transmission effects and higher stability.

[0056] Run data processing algorithms on the constructed edge nodes and deploy trained machine learning models to the edge nodes to achieve line loss prediction and fault diagnosis;

[0057] An alarm device is installed on the edge node. The alarm device has a monitoring function as well as data encryption and access control to protect the security of data during transmission and storage.

[0058] It should be noted that the numbering mode of the partitions is based on city + street + partition + monitoring point within the partition, such as (city A + street B + partition C + section C1, city A + street B + partition C + section C2, city A + street B + partition C + section Cn). Pre-training is carried out through all the above steps, and the trained data is input into the machine learning model. The machine learning model is then installed in each partition to monitor the line loss situation in each partition. When abnormal line loss occurs within the partition, the abnormal data is marked in red and sent to the main control in real time. Compared with the data without abnormality, the abnormal data is sent more quickly and at the same time The monitoring of abnormal parts is sent through the alarm device in the partition, and below the abnormal data, it is attached what kind of detection, what kind of abnormal data is generated, and a solution suggestion. If it can be automatically repaired, the time of automatic repair and the progress of the repair are attached. The normal monitoring data is then transmitted to the main control, which integrates it and determines the situation of line loss in the same period. If manual processing is required, it will be handled as soon as possible. When the abnormal data monitoring is sent, the data in the partition will be tested again. Continuous and uninterrupted detection will be carried out until the automatic repair is completed or the manual intervention repair is completed. When the subsequent detection is normal, regular detection will be carried out.

[0059] Technical effects and advantages of the present invention:

[0060] By deploying smart sensors in different partitions of the power grid, not only traditional electrical parameters are monitored, but also environmental variables and specific regional status, to achieve accurate regional data collection, more accurately analyze and optimize power grid performance, and use software-defined network (SDN) technology and error correction algorithms to optimize data routing and transmission reliability, reduce data loss and errors, and be used in power grid systems with large or complex terrains. Through the edge computing architecture, local processing nodes are set up in each power grid partition for preliminary analysis, which reduces the burden on the central server, improves data processing speed and system response time, introduces blockchain technology to record equipment maintenance history, realizes transparent information sharing and traceability, and adopts role-based access control to ensure data security. The adaptive energy management system adjusts energy configuration according to regional energy consumption patterns, cooperates with renewable energy technologies, improves energy efficiency and achieves energy self-sufficiency, and the system's self-diagnosis and repair functions guide local operators to execute or automatically execute repair plans, reduce dependence on central services, and improve maintenance efficiency and accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 It is a schematic diagram of the automatic acquisition system of synchronous line loss electricity data of the present invention.

[0062] Figure 2 It is a schematic diagram of the modules of the automatic acquisition system for synchronous line loss electricity data of the present invention.

[0063] Figure 3 It is a schematic diagram of an electronic device of the present invention. DETAILED DESCRIPTION

[0064] The technical solution of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the present invention. In addition, the forms of the various structures recorded in the following embodiments are only examples. The automatic collection system of synchronous line loss power data involved in the present invention is not limited to the various structures recorded in the following embodiments. All other implementations obtained by ordinary technicians in this field without making creative work belong to the scope of protection of the present invention.

[0065] Reference Figure 1-3 The present invention provides a synchronous line loss power data automatic collection system, comprising:

[0066] S1, data acquisition module, collects comprehensive data of line loss and power and environmental data;

[0067] S2, data comprehensive processing module, establishes the operation status function according to the comprehensive data of line loss and electricity; obtains the environmental impact factor according to the environmental data;

[0068] S3, partition processing output module, builds a smart sensor network deployed with the concept of microgrid in the power grid partition, builds software-defined grid (SDN) technology, optimizes the routing and efficient transmission of partition data, builds an edge computing architecture, preliminarily analyzes and processes data, and summarizes and transmits the data to the central server. It builds alarm logic in each partition, and can automatically adjust or recommend local power grid configuration to optimize performance while issuing targeted alarms. It builds intelligent analysis functions in each partition, performs self-diagnosis of faults, automatically executes repair plans, and guides operators for those that cannot be automatically repaired.

