Industrial internet monitoring system based on networking SVG

Abstract

The invention relates to the technical field of power supply systems, in particular to an industrial internet monitoring system based on a networking SVG (Static Var Generator), which comprises a data transmission module for transmitting power grid data through an industrial internet gateway; the data calculation module is used for selecting a calculation mode according to the power grid data and calculating the power grid data according to the calculation mode to obtain a calculation result; the topology modeling module is used for generating a power distribution network extension diagram according to a calculation result and pushing the power distribution network extension diagram to the network construction SVG control terminal; the topology monitoring module is used for monitoring the state of the power grid according to the calculation result and updating the power distribution network extension diagram according to the monitoring result; and the topology control module is used for performing topology control on the networking SVG equipment according to the monitoring result. According to the invention, through modularized cooperation, power grid data full-process efficient processing is realized, a power distribution network extension diagram is accurately generated and dynamically updated, the state of a power grid can be monitored in real time, network construction SVG equipment can be controlled, and the intelligence and reliability of power grid management are greatly improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power supply systems, in particular to an industrial internet monitoring system based on networked SVG. BACKGROUND

[0002] In the current power grid operation management, there are multiple pain points in data processing, topology visualization, state control and resource allocation, resulting in low monitoring feedback efficiency. On the one hand, the connection between power grid data collection, transmission and calculation is not smooth, which is prone to data processing faults and difficult to provide continuous and effective data support for monitoring. On the other hand, the distribution topology relies on manual drawing or lagging update, and the visualization presentation cannot match the actual structure of the power grid in real time. At the same time, the state monitoring and networked SVG device control lack linkage, abnormal conditions are difficult to respond in time, and the calculation resource allocation is unreasonable, resulting in resource waste or insufficient computing power. Under this background, it is urgent to build an industrial internet monitoring system based on networked SVG to solve the existing difficulties and improve the efficiency of power grid monitoring feedback and management level.

[0003] Chinese patent with publication number CN104375488A discloses a high-voltage SVG remote monitoring system, which includes a local area network, a router, a monitoring computer, a public digital mobile communication system, a high-voltage static reactive power generation device and a wireless communication device. Multiple monitoring computers form a local area network through a router and a network cable. The router is connected with the public digital mobile communication system through a network cable, the public digital mobile communication system is wirelessly connected with the wireless communication device, and the high-voltage static reactive power generation device and the wireless communication device are connected through an RS485 bus to form a field information acquisition unit. However, this scheme still has the problems of power grid data processing fault, topology visualization lag, state monitoring and device control disconnection, and unreasonable calculation resource allocation, resulting in low efficiency of networked SVG monitoring data feedback. SUMMARY

[0004] Therefore, the present application provides an industrial internet monitoring system based on networked SVG to overcome the problems of power grid data processing fault, topology visualization lag, state monitoring and device control disconnection, and unreasonable calculation resource allocation in the prior art, resulting in low efficiency of networked SVG monitoring data feedback.

[0005] To achieve the above-mentioned purpose, the present application provides an industrial internet monitoring system based on networked SVG, which comprises: a data acquisition module for acquiring power grid data; a data transmission module for transmitting power grid data through an industrial internet; a data calculation module for judging the data calculation volume according to the power grid data, selecting a calculation method according to the judgment result, calculating the power grid data according to the calculation method, and obtaining a calculation result. a topology modeling module configured to generate a power distribution topology diagram according to the calculation result and push the power distribution topology diagram to a network construction SVG control terminal; a topology monitoring module configured to monitor a power grid state according to the calculation result and update the power distribution topology diagram according to the monitoring result; a topology control module configured to perform topology control on the network construction SVG device according to the monitoring result.

[0006] Further, the data transmission module receives the power grid data collected by the data collection module and transmits the power grid data to the data calculation module through a wide area network.

