Zone area low-voltage topological relation identification and correction method and system based on cloud edge collaboration

Abstract

The invention discloses an area low-voltage topological relation identification and correction method and system based on cloud-edge collaboration. The method comprises the steps that an edge side obtains voltage curve data, current curve data and phase information in an area; performing cleaning and phase marking on the acquired data information to obtain effective data; the cloud analyzes the effective data to obtain a transformer area low-voltage topological relation; the edge-side intelligent fusion terminal verifies and corrects the transformer area low-voltage topological relation by using the power failure information reported by the electric energy meter, and transmits the corrected transformer area low-voltage topological relation back to the cloud; and the cloud end uses the corrected transformer area low-voltage topological relation and the corresponding original data information to train, identify and correct a model, and uses the trained model to identify and correct the transformer area low-voltage topological relation. According to the method, the investment cost is low, the result is automatically corrected, the accuracy is greatly improved, and the checking burden of operation and maintenance personnel is relieved.

Description

Technical Field

[0001] The present invention belongs to the technical field of power grid topology identification, and relates to a method and system for identifying and correcting low-voltage topology relationships in a substation based on cloud-edge collaboration. Background Art

[0002] Grid topology identification refers to the process of determining the circuit topology structure by collecting and analyzing data on the distribution network. It is also an important task to ensure the safe and stable operation of the power system. It can provide electrical network structure data for grid flow calculation, load capacity allocation, fault range diagnosis, grid state estimation and other operations, and is the basis for grid system analysis. Topology identification technology has developed from manual identification to system automatic identification. The main topology identification methods include station identification, power outage identification, advanced metering infrastructure (AMI) data analysis and identification, and installation of intelligent branch switch identification.

[0003] The manual identification method is accurate and can identify level 3 topology, but it is inefficient and has high manpower input costs. Topology changes require re-identification, which cannot meet real-time requirements. The use of substation identification instruments requires additional equipment, and the use of power outage identification will affect normal electricity consumption. For example, if there are key users and sensitive users in the substation, power outage identification will affect the normal production and life of users, and more costs will be required. AMI data analysis and identification mainly collects data such as voltage, active power and reactive power at the end nodes of the system branches through smart energy meters. These data reflect the state and flow distribution of the distribution network at different corresponding times, and the branch impedance estimation calculation is carried out to form the topology. The AMI data analysis method is still in the theoretical research stage, mainly for the estimation of the distance between energy meters in the ideal wiring method under the correct topology. In theory, it can identify the level 0 topology and add distance parameters, but this method does not consider the influence of equipment and wiring methods, and is too idealistic.

[0004] The installation of intelligent branch switch identification can effectively identify the second-level topology relationship. If an intelligent switch is added to the meter box main, it can even identify the third-level topology and real-time performance can be guaranteed. However, on the one hand, the investment cost is huge, and on the other hand, it increases the complexity and instability of the distribution network architecture, and there is a large operation and maintenance workload and risk. From the actual case, it is not suitable for large-scale promotion and use. Summary of the invention

[0005] In order to solve the shortcomings existing in the prior art, the present invention provides a method and system for identifying and correcting low-voltage topology relationships in substations based on cloud-edge collaboration. Taking into account the feasibility of data transmission and the economic efficiency of investment costs, low-voltage topology relationships in substations are identified and corrected based on cloud-edge collaboration, which improves the real-time and feasibility of topology identification, reduces investment costs, and reduces the workload of subsequent operation and maintenance.

[0006] The present invention adopts the following technical solution.

[0007] A first aspect of the present invention provides a method for identifying and correcting low-voltage topological relationships in a substation area based on cloud-edge collaboration, the method comprising:

[0008] S1: The low-voltage area intelligent fusion terminal at the edge side obtains the voltage curve data, current curve data and phase information of each electric energy meter in the area;

[0009] S2: The intelligent fusion terminal cleans and phase-marks the acquired data information, obtains valid data and uploads it to the cloud;

[0010] S3: The cloud uses the data measured by the residual current action protectors of different circuits in the low-voltage substation to analyze the valid data, obtain the low-voltage topology relationship of the substation, and send it to the intelligent fusion terminal on the edge side;

[0011] S4: The intelligent fusion terminal verifies and corrects the low-voltage topology relationship of the substation using the power outage information reported by the electric energy meter, and transmits the corrected low-voltage topology relationship of the substation back to the cloud;

[0012] S5: The cloud uses the corrected low-voltage topological relationship of the substation area and the corresponding original data information to train the recognition and correction model, and uses the trained model to identify and correct the low-voltage topological relationship of the substation area.

[0013] Preferably, in S1, the smart fusion terminal and the electric energy meter in the low voltage area perform clock synchronization, and then the smart fusion terminal periodically collects and stores the 96-point voltage curve data and current curve data and phase information frozen by the electric energy meter through HPLC communication.