[0069] Furthermore, the line loss power comprehensive data includes power parameter data and power archive data;

[0070] The electric energy parameter data include real-time line loss quantity, current, voltage, power factor, grid frequency, and changes in loss circuits before and after line loss control; the electric energy archive data include the physical parameters and configuration information of substations, distribution lines, distribution transformers, historical electricity quantity data, and abnormal record data.

[0071] The environmental data include meteorological environmental data and electromagnetic environmental data; the meteorological environmental data include temperature, humidity, environmental dust concentration and atmospheric pressure; the electromagnetic environmental data include the geomagnetic value and industrial frequency magnetic field strength of the area where the electric energy meter is located.

[0072] The geomagnetic value and the industrial frequency magnetic field strength may affect the changes in the magnetic field in the area where the electric energy meter is located, which may cause changes in the sensitivity of the current sensor and thus affect the accuracy of current measurement; changes in the magnetic field may cause the generation of induced voltage in the electric energy meter; the induced voltage may be related to the frequency and strength of the external magnetic field, thereby introducing measurement errors; and strong magnetic fields may interfere with the electronic components in the electric energy meter, causing circuit instability and failure.

[0073] Furthermore, in the smart sensor network deployed by the microgrid concept, the microgrid is a small, self-made power supply network that can operate independently or connect to a larger main grid, using distributed voltage transformers to obtain voltage; using passive current transformers to obtain current; and the grid frequency is obtained in the following ways:

[0074] A digital control oscillation structure is arranged on a circuit for partitioning line loss; the digital control oscillation structure comprises an analog output terminal, a preamplifier filter terminal, an analog-to-digital conversion terminal, a digital signal processing terminal and a control circuit terminal; the digital signal processing terminal comprises an oscillation generating unit;

[0075] The analog output terminal outputs the analog input signal; the preamplifier filter terminal processes the analog input signal, eliminates noise and interference, and obtains an analog signal; the analog signal is converted into a digital signal through the analog-to-digital conversion terminal; the digital signal enters the digital signal processing terminal, and the oscillation generating unit generates a reference signal; the digital signal processing terminal is used to compare the reference signal with the digital signal, and the control circuit terminal adjusts the frequency and phase of the reference signal to make the reference signal match the digital signal; when the reference signal and the digital signal match, the output frequency of the oscillation generating unit is the grid frequency;

[0076] The temperature, humidity and atmospheric pressure are obtained by corresponding sensors installed in the working area of ​​the power grid;

[0077] The power factor is obtained by obtaining the phase difference between current and voltage; the power factor ;in, is the phase difference between current and voltage;

[0078] The method for obtaining the geomagnetic value includes:

[0079] Install the Hall sensor at the edge of the partition, and the detection surface of the Hall sensor should be parallel to the ground during installation;

[0080] The Hall sensor obtains the magnetic induction intensity in the horizontal x direction respectively , horizontal y direction magnetic induction intensity and the magnetic induction intensity perpendicular to the z direction ; Synthetic horizontal x-direction magnetic induction intensity and the horizontal y-direction magnetic induction intensity Get the horizontal magnetic induction intensity ; Synthesize the horizontal magnetic induction intensity and the vertical z-direction magnetic induction intensity to obtain the geomagnetic value ;

[0081] The method for obtaining the power frequency magnetic field strength includes:

[0082] Set at a distance of v meters from the energy meter The b magnetic sensors are arranged in a triangle or a rectangle;

[0083] Apply the fast Fourier transform algorithm to the data collected by the magnetic sensor to obtain the complex magnetic induction intensity of the power frequency component and its amplitude and phase;

[0084] Then the power frequency magnetic field strength at the electric energy meter is ;in, is the magnetic permeability; is the amplitude of the complex magnetic induction intensity of the power frequency component; is the phase of the complex magnetic induction intensity of the power frequency component; For energy meters to magnetic sensors The vector distance of represents an imaginary unit;