[0007] Further, the data calculation module constructs a calculation volume estimation model by a calculation volume estimation model construction method to obtain the calculation volume estimation model, inputs the power grid data into the calculation volume estimation model, and obtains a calculation volume output by the calculation volume estimation model. When the calculation volume judgment result is the light calculation attribute, edge calculation is selected as the calculation method, the power grid data is downloaded to an edge server close to a power grid data collection point, the edge server is used to calculate the contaminated data, and a calculation result is obtained. When the calculation volume judgment result is the heavy calculation attribute, cloud calculation is selected as the calculation method, the power grid data is uploaded to a cloud data center, the cloud data center is used to calculate the power grid data, and a calculation result is obtained.

[0008] Further, the calculation volume estimation model construction method comprises: The calculation volume estimation data set in the power grid data is divided into a 70% estimation training set, a 15% estimation verification set, and a 15% estimation test set. The estimation training set is input into a decision tree model to train the decision tree model. The estimation verification set is input into the trained decision tree to perform hyperparameter iterative optimization on the trained decision tree. The estimation test set is input into the iterative optimization decision tree to perform conflict analysis testing on the iterative optimization decision tree model. An estimation test result is obtained. The total number of estimation test samples is b0, the number of correctly estimated test samples is b, the estimation test accuracy is B, B = b / b0. The estimation test accuracy B is compared with a preset estimation test accuracy B0. According to the comparison result, the training of the iterative optimization decision tree model is judged, and the judgment result is output. Wherein: When B≥B0, the data calculation module determines that the training of the iteratively optimized decision tree model has reached the target, outputs the iteratively optimized decision tree model as the computational volume estimation model, and provides the computational volume estimation model to the layout optimization module. When B < B0, the data calculation module determines that the training of the iteratively optimized decision tree model is not up to standard, updates the computational volume estimation dataset to obtain the updated computational volume estimation dataset, and trains, iteratively optimizes hyperparameters, and analyzes and tests the decision tree model based on the updated computational volume estimation dataset until the training of the decision tree model is up to standard.

[0009] Furthermore, the topology modeling module constructs a distribution topology generation model using a distribution topology generation model construction method, which includes: Step A1: Select a convolutional neural network model as the basic framework for the distribution topology generation model, and divide 70% of the distribution topology generation sample dataset in the power grid data into a topology generation training set, 15% into a topology generation validation set, and 15% into a topology generation test set. Step A2: Train the convolutional neural network model according to the topology generation training set, update the weights of the convolutional neural network model through the backpropagation algorithm, and after each epoch, validate the convolutional neural network model using the topology generation validation set to obtain the topology generation validation loss value and the topology generation validation accuracy. Step A3: When the topology generation verification loss value and the topology generation verification accuracy meet the preset verification conditions, the convolutional neural network model is tested according to the topology generation test set to obtain the topology generation test accuracy. When the topology generation test accuracy reaches the preset accuracy, the convolutional neural network model is output as the distribution topology generation model.

[0010] Furthermore, the topology modeling module extracts the power grid connection relationship information from the calculation results, constructs a real-time connection state logic diagram of the power grid using graph theory and the power grid connection relationship information, inputs the connection state logic diagram into the distribution topology generation model, obtains the distribution network topology diagram output by the distribution topology generation model, and pushes the distribution network topology diagram to the network SVG control terminal.

[0011] Furthermore, the topology monitoring module constructs a threshold database based on preset safety values ​​in the power grid data. The threshold database includes calculation items, thresholds corresponding to the calculation items, and comparison results of the calculation items.

[0012] Furthermore, the topology monitoring module compares the item values ​​of the calculated items in the calculation results with the corresponding thresholds in the threshold database to obtain the comparison results of the calculated items. These comparison results are then output as monitoring results, and the distribution network topology diagram is updated based on the monitoring results. When the value of the calculated item in the calculation result is less than or equal to the threshold corresponding to the calculated item in the threshold database, the topology monitoring module determines that the comparison result of the calculated item is normal and does not update the power distribution network diagram. When the value of a calculated item in the calculation result is greater than the threshold corresponding to the calculated item in the threshold database, the topology monitoring module determines that the comparison result of the calculated item is abnormal, updates the power distribution network topology diagram, and marks the calculated items with abnormal monitoring results in the power distribution network diagram.