[0014] Preferably, in S2, the phase mark is specifically:

[0015] For the data information of single-phase electric energy meters, the voltage and current curve data collected by the same electric energy meter for a set number of consecutive days are connected to form a curve that aggregates the voltage and current data for multiple days, and the corresponding phase of the electric energy meter is marked;

[0016] For the data information of the three-phase electric energy meter, the voltage and current curve data of the three-phase electric energy meter are disassembled into three-phase voltage and current curve data and the corresponding phases are marked. The voltage and current curve data of the same electric energy meter for a set number of days are connected respectively by phase to form a voltage and current curve that aggregates the voltage and current data of multiple days.

[0017] Preferably, in S3, the cloud uses the data measured by the residual current action protectors of different circuits in the low-voltage substation to analyze the valid data, obtain the low-voltage topology relationship of the substation and send it to the intelligent fusion terminal on the edge side, specifically including:

[0018] S3-1: Determine the attributes of the low-voltage area. If it is an area without distributed power sources, go to S3-2, otherwise go to S3-5;

[0019] S3-2: Separate the voltage curve data measured by the residual current operated protectors of different circuits in the substation area into phases and use them as the initial three-phase reference voltage curve of the corresponding circuit;

[0020] Calculate the correlation between the initial reference voltage curve of each phase of each circuit and the voltage curve of each aggregated multi-day voltage data of the corresponding phase, and determine the electric energy meter with the highest positive correlation for each phase of each circuit based on the voltage curve of the aggregated multi-day voltage data with the highest positive correlation with the initial reference voltage curve, and establish association with the corresponding circuit;

[0021] S3-3: Taking the voltage curve of the aggregated multi-day voltage data corresponding to the determined electric energy meter as the correlation between the reference voltage curve and each voltage curve of the aggregated multi-day voltage data of the corresponding phase, determine the electric energy meter with the highest positive correlation with the electric energy meter and establish an association between the two electric energy meters, and so on, until all the electric energy meters have the association, at this time, obtain the electric energy meter set on each phase line of each circuit;

[0022] S3-4: According to the characteristics of the voltage drop on the line after the load, the front and rear relationship of each electric energy meter in the electric energy meter set on each phase line of each loop is identified, and a preliminary low-voltage topology relationship of the substation area is formed based on the voltage curve, and then enter S3-6;

[0023] S3-5: Select the voltage curve of the aggregated multi-day voltage data of the distributed power source in the silent stage in the substation to execute S3-2 and S3-3, obtain the set of electric energy meters on each phase line of each loop, and identify the front and back relationship of each electric energy meter in the electric energy meter set on the line according to the characteristics of voltage drop on the line after the load and the characteristics of voltage boost during the working period of the equipment, form a preliminary low-voltage topology relationship of the substation based on the voltage curve, and enter S3-6;

[0024] S3-6: Replace the voltage curves in S3-1 to S3-5 with corresponding current curves, re-execute S3-1 to S3-5, form a preliminary low-voltage topology relationship of the substation based on the current curve, and enter S3-7;

[0025] S3-7: Compare the preliminary low-voltage topology relationship of the substation area obtained based on the voltage curve with the preliminary low-voltage topology relationship of the substation area obtained based on the current curve. If the error between the two is less than 1%, select one of them as the substation area low-voltage topology relationship, otherwise return to S3-1.

[0026] Preferably, the error in S3-7 is:

[0027]

[0028] in is the rth topological position data in the topological relationship obtained by S3-5, It is the rth topological position data in the topological relationship obtained by S3-6, and R represents the number of topological position data in the topological relationship.

[0029] Preferably, the calculation formula of the correlation is:

[0030]

[0031] In the formula, and Respectively represent the voltage mean values ​​of voltage curve x and voltage curve y;

[0032] The value of the correlation p is in the interval [–1,1];

[0033] If p>0, it means that the voltage curve x and the voltage curve y are positively correlated;

[0034] If p < 0, it means that the voltage curve x and the voltage curve y are negatively correlated;

[0035] The larger the absolute value of p is, the stronger the positive and negative correlation between the voltage curve x and the voltage curve y is.

[0036] Preferably, the correlation p is also verified in the following manner:

[0037] 1) Obtain the other two-phase voltage curves x', x", and the three-phase current curves I, I', and I", corresponding to the voltage curve x;

[0038] 2) Obtain the three-phase voltage data x1, y1, z1 of the integer point k from x, x', x", and obtain the three-phase current data x2, y2, z2 of the integer point k from I, I', I", calculate the Euclidean distance between x1, y1, z1 and x2, y2, z2, and obtain the voltage and current correlation of the integer point k, k is an integer ranging from 1 to 24, take the average of the voltage and current correlation of all integer points, obtain D1, and combine it with p to obtain the first verification value

[0039]

[0040] 3) Apply the same method as 1)-2) to the voltage curve y to obtain a second verification value Q2;

[0041] 4) If Q1 and Q2 are both within the set range, the correlation p check passes, otherwise it prompts that the correlation p is incorrect.