[0085] The establishment methods of the line loss power operation state function include:

[0086] Get voltage stability index based on voltage ;in, is the statistical cycle number of voltage; and is the voltage of two adjacent sampling points; is the average voltage within the statistical period; is the index of the sampling point voltage within the statistical period;

[0087] Get current stability index based on current ;in, is the first-order derivative of the original current curve; is the second-order derivative of the original current curve; the original current curve is obtained by dividing the current value in the current acquisition period by the time in the acquisition period. Perform curve fitting to obtain;

[0088] Obtain frequency stability indicators based on power grid frequency ,in, is the statistical cycle number of the power grid frequency; For the The grid frequency at the sub-sampling point; is the average grid frequency during the statistical period; is the index of the grid frequency of the sampling point within the statistical period;

[0089] Obtain electrical performance sub-functions based on voltage stability index, current stability index, frequency stability index, power factor, current harmonics and accumulated measurement values .

[0090] ;in, is the power factor; is the current harmonic; is the phase angle stability index; is the cumulative measurement error; and is the performance weight parameter;

[0091] ;in, It is the function of the phase angle changing with time, indicating the instantaneous phase difference between the input voltage and current. The changing function is about time. Function of

[0092] ;in, is the cumulative measurement value, It is the active electric energy measured by the metering chip.

[0093] Further, the edge computing architecture is set up at the local processing nodes of the power grid partition to perform preliminary analysis and process data;

[0094] The preliminary analysis steps include: data collection, preprocessing, anomaly detection and data cleaning;

[0095] The data detected by the intelligent sensor network is preprocessed, including filtering, denoising and normalization, to eliminate random errors and systematic errors in the data, improve the availability and consistency of the data, and ensure the accuracy and consistency of subsequent analysis through preprocessing;

[0096] The filtering process step comprises:

[0097] The collected current, voltage, power and other data are preliminarily cleaned to remove obvious outliers and noise for normalization;

[0098] According to the characteristics of different partitions and noise types, select appropriate filtering algorithms, including but not limited to sliding average filtering, median filtering and low-pass filtering;

[0099] Set filtering parameters and conduct filtering tests using simulated data;

[0100] The data of each partition is filtered separately, and the power data of each monitoring point in the partition is fused;

[0101] Use the filtered data to perform outlier detection and identify data points that deviate significantly from the normal range;

[0102] Integrate the filtered data of each partition, and compare and analyze the data before and after to form a complete power grid power analysis model;

[0103] It should be noted that the collected data needs to be preliminarily cleaned to remove obvious outliers and noise, and normalized to meet the requirements of subsequent filtering algorithms. Sliding average filtering is suitable for high-frequency oscillation systems and can effectively suppress periodic interference; while median filtering is suitable for eliminating impulse noise caused by accidental factors, and determining the parameters required for the filtering algorithm, such as window size, threshold, etc. These parameters directly affect the filtering effect and need to be optimized according to actual conditions. Through the simulation data, the filtering test is carried out to observe the filtering effect and adjust the parameters to optimize the performance. In each partition, the power data of different monitoring points in each partition are fused to form a unified data representation of the partition. For the detected outliers, they are corrected or eliminated to ensure the quality of the data set. At the same time, the abnormal information is fed back to the operation and inspection department for further investigation and processing. Through intuitive observation of the data comparison before and after filtering, quantitative evaluation is carried out in combination with specific indicators (such as signal-to-noise ratio, mean square error, etc.), the filtering effect is fully understood, the data of the entire line is integrated, the integrated data is analyzed, the cause of the abnormal line loss is found, and an alarm is issued. Targeted loss reduction measures are taken according to the cause to continuously optimize the operation efficiency of the power grid;

[0104] Furthermore, the environmental impact factor is obtained by:

[0105] Map temperature, humidity, atmospheric pressure, geomagnetic value and power frequency magnetic field strength to a numerical range of 0-1 to obtain corresponding normalized parameters;

[0106] Environmental impact factors ;

[0107] in, ; , , and is the meteorological parameter; is the temperature normalization parameter; Normalized parameters for humidity; is the normalized parameter for atmospheric pressure;

[0108] ; In the formula, and is the electromagnetic parameter; Normalization parameters for geomagnetic values; It is the normalized parameter of power frequency magnetic field strength.

