Transformer area intelligent management device

Abstract

The invention discloses a transformer area intelligent management device, which comprises a transformer area AI management unit and a field concentrator used for replacing a transformer area, and is characterized in that the transformer area AI management unit specifically comprises a high-speed power line carrier HPLC application and data storage module which is configured to be interconnected and intercommunicated with an external HPLC chip; the intelligent diagnosis module is configured to perform positioning of electricity stealing and metering abnormity by utilizing transformer area data acquired by an external HPLC chip at high frequency and combining a local electricity stealing prevention big data model and a line loss abnormity rule model; the line loss analysis module is used for carrying out statistical analysis on the line loss of the transformer area; and the topology identification module is used for realizing topology identification of branches in the transformer area. According to the method, the management efficiency and precision of the HPLC transformer area are remarkably improved, accurate positioning of electricity stealing behaviors and metering abnormity is achieved, illegal electricity utilization is effectively attacked, channel resources are utilized through efficient edge calculation, multiple functions such as clock error analysis and topology recognition are completed, and powerful support is provided for lean management of transformer area line loss.

Description

Technical Field

[0001] The present invention relates to the technical field of power systems, and in particular to an intelligent management device for a substation area. Background Art

[0002] In the power transmission system, the technical line loss of the power grid is a key issue. It mainly comes from the fact that when the power is transmitted, the current flows through the wires and transformers and other equipment. Due to the resistance of these equipment, it is inevitable to generate power loss. This part of the loss includes wire resistance loss and transformer loss (such as copper loss, etc.). In addition, factors such as line insulation aging and poor contact of equipment connection parts will also cause fault loss, further increasing the ineffective consumption of electricity. The main causes of technical line loss can be summarized as follows:

[0003] Equipment aging and performance degradation: The physical properties of power equipment such as transformers and conductors that have been in operation for a long time will gradually decay over time. Aging causes equipment resistance to increase, magnetic permeability to change, etc., which in turn causes electrical energy to be converted into other forms of energy such as heat during transmission, resulting in power loss.

[0004] Voltage drop leads to reduced transmission efficiency: According to Ohm's law, wire resistance will cause voltage drop in the transmission line, which not only affects the effective transmission of electricity, but also requires a higher voltage to be output at the source end to ensure the stability of the terminal voltage, thereby increasing the energy loss in the entire transmission process.

[0005] In recent years, with the construction of advanced measurement systems and the widespread application of smart meters, the field of data-driven power consumption anomaly detection has ushered in new development opportunities. Smart meters can accurately record user power consumption data at a high frequency and realize two-way real-time communication with data centers, which provides managers with rich power consumption information, enabling them to conduct in-depth analysis of the collected data to explore potential power consumption anomalies, thereby effectively reducing line loss risks. However, the current abnormal power consumption detection methods for high-frequency electric energy metering data generally have defects such as poor accuracy, weak versatility, and mismatch between operating procedures and actual complex working conditions, which makes it difficult to meet the actual needs of refined management and efficient operation of power systems. Summary of the invention

[0006] The technical problem to be solved by the present invention is to provide an intelligent management device for a substation area to avoid the problem of low efficiency in identifying and processing household transformer abnormalities caused by inconsistency between the user's power access point and the power consumption situation, and to improve the accuracy of abnormal identification and processing efficiency.

[0007] In order to solve the above technical problems, the present invention provides an intelligent management device for a substation area, including an AI management unit for the substation area, which is used to replace the on-site concentrator of the substation area. The AI ​​management unit for the substation area specifically includes:

[0008] A high-speed power line carrier HPLC application and data storage module configured to interconnect with an external HPLC chip;

[0009] The intelligent diagnosis module is configured to utilize the area data collected at high frequency through an external HPLC chip, combined with the local anti-electricity theft big data model and the line loss anomaly rule model, to locate the electricity theft and metering anomalies;

[0010] Line loss analysis module, used for statistical analysis of line loss in the substation area;

[0011] The topology recognition module is used to realize the topology recognition of branches within the substation area.

