Abnormal point identification method and system based on power distribution network voltage data cleaning

Abstract

The invention discloses an abnormal point identification method and system based on power distribution network voltage data cleaning, and the method comprises the steps: obtaining the three-phase voltage data of each metering point in a power distribution network according to a preset sampling frequency, and marking the data within 30% of the upper and lower limits of the voltage as effective; the average voltage of each phase of each metering point is calculated, and the rated voltage grade of each phase of voltage is speculated; speculating the rated voltage grade of a metering point according to the rated voltage grade of each phase, and marking data within 30% of the upper and lower limits of the rated voltage of the metering point as effective; calculating the voltage qualified rate of each phase according to the effective data, and calculating the voltage qualified rate of the metering point; and finding out the metering points with abnormal data, and marking the metering points. On the basis of effectively determining the rated voltage grade, the effective voltage data are confirmed again, the accuracy of the voltage qualification rate can be improved, an accurate and effective judgment basis is provided for voltage monitoring management, abnormal metering points in the power distribution network can be identified, and a reliable judgment basis is provided for dispatchers.

Description

Technical Field

[0001] The present invention relates to the technical field of power monitoring, and in particular to an abnormal point identification method and system based on power distribution network voltage data cleaning. Background Art

[0002] As an important part of the power system, the distribution network is responsible for distributing electric energy to various users. The distribution network consists of overhead lines, cables, poles, distribution transformers, disconnectors, reactive compensation capacitors, metering devices, and some ancillary facilities. It generally adopts a closed-loop design and open-loop operation, and its structure is radial. The distribution network is large in scale, with many operating points, a wide operating area, and a relatively poor safety environment. Therefore, there are relatively more safety risk factors in the distribution network. How to accurately and quickly identify abnormal conditions in the distribution network, and formulate appropriate inspection or maintenance plans based on the abnormal conditions to ensure that the suspected fault location can be quickly and accurately located and isolated has become a problem we need to solve at present.

[0003] The voltage monitoring and management system can realize the data collection and reporting work of the voltage monitoring points, but the massive data it accesses may have data quality problems, affecting the correctness of data statistics such as the voltage qualification rate. Before data calculation, these problem data need to be preprocessed, otherwise it is difficult to provide accurate and effective judgment basis for voltage monitoring management.

[0004] To this end, the present invention provides an abnormal point identification method based on voltage data cleaning to improve the accuracy of voltage qualification rate, identify abnormal voltage data in the distribution network, mark abnormal points, and provide a reliable judgment basis for dispatchers. Summary of the invention

[0005] In view of the above-mentioned problems, the present invention is proposed.

[0006] Therefore, the technical problem solved by the present invention is: there are a large number of metering points in the distribution network, and the voltage data will be disturbed by various factors, including grid load fluctuations, equipment failures or interference, etc., resulting in noise and abnormal values ​​in the collected data. If these data are not effectively cleaned, it will directly affect the accuracy of voltage monitoring, and then lead to incorrect voltage qualification rate assessment, and fail to provide a reliable basis for grid status judgment. Since the voltage fluctuation range of each metering point in the distribution network is large, and the nominal voltage levels of different devices are also different, it is difficult for traditional methods to accurately infer the rated voltage level of the metering point through simple sampling and calculation, which can easily lead to inaccurate voltage status judgment, especially when facing a complex, multi-phase grid environment. The risk of such misjudgment will be higher. Existing voltage monitoring systems usually rely on manual experience or simple algorithms to process voltage data, and it is difficult to automatically identify and mark abnormal data points, resulting in potential grid failures or abnormal conditions that cannot be discovered in time, increasing the risk of grid operation.

[0007] In order to solve the above technical problems, the present invention provides the following technical solutions: a method for identifying abnormal points based on power distribution network voltage data cleaning, comprising:

[0008] Acquire the three-phase voltage data of each metering point in the distribution network according to the preset sampling frequency, and mark the data within 30% of the upper and lower limits of the voltage as valid;

[0009] Calculate the average voltage of each phase at each metering point and infer the rated voltage level of each voltage item;

[0010] The rated voltage level of the metering point is estimated based on the rated voltage level of each phase, and the data within 30% of the upper and lower limits of the rated voltage of the metering point is marked as valid;

[0011] Calculate the voltage qualification rate of each phase according to the effective data, and calculate the voltage qualification rate of the metering point;

[0012] Find the measurement points where the data is abnormal and mark them.

