A phase identification and topology identification method for substations based on intelligent lock positioning of meter boxes

Abstract

The present invention relates to the technical field of phase topology identification of substations, and specifically discloses a substation phase identification and topology identification method based on the positioning of intelligent locks of meter boxes. The method first collects the location information of the intelligent locks of each meter box and the location information of each substation, obtains the corresponding substation of the intelligent locks of each meter box, and monitors the substation phase identification process of the intelligent locks of each meter box to determine whether the substation phase identification of the intelligent locks of each meter box is accurate. Therefore, for each topological substation whose substation phase identification accuracy is greater than the substation phase identification accuracy threshold, multi-dimensional intelligent lock behavior data is identified and analyzed, and a behavior topology model of each topological substation with accurate identification is constructed. Finally, the lock behavior topology process of each topological substation is fed back to obtain a lock substation with more accurate phase and topology, thereby accurately measuring the electric energy of the meter box and effectively managing the equipment, so as to improve the operating stability of the power system.

Description

Technical Field

[0001] The present invention relates to the technical field of phase topology identification of a transformer substation, and in particular to a phase identification and topology identification method of a transformer substation based on positioning of an intelligent lock of a meter box. Background Art

[0002] With the development of smart grids, the management and maintenance of power equipment has become increasingly important. The meter box is an important component of the power system. However, traditional meter boxes are protected by mechanical locks, which poses certain safety hazards and inconvenient management. Therefore, a more intelligent and real-time solution is needed to protect the meter box. Through the comprehensive operation of phase identification and topology identification, accurate metering of electric energy can be performed to facilitate effective equipment management and maintenance.

[0003] For example, the invention patent with announcement number CN118017506B announces a low-voltage substation topology identification method and system, which involves the field of electric energy metering technology, including obtaining phase-separated voltage data collected by HPLC, based on Kirchhoff's law, combining the similarity of the voltage change trend of each phase electric energy meter under the substation with the relationship between the phase line topology, generating a tree structure to reflect the hierarchical relationship of the power supply topology; structurally processing the user's power consumption address, and analyzing and counting the address; distinguishing the substation power supply topology type according to the address distribution characteristics, obtaining the branch segmentation threshold and segmenting, and merging the branch segmentation results of each phase to obtain the main topology structure of the substation. The low-voltage substation topology identification method provided can accurately determine the main topology of the substation by identifying the main topology structure of the substation, improve the management and control efficiency of the power network, and improve the accuracy of topology identification and the robustness of the algorithm, and achieve better results in terms of control efficiency, power consumption address and identification accuracy.

[0004] For example, the invention patent with the announcement number CN112886582B announces a method for identifying the phase of the substation based on the voltage correlation of the load rate and topology. It includes the following steps: S1: Obtain the time series data of the assessment total meter and each user meter at the transformer under the substation and the user branch data of the substation; S2: Establish a phase identification model based on user branches; S3: Use the above model to calculate the voltage correlation coefficient between meters within the user branch, between reference meters between branches, and between branch reference meters and total meters, diagnose the phase consistency between branches, and complete the phase identification of users of single-phase meters in the substation; S4: Output phase. Based on the known substation topology, the user meter voltage is mainly determined by the total meter voltage when the substation load rate is low. A suitable point selection strategy is formulated for the total meter and user meter voltage data, and the phase convergence principle is proposed. The Pearson correlation coefficient is used to perform voltage correlation analysis to achieve substation phase identification.

[0005] In combination with the above technical solutions, it is found that in the technical solutions for phase identification or topology identification, for a single identification process, the dimension of the identification analysis is relatively one-sided, and fewer influencing factors are considered, which cannot improve the identification accuracy and will lead to unstable operation of the power system. Summary of the invention

[0006] In view of the deficiencies in the prior art, the present invention provides a phase identification and topology identification method for a substation based on the positioning of an intelligent lock of a meter box, which can effectively solve the problems involved in the above-mentioned background technology.

[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions: a substation phase identification and topology identification method based on the positioning of meter box smart locks, including: substation phase identification evaluation: collecting the location information of the smart locks belonging to each meter box and the location information of each substation, analyzing to obtain the corresponding substation of the smart locks belonging to each meter box, and monitoring the substation phase identification process of the smart locks belonging to each meter box, evaluating the substation phase identification accuracy of the smart locks belonging to each meter box, and comparing it with the preset substation phase identification accuracy threshold; substation topology process feedback: marking the corresponding substations of the smart locks belonging to each meter box whose substation phase identification accuracy is greater than the preset substation phase identification accuracy threshold as each topology substation, acquiring and preprocessing the smart lock behavior data of each topology substation, thereby constructing a behavior topology model of each topology substation, and finally feedback the lock behavior topology process of each topology substation.

