Power grid real-time monitoring and early warning system using cloud technology

Abstract

The invention provides a power grid real-time monitoring and early warning system using a cloud technology. The system comprises an edge data acquisition preprocessing module, a cloud edge data synchronization consistency module, a heterogeneous data fusion analysis module and a real-time early warning response module. The edge device end collects data of various sensors in a power grid through an edge data collecting and preprocessing module and preprocesses the data. The edge device end also copies and synchronizes the state of the preprocessed data among the nodes through the cloud edge data synchronization consistency module; the edge device end carries out fusion processing on data of different data sources through a heterogeneous data fusion analysis module; and the cloud end carries out monitoring and early warning on the fused data and the real-time sensor data through a real-time early warning response module. Real-time monitoring and early warning of the power grid are achieved through the cloud technology, the consistency of edge equipment and cloud data is ensured, heterogeneous data can be effectively processed, and the safety and reliability of power grid operation are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of power grid early warning, and in particular relates to a power grid real-time monitoring and early warning system using cloud technology. Background Art

[0002] With the rapid development of information technology, cloud technology, as an innovative form of information processing, is changing the way traditional data processing and service provision are provided. The core of cloud technology is to decompose large computing tasks into many small subroutines through the network. These subroutines are executed by distributed server clusters and eventually return the processing results to users. This process greatly simplifies the operational needs of end users - only a device that can connect to the Internet is needed to access powerful cloud computing capabilities and services. Cloud computing is a service model developed on this basis, which allows users to obtain IT infrastructure, software platforms and even business processes in an on-demand and easy-to-expand manner. Users can enjoy convenient and efficient cloud services without having a deep understanding of the details of the underlying architecture or having advanced technical knowledge.

[0003] In the power industry, especially for real-time monitoring systems for power grids, traditional monolithic architectures face a series of challenges. First, in the process of ensuring efficient data synchronization between edge devices and the central cloud platform, how to effectively solve potential data delays and loss issues has become a technical problem that needs to be solved urgently; second, since the various monitoring devices used in the power grid often come from different manufacturers and have diverse product specifications and technical parameters (such as different data formats, large differences in accuracy, and different sampling frequencies), it makes it extremely difficult to effectively integrate heterogeneous data, thereby affecting the ability to accurately analyze and respond to the entire power grid status in a timely manner. Summary of the invention

[0004] In view of this, the present invention provides a real-time monitoring and early warning system for power grids using cloud technology, which aims to utilize the distributed storage and parallel processing capabilities in a cloud computing environment to optimize data flow management and improve information transmission efficiency. At the same time, it uses advanced big data analysis algorithms to achieve intelligent integration and deep mining of heterogeneous data generated by different types of equipment, thereby supporting more accurate and reliable power grid status assessment and fault prediction functions.

[0005] In order to achieve the above object, the technical solution provided by the present invention is as follows:

[0006] A real-time monitoring and early warning system for power grids using cloud technology, comprising:

[0007] The edge data collection and preprocessing module, cloud-edge data synchronization consistency module and heterogeneous data fusion analysis module deployed on the edge device side and the real-time warning response module deployed on the cloud side;

[0008] The edge device collects and preprocesses data from various sensors in the power grid through the edge data acquisition and preprocessing module;

[0009] The edge device also replicates and synchronizes the status of pre-processed data between nodes through the cloud-edge data synchronization consistency module to ensure that all nodes between the edge device and the cloud platform have the same data view;

[0010] The edge device also integrates data from different data sources through the heterogeneous data fusion analysis module;

[0011] The cloud monitors and issues warnings on the fused data and real-time sensor data through the real-time warning response module.

[0012] Further, the edge data acquisition preprocessing module includes: an edge device data acquisition unit and a data preprocessing unit;

[0013] The edge device data acquisition unit is used to collect data from various sensors in the power grid using a dynamic sampling algorithm, which defines the relationship between different data change speeds and corresponding sampling rates;

[0014] The data preprocessing unit is used to preprocess the collected data in order to improve the data quality.