[0013] As a preferred solution of the abnormal point identification method based on distribution network voltage data cleaning described in the present invention, wherein: the acquisition of three-phase voltage data of each metering point in the distribution network according to the preset sampling frequency includes: s Collect the A, B, and C three-phase voltage data U of each metering point A (t), U B (t), U C (t), and store these data in the database in real time;

[0014] For each collected voltage data, if its value is within 30% of the upper and lower limits of the rated voltage of the current metering point, it will be marked as valid data; otherwise, it will be marked as invalid data;

[0015] All labeled data are stored in the database in time series.

[0016] As a preferred solution of the abnormal point identification method based on distribution network voltage data cleaning described in the present invention, wherein: the calculation of the average voltage of each phase of each metering point includes: for each metering point three-phase voltage data U A (t),U B (t),U C (t) is averaged over time to obtain the average voltage of each phase

[0017] The estimated rated voltage level of each voltage includes: Estimate the rated voltage level V within the preset voltage level range nominal,X ,like If it falls within the upper and lower limits of the rated voltage level, it is inferred that the rated voltage of the current phase is the current level. The basis for the inference is:

[0018] If the average voltage of one phase If it falls within the range of [46.192V, 69.288V], it is inferred that the rated voltage level of this phase is 57.74V;

[0019] If the average voltage of one phase If it falls within the range of [80V, 120V], it is inferred that the rated voltage level of this phase is 100V;

[0020] If the average voltage of one phase If it falls within the range of [176V, 264V], it is inferred that the rated voltage level of this phase is 220V;

[0021] If the average voltage of one phase If it falls within the range of [304V, 456V], it is inferred that the rated voltage level of this phase is 380V;

[0022] If the average voltage of one phase If it is not within any range, the voltage data of that phase is marked as invalid.

[0023] As a preferred solution of the abnormal point identification method based on distribution network voltage data cleaning described in the present invention, wherein: the inference of the rated voltage level of the metering point according to the rated voltage level of each phase includes: if the three phases of the metering point have rated voltages and the three-phase rated voltages are the same, then the rated voltage level is set as the rated voltage level of the metering point, otherwise the rated voltage level of the metering point is set to unknown;

[0024] If the metering point has two-phase rated voltages, and the two-phase rated voltages are the same, then the rated voltage level is set as the rated voltage level of the metering point, otherwise the rated voltage level of the metering point is set to unknown;

[0025] If the metering point has only one phase rated voltage, the rated voltage level of the metering point is set to the rated voltage level of the phase;

[0026] If there is no valid rated voltage at the metering point, set the rated voltage level of the metering point to unknown.

[0027] As a preferred solution of the abnormal point identification method based on distribution network voltage data cleaning described in the present invention, wherein: the data within 30% of the upper and lower limits of the rated voltage of the metering point is marked as valid, including calculating the valid voltage data range of the metering point according to the estimated rated voltage of each metering point It is defined as:

[0028]

[0029] in, is the estimated rated voltage at the metering point;

[0030] For each voltage data point, if its value falls within the valid voltage data range If the data is within the range, the data is marked as valid data; otherwise, it is marked as invalid data;

[0031] All data marked as valid are stored in the database in time series.

[0032] As a preferred solution of the abnormal point identification method based on distribution network voltage data cleaning described in the present invention, wherein: the voltage qualification rate of each phase is calculated according to the effective data, including: calculating the voltage qualification rate P of each phase X according to the effective voltage data 合格,X , which is defined as:

[0033]

[0034] Among them, N 合格,X In the valid data range The number of qualified voltage data in N 总,X is the number of all valid voltage data of this phase;

[0035] For the phases set to invalid voltage, the voltage qualification rate is not calculated and the data of the phases is directly excluded;

[0036] The calculation of the voltage qualification rate of the metering point includes summing up the voltage qualification rates of all effective phases and calculating the comprehensive voltage qualification rate P of the metering point. 合格，计量点 , which is defined as:

[0037]

[0038] Among them, N 有效相 is the number of effective phases involved in the calculation, ∑ 有效相 P 合格,X It is the sum of the qualified rates of the phase voltages involved in the calculation.

