Electric power Internet of Things sensor data processing method

Abstract

The invention relates to the technical field of energy data processing, in particular to an electric power Internet of Things sensor data processing method, which comprises the following steps: a sensor transmits real-time data to a nearest edge computing node; the edge computing node processes the real-time data to obtain first data and sends the first data to a data center; and the data center processes the first data to obtain final decision data and sends the final decision data to the power network control system. According to the method, the computing resources of the edge computing nodes and the data center are reasonably distributed, and the processing pressure of the data center is relieved by the edge computing nodes, so that the data center can concentrate on more complex analysis tasks, the processing efficiency is improved, and the real-time performance of data processing is ensured. Meanwhile, long-term analysis and prediction can be carried out, and safe and stable operation of the power grid is further guaranteed.

Description

Technical Field

[0001] The present application relates to the technical field of energy data processing, and in particular to a method for processing data of power Internet of Things sensors. Background Art

[0002] Traditional central processing methods face problems such as processing delays, bandwidth limitations, and insufficient processing capacity of data centers when processing data in power networks, making it difficult to meet the demand for real-time data processing. Edge computing provides a viable solution. The core of edge computing is to transfer data processing tasks from the central data center to the edge of the network, that is, closer to the data source. This approach can significantly reduce data transmission delays and increase data processing speed, making it more suitable for scenarios that require real-time analysis and rapid response, such as monitoring and analyzing data in power networks.

[0003] However, in actual applications, especially in the field of power networks, when processing large-scale, high-frequency sensor data, there is an unreasonable allocation of computing resources in edge computing nodes, which leads to the inability to quickly complete data preprocessing and analysis when faced with massive data, and there are problems such as low data processing efficiency and lack of real-time performance.

[0004] Therefore, a new solution is urgently needed to overcome the problems of the existing technology and improve the efficiency and real-time performance of data processing. Summary of the invention

[0005] In view of this, it is necessary to provide a data processing method for power Internet of Things sensors to solve the problems of low data processing efficiency and lack of real-time performance caused by unreasonable allocation of computing resources in edge computing nodes.

[0006] A method for processing data of a power Internet of Things sensor, comprising: Sensors transmit real-time data to the nearest edge computing node; The edge computing node processes the real-time data, obtains the first data and sends it to the data center; The data center processes the first data to obtain final decision data, and sends the final decision data to the power network control system.

[0007] Preferably, the data center processes the first data to obtain final decision data, including: The data center processes the first data through a machine learning model to obtain final decision data; wherein the final decision data includes load balancing instruction data and maintenance scheduling instruction data.

[0008] Preferably, the machine learning model includes: Linear regression models, decision tree and random forest models, support vector machines, deep neural networks, and clustering algorithms.

[0009] Preferably, the edge computing node processes the real-time data, further comprising: obtaining second data, the second data comprising sensor abnormality data and alarm information; The edge computing node sends the second data to the data center and the operation and maintenance center; the data center and the operation and maintenance center maintain the abnormal sensor according to the second data.

[0010] Preferably, the data center sends the first data and the final decision data to a database for storage as historical data.

[0011] Preferably, the data center obtains final decision data based on the first data and the historical data.

[0012] Preferably, the edge computing node sends the first data to the data center after a preset time interval.

[0013] Preferably, access control is used in the data processing process, and data encryption is used in the data transmission process.

[0014] Compared with the prior art, the beneficial effect of the present application is that when executing the above-mentioned power Internet of Things sensor data processing method, especially when processing large-scale, high-frequency sensor data, the computing resources of edge computing nodes, data centers, etc. are reasonably allocated, and the edge computing nodes reduce the processing pressure of the data center, allowing it to focus on more complex analysis tasks. The processing efficiency is improved and the real-time nature of data processing is guaranteed. At the same time, long-term analysis and prediction can also be carried out, further ensuring the safe and stable operation of the power grid. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is a flow chart of a method for processing electric power Internet of Things sensor data provided in an embodiment of the present application.

[0016] Figure 2 It is a schematic diagram of the power network structure provided in an embodiment of the present application.

[0017] Figure 3 It is a schematic diagram of the data processing flow provided in the embodiment of the present application.