[0015] Furthermore, in the dynamic sampling algorithm, the data change speed and the sampling rate satisfy the following relationship:

[0016]

[0017] In the formula, f s is the adjusted sampling frequency, f base is the basic sampling frequency, α is the adjustment coefficient, x t and are the data values ​​of the current and previous sampling points, respectively, x max and x min are the maximum and minimum values ​​of the data.

[0018] Further, the cloud-edge data synchronization consistency module includes: a data synchronization control unit and a data buffer retransmission mechanism unit;

[0019] The data synchronization control unit is used to replicate and synchronize data status between nodes using a distributed consistency algorithm;

[0020] The data buffer retransmission mechanism unit is used to trigger a retransmission mechanism when data is lost or delayed, so as to retransmit the corresponding data.

[0021] Furthermore, the data state is determined as follows:

[0022]

[0023] Where V(t) represents the data state vector of all nodes at time t, ΔV i (t) represents the amount of data state change of node i at time t, and n represents the total number of nodes.

[0024] Furthermore, in the retransmission mechanism, the retransmission timeout threshold is determined according to the following formula:

[0025]

[0026] Where, T timeout is the timeout threshold, T base is the basic timeout, γ is the adjustment factor, RTT is the round-trip time, min is the minimum round trip time.

[0027] Furthermore, the heterogeneous data fusion analysis module includes: a data standardization unit and a fusion analysis unit;

[0028] The data standardization unit is used to standardize the data;

[0029] The fusion analysis unit is used to fuse data from different data sources and obtain comprehensive analysis results based on the fused data.

[0030] Furthermore, the data from different data sources are fused and processed, and comprehensive analysis results are obtained based on the fused data, including:

[0031] Based on the information from the current data source, Bayesian reasoning is used to determine the current operating status of the power grid;

[0032] Calculate the correlation weight between each current data source and the current operating state according to the relationship between different data sources and different operating states;

[0033] The evaluation results of all current data sources are integrated with the relevance weights to obtain a comprehensive evaluation result.

[0034] Furthermore, the real-time warning response module includes: a real-time warning unit and an automatic response unit;

[0035] The real-time warning unit is used to predict the state of the power grid based on the real-time sensor data, compare the predicted state result with the comprehensive evaluation result, and issue a warning signal based on the comparison result;

[0036] The automatic response unit is used to execute corresponding emergency response measures according to predefined rules after receiving the early warning signal.

[0037] Furthermore, it also includes an edge intelligent deployment management module, which includes: an intelligent algorithm deployment unit and an edge intelligent management unit;

[0038] The intelligent algorithm deployment unit is used to deploy the intelligent algorithm to the edge device to realize localized computing;

[0039] The edge intelligence management unit is used to monitor and manage intelligent algorithms on edge devices.

[0040] In summary, the present invention provides a real-time monitoring and early warning system for power grids using cloud technology, including: an edge data acquisition and preprocessing module deployed on the edge device side, a cloud-edge data synchronization consistency module and a heterogeneous data fusion analysis module, and a real-time early warning response module deployed on the cloud side; the edge device side collects and preprocesses the data of various sensors in the power grid through the edge data acquisition and preprocessing module; the edge device side also copies and synchronizes the status of the preprocessed data between each node through the cloud-edge data synchronization consistency module to ensure that all nodes between the edge device and the cloud platform have the same data view; the edge device side also fuses the data of different data sources through the heterogeneous data fusion analysis module; the cloud side monitors and warns the fused data and real-time sensor data through the real-time early warning response module. The present invention realizes real-time monitoring and early warning of power grids through cloud technology, improves the efficiency and accuracy of data acquisition, synchronization, fusion analysis and early warning response, ensures the consistency of edge device and cloud data, and can effectively process heterogeneous data, improving the safety and reliability of power grid operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] 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.