[0039] The qualified rate calculation method in the present invention provides a more detailed voltage quality assessment method by calculating the single-phase and overall voltage data separately. Especially when dealing with the situation where invalid voltage phases may exist, the present invention can flexibly adjust the calculation strategy to avoid the overall misjudgment caused by invalid data in the traditional method. In addition, by calculating the average qualified rate of the effective phase, the real situation of the voltage quality can be more accurately reflected. In practical applications, this method can effectively improve the power grid monitoring system's ability to analyze voltage data, thereby providing more reliable support for power grid operation and maintenance.

[0040] As a preferred solution of the abnormal point identification method based on distribution network voltage data cleaning described in the present invention, wherein: the step of finding the metering point with abnormal data includes: if the rated voltage level of the three-phase voltage of the metering point cannot be determined, then the metering point is an abnormal metering point and is marked as a metering point with unknown rated voltage;

[0041] If a metering point does not have any voltage data within the statistical period, the metering point is an abnormal metering point and is marked as a metering point with no data;

[0042] If the proportion of valid voltage data points of a metering point in the statistical period to the total data points is less than 50%, the metering point is an abnormal metering point and is marked as a metering point with insufficient data;

[0043] For the three-phase voltage of a metering point, if its three-phase variance is greater than the set threshold m, the metering point is an abnormal metering point and is marked as a metering point with large three-phase voltage variance;

[0044] If there are only two-phase voltages, the voltage difference ΔU between the two phases is calculated. If ΔU exceeds the set threshold n, the metering point is an abnormal metering point and is marked as a metering point with a large difference in two-phase voltages.

[0045] The abnormal metering point identification method can not only identify abnormal points that are easily overlooked in traditional methods (such as metering points with no data or insufficient data), but also discover potential power grid problems through detailed analysis of the three-phase voltage and two-phase voltage differences. This refined abnormal point marking strategy significantly improves the sensitivity and accuracy of the power grid monitoring system, and effectively reduces the risk of power grid failure caused by the failure to identify abnormal points in time. Especially in the processing of multi-phase voltage data, by introducing quantitative standards for variance and voltage differences, the present invention can maintain a high degree of judgment accuracy in a complex power grid environment.

[0046] An abnormal point identification system based on distribution network voltage data cleaning, characterized in that: it includes:

[0047] Data acquisition and preprocessing module: collects data on nodes and devices in the distribution network and performs preprocessing;

[0048] Dynamic simulation and modeling module: Establish a photovoltaic power generation model, a new energy power generation model, a special load model, a distributed energy storage model, and build a simulation environment for simulating the dynamic interaction between the new energy power generation system and the distribution network;

[0049] Fault identification and processing module: divides the distribution network into multiple areas, builds a hierarchical positioning model, identifies and locates different types of faults, and processes and repairs the faults through an intelligent dispatching system;

[0050] Access point optimization and control module: optimizes the access point location of the new energy power generation system in real time based on the fault handling results and the adjustment of the energy storage system.

[0051] The present invention adopts clear upper and lower limit rules, that is, the range of valid data is defined according to the upper and lower limits of 30% of the rated voltage. The formulation of this standard first ensures the rationality and consistency of the data; secondly, it effectively filters out abnormal data that deviates from the normal range to prevent these abnormal data from affecting the judgment of the power grid status. This strategy is particularly suitable for complex power grid environments, helping to improve the system's sensitivity to abnormal conditions, thereby more effectively monitoring the voltage status.

[0052] Beneficial effects of the present invention: The present invention can effectively filter out invalid data that does not meet the 30% range of the upper and lower limits of voltage, thereby ensuring the validity of the collected data. This method greatly improves the quality of voltage data and prevents the influence of noise data on the judgment of power grid status. Compared with traditional technologies, the present invention can automatically process massive voltage data and filter abnormal values ​​in time. This cleaning method significantly improves the accuracy and effectiveness of voltage monitoring.

[0053] The present invention can accurately estimate the rated voltage level of the metering point by calculating the average value of each phase voltage and combining it with the preset voltage level interval. When estimating, not only the single-phase voltage is considered, but also the three-phase voltage is comprehensively analyzed to ensure the consistency of voltages in different phases and avoid misjudgment caused by single-phase fluctuations. Compared with the traditional single-phase estimation method, the multi-phase joint estimation method of the present invention is more robust, especially suitable for multi-phase power grid environments, and can more accurately reflect the actual status of power grid operation.