[0012] Preferably, the HPLC application and data storage module is also used to perform phase identification, user change identification, chip ID management and clock management on the collected data.

[0013] Preferably, the intelligent diagnosis module further comprises:

[0014] Data preprocessing unit, used to clean, denoise and reduce the dimension of the station area data collected by the external HPLC chip at high frequency;

[0015] A feature extraction unit extracts features related to electricity theft and metering anomalies from the preprocessed data;

[0016] Model training and prediction unit, which builds and trains a local anti-electricity theft big data model based on the extracted feature data to predict future electricity consumption;

[0017] The anomaly detection unit calculates the power consumption anomaly rate based on the deviation between the predicted power consumption and the actual power consumption, and compares it with the set threshold to determine whether there is power theft or metering anomaly;

[0018] Among them, the prediction method adopted by the model training and prediction unit is a prediction method based on the RNN-LSTM network or a prediction method based on the RNN-Markov combination.

[0019] Preferably, the abnormality detection unit is further configured as follows:

[0020] The weight of the deviation rate of electricity consumption in different periods is calculated by using the analytic hierarchy process;

[0021] The power consumption anomaly rate is calculated based on the weight and power consumption deviation rate, providing a quantitative basis for locating power theft and metering anomalies.

[0022] Preferably, the calculation method of the abnormal power consumption rate is:

[0023] Abnormal power consumption rate = φ1×n1+φ2×n2+φ3×n3

[0024] Among them, φ1 is the power consumption deviation rate of the period before the measured period, φ2 is the power consumption deviation rate of the period before the measured period, φ3 is the power consumption deviation rate of the measured period, and n1, n2, and n3 are the corresponding weights respectively.

[0025] Preferably, the analytic hierarchy process is used to calculate the weights n1, n2, n3 of the power consumption deviation rate between the time period to be measured and the two adjacent time periods of the same length before the time period to be measured.

[0026] Preferably, the intelligent management device for the substation area further includes an intelligent low-voltage branch monitoring terminal LTU, and the LTU is configured as follows:

[0027] Collect voltage, current, power, frequency, electricity and area topology data of low-voltage area branch lines;

[0028] The collected data is reported to the AI ​​management unit of the station area through the HPLC communication module.

[0029] Preferably, the LTU specifically includes a controller, a GPS module, a topology transceiver module and an AC data acquisition module; wherein,

[0030] The controller is used to receive and process the area topology information sent by the topology transceiver module, the power-related data sent by the AC data acquisition module, and the location information sent by the GPS module according to the instructions issued by the area AI management unit, and then send the data to the area AI management unit through the HPLC communication module;

[0031] The topology transceiver module is used to send and identify characteristic current signals on the low-voltage line, determine the ownership of the branch line according to the characteristic signals, and send the data to the controller;

[0032] The AC data acquisition module is used to collect and store electricity-related data of the low-voltage branch line according to instructions, and upload the collected data to the controller.

[0033] Preferably, the line loss analysis module is further configured as follows:

[0034] Based on the synchronous collection of 96 points of electricity meter data in the entire area, the synchronous line loss calculation at 24:00 / 96:00 every day is realized to find abnormal line loss periods and analyze the user's electricity theft behavior trajectory;

[0035] Based on the phase identification results of single-phase / three-phase electricity meters and phase measurement data of HPLC, the phase line loss calculation can be realized at 24 / 96 o'clock every day;

[0036] Calculate branch line loss based on the area topology and LTU of branch nodes;

[0037] Based on the clustering results of the electric meter, combined with the calculation and analysis of the line loss of the LTU meter box, the meter box with abnormal line loss is located.

[0038] Preferably, the intelligent management device for the substation area further includes a power wireless communication data transmission unit DTU, and the DTU is configured as follows:

[0039] After receiving the data of the area AI management unit via Bluetooth, upload it to the remote management platform via 4G or 5G network;

[0040] Realize remote communication between the management platform and the area AI management unit.

