Method, device and equipment for monitoring running state of power distribution network and medium

Abstract

The invention discloses a power distribution network operation state monitoring method, device and equipment and a medium, and the method comprises the steps: constructing a power distribution network operation state monitoring model based on historical operation data and simulated operation data of a target power distribution network region; real-time operation data of each monitoring node in the target power distribution network area are acquired, and the real-time operation data comprise electric power parameters, equipment states and environmental parameters; and inputting the real-time operation data into the power distribution network operation state monitoring model to obtain an operation state evaluation result of the target power distribution network region. Through the above mode, the multi-dimensional operation data in the target power distribution network area are collected in real time, and the multi-source data are input into the pre-trained intelligent monitoring model to carry out comprehensive analysis on various operation characteristics, so that the real operation state of the power distribution network is accurately evaluated, the accuracy of state identification is improved, the possibility of misjudgment and missed judgment is reduced, and the real operation state of the power distribution network is accurately evaluated. And the adaptability to complex operation working conditions is also obviously enhanced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power grid, and particularly relates to a power distribution network operation state monitoring method, device, equipment and medium. BACKGROUND

[0002] With the rapid growth of power demand and the continuous expansion of power grid scale, as the final link of power transmission, the operation stability of the power distribution network directly affects the power quality of users. The related technology mainly relies on the voltage, current and other electrical quantity data collected by the SCADA (Supervisory Control And Data Acquisition, data acquisition and monitoring control) system to monitor the operation state. However, the actual operation characteristics of the power distribution network are not only determined by the electrical parameters, but also affected by various external factors such as environmental temperature, humidity and equipment aging. This single-dimensional monitoring method not only cannot fully reflect the real operation state of the power distribution network, but also significantly increases the risk of misjudgment and omission, ultimately affecting the power supply reliability. SUMMARY

[0003] The present application provides a power distribution network operation state monitoring method, device, electronic equipment and medium to solve the technical problem that the single-dimensional monitoring method not only cannot fully reflect the real operation state of the power distribution network, but also significantly increases the risk of misjudgment and omission, ultimately affecting the power supply reliability.

[0004] In a first aspect, a power distribution network operation state monitoring method is provided, comprising:

[0005] constructing a power distribution network operation state monitoring model based on historical operation data and simulation operation data of a target power distribution network region;

[0006] obtaining real-time operation data of each monitoring node in the target power distribution network region, wherein the real-time operation data includes power parameters, equipment states and environmental parameters;

[0007] inputting the real-time operation data into the power distribution network operation state monitoring model to obtain an operation state evaluation result of the target power distribution network region.

[0008] In a second aspect, a power distribution network operation state monitoring device is provided, comprising:

[0009] a construction module configured to construct a power distribution network operation state monitoring model based on historical operation data and simulation operation data of a target power distribution network region;

[0010] an acquisition module configured to obtain real-time operation data of each monitoring node in the target power distribution network region, wherein the real-time operation data includes power parameters, equipment states and environmental parameters;

[0011] The generating module inputs the real-time operation data into the power distribution network operation state monitoring model to obtain an operation state evaluation result of the target power distribution network region.

[0012] In a third aspect, an electronic device is provided, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the steps of the power distribution network operation state monitoring method when executing the computer program.

[0013] In a fourth aspect, a computer readable storage medium is provided, which stores a computer program, and the computer program implements the steps of the power distribution network operation state monitoring method when executed by a processor.

[0014] In the scheme implemented by the power distribution network operation state monitoring method, device, electronic device and storage medium, the state monitoring model is constructed by combining historical operation data and simulation operation data, effectively solving the dual dilemma of insufficient historical fault samples and overly idealized simulation data in traditional methods. Then, multi-dimensional operation data in the target power distribution network region are collected in real time, and these multi-source data are input into the pre-trained intelligent monitoring model to comprehensively analyze various operation characteristics, so as to accurately evaluate the real operation state of the power distribution network, not only improving the accuracy of state recognition and reducing the possibility of misjudgment and omission, but also significantly enhancing the adaptability to complex operation conditions. BRIEF DESCRIPTION OF DRAWINGS

[0015] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the description of the embodiments of the present application. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0016] Figure 1 is a flowchart of the power distribution network operation state monitoring method in an embodiment of the present application;

[0017] Figure 2 is a structural diagram of the power distribution network operation state monitoring device in an embodiment of the present application. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical scheme and advantages of the embodiments of the present application more clear, the technical scheme in the embodiments of the present application will be clearly and completely described below in combination with the drawings in the embodiments of the present application. It should be understood that the drawings in the present application only play the purpose of illustration and description, and do not limit the protection scope of the present application.