[0054] The present invention can automatically identify abnormal data points and classify and mark them (such as insufficient data points, points with large three-phase voltage differences, etc.) by calculating quantitative standards such as the three-phase voltage variance and the phase-to-phase voltage difference. This refined abnormal point identification method can timely discover potential problems in the power grid and provide early warnings and processing. Compared with traditional manual identification or simple voltage difference judgment methods, the present invention can realize automatic processing in a large-scale data environment of the power grid, greatly improving the efficiency and accuracy of abnormal point identification. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the figures required for use in the description of the embodiments will be briefly introduced below. Obviously, the figures described below are only some embodiments of the present invention. For ordinary technicians in this field, other figures can be obtained based on these figures without paying creative work.

[0056] Figure 1 A flowchart of an abnormal point identification method based on distribution network voltage data cleaning is provided for the first embodiment of the present invention. DETAILED DESCRIPTION

[0057] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the specification drawings. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0058] Example 1

[0059] Reference Figure 1 , is an embodiment of the present invention, and provides an abnormal point identification method based on power distribution network voltage data cleaning, comprising:

[0060] S1: Obtain the three-phase voltage data of each metering point in the distribution network according to the preset sampling frequency, and mark the data within 30% of the upper and lower limits of the voltage as valid.

[0061] With the preset sampling frequency f s Collect the A, B, and C three-phase voltage data U of each metering point A (t), U B (t), U C (t), and store these data in the database in real time.

[0062] Sampling frequency f s Dynamic adjustment is performed according to the load changes and voltage fluctuations of the power grid. The specific adjustment strategies are as follows:

[0063]

[0064] Where f0 is the reference sampling frequency, V nominal is the rated voltage, L(t) is the current load, and L0 is the reference load value; β and γ are adjustable parameters used to adjust the different influence weights of voltage and load on the sampling frequency.

[0065] For each collected voltage data, if its value is within 30% of the upper and lower limits of the rated voltage of the current metering point, it is marked as valid data; otherwise, it is marked as invalid data.

[0066] The range of valid data is defined as:

[0067] V effective =[0.7×V nominal ,1.3×V nominal ]

[0068] All labeled data are stored in the database in time series.

[0069] It should be noted that the reference sampling frequency f0 is the initial setting of the system in a stable state to ensure the basic accuracy of the voltage data. The present invention realizes the dynamic adjustment of the sampling frequency, so that it can respond to the load changes and voltage fluctuations of the power grid in real time. This adjustment strategy enables the system to increase the sampling frequency when the load of the power grid increases or the voltage fluctuation intensifies, so as to capture more subtle voltage changes. In traditional technology, fixed sampling frequencies are usually used, which makes it difficult to quickly respond to dynamic changes in the power grid. Therefore, the present invention has significant advantages in adaptability and accuracy.

[0070] Character V nominal (rated voltage) and L(t) (current load) are the key parameters for dynamically adjusting the sampling frequency. Rated voltage V nominal(determines the reference point of the voltage data, while the current load L(t) reflects the real-time operation status of the power grid. By combining these two parameters, the sampling frequency can not only be adjusted in time when the voltage fluctuates abnormally, but also be adaptively adjusted according to the load changes, thereby ensuring the accuracy and integrity of the data. This dual adjustment mechanism is relatively rare in the prior art. Usually, the power grid monitoring system only considers the adjustment of a single factor. The present invention improves the representativeness and reliability of the collected data by considering the joint influence of voltage and load.

[0071] Furthermore, the present invention adopts clear upper and lower limit rules, that is, the range of valid data is defined according to the upper and lower limits of 30% of the rated voltage. The formulation of this standard first ensures the rationality and consistency of the data; secondly, it effectively filters out abnormal data that deviates from the normal range to prevent these abnormal data from affecting the judgment of the power grid status. This strategy is particularly suitable for complex power grid environments, helping to improve the system's sensitivity to abnormal conditions, thereby more effectively monitoring the voltage status.

[0072] S2: Calculate the average voltage of each phase at each metering point and infer the rated voltage level of each voltage item.

[0073] For each metering point, the three-phase voltage data U A (t),U B (t),U C (t) is averaged over time to obtain the average voltage of each phase The calculation formula is:

[0074]

[0075] Where N is the number of sampling points, t i is the time point of the i-th sampling moment; X is the voltage phase, X=A, B, C.