[0042] Figure 1 A schematic diagram of the architecture of a power grid real-time monitoring and early warning system using cloud technology provided by an embodiment of the present invention;

[0043] Figure 2 A flowchart of a method for constructing a real-time monitoring and early warning system for a power grid provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0044] In order to make the purpose, features and advantages of the present invention more obvious and easy to understand, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described below are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0045] See also Figure 1 , this embodiment provides a real-time monitoring and early warning system for a power grid using cloud technology, including:

[0046] The edge data collection and preprocessing module, cloud-edge data synchronization consistency module and heterogeneous data fusion analysis module deployed on the edge device side and the real-time warning response module deployed on the cloud side;

[0047] The edge device collects and preprocesses data from various sensors in the power grid through the edge data acquisition and preprocessing module;

[0048] The edge device also replicates and synchronizes the status of pre-processed data between nodes through the cloud-edge data synchronization consistency module to ensure that all nodes between the edge device and the cloud platform have the same data view;

[0049] The edge device also integrates data from different data sources through the heterogeneous data fusion analysis module;

[0050] The cloud monitors and issues warnings on the fused data and real-time sensor data through the real-time warning response module.

[0051] It should be noted that edge devices refer to devices located at the power grid site, such as smart meters and sensors. These devices are directly connected to the physical components in the power grid and are responsible for collecting data and performing preliminary processing.

[0052] The edge data collection preprocessing module is a software or hardware component running on the edge device, which is used to collect data from various sensors and perform preliminary cleaning and formatting of the data for subsequent processing.

[0053] The cloud-edge data synchronization consistency module ensures data consistency and synchronization between edge devices and the cloud, as well as between different edge devices. It achieves this by replicating and synchronizing data states.

[0054] The heterogeneous data fusion analysis module integrates data from different sources (such as different manufacturers and different models of equipment) and in different formats for unified analysis.

[0055] The real-time early warning response module is a system component deployed in the cloud. It is responsible for monitoring the processed data, issuing alarms according to preset rules, and triggering corresponding response measures.

[0056] When the system proposed in this embodiment is implemented, the edge data acquisition preprocessing module is first run on the edge device to collect raw data from various sensors (such as voltage sensors, current sensors, etc.). Then the cloud-edge data synchronization consistency module is responsible for synchronizing the preprocessed data to the cloud and other related edge devices. The heterogeneous data fusion analysis module receives standardized data from different data sources and uses specific algorithms (such as Bayesian reasoning) for fusion analysis. The real-time warning response module runs in the cloud and continuously monitors the fused data and the sensor data transmitted in real time.

[0057] The system proposed in this embodiment realizes real-time monitoring and early warning of the power grid through cloud technology, improves the efficiency and accuracy of data collection, synchronization, fusion analysis and early warning response, ensures the consistency of edge device and cloud data, and can effectively process heterogeneous data, thereby improving the safety and reliability of power grid operation.

[0058] In one embodiment, the edge data collection and preprocessing module includes: an edge device data collection unit and a data preprocessing unit;

[0059] The edge device data acquisition unit is used to collect data from various sensors in the power grid using a dynamic sampling algorithm, which defines the relationship between different data change speeds and corresponding sampling rates;

[0060] The data preprocessing unit is used to preprocess the collected data in order to improve the data quality.

[0061] In a further embodiment, the sampling frequency is adaptively adjusted according to the grid operation state and the data change rate, as shown in formula (1):

[0062]

[0063] In the formula, f s is the adjusted sampling frequency, f base is the basic sampling frequency, α is the adjustment coefficient, x t and are the data values ​​of the current and previous sampling points, respectively, x max and x min It is the maximum and minimum value of the data. When the data change rate is large, the sampling frequency is increased accordingly to capture more key information. When the data changes slowly, the sampling frequency is reduced to reduce the amount of data transmission.

[0064] Through the relationship, the dynamic sampling algorithm adaptively adjusts the sampling frequency according to the grid operation status and data change rate, ensuring that key information is captured while reducing the amount of data transmission.

[0065] In another embodiment, the data preprocessing unit applies an adaptive noise suppression algorithm and a TinyML edge intelligence model for real-time processing.