[0019] In addition, it should be understood that the illustrative drawings are not drawn to scale. The flow charts used in the present disclosure show operations as implemented in some embodiments according to the present disclosure. It will be understood that the operations of the flow charts can be implemented in an order other than that shown, that steps can be added or removed, and that steps can be performed concurrently in different manners. Additionally, one or more other operations can be added to the flow charts or one or more operations can be removed from the flow charts under the direction of a person of skill in the art, based on the teachings of the present disclosure.

[0020] In addition, the embodiments described herein are only a part of the embodiments of the present disclosure, rather than all the embodiments. The components of the embodiments of the present disclosure described and shown in the accompanying drawings can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present disclosure provided in the accompanying drawings is not intended to limit the scope of the claimed present disclosure, but only represents selected embodiments of the present disclosure. Based on the embodiments of the present disclosure, all other embodiments obtained by a person of skill in the art without creative labor fall within the scope of the present disclosure.

[0021] It should be noted that the term "comprising" will be used in the embodiments of the present disclosure to indicate the presence of the features stated thereafter, but does not exclude the addition of other features. It should also be noted that similar reference numerals and letters refer to similar items in the following drawings, and therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings.

[0022] The present disclosure will be described in detail in conjunction with the accompanying drawings.

[0023] Referring to Figure 1 The embodiments of the present disclosure provide a power distribution network operation state monitoring method, which specifically comprises the following steps:

[0024] S10: Constructing a power distribution network state monitoring model based on historical operation data and simulation operation data of a target power distribution network region.

[0025] It can be understood that the execution subject of the present disclosure can be a power distribution network operation state monitoring device, and can also be a terminal or a server, which is not limited here. The embodiments of the present disclosure take the server as an example for description.

[0026] In this step, the current power distribution network operation state monitoring mainly adopts a machine learning method based on historical operation data. However, with the rapid development of smart grid technology and the continuous advancement of strong grid construction, the reliability of the power distribution network operation has been significantly improved, resulting in the system being in a normal operation state most of the time, and the number of fault samples actually collected is extremely limited. This phenomenon of severely unbalanced samples reduces the value of the massive normal state data, causes the distribution of positive and negative samples to be unbalanced during model training, and makes it difficult to achieve ideal fault recognition results, thereby restricting the accuracy of the reliability evaluation of the power distribution network. Based on the above problems, the application proposes to fuse the actual historical operation data of the target power distribution network region and the operation scene samples generated by simulation to construct a training data set, thereby training a high-precision power distribution network state monitoring model.

[0027] In the above manner, the reliability features of the real operation data are retained, and the simulation data effectively supplements the rare abnormal conditions in the actual samples, significantly improving the generalization ability and monitoring accuracy of the model.

[0028] In an embodiment of the application, a specific model training scheme is provided, in S10, that is, based on the historical operation data and the simulation operation data of the target power distribution network region, a power distribution network operation state monitoring model is constructed, specifically including the following steps S11-S15:

[0029] S11: Obtain the historical operation data of each monitoring node in the target power distribution network region, wherein the historical operation data includes historical operation data under normal state and historical operation data under abnormal state.

[0030] In this step, the historical operation data of each monitoring node in the target power distribution network region under different conditions such as normal operation and fault abnormality is collected, providing data support for subsequent system operation state evaluation.

[0031] S12: Simulate the simulation scenes of normal state and abnormal state by a time domain simulation method to obtain simulation operation data, wherein the simulation operation data includes normal simulation data under normal state and abnormal simulation data under multiple abnormal states.

[0032] In this step, a time-domain simulation method is used to simulate the working state of the power distribution network under design specifications and normal operating conditions. Normal conditions include smooth load, fault-free operation of equipment, and fluctuations of power grid frequency and voltage within the allowed range. Through normal condition simulation, the steady-state operating characteristics and key performance parameters of the system can be obtained, providing a benchmark reference for subsequent abnormal condition analysis. At the same time, various scenarios of the simulation of the analog power distribution network under abnormal conditions or deviating from the standard operating mode are simulated. Abnormal conditions include various types, such as power distribution network short-circuit fault, equipment overload, voltage out-of-limit electrical abnormality, and mechanical system component damage, lubrication failure mechanical fault. Through simulation of various abnormal conditions, the dynamic response characteristics under different fault modes can be comprehensively mastered, providing a basis for fault identification. On this basis, simulation operation data are collected, which accurately represent the dynamic characteristics of the system under different operating conditions.