[0018] Figure 4 It is a schematic diagram of the data center in-depth analysis process provided in an embodiment of the present application. DETAILED DESCRIPTION

[0019] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0020] See also Figures 1 to 3 , Figure 1 It is a flow chart of a method for processing data of a power Internet of Things sensor provided in an embodiment of the present application; Figure 2 It is a schematic diagram of the power network structure provided by an embodiment of the present application; Figure 3 The data processing flow diagram provided by the embodiment of the present application is shown in FIG. The data processing method of the power Internet of Things sensor includes the following steps: S101: The sensor transmits real-time data to the nearest edge computing node.

[0021] Sensors are usually deployed at key nodes of the power network, such as substations and important distribution points, and can measure and record current, voltage, frequency and temperature. ， Each sensor may have certain data calculation, processing and data communication capabilities. Different nodes may need to focus on different types of monitoring. For example, a substation may only need to monitor voltage, and the sensor here may be a voltage sensor. Therefore, the sensor can be a single monitoring type sensor or a sensor with multiple monitoring capabilities.

[0022] Since sensors have certain data calculation and processing capabilities, when they send real-time data to the nearest edge computing node, they can perform preliminary data preprocessing, such as data formatting, normalization, and preliminary anomaly detection. Among them, the preliminary anomaly detection of real-time data can be to preset a threshold for the data monitored by the sensor. When the monitored data exceeds the threshold, it indicates that the data is abnormal. The sensor can send an alarm message to the nearest edge computing node. At the same time, the sensor can also send an alarm message to the operation and maintenance center, so that the operation and maintenance personnel can quickly take maintenance measures to ensure the safety and stability of the system.

[0023] When the sensor transmits data, data encryption can also be performed to ensure the accuracy and security of the data.

[0024] S102: The edge computing node processes the real-time data, obtains first data, and sends it to the data center.

[0025] Each edge computing node processes the received real-time data to obtain the first data. Among them, the edge computing node processes the received real-time data including data cleaning, preliminary trend analysis and anomaly detection to obtain the first data. Through data cleaning, such as removing noise, the quality of subsequent preliminary trend analysis can be improved. Through real-time preliminary trend analysis based on real-time data, anomalies and potential problems can be quickly identified, which helps to take timely measures. For example, if the voltage or current in the power system suddenly deviates from the normal range, such as the voltage drops from the standard 220V±5% range to below 200V or rises to above 240V, or the current increases abnormally under non-high load conditions, this will be identified as an abnormality. Similarly, significant fluctuations in frequency, such as deviations from the standard 50Hz by more than 0.5Hz, may indicate an unbalanced grid load or a large equipment failure. In addition, if the device temperature recorded by the sensor rises rapidly to an abnormal value, such as the transformer temperature soaring from the normal 40°C to 60°C in a short period of time, it may be a precursor to an overheating failure. In addition, in terms of data continuity, if critical data is discontinuous or lost, the node will immediately alarm, as this may indicate a sensor failure or data transmission problem.

[0026] Furthermore, the edge computing node processes the received real-time data and obtains second data, which includes alarm information of abnormalities detected through the above-mentioned preliminary trend analysis. If the edge computing node detects an abnormality after preliminary trend analysis, it will immediately generate an alarm message and send the second data containing the alarm information to the data center. The second data can also be sent to the data center and the operation and maintenance center at the same time, which effectively improves the rapid coordination between the various levels of the power grid, and the operation and maintenance center can also respond quickly and maintain abnormal sensors.

[0027] Furthermore, the edge computing node sends the first data to the data center after a preset time interval. If the result obtained from the preliminary trend analysis is normal, or the error is within the normal range, the preset time interval or frequency for the edge computing node to send data to the data center can be set. Excessive occupation of communication bandwidth and computing power of the data center can be avoided. In this case, if an edge node reports frequently, the central data center may recommend adjusting the data filter or enhancing the sensitivity of the anomaly detection algorithm to reduce false positives and improve accuracy.

[0028] S103: The data center processes the first data to obtain final decision data, and sends the final decision data to the power network control system.

[0029] Please refer to Figure 4 , Figure 4 It is a schematic diagram of the data center in-depth analysis process provided in an embodiment of the present application.

[0030] The first data is processed by the machine learning model to obtain the final decision data, which includes load balancing instruction data and maintenance scheduling instruction data, etc., to identify long-term trends and potential problems. The data center generates the final decision data based on the in-depth analysis results, and sends these instructions back to the corresponding edge computing nodes or directly to the power grid control system according to the needs. These instructions are precisely adjusted and optimized for the current network status and forecast data.