[0013] Furthermore, the topology control module constructs the topology scheme model using the topology scheme model construction method to obtain the topology scheme model.

[0014] Furthermore, the topology control module inputs the monitoring results into the topology scheme model, obtains the topology control scheme output by the topology scheme model, and performs topology control on the network SVG devices according to the topology control scheme.

[0015] Compared with existing technologies, the beneficial effects of this invention are as follows: the system accurately collects power grid data through a data acquisition module, providing real and complete basic data support for subsequent data transmission and calculation. The system also relies on an industrial internet gateway through a data transmission module to stably transmit data, ensuring efficient and secure transmission of power grid data and preventing data loss. Furthermore, the system selects calculation methods as needed through a data calculation module to efficiently process power grid data, providing accurate calculation results for topology modeling and monitoring. The system also quickly generates distribution network topology diagrams and pushes them to terminals through a topology modeling module, helping staff intuitively grasp the power grid topology structure. Additionally, the system monitors the power grid status in real time through a topology monitoring module, updating the topology diagram promptly, allowing staff to accurately grasp dynamic changes in the power grid. Finally, the system controls the network SVG equipment based on monitoring results through a topology control module, ensuring the power grid operates in optimal condition and improving power supply stability. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the industrial internet monitoring system based on SVG in this embodiment. Detailed Implementation

[0017] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0018] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0019] It should be noted that in the description of this invention, the terms "upper", "lower", "left", "right", "inner", "outer", etc., which indicate directions or positional relationships, are based on the directions or positional relationships shown in the accompanying drawings. This is only for the convenience of description and is not intended to indicate or imply that the device or element must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, it should not be construed as a limitation of this invention.

[0020] Furthermore, it should be noted that, in the description of this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0021] Please see Figure 1 As shown, this is a schematic diagram of the industrial internet monitoring system based on SVG in this embodiment. The system includes: The data acquisition module is used to collect power grid data; The data transmission module is used to transmit power grid data through the Industrial Internet gateway; The data calculation module is used to determine the data calculation volume based on the power grid data, select the calculation method according to the determination result, calculate the power grid data according to the calculation method, and obtain the calculation result. The topology modeling module is used to generate the power distribution network topology diagram based on the calculation results and push the power distribution network topology diagram to the network SVG control terminal. The topology monitoring module is used to monitor the power grid status based on the calculation results and update the distribution network topology diagram based on the monitoring results. The topology control module is used to perform topology control on the networked SVG devices based on monitoring results.

[0022] Specifically, the industrial internet monitoring system based on SVG (Static Var Generator) is applied to the SVG control terminal. Through modular collaboration, it achieves efficient processing of power grid data throughout the entire process, accurately generates and dynamically updates the distribution network topology, and can monitor the power grid status and control the SVG equipment in real time, significantly improving the intelligence and reliability of power grid management. The system accurately collects power grid data through the data acquisition module, providing real and complete basic data support for subsequent data transmission and calculation. The system also transmits data stably through the data transmission module, relying on the industrial internet gateway, ensuring efficient and secure transmission of power grid data and preventing data loss. The system also processes power grid data efficiently through the data calculation module, selecting calculation methods as needed, providing accurate calculation results for topology modeling and monitoring. The system also quickly generates the distribution network topology and pushes it to the terminal through the topology modeling module, helping staff to intuitively grasp the power grid topology. The system also monitors the power grid status in real time through the topology monitoring module, updating the topology map in a timely manner, allowing staff to accurately grasp the dynamic changes in the power grid. The system also controls the SVG equipment based on the monitoring results through the topology control module, ensuring the power grid operates in the best condition and improving power supply stability.