[0033] S13: Based on the historical operation data and the simulation operation data, a sample data set is constructed.

[0034] In this step, with the rapid development of smart grid technology and the in-depth promotion of strong grid construction, the reliability of power distribution network operation continues to improve, resulting in that the system is in a normal operating state in most time periods, and the actual available historical fault sample data is extremely scarce. Simply relying on the massive data samples generated by simulation has the problem of over-idealization, which is significantly different from the actual power distribution network operation data (usually containing significant noise interference), causing the model trained based on simulation data to be easily affected by noise in actual power distribution network online state monitoring application, and it is difficult to achieve the expected effect. Based on the above problems, the present application proposes a method of fusing historical operation data and simulation operation data to construct a comprehensive sample data set, effectively solving the problem of interference of actual operation data noise on the idealized simulation data training model, and avoiding the accuracy deficiency caused by relying solely on idealized simulation data for state identification.

[0035] S14: The sample data set is labeled with operating state labels, and the feature variables in the sample data set are extracted as a training set.

[0036] S15: The training set is used to train the convolutional neural network model to obtain a power distribution network operating state monitoring model.

[0037] For steps S14-S15, the sample data set is classified and labeled with operating conditions, and key feature parameters are extracted to construct a training sample set. Based on the training set, a convolutional neural network is used for deep feature learning and model training, and finally a high-precision power distribution network operating state intelligent monitoring model is established.

[0038] Optionally, the abnormal state types include line overload state, transformer overheating state, frequency abnormal state, etc.

[0039] S20: Obtain real-time operation data of each monitoring node in the target power distribution network region, wherein the real-time operation data includes power parameters, device states, and environmental parameters.

[0040] In this step, the target power distribution network region refers to a power distribution network subsystem that is specifically selected as a key monitoring object in the process of power distribution network operation monitoring and management. The monitoring nodes reasonably arranged in this region cover key positions including busbars, tie switches, etc., ensuring that power distribution network operation information can be collected in all directions with high accuracy.

[0041] In actual applications, power parameters can only reflect basic electrical characteristics of the power system, such as voltage, current, power, etc. However, the actual operation state of the power distribution network is often significantly affected by environmental factors and device states. Specifically, in high-temperature weather, the increased resistance of the cable conductor leads to increased heating, which may cause insulation deterioration; in humid environments, the insulation of electrical equipment decreases significantly, greatly increasing the risk of dielectric loss and discharge. Relying solely on electrical quantity monitoring cannot capture the potential impact of these environmental factors and device state factors on the power distribution network. In addition, abnormal fluctuations in electrical quantity data do not necessarily indicate real faults. Environmental interference or device state changes can cause transient abnormalities in electrical parameters, such as parameter fluctuations caused by overhead line swinging due to strong winds. This single-dimensional monitoring method may misjudge normal disturbances as faults and may ignore early signs of device hidden defects on electrical parameters, significantly increasing the probability of misjudgment and omission. Based on the above reasons, the present application proposes to collect multi-dimensional operation data of each monitoring node in real time, specifically including: power parameters (voltage, current, power, frequency, etc.), device states (switch opening and closing position, transformer oil temperature, etc.), and environmental parameters (environmental temperature, relative humidity, wind speed, precipitation, etc.), thereby achieving real-time dynamic monitoring of the actual operation state of the power distribution network.

[0042] In the above manner, monitoring data from different dimensions is efficiently integrated, achieving full-factor coverage of operation parameters, improving monitoring sensitivity, and ensuring real-time and accurate identification of various abnormal states and dynamic changes during power distribution network operation.

[0043] In an embodiment of the present application, a specific target power distribution network region optimization scheme is provided, i.e., before obtaining real-time operation data of each monitoring node in the target power distribution network region, the following steps are further included:

[0044] Obtain the new energy penetration rate of the target power distribution network region.

[0045] If the new energy penetration rate is less than the preset penetration rate threshold, adjust the target power distribution network region.

[0046] In this embodiment, in the context of new power system construction, new energy dominated by photovoltaic and wind power continues to rise in the penetration rate in the power system due to its core advantages of zero carbon emission and renewable. For high proportion of new energy access area (usually defined as new energy penetration rate ≥ 30% of distribution network), large-scale renewable energy grid connection makes the accuracy of traditional voltage limit judgment decrease, and the coverage of diagnosis of new power grid problems such as harmonic resonance and reverse power flow is low. In order to improve the adaptability of online monitoring system to high penetration rate of new energy, the present application proposes to obtain the new energy penetration rate (i.e. photovoltaic / wind power penetration rate) of the target distribution network area, and when the penetration rate is lower than the preset penetration rate threshold, the target distribution network area is adjusted to solve the problems of reverse power flow, harmonic resonance, voltage fluctuation and improve the new energy consumption capacity.