[0008] As a further method, the analysis obtains the corresponding area of ​​the smart lock to each meter box, and the specific analysis process is:

[0009] Perform difference processing on the longitude coordinates, latitude coordinates and altitude of the smart locks belonging to each meter box and the longitude coordinates, latitude coordinates and altitude of each substation respectively, and obtain the longitude difference, latitude difference and altitude difference between the smart locks belonging to each meter box and each substation, and then record the accumulated values ​​as the position distance between the smart locks belonging to each meter box and each substation; arrange the position distances between the smart locks belonging to a meter box and each substation in order from small to large, and extract the substation corresponding to the first-ranked position distance, which is the corresponding substation of the smart lock belonging to the meter box; thus, the corresponding substation of the smart locks belonging to each meter box is obtained according to the above operations.

[0010] As a further method, the phase identification accuracy threshold of the preset station area is compared, and the specific comparison process is:

[0011] If the area phase recognition accuracy of the smart lock belonging to a certain meter box is greater than the preset area phase recognition accuracy threshold, the corresponding area of ​​the smart lock belonging to the meter box is located, and the corresponding areas of the smart locks belonging to each meter box are obtained by statistics; if the area phase recognition accuracy of the smart lock belonging to a certain meter box is less than or equal to the preset area phase recognition accuracy threshold, an early warning is issued for the location of the smart lock belonging to the meter box, and the smart locks belonging to each meter box whose area phase recognition accuracy is less than or equal to the preset area phase recognition accuracy threshold are recorded as abnormal smart locks belonging to each meter box, and an early warning is issued for the location of the abnormal smart locks belonging to each meter box.

[0012] As a further method, the behavioral topological model of each topological area is constructed, and the specific process is as follows:

[0013] The topological structure diagram of each topological substation is obtained, and the lock behavior utilization data of each topological substation is integrated as the input parameters of the power system analysis tool. Through the model building function of the power system analysis tool, the output model is recorded as the behavioral topological model of each topological substation.

[0014] As a further method, the lock behavior topology process of each topology area is fed back, and the specific feedback process is:

[0015] The real-time behavioral topological model of each topological area is collected, and the model is overlapped and compared with the behavioral topological model of each topological area to obtain the overlapping volume of the behavioral topological model of each topological area, and compared with the preset overlapping boundary volume. If the overlapping volume of the behavioral topological model of a certain topological area is smaller than the preset overlapping boundary volume, the lock behavioral topological process of the topological area is fed back. Thus, the topological areas corresponding to the overlapping volumes smaller than the preset overlapping boundary volume are recorded as abnormal areas, and the lock behavioral topological process of each abnormal area is fed back.

[0016] Compared with the prior art, the embodiments of the present invention have at least the following advantages or beneficial effects:

[0017] (1) The present invention provides a substation phase identification and topology identification method based on the positioning of the meter box smart lock. First, the location information of the smart locks belonging to each meter box and the location information of each substation are collected. The corresponding substation of the smart locks belonging to each meter box can be obtained by comparing the location information, and the substation phase identification process of the smart locks belonging to each meter box is monitored to determine whether the substation phase identification of the smart locks belonging to each meter box is accurate. Therefore, for each topological substation whose substation phase identification accuracy is greater than a preset substation phase identification accuracy threshold, multi-dimensional smart lock behavior data is identified and analyzed, and a behavior topology model of each topological substation with accurate identification is constructed. Finally, the lock behavior topology process of each topological substation is fed back to obtain a lock substation with more accurate phase and topology, thereby accurately measuring the electric energy of the meter box and effectively managing and maintaining the equipment, so as to improve the operating stability of the power system.

[0018] (2) The present invention collects the location information of the smart locks belonging to each meter box and the location information of each substation, and analyzes the corresponding substation of the smart locks belonging to each meter box, which can provide a reference for location information for the subsequent monitoring of the substation phase identification process of the smart locks belonging to each meter box, evaluate the substation phase identification accuracy of the smart locks belonging to each meter box, and compare it with the preset substation phase identification accuracy threshold, so as to obtain the corresponding substation of the smart locks belonging to each meter box that has been accurately identified, so as to improve the accuracy of phase identification.

[0019] (3) The present invention obtains and pre-processes the intelligent lock behavior data of each topological substation, thereby constructing a behavioral topological model of each topological substation. Through this model, the topological structure of each substation can be accurately understood, and the lock behavior topological process of each topological substation can be fed back, so as to timely discover and deal with potential problems of improper management and maintenance of each substation in the power grid, reduce fault losses and power outage time, and ultimately improve the safety of the power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The present invention is further described using the accompanying drawings, but the embodiments in the accompanying drawings do not constitute any limitation to the present invention. A person skilled in the art can obtain other drawings based on the following drawings without creative work.