[0076] The rated voltage level of each voltage is estimated based on the average voltage of each phase. Estimate the rated voltage level V within the preset voltage level range nominal,X ,like If it falls within the upper and lower limits of the rated voltage level, it is inferred that the rated voltage of the current phase is the current level. The basis for the inference is:

[0077] If the average voltage of one of the phases falls within the range of [46.192V, 69.288V], it is inferred that the rated voltage level of the phase is 57.74V; if the average voltage of one of the phases falls within the range of [80V, 120V], it is inferred that the rated voltage level of the phase is 100V; if the average voltage of one of the phases falls within the range of [176V, 264V], it is inferred that the rated voltage level of the phase is 220V; if the average voltage of one of the phases falls within the range of [304V, 456V], it is inferred that the rated voltage level of the phase is 380V; if the average voltage of one of the phases is not within any range, the voltage data of the phase is marked as invalid.

[0078] It should be noted that the settings of different voltage levels (such as 57.74V, 100V, 220V, 380V) are determined according to the actual operation standards of the power grid and the design requirements of electrical equipment. These levels provide a reliable basis for inferring the rated voltage of each phase. By comparing the average voltage with these intervals, it is possible to reasonably infer the level to which the current voltage belongs.

[0079] The voltage level estimation step in the present invention not only provides clear technical guidance, but also has good adaptability. Especially in a complex power grid environment, by using a preset voltage interval, it can be ensured that the voltage level can still be accurately judged even in the case of large voltage fluctuations. In addition, for those voltage data that are not within any preset interval, the present invention explicitly marks them as invalid, thereby avoiding the impact of abnormal data on the entire system. In practical applications, this processing method can significantly improve the accuracy and effectiveness of data analysis, avoid misjudgment caused by abnormal data interference, and ultimately improve the stability of power grid monitoring and control.

[0080] S3: The rated voltage level of the metering point is estimated based on the rated voltage level of each phase, and the data within 30% of the upper and lower limits of the rated voltage of the metering point is marked as valid.

[0081] If the three phases of the metering point have rated voltages, the three-phase rated voltages must be the same, otherwise the rated voltage level of the metering point is set to unknown; if the metering point has two-phase rated voltages, the two-phase rated voltages must be the same, otherwise the rated voltage level of the metering point is set to unknown; if the metering point has only one-phase rated voltage, the rated voltage level of the metering point is set to the rated voltage level of the phase; if the metering point has no valid rated voltage, the rated voltage level of the metering point is set to unknown.

[0082] Calculate the effective voltage data range of each metering point based on the estimated rated voltage of each metering point It is defined as:

[0083]

[0084] in, is the estimated rated voltage at the metering point.

[0085] For each voltage data point, if its value falls within the valid voltage data range If the value is within the range, the data is marked as valid data; otherwise, it is marked as invalid data.

[0086] All data marked as valid are stored in the database in time series.

[0087] It should be noted that the voltage level estimation method in the present invention has significant advantages in dealing with situations where voltage fluctuations are large or multi-phase inconsistencies occur. By checking the consistency of the three-phase voltage levels, the system can quickly and accurately identify abnormal situations and mark the voltage data that does not meet the conditions as invalid. This method effectively reduces misjudgments caused by inconsistent phase voltages, thereby improving the accuracy of data in power grid monitoring. In addition, by setting the effective voltage data range, the present invention can better screen out voltage data within a reasonable fluctuation range, so that the system remains efficient and accurate when dealing with complex power grid environments.

[0088] Furthermore, after the rated voltage level of each phase is inferred, the voltage data is compared with the newly determined upper and lower limits of 30% based on the inferred rated voltage level. This process is to verify the rationality of the voltage data more accurately, especially for inconsistencies between different phases. Further improve the accuracy of the data and ensure that only valid data within the new range is retained when the rated voltages of different phases are inconsistent. This helps to discover potential anomalies that may not have been identified in the first round of screening, especially when there are differences or uncertainties between phases.

[0089] S4: Calculate the voltage qualification rate of each phase according to the effective data, and calculate the voltage qualification rate of the metering point.

[0090] According to the effective voltage data, calculate the voltage qualification rate P of each phase X 合格,X , which is defined as:

[0091]

[0092] Among them, N 合格,X In the valid data range The number of qualified voltage data in N 总,X It is the number of all valid voltage data of this phase.

[0093] For a phase that is set to an invalid voltage, the voltage qualification rate is not calculated and the data of the phase is directly excluded.