[0047] In an embodiment of the present application, a specific real-time operation data acquisition scheme is provided, S20, that is, the real-time operation data of each monitoring node in the target distribution network area is acquired, specifically including the following steps S21-S23:

[0048] S21: Based on the topology structure of the target distribution network area, the monitoring nodes in the target distribution network area are divided into first-level monitoring nodes and second-level monitoring nodes.

[0049] In this step, in the distribution network monitoring system, the importance of each monitoring node is significantly different. Key nodes (such as bus, tie switch, etc.) need to use a higher data acquisition frequency to ensure that transient abnormalities and important operating parameter changes can be captured; while ordinary nodes will produce a large amount of redundant data if the same high-frequency acquisition strategy is used, which not only increases the storage and computing resource consumption, but also reduces the overall system operation efficiency. Therefore, the present application proposes a device importance grading method based on the topology structure of the distribution network: first, according to the network structure characteristics to evaluate the criticality of the device, the nodes corresponding to the key devices are defined as first-level monitoring nodes, and the nodes corresponding to the ordinary devices are defined as second-level monitoring nodes, thereby realizing the differentiated configuration of data acquisition strategy.

[0050] In practical application scenarios, edge nodes are deployed as monitoring nodes in the vicinity of data sources in the target power distribution network area (the starting point and end point of the distribution line, etc.). The monitoring nodes are installed with sensors (current / voltage transformers, temperature sensors, fault indicators), SCADA systems, transformer substation monitoring terminals (TTU), and feeder terminal units (FTU), which can collect real-time data such as voltage, current, power factor, and temperature. The collected data is uploaded to the server through a communication network (4G / 5G, optical fiber), and the server performs unified management and alarm pushing. The server has a distributed database (such as Hadoop, InfluxDB) that supports massive time series data storage; analysis platform: Python (Pandas, Scikit-learn), MATLAB, and power system simulation software (DIgSILENT, PSCAD). The server receives and parses multi-source heterogeneous data (such as CSV, JSON, binary), and unifies the data format (such as standardized timestamps and device ID encoding). In this way, the actual operating state of the power distribution network can be monitored in real time, and the data can be transmitted to the server for processing, reducing monitoring costs.

[0051] S22: Obtain real-time operating data of the first-level monitoring node based on the first preset frequency.

[0052] S23: Obtain real-time operating data of the second-level monitoring node based on the second preset frequency.

[0053] For steps S22-S23, the first-level monitoring node is configured to collect real-time data at a first preset frequency (higher sampling rate) to ensure complete acquisition of operating parameters of critical equipment; at the same time, the second-level monitoring node is configured to collect periodic data at a second preset frequency (relatively lower sampling rate) to ensure monitoring accuracy of critical equipment while effectively optimizing system resource utilization.

[0054] Optionally, the first-level and second-level monitoring nodes are set with differentiated data collection frequencies according to the importance of the monitoring equipment and the operating monitoring requirements of the target power distribution network area. The first preset frequency (first-level monitoring node) is set to a high sampling rate of 1 time per second, and the second preset frequency (second-level monitoring node) is set to a basic sampling rate of 1 time per minute, forming a significant level difference. This hierarchical collection mechanism ensures the real-time and completeness of operating data of critical equipment, and effectively controls the total amount of data by reducing the sampling frequency of ordinary nodes, achieving the best balance between monitoring accuracy and system load.

[0055] S30: Input the real-time operating data into the power distribution network operating state monitoring model to obtain the operating state monitoring result of the target power distribution network area.

[0056] In this step, the multi-dimensional operation data collected in real time is input into the trained power distribution network operation state monitoring model, and the operation state evaluation result of the target power distribution network region is output in real time through intelligent algorithm analysis and processing.

[0057] In an embodiment of the present application, a power distribution network operation state monitoring scheme is provided, and in S30, real-time operation data is input into the power distribution network operation state monitoring model to obtain the operation state monitoring result of the target power distribution network region, specifically including steps S31-S33:

[0058] S31: Preprocessing the real-time operation data.