[0041] The implementation of the present invention has the following beneficial effects: the present invention significantly improves the management efficiency and accuracy of HPLC substations, not only realizes the "point-to-point" precise positioning of electricity theft and metering anomalies, effectively combating illegal electricity use, but also utilizes channel resources through efficient edge computing, completes clock error analysis, topology identification and other functions, and provides strong support for the lean management of substation line losses. In addition, the present invention can seamlessly connect with various types of concentrators and electric energy meter modules, and quickly output abnormal analysis results, which is convenient for immediate verification and processing. While maintaining high performance, it also greatly reduces equipment investment and installation complexity. Only one substation AI management unit needs to be replaced to realize advanced functions such as intelligent diagnosis and time-sharing and phase-sharing line loss calculation. In conjunction with LTU, more refined segmentation and box management can be realized, which greatly promotes the intelligence and efficiency of power grid management, and lays a solid foundation for the access and efficient utilization of new energy such as distributed photovoltaics in the future. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0043] Figure 1 The present invention is a logic block diagram of an intelligent management device for a substation area according to an embodiment of the present invention.

[0044] Figure 2 The figure is a schematic diagram of the connection between the substation AI management unit and the intelligent low-voltage branch monitoring terminal LTU in an embodiment of the present invention. DETAILED DESCRIPTION

[0045] The following descriptions of the embodiments are with reference to the accompanying drawings to illustrate specific embodiments in which the present invention can be implemented. In the description of the present invention, it should be understood that the terms "longitudinal", "length", "circumferential", "front", "rear", "left", "right", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention.

[0046] Please refer to Figure 1 As shown, an embodiment of the present invention provides a substation intelligent management device, including a substation AI management unit, which is used to replace a field concentrator in the substation, and the substation AI management unit specifically includes:

[0047] A high-speed power line carrier HPLC application and data storage module configured to interconnect with an external HPLC chip;

[0048] The intelligent diagnosis module is configured to utilize the area data collected at high frequency through an external HPLC chip, combined with the local anti-electricity theft big data model and the line loss anomaly rule model, to locate the electricity theft and metering anomalies;

[0049] Line loss analysis module, used for statistical analysis of line loss in the substation area;

[0050] The topology recognition module is used to realize the topology recognition of branches within the substation area.

[0051] Specifically, in the embodiment of the present invention, the HPLC application and data storage module is used to achieve interconnection with an external HPLC chip, and supports phase recognition, household change recognition, high-frequency acquisition, chip ID management and clock precision management.

[0052] The intelligent diagnosis module of the substation specifically uses HPLC high-frequency data collection to calculate the local anti-electricity theft big data model and line loss anomaly rule model on a substation basis, and accurately locate electricity theft and metering anomalies. By utilizing the high-frequency acquisition performance of HPLC, the curve data, event records, zero-wire current and other information of the electricity meter are collected, and the NTB and SNR feature data of HPLC communication are collected and comprehensively analyzed to locate the clock anomaly, suspected electricity theft, metering anomaly and household variation analysis of the household meter.

[0053] Furthermore, the construction process of the local anti-electricity theft big data model adopted by the intelligent diagnosis module of the substation area is as follows:

[0054] Step S1: Obtain historical data of the branch line from the on-site metering equipment, and preprocess the obtained historical data; the preprocessing includes but is not limited to data denoising and filling in missing values. In this embodiment, the preprocessing includes:

[0055] Step S11: Processing missing values ​​of the collected data;

[0056] Step S12: Based on the outliers in the sample data collected in S11, data preprocessing is required before training the model;

[0057] Step S13: Reduce the dimension of the data through linear transformation, eliminate redundant feature variables, and then construct a prediction model data set;

[0058] Step S2: extracting characteristic data of the electric energy meter from the preprocessed data;

[0059] Step S3: The extracted feature data is used as training samples. In data sample division, a part of the data is randomly selected as a training set and another part of the data is randomly selected as a test set. A big data model for predicting power consumption is constructed to predict the power consumption of users in a specific period of time in the future.