[0031] Specifically, the data center receives the data sent by the edge computing node, and after verification, it enters multi-layer analysis. The first is basic data processing, such as normalization and denoising. Then, through machine learning models, advanced data analysis and trend prediction are performed. For example, when data analysis reveals that there may be an imbalance in power supply and demand in a certain area or time period, the data center will send a load balancing instruction. These instructions are designed to balance the load of the entire power grid by adjusting the power output of each part of the power grid or initiating demand-side management measures to ensure the stable operation of the power system.

[0032] Demand response measures are activated when peak loads or increased system stress are predicted. This includes reducing electricity consumption for non-critical purposes or encouraging users to reduce electricity consumption by implementing dynamic electricity pricing policies to reduce the burden on the grid.

[0033] Maintenance scheduling instructions: When long-term data analysis shows that certain equipment or lines have performance degradation or potential failure risks, the data center will generate scheduling instructions for maintenance or replacement of equipment. These instructions ensure that maintenance work on the power system can be planned and implemented in advance, thus avoiding sudden failures and service interruptions.

[0034] Machine learning models that can be used in data centers include: linear regression models, decision trees and random forest models, support vector machines, deep neural networks, and clustering algorithms. Among them, linear regression models can be used to analyze and predict the load demand of the power grid, and predict possible future load changes based on past trends; decision trees and random forest models can be used to identify and classify different states in power grid operation, such as the transition from normal operation to unstable state; support vector machines (SVM) can be used to accurately classify abnormal conditions in power grid data and help identify data patterns that may cause system failures; deep neural networks can be used to process complex data structures, analyze time series data, and identify long-term trends and periodic changes in the power grid; clustering algorithms can analyze data in an unsupervised learning environment and identify natural groups or patterns in power grid operation to optimize the response strategy and operation and maintenance of the power grid.

[0035] Different models are suitable for different types of problems, and the most appropriate model can be selected for analysis according to the actual situation. At the same time, multiple machine learning models can be used in combination to analyze data from different angles and improve the comprehensiveness and accuracy of decision-making.

[0036] Furthermore, the data center sends the first data and the final decision data to the database for storage as historical data. The data center obtains the final decision data based on the first data and the historical data. All collected and analyzed data can be stored regularly in a secure database to achieve data management and long-term trend analysis. In addition, it is also conducive to the training and updating of machine learning models, to achieve continuous monitoring and self-optimization of data centers and power grid systems, to learn and predict changes in power loads within a specific period of time, to automatically adjust power distribution or to warn of potential system pressures, so that quick and effective measures can be taken when problems arise.

[0037] Furthermore, strict security measures are implemented throughout the entire data processing process, including data encryption during data transmission, access control and continuous monitoring during data processing, to ensure data privacy and compliance with relevant regulatory requirements.

Claims

1. A method for processing data of electric power Internet of Things sensors, characterized in that: include: Sensors transmit real-time data to the nearest edge computing node; The edge computing node processes the real-time data to obtain first data, and sends the first data to the data center; The data center processes the first data to obtain final decision data, and sends the final decision data to the power network control system.

2. The method according to claim 1, characterized in that: The data center processes the first data to obtain final decision data, including: The data center processes the first data through a machine learning model to obtain final decision data; wherein the final decision data includes load balancing instruction data and maintenance scheduling instruction data.

3. The method according to claim 2, characterized in that: The machine learning model includes: Linear regression models, decision tree and random forest models, support vector machines, deep neural networks, and clustering algorithms.

4. The method according to claim 1, characterized in that: The edge computing node processes the real-time data, further comprising: obtaining second data, the second data comprising sensor abnormality data and alarm information; The edge computing node sends the second data to the data center and the operation and maintenance center; the data center and the operation and maintenance center maintain the abnormal sensor according to the second data.

5. The method according to claim 1, characterized in that: include: The data center sends the first data and the final decision data to a database for storage as historical data.

6. The method according to claim 5, characterized in that: include: The data center obtains the final decision data according to the first data and the historical data.

7. The method according to claim 1, characterized in that: include: The edge computing node sends the first data to the data center after a preset time interval.

8. The method according to any one of claims 1 to 7, characterized in that: Access control is used in the data processing process, and data encryption is used in the data transmission process.

Images