[0021] Figure 1 It is a schematic diagram of the method flow of the present invention. DETAILED DESCRIPTION

[0022] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0023] Reference Figure 1 As shown, the present invention provides a substation phase identification and topology identification method based on the positioning of the smart locks of the meter boxes, including: substation phase identification evaluation: collecting the location information of the smart locks belonging to each meter box and the location information of each substation, analyzing to obtain the corresponding substation of the smart locks belonging to each meter box, and monitoring the substation phase identification process of the smart locks belonging to each meter box, evaluating the substation phase identification accuracy of the smart locks belonging to each meter box, and comparing it with the preset substation phase identification accuracy threshold.

[0024] In a specific embodiment, the present invention collects the location information of the smart locks belonging to each meter box and the location information of each substation, and analyzes to obtain the corresponding substation of the smart locks belonging to each meter box, which can provide a reference for location information for the subsequent monitoring of the substation phase identification process of the smart locks belonging to each meter box, evaluate the substation phase identification accuracy of the smart locks belonging to each meter box, and compare it with the preset substation phase identification accuracy threshold, so as to obtain the corresponding substation of the smart locks belonging to each meter box that is accurately identified, so as to improve the accuracy of phase identification.

[0025] Specifically, the comparison is performed with the preset phase recognition accuracy threshold of the station area, and the specific comparison process is:

[0026] If the area phase recognition accuracy of the smart lock belonging to a meter box is greater than the area phase recognition accuracy threshold preset by the identification feature data cloud platform, it means that the smart lock belonging to the meter box has a high accuracy in identifying the area phase and can meet the accuracy threshold requirements preset by the identification feature data cloud platform; the area phase recognition accuracy threshold refers to a standard pre-set in the identification feature data cloud platform, which is used to determine whether the area phase recognition accuracy of the smart lock meets the expected requirements.

[0027] It should be added that the above-mentioned identification feature data cloud platform is a cloud computing platform for storing and managing identification feature data generated by devices such as smart locks. In this example, the stored data includes the influencing factor of the preset phase deviation unit value, the preset station phase identification accuracy threshold, the preset overlap boundary volume and the preset data cleaning threshold.

[0028] The corresponding area of ​​the smart lock to which the meter box belongs is located. The area to which the meter box belongs can be determined through a geographic information system (GIS), thereby obtaining the corresponding area of ​​the smart lock to which each meter box belongs.

[0029] If the phase recognition accuracy of the smart lock of a meter box is less than or equal to the phase recognition accuracy threshold of the area preset by the identification feature data cloud platform, it means that the smart lock may have certain accuracy problems when identifying the phase of the area, and cannot fully meet the threshold requirements of the identification feature data cloud platform. Then, an early warning is issued for the location of the smart lock of the meter box. The smart locks of each meter box corresponding to the phase recognition accuracy of the area less than or equal to the preset phase recognition accuracy threshold of the area are counted and recorded as abnormal smart locks of each meter box. The location of the abnormal smart lock of each meter box is warned. The specific early warning process is as follows:

[0030] The identification feature data cloud platform notifies the data processor of the abnormal smart locks belonging to each meter box, and the data processor re-collects and analyzes the location information of the abnormal smart locks belonging to each meter box, obtains the secondary corresponding substations of the abnormal smart locks belonging to each meter box, and then monitors the substation phase identification process. If the substation phase identification accuracy of the abnormal smart lock belonging to a meter box is greater than the preset substation phase identification accuracy threshold, the substation corresponding to the abnormal smart lock belonging to the meter box can be marked as each topological substation. If the substation phase identification accuracy of the abnormal smart lock belonging to a meter box is equal to or less than the preset substation phase identification accuracy threshold, the abnormal smart lock belonging to the meter box needs to be notified to the equipment maintenance personnel for quality inspection. The subsequent quality inspection is not described in detail in this example, and the above-mentioned early warning process is completed here.

[0031] Furthermore, the phase recognition accuracy of each meter box's smart lock is evaluated, and the specific evaluation process is as follows:

[0032] By monitoring the phase identification process of the smart locks of each meter box, the phase data of the smart locks of each meter box and the phase data of the corresponding area of ​​the smart locks of each meter box can be obtained. The phase identification accuracy of the smart locks of each meter box is obtained by data processing. The data processing expression is as follows:

[0033]

[0034] The phase data of the smart locks belonging to each meter box include the average voltage corresponding to the first phase of the smart locks belonging to each meter box during the identification period, the average voltage corresponding to the second phase of the smart locks belonging to each meter box during the identification period, and the average voltage corresponding to the third phase of the smart locks belonging to each meter box during the identification period; the phase data of the corresponding substations of the smart locks belonging to each meter box include the average voltage corresponding to the first phase of the corresponding substations of the smart locks belonging to each meter box during the identification period, the average voltage corresponding to the second phase of the corresponding substations of the smart locks belonging to each meter box during the identification period, and the average voltage corresponding to the third phase of the corresponding substations of the smart locks belonging to each meter box during the identification period.