[0059] In this step, since the real-time operation data may have problems such as noise interference, data missing, and abnormal values, which will affect the accuracy and reliability of subsequent analysis, data preprocessing techniques are used to effectively eliminate these adverse factors and improve data quality.

[0060] S32: Inputting the preprocessed real-time operation data into the power distribution network operation state monitoring model to obtain the confidence probability of various operation states.

[0061] In this step, the preprocessed real-time operation data is input into the trained power distribution network operation state monitoring model. After obtaining the real-time data, the model performs multi-state joint evaluation and outputs the confidence probability distribution of the power distribution network in various operation states. The confidence probability represents the likelihood of the target power distribution network region being in a specific state, and its numerical range is strictly limited to [0, 1]. For example, when the model outputs a normal state confidence probability of 0.8 and an abnormal state confidence probability of 0.2, it means that the current power distribution network is in a normal operation state with a probability of 80%, and there is a probability of 20% of being in an abnormal state.

[0062] S33: Determining the operation state of the target power distribution network region based on the confidence probability.

[0063] In this step, the confidence probabilities of various operation states are arranged in descending order, and the state corresponding to the probability peak is selected as the current operation state of the target power distribution network region. Specifically, when the confidence probability of the normal operation state is the highest, it is determined that the system is in a steady-state operation condition; if the confidence probability of a specific abnormal state is significantly higher than that of other states, it is diagnosed that the target power distribution network region has an abnormal condition of the corresponding type.

[0064] In actual application scenarios, the distribution characteristics of the confidence probability need to be comprehensively evaluated in the state determination process. When the confidence probability values of each state present approximate distribution (such as the maximum probability difference being less than 0.2), it indicates that there is significant uncertainty in the system operating state, at which time a supplementary data acquisition mechanism or a multi-modal fusion analysis method should be started to assist in judgment. As an enhancement strategy, a dynamic confidence threshold (recommended value 0.7-0.9) can be preset, and when the highest confidence probability of the state fails to exceed the set threshold, the system will trigger a hierarchical warning mechanism to prompt the operation and maintenance personnel to intervene in expert diagnosis and in-depth analysis.

[0065] In the above manner, by monitoring the operating state of the target power distribution network region in real time, potential faults and abnormal conditions are found in time, corresponding measures are taken for processing, the expansion and spread of faults are avoided, and the power outage time and loss are reduced.

[0066] In an embodiment of the present application, a specific data preprocessing scheme is provided, in S31, that is, the real-time operating data is preprocessed, specifically including the following steps S311-S314:

[0067] S311: Cleaning the real-time operating data.

[0068] In this step, due to factors such as device failure, channel interference, and abnormal human operation of the data acquisition terminal, there are often problems such as noise pollution and data redundancy in the real-time operating data. Data cleaning can effectively identify and eliminate these records that do not meet the data quality standards, thereby ensuring that the data set meets the analysis requirements in terms of accuracy, consistency, and completeness.

[0069] In actual application scenarios, repeated data elimination is implemented, and for the phenomenon of multiple records of the same data item caused by abnormal acquisition system (such as data retransmission caused by communication interruption of the SCADA system), redundant records need to be accurately identified and deleted through methods such as timestamp comparison and feature value matching.

[0070] S312: Abnormal value correction of the cleaned real-time operating data.

[0071] In this step, an abnormal value refers to an individual value in the data that deviates significantly from other data, which may be caused by accidental factors (such as instantaneous device failure, external interference, etc.). For example, the current value monitored at a certain moment is suddenly much higher than during normal operation, and this excessively high current value may be an abnormal value. The existence of abnormal values may have a great impact on subsequent data analysis, leading to a decrease in the accuracy of the model. Therefore, the abnormal values need to be corrected to make them more consistent with the overall distribution of the data, thereby improving the reliability of data analysis.

[0072] In practical application scenarios, the commonly used method is statistical analysis method, which calculates the mean, standard deviation and other statistical quantities of data, identifies abnormal values according to certain rules (such as setting threshold, regarding data exceeding the mean by a certain multiple of standard deviation as abnormal value), and then can use interpolation method (such as linear interpolation, spline interpolation, etc.), smoothing method (such as moving average method) and the like to correct the abnormal values; it can also be combined with business knowledge and historical data to judge and correct, for example, according to the normal operation range of the equipment and the change trend of the historical data to determine the reasonable replacement value of the abnormal value.

[0073] S313: filling the missing values in the corrected real-time operation data.