[0094] Calculating the voltage qualification rate of the metering point includes summing up the voltage qualification rates of all effective phases and calculating the comprehensive voltage qualification rate P of the metering point. 合格，计量点 , which is defined as:

[0095]

[0096] Among them, N 有效相 is the number of effective phases involved in the calculation, ∑ 有效相 P 合格,X It is the sum of the qualified rates of the phase voltages involved in the calculation.

[0097] It should be noted that the qualified rate calculation method in the present invention provides a more detailed voltage quality assessment method by calculating the single-phase and overall voltage data separately. Especially when dealing with the situation where invalid voltage phases may exist, the present invention can flexibly adjust the calculation strategy to avoid the overall misjudgment caused by invalid data in the traditional method. In addition, by calculating the average qualified rate of the effective phase, the real situation of the voltage quality can be more accurately reflected. In practical applications, this method can effectively improve the ability of the power grid monitoring system to analyze voltage data, thereby providing more reliable support for power grid operation and maintenance.

[0098] S5: Find the measurement points with abnormal data and mark them.

[0099] If the rated voltage level of the three-phase voltage of the metering point cannot be determined, the metering point is an abnormal metering point and is marked as a metering point with unknown rated voltage; if there is no voltage data at the metering point within the statistical period, the metering point is an abnormal metering point and is marked as a metering point with no data; if the number of valid voltage data points at the metering point within the statistical period accounts for less than 50% of the total number of data points, the metering point is an abnormal metering point and is marked as a metering point with insufficient data; for the three-phase voltage of the metering point, if its three-phase variance is greater than the set threshold m (generally 16), the metering point is an abnormal metering point and is marked as a metering point with large three-phase voltage difference; if there are only two-phase voltages, the voltage difference ΔU between the two phases is calculated. If ΔU exceeds the set threshold n (it is recommended to be set as the maximum value of the two-phase voltage difference in the historical data plus an appropriate margin, and the specific value can be adjusted according to the equipment specifications and grid operation characteristics), the metering point is an abnormal metering point and is marked as a metering point with large two-phase voltage difference.

[0100] It should be noted that the abnormal metering point identification method can not only identify abnormal points that are easily overlooked in traditional methods (such as metering points with no data or insufficient data), but also discover potential power grid problems through detailed analysis of the three-phase voltage and two-phase voltage differences. This refined abnormal point marking strategy significantly improves the sensitivity and accuracy of the power grid monitoring system, and effectively reduces the risk of power grid failure caused by the failure to identify abnormal points in time. Especially in the processing of multi-phase voltage data, by introducing quantitative standards for variance and voltage differences, the present invention can maintain a high degree of judgment accuracy in a complex power grid environment.

[0101] Embodiment 2 is the second embodiment of the present invention, which is different from the previous embodiment in that:

[0102] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the current technical solution, can be embodied in the form of a software product. The current computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0103] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0104] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0105] Embodiment 3 is an embodiment of the present invention, which provides an abnormal point identification system based on distribution network voltage data cleaning, specifically including: a data acquisition module, a voltage average calculation and rated voltage estimation module, a metering point rated voltage estimation module and a voltage qualification rate calculation module.

[0106] Data acquisition module: obtains the three-phase voltage data of each metering point in the distribution network according to the preset sampling frequency, and marks the data within 30% of the upper and lower limits of the voltage as valid.

[0107] Voltage average calculation and rated voltage estimation module: calculates the average voltage of each phase at each metering point and estimates the rated voltage level of each voltage item.

[0108] Metering point rated voltage estimation module: estimates the rated voltage level of the metering point based on the rated voltage level of each phase, and marks the data within 30% of the upper and lower limits of the metering point rated voltage as valid.

[0109] Voltage qualification rate calculation module: Calculate the voltage qualification rate of each phase according to the valid data, and calculate the voltage qualification rate of the metering point.

[0110] Anomaly identification and marking module: finds measurement points with data anomalies and marks them.

[0111] Example 4 is an embodiment of the present invention, which provides an abnormal point identification method based on distribution network voltage data cleaning. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation / comparative experiments.

[0112] In order to verify the effectiveness and innovation of the invention, a systematic experiment was conducted, and six distribution network metering points (named A, B, C, D, E, and F) were selected for the experiment. The voltage data of each metering point is collected through a preset sampling frequency, and the sampling frequency is dynamically adjusted according to the load changes and voltage fluctuations of the power grid to ensure the real-time and accuracy of the data. The voltage data is collected as phase A, phase B, and phase C voltages, and stored in the database respectively.