[0023] Specifically, the power grid data includes electrical data, preset safety values, a calculated volume estimation dataset, and a distribution topology generation sample dataset. The electrical data refers to all operational data generated during power grid operation, such as three-phase voltage and three-phase current. The data acquisition module collects electrical data through sensor clusters set at key nodes of the power grid. Key nodes of the power grid refer to key operational nodes in the power grid system, such as substations, distribution rooms, and line switches. The sensor cluster refers to a set of sensors used to collect electrical data, such as voltage transformers and current transformers. The preset safety values ​​are preset values ​​used to judge the comparison results of the calculated items. Thresholds corresponding to the calculated items can be set based on the experience of power grid experts. The data acquisition module obtains the preset safety values ​​by having staff input them in advance. The data acquisition module also collects the calculated volume estimation dataset and the distribution topology generation sample dataset through manual input.

[0024] Specifically, the data transmission module uses the industrial internet gateway as the aggregation point of the local network, receives the power grid data collected by the data acquisition module, and transmits the power grid data to the data computing module through the wide area network.

[0025] Specifically, the Industrial Internet gateway is a hardware device deployed in an industrial site, which has multiple interfaces and protocol conversion capabilities. It is responsible for converting local heterogeneous industrial equipment data into a unified format and sending it to a remote server through an uplink network. The wide area network refers to a computer network that spans a wide geographical area.

[0026] Specifically, the data calculation module constructs a computational volume estimation model using a computational volume estimation model construction method, obtains the computational volume estimation model, inputs the power grid data into the computational volume estimation model, and obtains the computational volume output by the computational volume estimation model. The computational volume judgment result includes lightweight computational attributes and weight computational attributes, wherein: When the calculation volume assessment result is a lightweight calculation attribute, edge computing is selected as the calculation method. The power grid data is sent to the edge server close to the power grid data acquisition point. The edge server calculates the contaminated data and obtains the calculation result. When the volume calculation result is a weight calculation attribute, cloud computing is selected as the calculation method. The power grid data is uploaded to the cloud data center, and the power grid data is calculated through the cloud data center to obtain the calculation result.

[0027] Specifically, the computational volume refers to the size and complexity of the data processing task, which is determined by the amount of data, computational complexity, and real-time requirements.

[0028] Specifically, the method for constructing the computational volume estimation model includes: The computational volume estimation dataset from the power grid data is divided into a 70% estimation training set, a 15% estimation validation set, and a 15% estimation test set. The estimation training set is input into a decision tree model for training. The estimation validation set is input into the trained decision tree for iterative hyperparameter optimization. The estimation test set is input into the optimized decision tree for conflict analysis testing, yielding estimation test results. The total number of samples in the estimation test set is set as b0, the number of correctly estimated test samples is b, and the estimation test accuracy is B, where B = b / b0. The estimation test accuracy B is compared with the preset estimation test accuracy B0. Based on the comparison results, the training status of the iteratively optimized decision tree model is judged, and the judgment result is output. Where: When B≥B0, the data calculation module determines that the training of the iteratively optimized decision tree model has reached the target, outputs the iteratively optimized decision tree model as the computational volume estimation model, and provides the computational volume estimation model to the layout optimization module. When B < B0, the data calculation module determines that the training of the iteratively optimized decision tree model is not up to standard, updates the computational volume estimation dataset to obtain the updated computational volume estimation dataset, and trains, iteratively optimizes hyperparameters, and analyzes and tests the decision tree model based on the updated computational volume estimation dataset until the training of the decision tree model is up to standard.

[0029] Specifically, the computational volume estimation model refers to a decision tree model that takes power grid data as input and outputs computational volume. The computational volume estimation dataset includes historically acquired power grid data and the corresponding computational volumes. The estimation training set refers to a subset of the computational volume estimation dataset used to train the decision tree model. The estimation validation set refers to a subset of the computational volume estimation dataset used to validate the training results of the decision tree model. The estimation test set refers to a subset of the computational volume estimation dataset used to test the decision tree model. The decision tree model refers to the basic architecture used to construct the computational volume estimation model. The estimation test accuracy... This refers to the ratio of the correctly estimated number of test samples to the estimated total number of test samples. The preset estimated test accuracy rate is a preset value used to judge the training achievement status of the iteratively optimized decision tree model. This embodiment does not limit the preset estimated test accuracy rate. Those skilled in the art can set it according to actual needs, as long as it meets the requirements for judging the training achievement status of the iteratively optimized decision tree model. For example, the preset estimated test accuracy rate B0 can be set to: 96%≤B0≤97%. The training achievement status of the iteratively optimized decision tree model refers to the accuracy achievement status of the trained decision tree model. The training achievement status of the iteratively optimized decision tree model includes achievement and non-achievement.