[0035] It should be noted that the above-mentioned identification cycle is the time period corresponding to the phase identification process of the substation, which can be obtained directly from the smart lock. The specific duration will be determined according to the number of smart locks and the phase identification complexity of the substation.

[0036] In the formula, is the phase recognition accuracy of the smart lock belonging to the u-th meter box. In this example, in order to accurately describe the phase recognition status of the smart lock belonging to the meter box, precise data processing is required. Through this quantitative data processing, the phase recognition accuracy of the smart lock belonging to the meter box can be obtained. Therefore, this example obtains the phase corresponding voltage of the smart lock belonging to the meter box and the phase corresponding voltage of the smart lock belonging to the meter box, and performs difference processing, thereby obtaining whether their phases are consistent through numerical analysis; through the precise quantitative evaluation of the recognition accuracy of the above expression, it is determined whether the recognition phase of the lock and the corresponding substation is the same. If different, subsequent processing can be performed. If the same, based on this precise phase recognition, the metering status of the meter box electric energy can be improved subsequently, and ultimately the operation stability of the overall power system can be improved.

[0037] In order to identify the influencing factors of the phase deviation unit value preset by the feature data cloud platform, a mapping set of phase deviation and its corresponding influencing factors is constructed according to the relationship between the historical phase deviation and the phase recognition accuracy of the station area, and the real-time phase deviation is input into the mapping set to obtain the influencing factors of the phase deviation unit value. At the same time, its value range in this example is [0,1].

[0038] thus, is the voltage average corresponding to the first phase of the smart lock belonging to the u-th meter box within the identification period. This example describes the first phase as phase A, and its corresponding voltage average is obtained by obtaining multiple real-time voltage values ​​of the smart lock within the identification period through a voltage sensor and then averaging them (the following voltage average of the smart lock within the identification period is obtained in the same way as this time, so the following method of obtaining the corresponding voltage average of the smart lock is not explained); this voltage average can be used to represent the voltage characteristics of phase A. Of course, the above-mentioned phase identification method may be affected by some factors, such as instantaneous fluctuations, noise, etc., so this example performs mean calculation in order to reduce the impact of these factors and improve the accuracy of the identification results; the average value calculation can average out the fluctuations and noise of multiple measured values ​​within a period of time to obtain a more stable and reliable value, which helps us better understand the voltage characteristics of phase A and more accurately evaluate the phase identification accuracy of the smart lock belonging to the meter box.

[0039] It is the voltage average corresponding to the first phase of the corresponding area of ​​the smart lock belonging to the u-th meter box during the identification period. It is obtained by obtaining multiple real-time voltage values ​​of the area during the identification period through the voltage sensor and then averaging them (the voltage average of the following area during the identification period is obtained in the same way as this time, so the method of obtaining the corresponding voltage average of the following area will not be explained).

[0040] is the voltage average value corresponding to the second phase of the smart lock belonging to the u-th meter box within the identification period. In this example, the second phase is described as phase B, which means the voltage average value corresponding to B obtained by the smart lock identification.

[0041] It is the voltage average value corresponding to the second phase of the corresponding area of ​​the smart lock belonging to the u-th meter box within the identification period, which represents the voltage average value corresponding to B of the area.

[0042] is the voltage average value corresponding to the third phase of the smart lock belonging to the u-th meter box within the identification period. In this example, the third phase is described as phase C, which means the voltage average value corresponding to C obtained by the smart lock identification.

[0043] It is the voltage mean value corresponding to the third phase of the corresponding area of ​​the smart lock belonging to the u-th meter box within the identification period, which represents the voltage mean value corresponding to C of the area.

[0044] It should be noted that this example only analyzes phase A, phase B and phase C because these three phases are the basic phases in the power system. The relationship and role between them are the basis of power system research. In most cases, the load, power generation and transmission processes in the power system involve these three phases, and in this example, it is necessary to maintain and ensure the safe use of the meter box. Therefore, analyzing the characteristics of these three phases is of great significance for understanding the operation and control of the meter box. In addition, there is a certain degree of mutual independence between phase A, phase B and phase C, and the voltages between them are independent of each other and can be measured and calculated separately. Therefore, when analyzing the phase recognition accuracy of the smart lock to which the meter box belongs, these three phases can be focused on. Of course, in actual applications, the power system contains other phases, such as the neutral line (N line) and the protective grounding line (PE line). If this phase is applied, its operation method is the same as the operation method in the above-mentioned phase recognition accuracy expression of the substation, but since this example requires universality and needs to be applied to multiple scenarios, this example only focuses on phase A, phase B and phase C, so other phases are not considered in this example.