[0113] Each collected voltage data is marked according to whether it is within the upper and lower limits of the rated voltage by 30%. All data marked as valid will continue to be used for subsequent analysis, while data marked as invalid will be excluded.

[0114] The three-phase voltage data of each metering point is averaged over time to calculate the average voltage of each phase, and its rated voltage level is estimated based on the preset voltage level range. If the average voltage of a phase falls within the specified level range, its rated voltage level is determined, otherwise the phase data is marked as invalid.

[0115] Based on the valid data, the voltage qualification rate of each phase is calculated, and the comprehensive voltage qualification rate of the metering point is calculated. If the data of a phase is marked as invalid, the phase will not be included in the qualification rate calculation.

[0116] According to the abnormal data points identified in the previous steps, including no data, insufficient data, excessive three-phase voltage difference, excessive two-phase voltage difference, etc., the abnormal metering points are marked one by one and stored in the database for further analysis. The specific data is shown in Table 1.

[0117] Table 1 Implementation data table

[0118]

[0119]

[0120] It can be seen from the data in the above table that the application effect of the invention in the voltage monitoring of the distribution network has obvious advantages. First, based on the dynamically adjusted sampling frequency and the reasonable voltage data marking strategy, the voltage data of each metering point is accurately collected and effectively marked. This can be proved from the voltage qualification rate data of each metering point: for example, the comprehensive voltage qualification rate of metering point A reaches 96.87%, indicating that in a complex power grid environment, this method can effectively ensure the accuracy and stability of voltage monitoring.

[0121] Secondly, by inferring and calculating the rated voltage level and pass rate of each phase, the system can comprehensively evaluate the voltage quality of different metering points. The table shows that although the three-phase voltage data of each metering point are different, the overall voltage situation of each metering point can be objectively reflected through reasonable pass rate calculation. In particular, when the difference between the two-phase voltages at some metering points is too large (such as the difference between phase B and phase C at metering point B), the system can quickly identify and mark the abnormality, thereby avoiding potential grid operation risks.

[0122] Finally, the recognition and marking of data anomalies further improves the monitoring capabilities of the system. For example, if in some cases the data at a certain metering point is insufficient or the three-phase voltage variance is too large, the system can promptly mark the abnormal situation and store this information in a database for further processing by grid maintenance personnel. Compared with the prior art, the present invention not only achieves a technological breakthrough in data collection and processing, but also significantly improves the accuracy and reliability of grid monitoring through innovative data marking and analysis methods.

[0123] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for identifying abnormal points based on distribution network voltage data cleaning, characterized in that: include: Acquire the three-phase voltage data of each metering point in the distribution network according to the preset sampling frequency, and mark the data within 30% of the upper and lower limits of the voltage as valid; Calculate the average voltage of each phase at each metering point and infer the rated voltage level of each voltage; The rated voltage level of the metering point is estimated based on the rated voltage level of each phase, and the data within 30% of the upper and lower limits of the rated voltage of the metering point is marked as valid; Calculate the voltage qualification rate of each phase according to the effective data, and calculate the voltage qualification rate of the metering point; Find the measurement points where the data is abnormal and mark them.

2. The abnormal point identification method based on distribution network voltage data cleaning according to claim 1 is characterized in that: The method of obtaining the three-phase voltage data of each metering point in the distribution network according to the preset sampling frequency includes: s Collect the A, B, and C three-phase voltage data U of each metering point A (t), U B (t), U C (t), and store these data in the database in real time; For each collected voltage data, if its value is within 30% of the upper and lower limits of the rated voltage of the current metering point, it will be marked as valid data; otherwise, it will be marked as invalid data; All labeled data are stored in the database in time series.

3. The abnormal point identification method based on distribution network voltage data cleaning according to claim 2 is characterized in that: The calculation of the average voltage of each phase at each metering point includes: A (t),U B (t),U C (t) is averaged over time to obtain the average voltage of each phase The estimated rated voltage level of each voltage includes: Estimate the rated voltage level V within the preset voltage level range nominal,X ,like If it falls within the upper and lower limits of the rated voltage level, it is inferred that the rated voltage of the current phase is the current level. The basis for the inference is: If the average voltage of one phase If it falls within the range of [46.192V, 69.288V], it is inferred that the rated voltage level of this phase is 57.74V; If the average voltage of one phase If it falls within the range of [80V, 120V], it is inferred that the rated voltage level of this phase is 100V; If the average voltage of one phase If it falls within the range of [176V, 264V], it is inferred that the rated voltage level of this phase is 220V; If the average voltage of one phase If it falls within the range of [304V, 456V], it is inferred that the rated voltage level of this phase is 380V; If the average voltage of one phase If it is not within any range, the voltage data of that phase is marked as invalid.