[0060] Step S4: Calculate the abnormal power consumption rate by combining the predicted power consumption with the user's recent actual power consumption;

[0061] Step S5: Compare the abnormal power usage rate with a set threshold value. If the abnormal power usage rate exceeds the threshold value, it is determined that the low-voltage user has abnormal power usage and may have committed electricity theft.

[0062] In step S3, there are many methods for constructing a big data model for predicting power consumption. In this embodiment, a prediction method based on an RNN-LSTM network is mainly used, and the steps are as follows:

[0063] Step S31: constructing an RNN-LSTM network structure, wherein the input of the RNN-LSTM network structure is the power usage from time 1 to time t-1, and the output is the power usage from time 2 to time t;

[0064] Step S32: using the training set data as the RNN-LSTM network structure input to train the RNN-LSTM network structure;

[0065] Step S33: Input the test set data into the trained RNN-LSTM network structure for verification, calculate the prediction accuracy, and fine-tune the parameters of the RNN-LSTM network structure model through the test set to improve the prediction accuracy, and use the final RNN-LSTM network structure model as the prediction model.

[0066] In another embodiment of the present invention, the prediction method based on RNN-Markov combination is used to construct the big data model for predicting power consumption, and the steps are as follows:

[0067] Step 31: Build the RNN network model, input the initial data set into the RNN network model, adjust the hyperparameters, perform training and prediction, and output the predicted values ​​of the indicators of the training set and the test set;

[0068] Step S32: Divide the state space of the Markov model according to the relative residual between the predicted value of the training set output by the RNN network model and the true value of the collected initial data;

[0069] Step S33: obtaining a state transition probability matrix of the Markov model according to the divided state space and the predicted value of the RNN network model training set;

[0070] Step S34: Correct the index prediction value of the RNN network model test set according to the state transition probability matrix and the divided state interval of the Markov model to obtain the index prediction value of the RNN-Markov combination model.

[0071] Further, in step S4, the step of calculating the abnormal power consumption rate is:

[0072] Step S41: obtaining the actual power consumption of the low-voltage user during the period to be detected, using the predicted power consumption during the period to be detected obtained in step S3, and then deriving the power consumption deviation rate.

[0073] Step S42: Use the hierarchical analysis method to calculate the weights n1, n2, and n3 of the power consumption deviation rate between the time period to be measured and the two adjacent time periods of the same length before the time period to be measured, wherein the preset hierarchical analysis method includes a target layer, a criterion layer, an indicator layer, and a scheme layer; the time period to be measured described in this embodiment is a set time period, such as one day, one week, or one month, which can be set according to actual needs.

[0074] The analytic hierarchy process is a decision-making method that decomposes decision-related elements into goals, criteria, plans, etc., and conducts qualitative and quantitative analysis on this basis. It mainly includes the following steps:

[0075] Analyze and evaluate the relationship between the basic elements in the system and establish the hierarchical structure of the system;

[0076] Compare the importance of each element in the same level with respect to a criterion in the previous level, construct a pairwise comparison judgment matrix, and perform consistency test;

[0077] The relative weight of the compared elements for the criterion is calculated by the judgment matrix;

[0078] Calculate the total weight of each layer's elements to the system's purpose and rank the alternatives.

[0079] Step S43: Obtain the power consumption deviation rates of the two adjacent time periods of equal length before the time period to be measured, which are φ1, φ2 and φ3 respectively, and calculate the power consumption abnormality rate according to the weights n1, n2 and n3, wherein the calculation formula of the power consumption abnormality rate is: Power consumption abnormality rate = φ1×n1+φ2×n2+φ3×n3, φ1 is the power consumption deviation rate of the time period before the time period to be measured, φ2 is the power consumption deviation rate of the time period before the time period to be measured, and φ3 is the power consumption deviation rate of the time period to be measured.