[0042] Preferably, in S3-3, if there is an electric energy meter whose correlation degree of the voltage curve of the aggregated multi-day voltage data of all other electric energy meters is lower than the set threshold, then the electric energy meter is judged to be an electric energy meter of a user in a non-local area, and for the electric energy meter of a user in a non-local area, the adjacent area is searched, and the correlation degree of the voltage curve of the aggregated multi-day voltage data of the electric energy meter of the user in the non-local area and the initial reference voltage curve of the adjacent area is calculated, so as to determine the correct area where the electric energy meter of the user in the non-local area is located.

[0043] Preferably, in S5, the neural network model is trained using the data information processed by S1-S2 used to obtain the substation low-voltage topology relationship after correction in S4, and the weights of the neural network model are updated using the back propagation algorithm using the corrected substation low-voltage topology relationship to obtain a trained model, and the trained model is used to identify and correct the substation low-voltage topology relationship.

[0044] Preferably, the neural network model is trained using an algorithm that combines BIRCH and Mini-Batch K-means.

[0045] A second aspect of the present invention provides a system for identifying and correcting low-voltage topological relationships in a substation area based on cloud-edge collaboration, comprising:

[0046] The data acquisition module is used for the low-voltage area intelligent fusion terminal on the edge side to obtain the voltage curve data, current curve data and phase information of each electric energy meter in the area;

[0047] The data processing module is used by the intelligent fusion terminal to clean and phase-mark the acquired data information, obtain valid data and upload it to the cloud;

[0048] A data analysis module is used to analyze the valid data in the cloud using the data measured by the residual current action protectors of different circuits in the low-voltage substation, obtain the low-voltage topology relationship of the substation and send it to the intelligent fusion terminal on the edge side;

[0049] Verification and correction module, used by the intelligent fusion terminal to verify and correct the low-voltage topology relationship of the substation using the power outage information reported by the electric energy meter, and transmit the corrected low-voltage topology relationship of the substation back to the cloud;

[0050] The model training module is used in the cloud to train the recognition and correction model using the corrected low-voltage topological relationship of the substation and the corresponding original data information, and uses the trained model to identify and correct the low-voltage topological relationship of the substation.

[0051] A terminal comprises a processor and a storage medium; the storage medium is used to store instructions; the processor is used to operate according to the instructions to execute the steps of the method.

[0052] A computer-readable storage medium stores a computer program, which implements the steps of the method when executed by a processor.

[0053] Compared with the prior art, the beneficial effects of the present invention include at least:

[0054] The present invention performs topology identification through the collaboration of the edge-side intelligent fusion terminal and the cloud-side master station, taking advantage of the cloud-edge collaboration to maximize the efficient use of computing resources. It does not require the installation of too many measuring devices, has a low investment cost, and can self-correct the results, greatly improving the accuracy and reducing the burden of verification on operation and maintenance personnel. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 It is the overall process step diagram of the method of the present invention;

[0056] Figure 2 It is a schematic diagram of voltage curve analysis in the present invention;

[0057] Figure 3 It is a schematic diagram of the topology recognition analysis results in the present invention. DETAILED DESCRIPTION

[0058] In order to make the purpose, technical scheme and advantages of the present invention clearer, the technical scheme of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. The embodiments described in this application are only embodiments of a part of the present invention, rather than all embodiments. Based on the spirit of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work belong to the protection scope of the present invention.

[0059] Embodiment 1 of the present invention provides a method for identifying and correcting low-voltage substation topological relationships based on cloud-edge collaboration, wherein the edge side of the cloud-edge collaboration is the intelligent fusion terminal in the substation, and the cloud is the main station server; the low-voltage substation topology is composed of different loops. In this embodiment, the selected low-voltage substation electric energy meter communication module needs to install the latest version of the HPLC module, which needs to support power outage event reporting, and all loop line outlets need to be installed with residual current operated protectors with telemetry functions. Figure 1 As shown, the method comprises the following steps:

[0060] S1: The low-voltage area intelligent fusion terminal at the edge side obtains the voltage curve data, current curve data and phase information of each electric energy meter in the area;

[0061] This step obtains various original electric energy data. Further preferably, before executing S1, it is ensured that the clocks of the terminals and the electric energy meters in the substation have been synchronized, which is the basis for ensuring the credibility of the data and the accuracy of subsequent calculations. In the said S1, on the basis of ensuring that the clocks of the terminals and the electric energy meters in the substation have been synchronized, the intelligent fusion terminal periodically collects the 96-point voltage curve data and current curve data and phase information frozen by the electric energy meter through HPLC communication, and the collected data is stored in the edge-side intelligent fusion terminal.