[0045] e is a natural constant.

[0046] Specifically, the location information of the smart locks to which each meter box belongs includes the longitude coordinates, latitude coordinates and altitude of the smart locks to which each meter box belongs.

[0047] Among them, the longitude coordinates and latitude coordinates can be obtained through GPS (Global Positioning System) devices. Smart locks are usually equipped with GPS modules for real-time recording and transmission of the location information of the locks. By reading these location information through GPS devices, the longitude coordinates and latitude coordinates of the smart locks can be obtained; the altitude can also usually be obtained through GPS devices. The GPS module can measure the altitude of the location of the lock. This information is usually recorded together with other location information (such as longitude and latitude).

[0048] Furthermore, the location information of each station area specifically includes the longitude coordinates, latitude coordinates and altitude of each station area.

[0049] It should be clarified that the above-mentioned longitude and latitude coordinates can be obtained through GPS (Global Positioning System) equipment. In the power system, there will be GPS receivers or other positioning devices installed in each substation to record and transmit the location information of the substation in real time. By reading this location information, the longitude and latitude coordinates of each substation can be obtained; and the altitude can also be obtained through GPS equipment. GPS receivers or other positioning devices can measure the altitude of the location of the substation. This information is usually recorded together with other location information (such as longitude and latitude).

[0050] Specifically, the analysis obtains the corresponding area of ​​the smart locks to which each meter box belongs, and the specific analysis process is as follows:

[0051] The longitude coordinates, latitude coordinates and altitude of the smart locks belonging to each meter box are subtracted from the longitude coordinates, latitude coordinates and altitude of each substation to obtain the longitude difference, latitude difference and altitude difference between the smart locks belonging to each meter box and each substation, and the accumulated values ​​are recorded as the location distance between the smart locks belonging to each meter box and each substation.

[0052] Arrange the location distances between the smart lock belonging to a meter box and each area in order from small to large, and extract the area corresponding to the first-ranked location distance, which is the corresponding area of ​​the smart lock belonging to the meter box; thus, according to the above operation, the areas corresponding to the first-ranked location distance are extracted, which are the corresponding areas of the smart locks belonging to each meter box.

[0053] It needs to be explained that the above-mentioned are arranged in order from small to large, and the substation corresponding to the first-ranked position distance is extracted to obtain the corresponding substation of the smart locks belonging to each meter box. The purpose is to select the lock closest to the substation as the matching object, which can improve the matching accuracy between the smart locks belonging to the meter box and the substation, thereby improving the operating efficiency of the entire power system, and more conveniently manage and maintain the smart locks belonging to the meter box, thereby improving its work efficiency.

[0054] Substation topology process feedback: The corresponding substations of the smart locks belonging to each meter box whose substation phase recognition accuracy is greater than the preset substation phase recognition accuracy threshold are marked as each topology substation, and the smart lock behavior data of each topology substation is obtained and preprocessed, thereby constructing a behavioral topology model of each topology substation, and finally feedback is given to the lock behavior topology process of each topology substation.

[0055] In a specific embodiment, the present invention obtains and pre-processes the intelligent lock behavior data of each topological substation, thereby constructing a behavioral topological model of each topological substation. Through this model, the topological structure of each substation can be accurately understood, and the lock behavior topological process of each topological substation can be fed back, so as to timely discover and deal with potential problems of improper management and maintenance of each substation in the power grid, reduce fault losses and power outage time, and ultimately improve the safety of the power system.

[0056] Specifically, the lock behavior topology process of each topology area is fed back, and the specific feedback process is:

[0057] The real-time behavioral topological model of each topological substation can be directly obtained through the substation identification system in this example. This substation identification system is a system used in this example to identify and distinguish different power supply areas (substations) in the power system. It is a highly integrated and complex identification management system, including but not limited to the identification feature data cloud platform and data processor mentioned above; and the substation identification system comes with algorithms that can build behavioral topological models, such as graph neural networks (GNN), which are deep learning models for processing graph structure data and are very suitable for building and analyzing topological structures. In the substation identification system, GNN can be used to analyze the connection relationship and mutual influence between substations, and build a more complex and sophisticated behavioral topological model; the specific construction process is not within the scope of consideration of this example.