[0030] Specifically, the topology modeling module constructs a distribution topology generation model using a distribution topology generation model construction method, which includes: Step A1: Select a convolutional neural network model as the basic framework for the distribution topology generation model, and divide 70% of the distribution topology generation sample dataset in the power grid data into a topology generation training set, 15% into a topology generation validation set, and 15% into a topology generation test set. Step A2: Train the convolutional neural network model according to the topology generation training set, update the weights of the convolutional neural network model through the backpropagation algorithm, and after each epoch, validate the convolutional neural network model using the topology generation validation set to obtain the topology generation validation loss value and the topology generation validation accuracy. Step A3: When the topology generation verification loss value and the topology generation verification accuracy meet the preset verification conditions, the convolutional neural network model is tested according to the topology generation test set to obtain the topology generation test accuracy. When the topology generation test accuracy reaches the preset accuracy, the convolutional neural network model is output as the distribution topology generation model.

[0031] Specifically, the distribution topology generation sample dataset refers to a learning dataset stored in the form of a connection state logic diagram-distribution topology diagram, used for training a convolutional neural network model. The validation loss value refers to the loss value of the loss function in the convolutional neural network model when the topology generation validation set is input into the convolutional neural network model for validation. The epoch refers to the process of the convolutional neural network model completing one forward and backward propagation on the topology generation training set. The validation accuracy refers to the difference between the distribution topology diagram output by the convolutional neural network model after inputting the topology generation validation set into the convolutional neural network model and the distribution topology diagram in the topology generation validation set. Figure 1 The ratio of the number of identical topology topology diagrams to the total number of samples in the topology generation validation set. The preset validation condition refers to the validation loss value not decreasing and the validation accuracy not improving for 8 consecutive epochs. The analytical test accuracy refers to the ratio of the number of distribution network diagrams output by the topology generation test set after inputting it into the convolutional neural network model to the total number of samples in the topology generation test set. The preset accuracy refers to the preset value of the topology generation test accuracy for judging that the convolutional neural network model has reached the output standard, such as setting the preset accuracy to 98%.

[0032] Specifically, the topology modeling module extracts the power grid connection relationship information from the calculation results, constructs a real-time connection state logic diagram of the power grid using graph theory and the power grid connection relationship information, inputs the connection state logic diagram into the distribution topology generation model, obtains the distribution network topology diagram output by the distribution topology generation model, and pushes the distribution network topology diagram to the network SVG control terminal.

[0033] Specifically, the power grid connection relationship refers to the physical electrical connection method between various electrical devices in the power grid. The graph theory is an independent branch of mathematics, belonging to the category of discrete mathematics. It systematically studies the mathematical structure, properties, attributes of graphs and the relationships between them. The real-time connection status diagram of the power grid refers to a logical model and visual graphic constructed in a computer based on the power grid connection relationship and using graph theory, which can dynamically reflect the actual wiring method of the power grid at the current moment. This embodiment does not limit the specific implementation method of pushing the distribution network topology diagram to the network SVG control terminal. Those skilled in the art can set it according to the actual situation, such as pushing the distribution network topology diagram to the network SVG control terminal through wireless signal transmission.

[0034] Specifically, the topology monitoring module constructs a threshold database based on preset safety values ​​in the power grid data. The threshold database includes calculation items, thresholds corresponding to the calculation items, and comparison results of the calculation items.