[0062] S2: The intelligent fusion terminal cleans and phase-marks the acquired data information, obtains valid data and uploads it to the cloud;

[0063] This step performs data screening, cleaning, and phase marking on the raw data at the edge side; further preferably, the edge side terminal computing power is used to screen and clean the collected data, and through data recognition, locations where mutations or invalid data exist are found and removed;

[0064] For the data information of single-phase electric energy meters, the voltage and current curve data of the same electric energy meter for a set number of days are connected separately to form a curve that aggregates the voltage and current data of multiple days, and the corresponding phase of the electric energy meter is marked;

[0065] For the data information of the three-phase electric energy meter, the voltage and current curve data of the three-phase electric energy meter are disassembled into three-phase voltage and current curve data and the corresponding phases are marked. The voltage and current curve data of the same electric energy meter for a set number of days are connected respectively by phase to form a voltage and current curve that aggregates the voltage and current data of multiple days.

[0066] Taking the voltage curve as an example, the voltage curve data of the corresponding electric energy meter for multiple days are connected to form a voltage curve that aggregates the data for multiple days, such as Figure 2As shown, in the specific implementation, the data date for the photovoltaic area is doubled for analysis, for example, 10 days for the non-photovoltaic area and 20 days for the photovoltaic area, and then the phase of the household meter is matched. For the three-phase meter, its data is disassembled into three-phase voltages and marked, and the voltage curve is formed by phase to complete the data preparation.

[0067] S3: The cloud uses the data measured by the residual current action protectors of different circuits in the low-voltage substation to analyze the valid data, obtain the low-voltage topology relationship of the substation, and send it to the intelligent fusion terminal on the edge side;

[0068] This step analyzes valid data in the cloud and obtains the topological structure of the substation, including:

[0069] S3-1: Determine the attributes of the low-voltage area. If it is an area without distributed power sources (including photovoltaic, energy storage, orderly charging and other area types), enter S3-2, otherwise enter S3-5;

[0070] S3-2: Separate the voltage curve data measured by the residual current operated protectors of different circuits in the substation area into phases and use them as the initial three-phase reference voltage curve of the corresponding circuit;

[0071] Calculate the correlation between the initial reference voltage curve of each phase of each circuit and the voltage curve of each aggregated multi-day voltage data of the corresponding phase, and determine the electric energy meter with the highest positive correlation for each phase of each circuit based on the voltage curve of the aggregated multi-day voltage data with the highest positive correlation with the initial reference voltage curve, and establish association with the corresponding circuit;

[0072] S3-3: Taking the voltage curve of the aggregated multi-day voltage data corresponding to the determined electric energy meter as the correlation between the reference voltage curve and each voltage curve of the aggregated multi-day voltage data of the corresponding phase, determine the electric energy meter with the highest positive correlation with the electric energy meter and establish an association between the two electric energy meters, and so on, until all the electric energy meters have the association, at this time, obtain the electric energy meter set on each phase line of each circuit;

[0073] S3-4: According to the characteristics of the voltage drop on the line after the load, the front and rear relationship of each electric energy meter in the electric energy meter set on each phase line of each loop is identified, and a preliminary low-voltage topology relationship of the substation area is formed based on the voltage curve, and then enter S3-6;

[0074] S3-5: Select the voltage curve of the aggregated multi-day voltage data of the distributed power source in the silent stage in the substation to execute S3-2 and S3-3, obtain the set of electric energy meters on each phase line of each loop, and identify the front and back relationship of each electric energy meter in the electric energy meter set on the line according to the characteristics of voltage drop on the line after the load and the characteristics of voltage boost during the working period of the equipment, form a preliminary low-voltage topology relationship of the substation based on the voltage curve, and enter S3-6;

[0075] S3-6: Replace the voltage curves in S3-1 to S3-5 with corresponding current curves, re-execute S3-1 to S3-5, form a preliminary low-voltage topology relationship of the substation based on the current curve, and enter S3-7;

[0076] S3-7: Compare the preliminary low-voltage topology relationship of the substation area obtained based on the voltage curve with the preliminary low-voltage topology relationship of the substation area obtained based on the current curve. If the error between the two is less than 1%, select one of them as the substation area low-voltage topology relationship, otherwise return to S3-1 to perform attribute judgment on the new low-voltage substation area. The error is:

[0077]

[0078] in is the rth topological position data in the topological relationship obtained by S3-5, It is the rth topological position data in the topological relationship obtained by S3-6, and R represents the number of topological position data in the topological relationship.

[0079] Based on the above steps, the specific implementation is as follows:

[0080] (1): First, we need to determine the correlation of the curve using the Pearson algorithm. The formula is:

[0081]

[0082] In the formula, and Represent the mean of the voltage curves x and y respectively. The value of the correlation coefficient p is in the interval [–1,1]; if p>0, it indicates that the two curves are positively correlated; if p<0, it indicates that the two curves are negatively correlated. The larger the absolute value of p, the stronger the correlation.