[0058] The overlap comparison is performed with the behavioral topological model of each topological area. The model in this example is a three-dimensional image model. Therefore, the shape matching coefficient built into the area recognition system can be used to obtain the overlap between the real-time model of each topological area and the corresponding model, that is, the overlap volume of the behavioral topological model of each topological area is compared with the overlap definition volume preset by the recognition feature data cloud platform, wherein the overlap definition volume represents the threshold used to determine whether the two models overlap in the three-dimensional image model matching process. Specifically, it is a preset volume range used to define the overlap degree between the two models. If the overlap volume of the behavioral topological model of a topological area is smaller than the preset overlap definition volume, it means that there is a large gap between the behavioral topological model of the topological area and the model in the actual scene, the model quality is low or the recognition result is inaccurate, then the lock behavior topological process of the topological area is fed back, thereby statistically recording each topological area corresponding to the overlap volume smaller than the preset overlap definition volume as each abnormal area, and the lock behavior topological process of each abnormal area is fed back, and the feedback process is as follows:

[0059] For these abnormal substations, feedback is given to their lock behavior topology process. Specifically, the information of the abnormal substation is recorded (for example, inaccurate data acquisition), and the relevant parameters are automatically adjusted according to the abnormal information of the abnormal substation. The power system analysis tool is used to build a more accurate topology behavior model to improve the accuracy of model construction; after feedback is given to the lock behavior topology process of the abnormal substation, the feedback results need to be evaluated (that is, compared with the real-time behavior topology model again) to determine whether the expected overlap volume has been achieved. If achieved, the phase recognition and topology recognition of the smart locks belonging to the meter box are completed, which can ultimately improve the accurate measurement of electric energy in the meter box; if the expected effect is not achieved, a deeper level of feedback may be required for the process (this example will not explain this in detail) to improve the subsequent stable operation of the power system.

[0060] It should be clarified that if the overlapping volume of the behavioral topological model of a certain topological area is greater than or equal to the preset overlapping boundary volume, it means that the behavioral topological model of the topological area has a high degree of overlap with the model in the actual scene, the model quality is high, and the recognition result is more accurate, so no feedback is required.

[0061] Furthermore, the behavior topology model of each topology area is constructed, and the specific process is as follows:

[0062] The topological structure diagram of each topological substation is obtained by drawing the topological structure diagram using the built-in electrical design software (such as AutoCAD (automatic computer-aided design), EPLAN (electrical planning software), SolidWorks (three-dimensional computer-aided design software), etc.) in the above-mentioned substation identification system, and the lock behavior utilization data of each topological substation is used as the input parameter of the power system analysis tool. Thus, the output model is recorded as the behavioral topological model of each topological substation through the model building function of the power system analysis tool. The specific model building process of the power system analysis tool is as follows:

[0063] The power system analysis tool extracts useful features from the preprocessed data (i.e., the lock behavior utilization data of each topological substation), such as the distribution of lock unlocking frequency, the mean and standard deviation of response time, the distribution of longitude and latitude coordinates, etc.; selects a suitable model for construction. In this example, a multivariate linear regression model is selected; the selected multivariate linear regression model is used to train the feature data selected after preprocessing and the topological structure diagram; during the training process, the parameters of the multivariate linear regression model need to be adjusted to optimize the performance of the multivariate linear regression model. The parameter adjustments that may need to be made include learning rate, number of iterations, etc.; after the training is completed, the multivariate linear regression model is evaluated to determine the accuracy and generalization ability of the model. Commonly used evaluation indicators include accuracy, precision, recall rate, F1 value, etc.; the trained model is applied to practical problems, such as predicting the number of components in a topological substation power grid, the number of connections between components, or determining the weight of edges according to the law of location changes, etc., thereby finally constructing a behavioral topological model of a topological substation based on the above-mentioned predicted data; according to the above operations, the behavioral topological model of each topological substation can be constructed.

[0064] Specifically, the smart lock behavior data of each topological area includes the unlocking frequency, response time of each unlocking, longitude coordinates, and latitude coordinates of each smart lock in each topological area within the behavior cycle, where the behavior cycle represents the unlocking record of the smart lock within a period of time (such as one day, one week, etc.), which is obtained by analyzing the usage records of the smart locks, such as unlocking logs, operation logs and other data.

[0065] It should be explained that the unlocking frequency, the response time of each unlocking, the longitude coordinates, and the latitude coordinates are obtained as follows:

[0066] By analyzing the usage records of smart locks, the number of times each lock is unlocked within the behavior cycle is counted. The unlocking frequency indicates the number of times the smart lock is unlocked within a unit time. By analyzing the usage records of smart locks, the start time and end time of each unlocking are obtained, and then the difference between the two times, that is, the response time, is calculated. The response time of each unlocking indicates the time required for each unlocking of the smart lock within the behavior cycle. The longitude coordinates of the smart lock within the behavior cycle are obtained through the built-in positioning function of the smart lock (such as GPS, Beidou, etc.), and the longitude coordinates indicate the longitude position of the smart lock on the earth's surface. Latitude coordinates: The latitude coordinates of the smart lock within the behavior cycle are obtained through the built-in positioning function of the smart lock (such as GPS, Beidou, etc.). The latitude coordinates indicate the latitude position of the smart lock on the earth's surface.