[0109] It should be noted that the performance weight parameters, temperature influence parameters, usage rate influence parameters, state weight parameters, electrical performance sub-function weight parameters, own state sub-function weight parameters, meteorological parameters, electromagnetic parameters and data adjustment parameters in the above formulas are all obtained through training and solving with a large amount of measured data.

[0110] Furthermore, the method for constructing an edge computing framework includes:

[0111] Based on the smart sensor network deployed based on the concept of building a microgrid, select appropriate edge computing technologies and devices, such as edge nodes, gateway devices, and servers;

[0112] Install edge node devices (such as smart gateways or dedicated edge computing servers) at key points of the power grid to form partitions and number them. These devices should have the ability to collect, process and communicate data.

[0113] Choose a suitable operating system, such as Linux or Windows, based on the selected hardware platform. The operating system should support multi-threading and multi-tasking to fully utilize the computing resources of edge nodes, so that the edge data in the partition can be processed separately with better effect and faster response speed;

[0114] Configure appropriate transmission protocols (such as MQTT, CoAP, etc.) for the meters and sensors in the partition to ensure that they can send data at the preset frequency and format. These transmission protocols should support lightweight and low-power data transmission, conform to the partition edge computing environment, and have better transmission effects and higher stability.

[0115] Run data processing algorithms on the constructed edge nodes and deploy trained machine learning models to the edge nodes to achieve line loss prediction and fault diagnosis;

[0116] An alarm device is installed on the edge node. The alarm device has a monitoring function as well as data encryption and access control to protect the security of data during transmission and storage.

[0117] It should be noted that the numbering mode of the partitions is represented by city + street + partition + monitoring point within the partition, such as (city A + street B + partition C + section C1, city A + street B + partition C + section C2, city A + street B + partition C + section Cn). Pre-training is carried out through all the above steps, and the trained data is input into the machine learning model. The machine learning model is then installed in the partitions to monitor the line loss conditions in each partition. When abnormal line loss occurs within the partition, the abnormal data is marked in red and sent to the main control in real time. Compared with the data without abnormality, the abnormal data is sent more quickly and at the same time The monitoring of abnormal parts is sent through the alarm device in the partition, and below the abnormal data, it is attached what kind of detection, what kind of abnormal data is generated, and a solution suggestion. If it can be automatically repaired, the time of automatic repair and the progress of the repair are attached. The normal monitoring data is then transmitted to the main control, which integrates it and determines the situation of line loss in the same period. If manual processing is required, it will be handled as soon as possible. When the abnormal data monitoring is sent, the data in the partition will be tested again. Continuous and uninterrupted detection will be carried out until the automatic repair is completed or the manual intervention repair is completed. When the subsequent detection is normal, regular detection will be carried out.