[0083] According to the correlation analysis results of the user voltage curve, the voltage curve data measured by the residual current action protector (with telemetry) of different circuits is used as the benchmark, and the meter with the highest correlation is found in phases. The found meter data is used as the benchmark to proceed in sequence until all electric energy meters are associated, and the circuit to which the household meter belongs is identified;

[0084] For example, the voltage curve data measured by residual current operated protectors (with telemetry) in different circuits are analyzed to obtain the baseline data of three circuit lines in the substation. There are 26 electric energy meters in Table A to Table Z in the file.

[0085] The voltage curves of phase A, E, F, B(A), D(A), M, N, P(A), U, V, and Z(A) and the data of loop phase IA, loop phase IIA, and loop phase IIIA were used for analysis. The most relevant table position data of loop phase IA, loop phase IIA, and loop phase IIIA were found respectively, and the analysis was carried out in sequence based on this data.

[0086] The voltage curves of phase B, Table C, Table G, Table H, Table B (B), Table D (B), Table K, Table L, Table O, Table P (B), Table W, Table X, Table Z (B) and the data of loop phase IB, loop phase IIB, and loop phase IIIB are fitted respectively. The most relevant table position data of loop phase IB, loop phase IIB, and loop phase IIIB are found respectively, and the analysis is carried out in sequence based on this data.

[0087] The voltage curves of phase C in Table I, Table J, Table B (C), Table D (C), Table Q, Table R, Table P (C), Table S, Table T, Table Y, Table Z (C) and the data of loop IC phase, loop IIC phase, and loop IIIC phase are fitted respectively. The most relevant table position data of loop IC phase, loop IIC phase, and loop IIIC phase are found respectively, and the analysis is carried out in sequence based on this data.

[0088] (2): According to the characteristics of voltage drop on the line after the load, the relationship between the household meter and the line is identified, and the preliminary area topology relationship is formed based on the voltage curve, such as Figure 3 As shown;

[0089] (3): For substations with voltage-boosting conditions such as distributed power sources, the data of the distributed power source during the silent phase (the inverter has not reached the threshold and is in a non-working state, such as at night) are selected for analysis according to methods (1) and (2). At the same time, based on the characteristics of voltage boost during the photovoltaic working period, the position of the photovoltaic power meter relative to the non-photovoltaic power meter is identified; for example, if the voltage at meter F of loop I during the photovoltaic working period is significantly higher than that at meter E, it can be determined that meter F is a photovoltaic meter.

[0090] (4): If the correlation between the fitted voltage curve of the electric energy meter and the other electric energy meters is too different, it is judged that the electric energy meter is not a user in this area. For users in other areas, find the adjacent area in the system, check the correlation between the user and the surrounding areas according to the method (1), and find the correct area where the user meter is located;

[0091] (5): After obtaining the topological relationship identification result based on the voltage curve, the Pearson algorithm is used to conduct preliminary verification of the topological relationship using the three-loop three-phase current data and the gateway current data. A certain allowable error value is set (such as within plus or minus 1%) to avoid the situation where the correct identification result is verified incorrectly due to errors.

[0092] Further preferably, the present invention also verifies the Pearson correlation p in the following manner:

[0093] 1) Obtain the other two-phase voltage curves x', x", and the three-phase current curves I, I', and I", corresponding to the voltage curve x;

[0094] 2) Obtain the three-phase voltage data x1, y1, z1 of the integer point k from x, x', x", and obtain the three-phase current data x2, y2, z2 of the integer point k from I, I', I", calculate the Euclidean distance between x1, y1, z1 and x2, y2, z2, and obtain the voltage and current correlation of the integer point k, k is an integer ranging from 1 to 24, take the average of the voltage and current correlation of all integer points, obtain D1, and combine it with p to obtain the first verification value

[0095]

[0096] The Euclidean distance is:

[0097]

[0098] 3) Apply the same method as 1)-2) to the voltage curve y to obtain a second verification value Q2;

[0099] 4) If Q1 and Q2 are both within the set range, the correlation p check passes, otherwise it prompts that the correlation p is incorrect.

[0100] In S3-6, the voltage curves in S3-1 to S3-5 are replaced with corresponding current curves. When S3-1 to S3-5 are executed again, the voltage and current are also correspondingly exchanged in the process of verifying the Pearson correlation p.

[0101] S4: The intelligent fusion terminal verifies and corrects the low-voltage topology relationship of the substation using the power outage information reported by the electric energy meter, and transmits the corrected low-voltage topology relationship of the substation back to the cloud;

[0102] This step verifies and corrects the topology identification results on the edge side using the power outage reporting information. It is further preferred that the electric energy meters in the substation support power outage event reporting. In the above S4, the topological relationship initially verified in the cloud is sent to the edge terminal. When a power outage occurs due to a fault in the electric energy meter, branch line circuit or substation, the substation topology identification results are finally verified on the edge terminal side using the power outage reporting information.