[0074] In this step, due to sensor failure, communication interruption, data storage error and other reasons in the data acquisition process, part of the data may be missing. For example, if a temperature sensor fails at a certain time, the temperature data at that time will be missing. The existence of missing values will affect the integrity and continuity of the data, so that some data analysis methods and models cannot be used normally. Therefore, it is necessary to fill the missing values to ensure the availability of the data and the accuracy of the analysis results.

[0075] In an embodiment of the present application, a specific data preprocessing scheme is provided, S313, that is, filling the missing values in the corrected real-time operation data, specifically including the following steps:

[0076] Obtaining the missing time of the missing values in the real-time operation data.

[0077] If the missing time is less than the missing time threshold, linear interpolation is used to fill the missing values.

[0078] If the missing time is greater than or equal to the missing time threshold, the LSTM algorithm is used to fill the missing values.

[0079] In this embodiment, in the process of filling the missing values, the time period with missing data needs to be located and marked first, and the duration of each missing period (i.e. missing time) is accurately recorded. According to the difference in missing time, different missing mechanisms can be inferred: short-time missing (such as <5 minutes) is usually caused by temporary factors such as communication delay or transient interference; while long-time missing (such as ≥5 minutes) often indicates continuous abnormalities such as failure of the acquisition device or interruption of the communication link. For missing data caused by different mechanisms, different interpolation algorithms should be used to ensure the accuracy and practicality of data filling.

[0080] Specifically, a pre-defined time threshold is used to distinguish different missing data situations. Its setting typically needs to be determined based on the specific business scenario and data characteristics. For example, setting the missing time threshold to 5 minutes means that missing values ​​less than 5 minutes are considered transient. When the missing time is short (<5 minutes), data changes usually exhibit a linear trend, and linear interpolation can reasonably estimate missing values; it is also computationally simple and efficient. When the missing time is long (≥5 minutes), data changes may be influenced by various complex factors, exhibiting non-linear and long-term trends, and linear interpolation may not accurately estimate missing values. The LSTM algorithm, however, can learn patterns and regularities in historical data to more accurately predict missing values.

[0081] By adopting the above-mentioned adaptive strategy based on the dynamic selection of interpolation methods according to the missing time, and through in-depth analysis of the differences in data characteristics with different missing durations, the computational efficiency of linear interpolation and the nonlinear prediction advantages of LSTM deep networks are organically integrated, thereby significantly improving the accuracy and reliability of missing value filling.

[0082] S314: Standardize and transform the populated real-time running data.

[0083] In this step, the filled real-time running data is normalized, and the conversion range can be [-1, 1] or [0, 1] to eliminate the influence of dimensions.

[0084] As can be seen, the above scheme provides a smart monitoring method for distribution networks based on multi-source data fusion. By combining historical operating data and simulated operating data to construct a condition monitoring model, it effectively solves the dual dilemmas of insufficient historical fault samples and overly idealized simulation data in traditional methods. Subsequently, multi-dimensional operating data within the target distribution network area are collected in real time, and this multi-source data is input into a pre-trained smart monitoring model to comprehensively analyze various operating characteristics, thereby accurately assessing the actual operating status of the distribution network. This not only improves the accuracy of condition identification and reduces the possibility of misjudgment and omission, but also significantly enhances the adaptability to complex operating conditions.

[0085] In one embodiment, a power distribution network operation status monitoring device is provided, which corresponds one-to-one with the power distribution network operation status monitoring method described in the above embodiments. For example... Figure 2 As shown, the power distribution network operation status monitoring device 100 includes: a construction module 101, an acquisition module 102, and a generation module 103. Detailed descriptions of each functional module are as follows:

[0086] Module 101 is used to build a power distribution network operation status monitoring model based on historical and simulated operation data of the target power distribution network area.

[0087] The acquisition module 102 is configured to acquire real-time operation data of each monitoring node in the target power distribution network region, wherein the real-time operation data comprises power parameters, device states and environmental parameters.

[0088] The generation module 103 is configured to input the real-time operation data into a power distribution network operation state monitoring model to obtain an operation state evaluation result of the target power distribution network region.

[0089] In an embodiment, the construction module 101 specifically comprises:

[0090] The third acquisition unit is configured to acquire historical operation data of each monitoring node in the target power distribution network region,

[0091] The historical operation data comprises historical operation data in a normal state and historical operation data in an abnormal state.

[0092] The fourth acquisition unit is configured to simulate simulation scenarios in the normal state and the abnormal state by a time-domain simulation method to acquire simulation operation data, wherein the simulation operation data comprises normal simulation data in the normal state and abnormal simulation data in multiple abnormal states.