[0035] Specifically, the calculation item refers to the item in the calculation result, the threshold corresponding to the calculation item refers to the preset value used to judge the comparison result of the calculation item, and the threshold corresponding to the calculation item can be set according to the experience of power grid experts. The comparison result of the calculation item refers to the comparison result obtained after comparing the calculation result with the threshold corresponding to the calculation item. The comparison result of the calculation item includes normal and abnormal.

[0036] Specifically, the topology monitoring module compares the item values ​​of the calculated items in the calculation results with the corresponding thresholds in the threshold database to obtain the comparison results. These comparison results are then output as monitoring results, and the distribution network topology diagram is updated based on these monitoring results. When the value of the calculated item in the calculation result is less than or equal to the threshold corresponding to the calculated item in the threshold database, the topology monitoring module determines that the comparison result of the calculated item is normal and does not update the power distribution network diagram. When the value of a calculated item in the calculation result is greater than the threshold corresponding to the calculated item in the threshold database, the topology monitoring module determines that the comparison result of the calculated item is abnormal, updates the power distribution network topology diagram, and marks the calculated items with abnormal monitoring results in the power distribution network diagram.

[0037] Specifically, the topology control module constructs the topology scheme model using the topology scheme model construction method to obtain the topology scheme model.

[0038] Specifically, this embodiment does not limit the specific implementation method of the topology scheme model construction method. Those skilled in the art can set it according to the actual situation, such as training the machine learning model through the scheme dataset to obtain the topology scheme model. The scheme dataset refers to the training data set used to construct the topology scheme model. The scheme dataset includes historically acquired monitoring results and the topology control schemes corresponding to the historically acquired monitoring results.

[0039] Specifically, the topology control module inputs the monitoring results into the topology scheme model, obtains the topology control scheme output by the topology scheme model, and performs topology control on the network SVG devices according to the topology control scheme.

[0040] Specifically, this embodiment does not limit the specific implementation method of topology control of the network SVG device according to the topology control scheme. Those skilled in the art can set it according to the actual situation, such as using PID control method to perform topology control of the network SVG device according to the topology control scheme.

[0041] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

Claims

1. An industrial internet monitoring system based on SVG (Static Variational Object) architecture, characterized in that, include: The data acquisition module is used to collect power grid data; The data transmission module is used to transmit power grid data through the Industrial Internet gateway; The data calculation module is used to determine the data calculation volume based on the power grid data, select the calculation method according to the determination result, calculate the power grid data according to the calculation method, and obtain the calculation result. The topology modeling module is used to generate the power distribution network topology diagram based on the calculation results and push the power distribution network topology diagram to the network SVG control terminal. The topology monitoring module is used to monitor the power grid status based on the calculation results and update the distribution network topology diagram based on the monitoring results. The topology control module is used to perform topology control on the networked SVG devices based on monitoring results.

2. The industrial internet monitoring system based on SVG network according to claim 1, characterized in that, The data transmission module uses the industrial internet gateway as the aggregation point of the local network, receives power grid data collected by the data acquisition module, and transmits the power grid data to the data computing module through the wide area network.

3. The industrial internet monitoring system based on SVG network according to claim 2, characterized in that, The data calculation module constructs a computational volume estimation model using a computational volume estimation model construction method, obtains the computational volume estimation model, and inputs the power grid data into the computational volume estimation model to obtain the computational volume output by the model. The computational volume judgment result includes lightweight computational attributes and weight computational attributes, wherein: When the calculation volume assessment result is a lightweight calculation attribute, edge computing is selected as the calculation method. The power grid data is sent to the edge server close to the power grid data acquisition point. The edge server calculates the contaminated data and obtains the calculation result. When the volume calculation result is a weight calculation attribute, cloud computing is selected as the calculation method. The power grid data is uploaded to the cloud data center, and the power grid data is calculated through the cloud data center to obtain the calculation result.