[0103] For example, if there is a power outage at the front of loop III, all the electric energy meters on this loop should have a power outage event. The electric energy meters that have not reported a power outage event can be judged as not belonging to this loop, and more daily data should be collected for re-analysis. If there is a power outage at meter Q in loop II, there should also be a power outage at meter R behind it. If there is no power outage at meter R, then meter R should be adjusted to the front of meter Q.

[0104] For verification errors, the edge computing terminal automatically corrects the topological relationship based on the power outage information and transmits the corrected results back to the cloud.

[0105] S5: The cloud uses the corrected low-voltage topological relationship of the substation area and the corresponding original data information to train the recognition and correction model, and uses the trained model to identify and correct the low-voltage topological relationship of the substation area.

[0106] This step transmits the corrected results back to the cloud to complete the model building and optimization of the analysis method. Preferably, the step specifically includes the following steps:

[0107] S5-1: Build a neural network model.

[0108] S5-2: Prepare a data set using the data information processed by the model aggregation S1-S2. It can be understood that the data in this process refers to the data obtained and processed through S1 and S2 in order to obtain the corrected low-voltage topological relationship of the substation area.

[0109] S5-3: Use the dataset to train the constructed neural network model, and use the corrected results to update the model weights using the back propagation algorithm.

[0110] Furthermore, the neural network model is trained by adopting the algorithm fused with the BIRCH algorithm and the Mini-Batch K-means, and a CF tree can be constructed and clustering operations can be performed. During the learning process, due to the large number of users in the substation area, the number of nodes will also be large. By integrating the two algorithms, the factors of hierarchical clustering can be effectively solved.

[0111] After obtaining the trained model, the raw data collected in real time is directly input into the model to automatically identify and correct the low-voltage topological relationship in the substation area.

[0112] Embodiment 2 of the present invention provides a system for identifying and correcting low-voltage topological relationships in a substation area based on cloud-edge collaboration, including:

[0113] The data acquisition module is used for the low-voltage area intelligent fusion terminal on the edge side to obtain the voltage curve data, current curve data and phase information of each electric energy meter in the area;

[0114] The data processing module is used by the intelligent fusion terminal to clean and phase-mark the acquired data information, obtain valid data and upload it to the cloud;

[0115] A data analysis module is used to analyze the valid data in the cloud using the data measured by the residual current action protectors of different circuits in the low-voltage substation, obtain the low-voltage topology relationship of the substation and send it to the intelligent fusion terminal on the edge side;

[0116] Verification and correction module, used by the intelligent fusion terminal to verify and correct the low-voltage topology relationship of the substation using the power outage information reported by the electric energy meter, and transmit the corrected low-voltage topology relationship of the substation back to the cloud;

[0117] The model training module is used in the cloud to train the recognition and correction model using the corrected low-voltage topological relationship of the substation and the corresponding original data information, and uses the trained model to identify and correct the low-voltage topological relationship of the substation.

[0118] A terminal comprises a processor and a storage medium; the storage medium is used to store instructions; the processor is used to operate according to the instructions to execute the steps of the method.

[0119] A computer-readable storage medium stores a computer program, which implements the steps of the method when executed by a processor.

[0120] Compared with the prior art, the beneficial effects of the present invention include at least:

[0121] The present invention performs topology identification through the collaboration of the edge-side intelligent fusion terminal and the cloud-side master station, taking advantage of the cloud-edge collaboration to maximize the efficient use of computing resources. It does not require the installation of too many measuring devices, has a low investment cost, and can self-correct the results, greatly improving the accuracy and reducing the burden of verification on operation and maintenance personnel.

[0122] The present disclosure may be a system, a method and / or a computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for causing a processor to implement various aspects of the present disclosure.

[0123] A computer-readable storage medium may be a tangible device that can hold and store instructions used by an instruction execution device. A computer-readable storage medium may be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing. More specific examples of computer-readable storage media (a non-exhaustive list) include: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a static random access memory (SRAM), a portable compact disk read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanical encoding device, such as a punch card or a raised structure in a groove on which instructions are stored, and any suitable combination of the foregoing. As used herein, a computer-readable storage medium is not to be interpreted as a transient signal per se, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagating through a waveguide or other transmission medium (e.g., a light pulse through a fiber optic cable), or an electrical signal transmitted through a wire.

[0124] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to each computing / processing device, or downloaded to an external computer or external storage device via a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network can include copper transmission cables, optical fiber transmissions, wireless transmissions, routers, firewalls, switches, gateway computers, and / or edge servers. The network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards the computer-readable program instructions for storage in the computer-readable storage medium in each computing / processing device.