[0067] Furthermore, the smart lock behavior data of each topology area is preprocessed, and the specific preprocessing process is:

[0068] The cleaning operation of each smart lock behavior data point in each topological area refers to processing the outliers, missing values, duplicate values, etc. in the data points to improve the data quality, and obtaining the data cleaning value of each smart lock behavior data point in each topological area, which means that the value retained by each data point after the cleaning operation is compared with the data cleaning threshold preset by the recognition feature data cloud platform, where the data cleaning threshold represents the judgment standard used in the data cleaning operation. If the data cleaning value of each smart lock behavior data point in a certain topological area is lower than the preset data cleaning threshold, that is, the data point is If there are still serious abnormal situations, such as too many missing values, too large numerical deviations, etc., the smart lock behavior data points of this topological area will be eliminated; if the data cleaning value of each smart lock behavior data point in a topological area is higher than or equal to the preset data cleaning threshold, it means that after the cleaning operation, the data point has reached the preset standard and the data quality is high, then there is no need to eliminate the smart lock behavior data point of this topological area. Therefore, the smart lock behavior data of each topological area corresponding to the statistical data cleaning value higher than or equal to the preset data cleaning threshold is recorded as the lock behavior utilization data of each topological area.

[0069] In a specific embodiment, the present invention provides a substation phase identification and topology identification method based on the positioning of the smart lock of the meter box. First, the location information of the smart locks belonging to each meter box and the location information of each substation are collected. The corresponding substation of the smart locks belonging to each meter box can be obtained by comparing the location information, and the substation phase identification process of the smart locks belonging to each meter box is monitored to determine whether the substation phase identification of the smart locks belonging to each meter box is accurate. Therefore, for each topological substation whose substation phase identification accuracy is greater than a preset substation phase identification accuracy threshold, multi-dimensional smart lock behavior data is identified and analyzed, and a behavioral topological model of each topological substation with accurate identification is constructed. Finally, the lock behavior topological process of each topological substation is fed back to obtain a lock substation with more accurate phase and topology, thereby accurately measuring the electric energy of the meter box and effectively managing and maintaining the equipment, so as to improve the operating stability of the power system.

[0070] The above contents are merely examples and explanations of the structure of the present invention. The technicians in this technical field may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the structure of the invention or exceed the scope defined by the present invention, they should all fall within the protection scope of the present invention.

Claims

1. A phase identification and topology identification method for a station area based on the positioning of a meter box intelligent lock, characterized in that: include: Phase identification evaluation of substations: collect the location information of the smart locks of each meter box and the location information of each substation, analyze and obtain the corresponding substation of the smart locks of each meter box, monitor the phase identification process of the substation of the smart locks of each meter box, evaluate the phase identification accuracy of the substation of the smart locks of each meter box, and compare it with the preset phase identification accuracy threshold of the substation; Feedback on the topological process of the substation: the corresponding substations of the smart locks of each meter box whose phase recognition accuracy of the substation is greater than the preset phase recognition accuracy threshold of the substation are marked as each topological substation, and the behavior data of the smart locks of each topological substation are obtained and preprocessed, thereby constructing the behavior topological model of each topological substation, and finally feedback the lock behavior topological process of each topological substation; The phase recognition accuracy of each meter box's smart lock is evaluated, and the specific evaluation process is as follows: By monitoring the phase identification process of the smart locks of each meter box, the phase data of the smart locks of each meter box and the phase data of the corresponding zones of the smart locks of each meter box are obtained; The phase data of the smart locks belonging to each meter box include the voltage average corresponding to the first phase of the smart locks belonging to each meter box in the identification period, the voltage average corresponding to the second phase of the smart locks belonging to each meter box in the identification period, and the voltage average corresponding to the third phase of the smart locks belonging to each meter box in the identification period; The phase data of the corresponding area of ​​the smart locks to which each meter box belongs includes the voltage average value corresponding to the first phase of the corresponding area of ​​the smart locks to which each meter box belongs within the identification period, the voltage average value corresponding to the second phase of the corresponding area of ​​the smart locks to which each meter box belongs within the identification period, and the voltage average value corresponding to the third phase of the corresponding area of ​​the smart locks to which each meter box belongs within the identification period; The data processing results in the phase recognition accuracy of each metering box's intelligent lock. The data processing expression is as follows: In the formula, is the phase recognition accuracy of the intelligent lock belonging to the u-th meter box, To identify the influencing factors of the phase deviation unit value preset by the feature data cloud platform, is the voltage mean value corresponding to the first phase of the smart lock belonging to the u-th meter box in the identification cycle, is the voltage mean value corresponding to the first phase of the corresponding area of ​​the intelligent lock of the u-th meter box in the identification period, is the voltage mean value corresponding to the second phase of the smart lock belonging to the u-th meter box during the identification period, is the average voltage corresponding to the second phase of the corresponding area of ​​the intelligent lock of the u-th meter box in the identification period, is the voltage mean value corresponding to the third phase of the smart lock belonging to the u-th meter box during the identification period, is the voltage mean value corresponding to the third phase of the corresponding area of ​​the intelligent lock of the u-th meter box in the identification cycle, and e is a natural constant; The specific process of constructing the behavioral topological model of each topological area is as follows: The topological structure diagram of each topological substation is obtained, and the lock behavior utilization data of each topological substation is integrated as the input parameters of the power system analysis tool. Through the model building function of the power system analysis tool, the output model is recorded as the behavioral topological model of each topological substation.