[0118] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A synchronous line loss power data automatic collection system, characterized in that: include: S1, data acquisition module, collects comprehensive data of line loss and power and environmental data; S2, data comprehensive processing module, establishes the operation status function according to the comprehensive data of line loss and electricity; obtains the environmental impact factor according to the environmental data; S3, partition processing output module, builds a smart sensor network with microgrid concept deployment in the power grid partition, builds software-defined grid technology, optimizes the routing and efficient transmission of partition data, builds an edge computing architecture, preliminarily analyzes and processes data, and aggregates and transmits the data to the central server. It builds alarm logic in each partition, and can automatically adjust or recommend local power grid configuration to optimize performance while issuing targeted alarms. It builds intelligent analysis functions in each partition, self-diagnoses faults, automatically executes repair plans, and guides operators for those that cannot be automatically repaired; In the smart sensor network deployed by the microgrid concept, the microgrid is a small, homemade power supply network that can operate independently or connect to a larger main grid, using distributed voltage transformers to obtain voltage; passive current transformers to obtain current; The line loss power comprehensive data includes power parameter data and power archive data; Power parameter data include real-time line loss quantity, current, voltage, power factor, grid frequency, and changes in loss circuits before and after line loss control; power archive data include physical parameters and configuration information of substations, distribution lines, and distribution transformers, historical power data, and abnormal record data; The environmental data includes meteorological environmental data and electromagnetic environmental data; the meteorological environmental data includes temperature, humidity, environmental dust concentration and atmospheric pressure; the electromagnetic environmental data includes the geomagnetic value and power frequency magnetic field strength of the area where the electric energy meter is located; The temperature, humidity and atmospheric pressure are obtained by corresponding sensors installed in the working area of ​​the power grid; The power factor is obtained by obtaining the phase difference between the current and the voltage; the power factor in, is the phase difference between current and voltage; The method for obtaining the geomagnetic value includes: Install the Hall sensor at the edge of the partition, and the detection surface of the Hall sensor should be parallel to the ground during installation; The Hall sensor obtains the horizontal x-direction magnetic induction intensity B x , horizontal y direction magnetic induction intensity B y and the magnetic induction intensity B in the perpendicular z direction z ; Synthetic horizontal x-direction magnetic induction intensity B x and the horizontal y-direction magnetic induction intensity B y Get the horizontal magnetic induction intensity The horizontal magnetic induction intensity and the vertical z-direction magnetic induction intensity are synthesized to obtain the geomagnetic value The method for obtaining the power frequency magnetic field strength includes: N magnetic sensors are arranged at a distance of v meters from the electric energy meter, and b magnetic sensors are arranged in a triangle or a rectangle; Apply the fast Fourier transform algorithm to the data collected by the magnetic sensor to obtain the complex magnetic induction intensity of the power frequency component and its amplitude and phase; Then the power frequency magnetic field strength at the electric energy meter is Where, μ is the magnetic permeability; b i is the amplitude of the complex magnetic induction intensity of the power frequency component; θ i is the phase of the complex magnetic induction intensity of the power frequency component; R i is the vector distance from the electric energy meter to the magnetic sensor i; j represents the imaginary unit; The establishment methods of the line loss power operation state function include: Get voltage stability index based on voltage Where, C is the statistical cycle number of voltage; U c and U c-1 is the voltage of two adjacent sampling points; is the average voltage in the statistical period; c is the index of the sampling point voltage in the statistical period; Get current stability index based on current Wherein, I'(t) is the first-order derivative of the original current curve; I"(t) is the second-order derivative of the original current curve; the original current curve is obtained by curve fitting the current value in the current acquisition period according to the time t in the acquisition period; Obtain frequency stability indicators based on power grid frequency Where A is the statistical cycle number of the power grid frequency; f a is the grid frequency at the ath sampling point; is the average grid frequency in the statistical period; a is the index of the grid frequency at the sampling point in the statistical period; Obtaining electrical performance sub-function F1 according to voltage stability index, current stability index, frequency stability index, power factor, current harmonics and accumulated measurement values; Among them, P is the power factor; G is the current harmonic; S φ is the phase angle stability index; W is the cumulative measurement error; k1 and k2 are performance weight parameters; S φ =1-∫|φ(t)| 2 dt; where φ(t) is the function of the phase angle changing with time, indicating the instantaneous phase difference angle between the input voltage and current, and the changing function is a function of time t; Among them, E0 is the accumulated measurement value, and E is the active electric energy measured by the metering chip.