[0125] The computer program instructions for performing the operation of the present disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-related instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages, such as Smalltalk, C++, etc., and conventional procedural programming languages, such as "C" language or similar programming languages. Computer-readable program instructions may be executed completely on a user's computer, partially on a user's computer, as an independent software package, partially on a user's computer, partially on a remote computer, or completely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., using an Internet service provider to connect via the Internet). In some embodiments, an electronic circuit, such as a programmable logic circuit, a field programmable gate array (FPGA), or a programmable logic array (PLA), may be customized by utilizing the state information of the computer-readable program instructions, and the electronic circuit may execute the computer-readable program instructions, thereby realizing various aspects of the present disclosure.

[0126] Finally, 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 above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents, and any modifications or equivalent replacements that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for identifying and correcting low-voltage topological relationships in a substation area based on cloud-edge collaboration, characterized in that: The method comprises: S1: The low-voltage area intelligent fusion terminal at the edge side obtains the voltage curve data, current curve data and phase information of each electric energy meter in the area; S2: The intelligent fusion terminal cleans and phase-marks the acquired data information, obtains valid data and uploads it to the cloud; S3: The cloud uses the data measured by the residual current action protectors of different circuits in the low-voltage substation to analyze the valid data, obtain the low-voltage topology relationship of the substation, and send it to the intelligent fusion terminal on the edge side; S4: The intelligent fusion terminal verifies and corrects the low-voltage topology relationship of the substation using the power outage information reported by the electric energy meter, and transmits the corrected low-voltage topology relationship of the substation back to the cloud; S5: The cloud uses the corrected low-voltage topological relationship of the substation area and the corresponding original data information to train the recognition and correction model, and uses the trained model to identify and correct the low-voltage topological relationship of the substation area.

2. According to the method for identifying and correcting low-voltage topological relationships in substations based on cloud-edge collaboration according to claim 1, it is characterized in that: In S1, the intelligent fusion terminal and the electric energy meter in the low-voltage area perform clock synchronization, and then the intelligent fusion terminal periodically collects and stores the 96-point voltage curve data and current curve data and phase information frozen by the electric energy meter through HPLC communication.

3. According to the method for identifying and correcting low-voltage topological relationships in substations based on cloud-edge collaboration according to claim 1, it is characterized in that: In S2, the phase mark is specifically: For the data information of single-phase electric energy meters, the voltage and current curve data collected by the same electric energy meter for a set number of consecutive days are connected to form a curve that aggregates the voltage and current data for multiple days, and the corresponding phase of the electric energy meter is marked; For the data information of the three-phase electric energy meter, the voltage and current curve data of the three-phase electric energy meter are disassembled into three-phase voltage and current curve data and the corresponding phases are marked. The voltage and current curve data of the same electric energy meter for a set number of days are connected respectively by phase to form a voltage and current curve that aggregates the voltage and current data of multiple days.

4. According to the method of claim 1, the method is characterized by: In S3, the cloud uses the data measured by the residual current action protectors of different circuits in the low-voltage substation to analyze the valid data, obtain the low-voltage topology of the substation and send it to the intelligent fusion terminal on the edge side, specifically including: S3-1: Determine the attributes of the low-voltage area. If it is an area without distributed power sources, go to S3-2, otherwise go to S3-5; S3-2: Separate the voltage curve data measured by the residual current operated protectors of different circuits in the substation area into phases and use them as the initial three-phase reference voltage curve of the corresponding circuit; Calculate the correlation between the initial reference voltage curve of each phase of each circuit and the voltage curve of each aggregated multi-day voltage data of the corresponding phase, and determine the electric energy meter with the highest positive correlation for each phase of each circuit based on the voltage curve of the aggregated multi-day voltage data with the highest positive correlation with the initial reference voltage curve, and establish association with the corresponding circuit; S3-3: Taking the voltage curve of the aggregated multi-day voltage data corresponding to the determined electric energy meter as the correlation between the reference voltage curve and each voltage curve of the aggregated multi-day voltage data of the corresponding phase, determine the electric energy meter with the highest positive correlation with the electric energy meter and establish an association between the two electric energy meters, and so on, until all the electric energy meters have the association, at this time, obtain the electric energy meter set on each phase line of each circuit; S3-4: According to the characteristics of the voltage drop on the line after the load, the front and rear relationship of each electric energy meter in the electric energy meter set on each phase line of each loop is identified, and a preliminary low-voltage topology relationship of the substation area is formed based on the voltage curve, and then enter S3-6; S3-5: Select the voltage curve of the aggregated multi-day voltage data of the distributed power source in the silent stage in the substation to execute S3-2 and S3-3, obtain the set of electric energy meters on each phase line of each loop, and identify the front and back relationship of each electric energy meter in the electric energy meter set on the line according to the characteristics of voltage drop on the line after the load and the characteristics of voltage boost during the working period of the equipment, form a preliminary low-voltage topology relationship of the substation based on the voltage curve, and enter S3-6; S3-6: Replace the voltage curves in S3-1 to S3-5 with corresponding current curves, re-execute S3-1 to S3-5, form a preliminary low-voltage topology relationship of the substation based on the current curve, and enter S3-7; S3-7: Compare the preliminary low-voltage topology relationship of the substation area obtained based on the voltage curve with the preliminary low-voltage topology relationship of the substation area obtained based on the current curve. If the error between the two is less than 1%, select one of them as the substation area low-voltage topology relationship, otherwise return to S3-1.