[0080] In the embodiment of the present invention, the line loss analysis module is based on the lean line loss management solution of time-sharing line loss + phase-sharing line loss + segment-sharing line loss + box-sharing line loss + meter intelligent diagnosis of the substation AI management unit, which solves the problem of high loss in the substation caused by meter failure, power theft at the meter end, private connection, black households, hidden cables, etc. Specifically including:

[0081] Time-based line loss analysis: Based on the synchronous collection of 96 points of electricity meter power data in the entire area, the synchronous line loss calculation is realized at 24:00 / 96:00 every day, the abnormal line loss period is found, and the user's electricity theft behavior trajectory is analyzed;

[0082] Phase line loss analysis: Based on the phase identification results of single-phase / three-phase meters and phase measurement data of HPLC, phase line loss calculation can be realized at 24 / 96 o'clock every day;

[0083] Segment line loss analysis: Based on the topological relationship of the substation area and the LTU equipment of the branch node, the branch line loss is calculated to investigate and punish illegal acts such as hidden private wiring and electricity theft;

[0084] Line loss analysis by box: Based on the clustering results of the electric meter, combined with the calculation and analysis of the line loss of the meter box of the LTU equipment, the meter box with abnormal line loss is located.

[0085] like Figure 2 As shown, the intelligent management device for the substation in the embodiment of the present invention also includes an intelligent low-voltage branch monitoring terminal LTU, which is used to collect voltage, current, power, frequency, electricity, and substation topology data of the low-voltage substation branch line and report it to the substation AI management unit.

[0086] The intelligent low-voltage branch monitoring terminal specifically includes: a controller, a GPS module, an HPLC communication module, a power module, a topology transceiver module and an AC data acquisition module.

[0087] According to the instructions issued by the substation AI management unit, the controller receives and processes the substation topology information sent by the topology transceiver module, the power-related data sent by the AC data acquisition module, and the location information sent by the GPS module, and then sends the data to the substation AI management unit through the HPLC communication module.

[0088] The topology transceiver module is used to send and identify the characteristic current signal on the low-voltage line, determine the ownership of the branch line according to the characteristic signal, and send the data to the controller.

[0089] The AC data acquisition module is used to collect and store electricity-related data including voltage, current, power, frequency, and power data of the low-voltage branch line according to instructions, and upload the collected data to the controller.

[0090] As a further improvement, the substation intelligent management device of the embodiment of the present invention also includes a power wireless communication data transmission unit (DTU), which is used to transmit the data of the substation AI management unit to the management platform to realize remote viewing; it communicates with the substation AI management unit via Bluetooth in the downlink and communicates with the platform in the uplink via 4G or 5G.

[0091] Compared with the prior art, the beneficial effect brought by the embodiments of the present invention is that the present invention significantly improves the management efficiency and accuracy of the HPLC substation, not only realizing the "point-to-point" precise positioning of electricity theft and metering anomalies, effectively combating illegal electricity use, but also utilizing channel resources through efficient edge computing to complete clock error analysis, topology identification and other functions, providing strong support for the lean management of substation line losses. In addition, the present invention can seamlessly connect with various types of concentrators and electric energy meter modules, and quickly output abnormal analysis results, which is convenient for immediate verification and processing. While maintaining high performance, it also greatly reduces equipment investment and installation complexity. Only one substation AI management unit needs to be replaced to realize advanced functions such as intelligent diagnosis and time-sharing and phase-sharing line loss calculation. In conjunction with LTU, more refined segmentation and box management can be realized, which greatly promotes the intelligence and efficiency of power grid management, and lays a solid foundation for the access and efficient utilization of new energy such as distributed photovoltaics in the future.

[0092] The above disclosure is only the preferred embodiment of the present invention, which certainly cannot be used to limit the scope of the present invention. Therefore, equivalent changes made according to the claims of the present invention are still within the scope of the present invention.

Claims

1. A smart management device for a substation, characterized in that: It includes a station area AI management unit, which is used to replace the field concentrator of the station area. The station area AI management unit specifically includes: A high-speed power line carrier HPLC application and data storage module configured to interconnect with an external HPLC chip; The intelligent diagnosis module is configured to utilize the area data collected at high frequency through an external HPLC chip, combined with the local anti-electricity theft big data model and the line loss anomaly rule model, to locate the electricity theft and metering anomalies; Line loss analysis module, used for statistical analysis of line loss in the substation area; The topology recognition module is used to realize the topology recognition of branches within the substation area.