2. The automatic acquisition system for synchronous line loss power data according to claim 1 is characterized in that: The edge computing architecture is set up in the local processing nodes of the power grid partition to perform preliminary analysis and process data; The preliminary analysis steps include: data collection, preprocessing, anomaly detection and data cleaning; By preprocessing the data detected by the smart sensor network, including filtering, denoising and normalization, random errors and systematic errors in the data can be eliminated, and the availability and consistency of the data can be improved. Through preprocessing, the accuracy and consistency of subsequent analysis can be ensured.

3. The automatic data collection system for synchronous line loss electricity according to claim 2 is characterized in that: The filtering process step comprises: The collected current, voltage, power and other data are preliminarily cleaned, and the obvious outliers and noise are removed for normalization.

4. The automatic data collection system for synchronous line loss electricity according to claim 3 is characterized in that: According to the characteristics of different partitions and noise types, select appropriate filtering algorithms, including but not limited to sliding average filtering, median filtering and low-pass filtering; Set filtering parameters and conduct filtering tests using simulated data; The data of each partition is filtered separately, and the power data of each monitoring point in the partition is fused; Use the filtered data to perform outlier detection and identify data points that deviate significantly from the normal range; The filtered data of each partition are integrated, and the data before and after are compared and analyzed to form a complete power grid electricity analysis model.

5. The automatic data collection system for synchronous line loss electricity according to claim 3 is characterized in that: The environmental impact factors are obtained by: Map temperature, humidity, atmospheric pressure, geomagnetic value and power frequency magnetic field strength to a numerical range of 0-1 to obtain corresponding normalized parameters; Environmental impact factors in, r1, r2, r3 and r4 are meteorological parameters; Tw is the temperature normalization parameter; Hw is the humidity normalization parameter; Pw is the atmospheric pressure normalization parameter; EM=r5·Dw+r6·Sw; wherein r5 and r6 are electromagnetic parameters; Dw is the normalized parameter of geomagnetic value; Sw is the normalized parameter of industrial frequency magnetic field strength.

6. The automatic acquisition system for synchronous line loss power data according to claim 1 is characterized in that: The method for constructing an edge computing framework includes: Based on the smart sensor network deployed according to the concept of building a microgrid, select appropriate edge computing technologies and equipment, such as edge nodes, gateway devices, and servers.

7. The automatic data collection system for synchronous line loss electricity according to claim 6 is characterized in that: Install edge node devices at the joint checkpoints of the power grid to form partitions and number them. These devices should have the ability to collect, process and communicate data; Choose a suitable operating system based on the selected hardware platform. The operating system should support multi-threading and multi-tasking to fully utilize the computing resources of edge nodes. Configure appropriate transmission protocols for the meters and sensors in the zone to ensure that they can send data at the preset frequency and format. These transmission protocols should support lightweight and low-power data transmission. Run data processing algorithms on the constructed edge nodes and deploy trained machine learning models to the edge nodes to achieve line loss prediction and fault diagnosis; An alarm device is installed on the edge node. The alarm device has a monitoring function, as well as data encryption and access control to protect the security of data during transmission and storage.

8. The automatic data collection system for synchronous line loss electricity according to claim 1 is characterized in that: The method for acquiring the grid frequency includes: setting a digital control oscillation structure on a circuit for partitioning line loss electricity; the digital control oscillation structure includes an analog output terminal, a preamplifier filter terminal, an analog-to-digital conversion terminal, a digital signal processing terminal and a control circuit terminal; the digital signal processing terminal contains an oscillation generating unit; the analog output terminal outputs an analog input signal; the preamplifier filter terminal processes the analog input signal, eliminates noise and interference, and obtains an analog signal; the analog signal is converted into a digital signal through the analog-to-digital conversion terminal; the digital signal enters the digital signal processing terminal, and the oscillation generating unit generates a reference signal; the digital signal processing terminal is used to compare the reference signal with the digital signal, and the control circuit terminal adjusts the frequency and phase of the reference signal to make the reference signal match the digital signal; when the reference signal and the digital signal match, the output frequency of the oscillation generating unit is the grid frequency.

Images