5. According to the method of claim 4, the method is characterized by: The error described in S3-7 is: in is the rth topological position data in the topological relationship obtained by S3-5, It is the rth topological position data in the topological relationship obtained by S3-6, and R represents the number of topological position data in the topological relationship.

6. According to the method of claim 4, the method is characterized by: The calculation formula of the correlation is: In the formula, and Respectively represent the voltage mean values ​​of voltage curve x and voltage curve y; The value of the correlation p is in the interval [–1,1]; If p>0, it means that the voltage curve x and the voltage curve y are positively correlated; If p < 0, it means that the voltage curve x and the voltage curve y are negatively correlated; The larger the absolute value of p is, the stronger the positive and negative correlation between the voltage curve x and the voltage curve y is.

7. According to the method of claim 6, the method is characterized by: The correlation p is also checked as follows: 1) Obtain the other two-phase voltage curves x', x", and the three-phase current curves I, I', and I", corresponding to the voltage curve x; 2) Obtain the three-phase voltage data x1, y1, z1 of the integer point k from x, x', x", and obtain the three-phase current data x2, y2, z2 of the integer point k from I, I', I", calculate the Euclidean distance between x1, y1, z1 and x2, y2, z2, and obtain the voltage and current correlation of the integer point k, k is an integer ranging from 1 to 24, take the average of the voltage and current correlation of all integer points, obtain D1, and combine it with p to obtain the first verification value 3) Apply the same method as 1)-2) to the voltage curve y to obtain a second verification value Q2; 4) If Q1 and Q2 are both within the set range, the correlation p check passes, otherwise it prompts that the correlation p is incorrect.

8. According to the method of claim 1, the method is characterized by: In S3-3, if there is an electric energy meter whose correlation degree of the voltage curve of the aggregated multi-day voltage data with all other electric energy meters is lower than the set threshold, then the electric energy meter is judged to be an electric energy meter of users in non-local area. For the electric energy meter of users in non-local area, the adjacent area is searched, and the correlation degree of the voltage curve of the aggregated multi-day voltage data of the electric energy meter of users in non-local area and the initial reference voltage curve of the adjacent area is calculated, so as to determine the correct area where the electric energy meter of users in non-local area is located.

9. According to the method of claim 1, the method is characterized by: In S5, the neural network model is trained using the data information processed by S1-S2 used to obtain the corrected low-voltage topological relationship of the substation in S4. The back propagation algorithm is used to update the weights of the neural network model using the corrected low-voltage topological relationship of the substation to obtain a trained model. The trained model is used to identify and correct the low-voltage topological relationship of the substation.

10. The method for identifying and correcting low-voltage topological relationships in a substation area based on cloud-edge collaboration according to claim 9, characterized in that: The neural network model is trained using an algorithm that combines BIRCH and Mini-Batch K-means.

11. A system for identifying and correcting low-voltage topological relationships in a substation area based on cloud-edge collaboration, using the method described in any one of claims 1 to 10, characterized in that: The system comprises: The data acquisition module is used for the low-voltage area intelligent fusion terminal on the edge side to obtain the voltage curve data, current curve data and phase information of each electric energy meter in the area; The data processing module is used by the intelligent fusion terminal to clean and phase-mark the acquired data information, obtain valid data and upload it to the cloud; A data analysis module is used to analyze the valid data in the cloud using the data measured by the residual current action protectors of different circuits in the low-voltage substation, obtain the low-voltage topology relationship of the substation and send it to the intelligent fusion terminal on the edge side; Verification and correction module, used by the intelligent fusion terminal to verify and correct the low-voltage topology relationship of the substation using the power outage information reported by the electric energy meter, and transmit the corrected low-voltage topology relationship of the substation back to the cloud; The model training module is used in the cloud to train the recognition and correction model using the corrected low-voltage topological relationship of the substation and the corresponding original data information, and uses the trained model to identify and correct the low-voltage topological relationship of the substation.

12. A terminal comprising a processor and a storage medium; characterized in that: The storage medium is used to store instructions; The processor is configured to operate according to the instructions to execute the steps of the method according to any one of claims 1-10.

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

Images