4. The industrial internet monitoring system based on SVG network according to claim 3, characterized in that, The method for constructing the computational volume estimation model includes: The computational volume estimation dataset from the power grid data is divided into a 70% estimation training set, a 15% estimation validation set, and a 15% estimation test set. The estimation training set is input into a decision tree model for training. The estimation validation set is input into the trained decision tree for iterative hyperparameter optimization. The estimation test set is input into the optimized decision tree for conflict analysis testing, yielding estimation test results. The total number of samples in the estimation test set is set as b0, the number of correctly estimated test samples is b, and the estimation test accuracy is B, where B = b / b0. The estimation test accuracy B is compared with the preset estimation test accuracy B0. Based on the comparison results, the training status of the iteratively optimized decision tree model is judged, and the judgment result is output. Where: When B≥B0, the data calculation module determines that the training of the iteratively optimized decision tree model has reached the target, outputs the iteratively optimized decision tree model as the computational volume estimation model, and provides the computational volume estimation model to the layout optimization module. When B < B0, the data calculation module determines that the training of the iteratively optimized decision tree model is not up to standard, updates the computational volume estimation dataset to obtain the updated computational volume estimation dataset, and trains, iteratively optimizes hyperparameters, and analyzes and tests the decision tree model based on the updated computational volume estimation dataset until the training of the decision tree model is up to standard.

5. The industrial internet monitoring system based on SVG network according to claim 4, characterized in that, The topology modeling module constructs a distribution topology generation model using a distribution topology generation model construction method, which includes: Step A1: Select a convolutional neural network model as the basic framework for the distribution topology generation model, and divide 70% of the distribution topology generation sample dataset in the power grid data into a topology generation training set, 15% into a topology generation validation set, and 15% into a topology generation test set. Step A2: Train the convolutional neural network model according to the topology generation training set, update the weights of the convolutional neural network model through the backpropagation algorithm, and after each epoch, validate the convolutional neural network model using the topology generation validation set to obtain the topology generation validation loss value and the topology generation validation accuracy. Step A3: When the topology generation verification loss value and the topology generation verification accuracy meet the preset verification conditions, the convolutional neural network model is tested according to the topology generation test set to obtain the topology generation test accuracy. When the topology generation test accuracy reaches the preset accuracy, the convolutional neural network model is output as the distribution topology generation model.

6. The industrial internet monitoring system based on SVG network according to claim 5, characterized in that, The topology modeling module extracts the power grid connection relationship information from the calculation results, constructs a real-time connection state logic diagram of the power grid using graph theory and the power grid connection relationship information, inputs the connection state logic diagram into the distribution topology generation model, obtains the distribution network topology diagram output by the distribution topology generation model, and pushes the distribution network topology diagram to the network SVG control terminal.

7. The industrial internet monitoring system based on SVG network according to claim 6, characterized in that, The topology monitoring module constructs a threshold database based on preset safety values ​​in the power grid data. The threshold database includes calculation items, thresholds corresponding to the calculation items, and comparison results of the calculation items.

8. The industrial internet monitoring system based on SVG network according to claim 7, characterized in that, The topology monitoring module compares the item values ​​of the calculated items in the calculation results with the corresponding thresholds in the threshold database to obtain the comparison results. These comparison results are then output as monitoring results, and the distribution network topology diagram is updated based on these results. When the value of the calculated item in the calculation result is less than or equal to the threshold corresponding to the calculated item in the threshold database, the topology monitoring module determines that the comparison result of the calculated item is normal and does not update the power distribution network diagram. When the value of a calculated item in the calculation result is greater than the threshold corresponding to the calculated item in the threshold database, the topology monitoring module determines that the comparison result of the calculated item is abnormal, updates the power distribution network topology diagram, and marks the calculated items with abnormal monitoring results in the power distribution network diagram.

9. The industrial internet monitoring system based on SVG network according to claim 8, characterized in that, The topology control module constructs the topology scheme model using the topology scheme model construction method to obtain the topology scheme model.

10. The industrial internet monitoring system based on SVG network according to claim 9, characterized in that, The topology control module inputs the monitoring results into the topology scheme model, obtains the topology control scheme output by the topology scheme model, and performs topology control on the network SVG devices according to the topology control scheme.

Images