[0093] The construction unit is configured to construct a sample data set based on the historical operation data and the simulation operation data.

[0094] The first generation unit is configured to label operation state labels of the sample data set and extract feature variables in the sample data set as a training set.

[0095] The training unit is configured to train the convolutional neural network model using the training set to obtain the power distribution network operation state monitoring model.

[0096] In an embodiment, the acquisition module 102 specifically comprises:

[0097] The distribution unit is configured to distribute the monitoring nodes in the target power distribution network region into first-level monitoring nodes and second-level monitoring nodes based on a topological structure of the target power distribution network region.

[0098] The first acquisition unit is configured to acquire real-time operation data of the first-level monitoring nodes based on a first preset frequency.

[0099] The second acquisition unit is configured to acquire real-time operation data of the second-level monitoring nodes based on a second preset frequency.

[0100] In an embodiment, the generation module 103 specifically comprises:

[0101] The processing unit is configured to pre-process the real-time operation data.

[0102] The second generation unit is configured to input the preprocessed real-time operation data into a power distribution network operation state monitoring model to obtain confidence probabilities of various operation states.

[0103] The determination unit is configured to determine an operation state of the target power distribution network region based on the confidence probabilities.

[0104] In an embodiment, the processing unit is specifically configured to:

[0105] cleaning the real-time operation data row data;

[0106] performing outlier correction on the cleaned real-time operation data;

[0107] performing missing value filling on the corrected real-time operation data;

[0108] performing standardization conversion on the filled real-time operation data.

[0109] In an embodiment, the processing unit is specifically further configured to:

[0110] obtaining a missing time of the missing value in the real-time operation data;

[0111] if the missing time is less than a missing time threshold, performing linear interpolation on the missing value;

[0112] if the missing time is greater than or equal to the missing time threshold, performing LSTM algorithm on the missing value.

[0113] In an embodiment, the obtaining module 102 further includes:

[0114] The fourth obtaining unit is configured to obtain a new energy penetration rate of the target power distribution network region.

[0115] In an embodiment, the device further includes:

[0116] The adjusting module is configured to, if the new energy penetration rate is less than a preset penetration rate threshold, adjust the target power distribution network region.

[0117] The present application provides a power distribution network operation state monitoring device 100, which effectively solves the dual dilemmas of insufficient historical fault samples and overly idealized simulation data in traditional methods by combining historical operation data and simulation operation data to construct a state monitoring model. Subsequently, multi-dimensional operation data within the target power distribution network region are collected in real time, and these multi-source data are input into a pre-trained intelligent monitoring model to comprehensively analyze various operation characteristics, thereby accurately evaluating the real operation state of the power distribution network, not only improving the accuracy of state recognition and reducing the possibility of misjudgment and missed judgment, but also significantly enhancing the adaptability to complex operation conditions.

[0118] The specific limitations of the power distribution network operation state monitoring device can refer to the limitations of the power distribution network operation state monitoring method described above, and will not be described again here. Each module in the power distribution network operation state monitoring device described above can be implemented by software, hardware, or a combination thereof, in whole or in part. The above-mentioned modules can be embedded in or independent of the processor in the electronic device in hardware form, or can be stored in the memory in the electronic device in software form, so as to be called and executed by the processor to perform the operations corresponding to each of the above-mentioned modules.

[0119] In one embodiment, an electronic device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the following steps when executing the computer program:

[0120] constructing a power distribution network operation state monitoring model based on historical operation data and simulation operation data of a target power distribution network region;

[0121] obtaining real-time operation data of each monitoring node in the target power distribution network region, wherein the real-time operation data includes power parameters, device states, and environmental parameters;

[0122] inputting the real-time operation data into the power distribution network operation state monitoring model to obtain an operation state evaluation result of the target power distribution network region.

[0123] In one embodiment, a computer readable storage medium is provided, which stores a computer program, and the computer program is executed by a processor to implement the following steps:

[0124] constructing a power distribution network operation state monitoring model based on historical operation data and simulation operation data of a target power distribution network region;

[0125] obtaining real-time operation data of each monitoring node in the target power distribution network region, wherein the real-time operation data includes power parameters, device states, and environmental parameters;

[0126] inputting the real-time operation data into the power distribution network operation state monitoring model to obtain an operation state evaluation result of the target power distribution network region.

[0127] It should be noted that the functions or steps that the computer readable storage medium or the electronic device can achieve described above can correspond to the related descriptions of the server side and the client side in the foregoing method embodiments. To avoid repetition, they will not be described one by one here.