2. The intelligent management device for substations according to claim 1, characterized in that: The HPLC application and data storage module is also used to perform phase identification, user change identification, chip ID management, and clock management on the collected data.

3. The intelligent management device for substations according to claim 1, characterized in that: The intelligent diagnosis module further comprises: Data preprocessing unit, used to clean, denoise and reduce the dimension of the station area data collected by the external HPLC chip at high frequency; A feature extraction unit extracts features related to electricity theft and metering anomalies from the preprocessed data; Model training and prediction unit, which builds and trains a local anti-electricity theft big data model based on the extracted feature data to predict future electricity consumption; The anomaly detection unit calculates the power consumption anomaly rate based on the deviation between the predicted power consumption and the actual power consumption, and compares it with the set threshold to determine whether there is power theft or metering anomaly; Among them, the prediction method adopted by the model training and prediction unit is a prediction method based on the RNN-LSTM network or a prediction method based on the RNN-Markov combination.

4. The intelligent management device for substations according to claim 3, characterized in that: The anomaly detection unit is further configured as: The weight of the deviation rate of electricity consumption in different periods is calculated by using the analytic hierarchy process; The power consumption anomaly rate is calculated based on the weight and power consumption deviation rate, providing a quantitative basis for locating power theft and metering anomalies.

5. The intelligent management device for substations according to claim 4, characterized in that: The calculation method of abnormal power consumption rate is: Abnormal power consumption rate = φ1×n1+φ2×n2+φ3×n3 Among them, φ1 is the power consumption deviation rate of the period before the measured period, φ2 is the power consumption deviation rate of the period before the measured period, φ3 is the power consumption deviation rate of the measured period, and n1, n2, and n3 are the corresponding weights respectively.

6. The intelligent management device for substations according to claim 5, characterized in that: The analytic hierarchy process is used to calculate the weights n1, n2, and n3 of the power consumption deviation rate between the time period to be tested and the two adjacent time periods of the same length before the time period to be tested.

7. The intelligent management device for substations according to claim 1, characterized in that: It also includes an intelligent low-voltage branch monitoring terminal LTU, and the LTU is configured as: Collect voltage, current, power, frequency, electricity and area topology data of low-voltage area branch lines; The collected data is reported to the AI ​​management unit of the station area through the HPLC communication module.

8. The intelligent management device for substations according to claim 7, characterized in that: The LTU specifically includes a controller, a GPS module, a topology transceiver module and an AC data acquisition module; wherein, The controller is used to receive and process the area topology information sent by the topology transceiver module, the power-related data sent by the AC data acquisition module, and the location information sent by the GPS module according to the instructions issued by the area AI management unit, and then send the data to the area AI management unit through the HPLC communication module; The topology transceiver module is used to send and identify characteristic current signals on the low-voltage line, determine the ownership of the branch line according to the characteristic signals, and send the data to the controller; The AC data acquisition module is used to collect and store electricity-related data of the low-voltage branch line according to instructions, and upload the collected data to the controller.

9. The intelligent management device for substations according to claim 7, characterized in that: The line loss analysis module is further configured as follows: Based on the synchronous collection of 96 points of electricity meter data in the entire area, the synchronous line loss calculation at 24:00 / 96:00 every day is realized to find abnormal line loss periods and analyze the user's electricity theft behavior trajectory; Based on the phase identification results of single-phase / three-phase electricity meters and phase measurement data of HPLC, the phase line loss calculation can be realized at 24 / 96 o'clock every day; Calculate branch line loss based on the area topology and LTU of branch nodes; Based on the clustering results of the electric meter, combined with the calculation and analysis of the line loss of the LTU meter box, the meter box with abnormal line loss is located.

10. The intelligent management device for substations according to claim 1, characterized in that: It also includes a power wireless communication data transmission unit DTU, and the DTU is configured as follows: After receiving the data of the area AI management unit via Bluetooth, upload it to the remote management platform via 4G or 5G network; Realize remote communication between the management platform and the area AI management unit.

Images