2. According to claim 1, a method for phase identification and topology identification of a substation based on positioning of intelligent locks of a meter box, characterized in that: The location information of the smart locks to which each meter box belongs specifically includes the longitude coordinates, latitude coordinates and altitude of the smart locks to which each meter box belongs.

3. According to claim 1, a method for phase identification and topology identification of a substation based on positioning of intelligent locks of a meter box, characterized in that: The location information of each station area specifically includes the longitude coordinates, latitude coordinates and altitude of each station area.

4. According to claim 3, a method for phase identification and topology identification of a substation based on positioning of intelligent locks of a meter box, characterized in that: The analysis obtains the corresponding area of ​​the intelligent locks to which each meter box belongs. The specific analysis process is as follows: The longitude coordinates, latitude coordinates and altitude of the smart locks of each meter box are respectively processed with the longitude coordinates, latitude coordinates and altitude of each station area to obtain the longitude difference, latitude difference and altitude difference between the smart locks of each meter box and each station area, and then the accumulated values ​​are recorded as the location distance between the smart locks of each meter box and each station area; Arrange the distances between the smart locks of a meter box and each area in ascending order, and extract the area corresponding to the first-ranked distance, which is the area corresponding to the smart lock of the meter box; According to the above operation, the corresponding area of ​​the smart lock belonging to each meter box is obtained.

5. According to claim 1, a method for phase identification and topology identification of a substation based on positioning of intelligent locks of a meter box, characterized in that: The comparison is performed with the preset phase recognition accuracy threshold of the substation area, and the specific comparison process is as follows: If the phase recognition accuracy of the area of ​​the smart lock belonging to a meter box is greater than the preset phase recognition accuracy threshold of the area, the corresponding area of ​​the smart lock belonging to the meter box is located, and the corresponding area of ​​the smart lock belonging to each meter box is obtained by statistics; If the area phase recognition accuracy of the smart lock belonging to a meter box is less than or equal to the preset area phase recognition accuracy threshold, an early warning will be issued for the location of the smart lock belonging to the meter box. Thus, the smart locks belonging to each meter box whose area phase recognition accuracy is less than or equal to the preset area phase recognition accuracy threshold are counted and recorded as abnormal smart locks belonging to each meter box, and an early warning will be issued for the location of the abnormal smart lock belonging to each meter box.

6. According to claim 1, a method for phase identification and topology identification of a substation based on positioning of intelligent locks of a meter box, characterized in that: The smart lock behavior data of each topological area specifically includes the unlocking frequency, response time of each unlocking, longitude coordinates, and latitude coordinates of each smart lock in each topological area within the behavior cycle.

7. According to claim 6, a method for phase identification and topology identification of a substation based on positioning of intelligent locks of a meter box, characterized in that: The specific preprocessing process of the smart lock behavior data of each topology area is as follows: A cleaning operation is performed on each smart lock behavior data point in each topological area to obtain a data cleaning value of each smart lock behavior data point in each topological area, and the data cleaning value is compared with a preset data cleaning threshold. If the data cleaning value of each smart lock behavior data point in a certain topological area is lower than the preset data cleaning threshold, the smart lock behavior data point in this topological area is eliminated. The smart lock behavior data of each topological area corresponding to the statistical data cleaning value being higher than or equal to the preset data cleaning threshold is recorded as the lock behavior utilization data of each topological area.

8. According to claim 1, a method for phase identification and topology identification of a substation based on positioning of intelligent locks of a meter box, characterized in that: The feedback of the lock behavior topology process of each topology area is as follows: The real-time behavioral topological model of each topological area is collected, and the model is overlapped and compared with the behavioral topological model of each topological area to obtain the overlapping volume of the behavioral topological model of each topological area, and compared with the preset overlapping boundary volume. If the overlapping volume of the behavioral topological model of a certain topological area is smaller than the preset overlapping boundary volume, the lock behavioral topological process of the topological area is fed back. Thus, the topological areas corresponding to the overlapping volumes smaller than the preset overlapping boundary volume are recorded as abnormal areas, and the lock behavioral topological process of each abnormal area is fed back.

Images