[0128] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer readable storage medium, and when executed, can include the processes of the above-mentioned embodiment methods. Any reference to memory, storage, database or other medium used in the embodiments provided by the present application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration but not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0129] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the division of the above-mentioned functional units and modules is exemplified, and in actual application, the above-mentioned functions can be completed by different functional units and modules according to needs, that is, the internal structure of the device is divided into different functional units or modules to complete all or part of the functions described above.

[0130] The above-mentioned embodiments are only used to illustrate the technical solutions of the present application, but not limit it. Although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part of the technical features. The modification or replacement does not make the essence of the corresponding technical solution deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application.

Claims

1. A method of monitoring the operating state of an electrical distribution network, characterized in that, The method comprises the following steps: constructing a power distribution network operation state monitoring model based on historical operation data and simulation operation data of a target power distribution network region; obtaining real-time operation data of each monitoring node in the target power distribution network region, wherein the real-time operation data comprises power parameters, equipment states and environmental parameters; inputting the real-time operation data into the power distribution network operation state monitoring model to obtain an operation state evaluation result of the target power distribution network region.

2. The method of claim 1, wherein, The step of constructing a power distribution network operation state monitoring model based on historical operation data and simulation operation data of a target power distribution network region comprises the following steps: obtaining historical operation data of each monitoring node in the target power distribution network region, wherein the historical operation data comprises historical operation data in a normal state and historical operation data in an abnormal state; simulating simulation scenarios in the normal state and the abnormal state by a time domain simulation method to obtain simulation operation data, wherein the simulation operation data comprises normal simulation data in the normal state and abnormal simulation data in multiple abnormal states; constructing a sample data set based on the historical operation data and the simulation operation data; performing operation state label annotation on the sample data set and extracting feature variables in the sample data set as a training set; training a convolutional neural network model using the training set to obtain the power distribution network operation state monitoring model.

3. The method of claim 1, wherein, The step of obtaining real-time operation data of each monitoring node in the target power distribution network region comprises the following steps: dividing monitoring nodes in the target power distribution network region into first-level monitoring nodes and second-level monitoring nodes based on a topological structure of the target power distribution network region; obtaining real-time operation data of the first-level monitoring nodes based on a first preset frequency; obtaining real-time operation data of the second-level monitoring nodes based on a second preset frequency.

4. The method of claim 1, wherein, The step of inputting the real-time operation data into the power distribution network operation state monitoring model to obtain an operation state evaluation result of the target power distribution network region comprises the following steps: preprocessing the real-time operation data; inputting the preprocessed real-time operation data into the power distribution network operation state monitoring model to obtain confidence probabilities of various operation states; determining an operation state of the target power distribution network region based on the confidence probabilities.

5. The method of claim 4, wherein, The step of preprocessing the real-time operation data comprises the following steps: cleaning the real-time operation data; performing outlier correction on the cleaned real-time operation data; performing missing value filling on the corrected real-time operation data; performing standardization conversion on the filled real-time operation data.

6. The method of claim 5, wherein, The step of performing missing value filling on the corrected real-time operation data comprises the following steps: obtaining a missing time of a missing value in the real-time operation data; if the missing time is less than a missing time threshold, performing linear interpolation on the missing value; if the missing time is greater than or equal to the missing time threshold, performing LSTM algorithm on the missing value.

7. The method according to any one of claims 1 to 6, characterized in that, Before the step of obtaining real-time operation data of each monitoring node in the target power distribution network region, the method further comprises the following steps: obtaining a new energy penetration rate of the target power distribution network region. If the new energy penetration rate is less than a preset penetration rate threshold, the target power distribution network region is adjusted.

8. A power distribution network operating state monitoring apparatus characterized by comprising: The method comprises the following steps: A construction module is configured to construct a power distribution network operation state monitoring model based on historical operation data and simulation operation data of a target power distribution network region. An acquisition module is configured to acquire real-time operation data of each monitoring node in the target power distribution network region, wherein the real-time operation data comprises power parameters, equipment states and environmental parameters. A generation module is configured to input the real-time operation data into the power distribution network operation state monitoring model to obtain an operation state evaluation result of the target power distribution network region.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, The processor executes the computer program to implement the steps of the power distribution network operation state monitoring method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, the computer program comprising instructions that, when executed by a computer, cause the computer to perform the method of any one of claims 1 to 9. The computer program is executed by the processor to implement the steps of the power distribution network operation state monitoring method according to any one of claims 1 to 7.

Images