4. The abnormal point identification method based on distribution network voltage data cleaning according to claim 3 is characterized in that: The inferring the rated voltage level of the metering point according to the rated voltage level of each phase includes: if the three phases of the metering point have rated voltages, the three-phase rated voltages must be the same, otherwise the rated voltage level of the metering point is set to be unknown; If the metering point has two-phase rated voltages, the two-phase rated voltages must be the same, otherwise the rated voltage level of the metering point is set to unknown; If the metering point has only one phase rated voltage, the rated voltage level of the metering point is set to the rated voltage level of the phase; If there is no valid rated voltage at the metering point, set the rated voltage level of the metering point to unknown.

5. The abnormal point identification method based on distribution network voltage data cleaning according to claim 4 is characterized in that: The marking of data within 30% of the upper and lower limits of the rated voltage of the metering point as valid includes calculating the valid voltage data range of the metering point based on the estimated rated voltage of each metering point. It is defined as: in, is the estimated rated voltage at the metering point; For each voltage data point, if its value falls within the valid voltage data range If the data is within the range, the data is marked as valid data; otherwise, it is marked as invalid data; All data marked as valid are stored in the database in time series.

6. The abnormal point identification method based on distribution network voltage data cleaning according to claim 5 is characterized in that: The method of calculating the voltage qualification rate of each phase according to the effective data includes calculating the voltage qualification rate P of each phase X according to the effective voltage data. 合格,X , which is defined as: Among them, N 合格,X In the valid data range The number of qualified voltage data in N 总,X is the number of all valid voltage data of this phase; For the phases set to invalid voltage, the voltage qualification rate is not calculated and the data of the phases is directly excluded; The calculation of the voltage qualification rate of the metering point includes summing up the voltage qualification rates of all effective phases and calculating the comprehensive voltage qualification rate P of the metering point. 合格，计量点 , which is defined as: Among them, N 有效相 is the number of effective phases involved in the calculation, ∑ 有效相 P 合格,X It is the sum of the qualified rates of the phase voltages involved in the calculation.

7. The abnormal point identification method based on distribution network voltage data cleaning according to claim 6 is characterized in that: The method of finding the metering point with abnormal data includes: if the rated voltage level of the three-phase voltage of the metering point cannot be determined, then the metering point is an abnormal metering point and is marked as a metering point with unknown rated voltage; If a metering point does not have any voltage data within the statistical period, the metering point is an abnormal metering point and is marked as a metering point with no data; If the proportion of valid voltage data points of a metering point in the statistical period to the total data points is less than 50%, the metering point is an abnormal metering point and is marked as a metering point with insufficient data; For the three-phase voltage of a metering point, if its three-phase variance is greater than the set threshold m, the metering point is an abnormal metering point and is marked as a metering point with large three-phase voltage variance; If there are only two-phase voltages, the voltage difference ΔU between the two phases is calculated. If ΔU exceeds the set threshold n, the metering point is an abnormal metering point and is marked as a metering point with a large difference in two-phase voltages.

8. An abnormal point identification system based on distribution network voltage data cleaning using the method according to any one of claims 1 to 7, characterized in that: Data acquisition module: obtains three-phase voltage data of each metering point in the distribution network according to the preset sampling frequency, and marks the data within 30% of the upper and lower limits of the voltage as valid; Voltage average calculation and rated voltage estimation module: calculates the average voltage of each phase at each metering point and estimates the rated voltage level of each voltage item; Metering point rated voltage estimation module: estimates the rated voltage level of the metering point based on the rated voltage level of each phase, and marks the data within 30% of the upper and lower limits of the metering point rated voltage as valid; Voltage qualification rate calculation module: calculates the voltage qualification rate of each phase according to the valid data, and calculates the voltage qualification rate of the metering point; Anomaly identification and marking module: finds measurement points with data anomalies and marks them.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images