[0066] In practical applications, the denoising parameters are dynamically adjusted according to the data characteristics, as shown in formula (2):

[0067]

[0068] In the formula, y t is the denoised data, x t and are the original data of the current and previous sampling points respectively, β is the denoising coefficient, and its value is dynamically adjusted according to the noise level of the data. Denoising is achieved by calculating the difference between the current data point and the previous data point, multiplying it by the denoising coefficient, and subtracting a part of the difference from the current data point;

[0069] Lightweight data filtering and normalization are performed on edge devices, see equations (3) and (4):

[0070]

[0071]

[0072] In the formula, y t is the filtered data, x i is the original data point, N is the size of the sliding window, z t is the normalized data, y t is the filtered data, y max and min are the maximum and minimum values ​​of the filtered data. Data filtering smoothes the data by calculating the average value of the data in the sliding window; data normalization scales the data to the range of [0,1] for subsequent processing;

[0073] Through the above steps, the adaptive noise suppression algorithm dynamically adjusts the denoising parameters according to the data characteristics, and the TinyML model is used to perform lightweight data filtering and normalization on edge devices to improve data processing efficiency.

[0074] In one embodiment, the cloud-edge data synchronization consistency module includes: a data synchronization control unit and a data buffer retransmission mechanism unit;

[0075] The data synchronization control unit is used to replicate and synchronize data status between nodes using a distributed consistency algorithm;

[0076] The data buffer retransmission mechanism unit is used to trigger a retransmission mechanism when data is lost or delayed, so as to retransmit the corresponding data.

[0077] In a further embodiment, data synchronization control uses a distributed consistency algorithm to ensure data synchronization between the edge device and the cloud platform.

[0078] In practical applications, data status is replicated and synchronized between network nodes to ensure that all nodes have the same data view, see formula (5):

[0079]

[0080] Where V(t) represents the data state vector of all nodes at time t, ΔV i (t) represents the amount of data state change of node i at time t, and n represents the total number of nodes.

[0081] At time t, each node updates its data status according to the amount of change in its data status, and broadcasts the change to other nodes. After receiving the change, other nodes also update their own data status accordingly to maintain data consistency. This algorithm avoids data inconsistency caused by network delays by replicating and synchronizing data status between network nodes.

[0082] In another further embodiment, a data buffer retransmission mechanism is constructed, based on an ACK transmission protocol, and the retransmission mechanism is automatically triggered when data loss or delay is detected during data transmission.

[0083] During data transmission, the sender waits for the receiver's ACK confirmation after sending the data packet. If the ACK is not received within the predetermined time, the sender considers that the data packet is lost and triggers the retransmission mechanism. At the same time, the sender maintains a data buffer to store data packets that have been sent but not yet confirmed. When data packet loss is detected, the lost data packet is resent from the buffer. A timeout threshold is set. If the ACK is not received within this time, the retransmission mechanism is triggered, see formula (6):

[0084]

[0085] Where, T timeout is the timeout threshold, T base is the basic timeout, γ is the adjustment factor, RTT is the round-trip time, min is the minimum round trip time.

[0086] The timeout threshold is dynamically adjusted according to the network conditions to adapt to different network delay conditions. When the network delay is large, the timeout threshold is increased accordingly to avoid frequent triggering of the retransmission mechanism; when the network delay is small, the timeout threshold is reduced to improve data transmission efficiency; through this step, the cloud-edge data synchronization consistency module can effectively ensure data synchronization between edge devices and the cloud platform, avoid data inconsistency problems caused by network delays, and provide a reliable retransmission mechanism during data transmission to ensure data integrity, thereby supporting effective decision-making and response measures.

[0087] In one embodiment, the heterogeneous data fusion analysis module includes: a data standardization unit and a fusion analysis unit;

[0088] The data standardization unit is used to standardize the data;

[0089] The fusion analysis unit is used to fuse data from different data sources and obtain comprehensive analysis results based on the fused data.

[0090] For data standardization, you can choose standardized conversion tools based on XML or JSON. These tools are cross-platform and easy to expand. They are suitable for processing heterogeneous data from different manufacturers and different types of equipment. By defining a unified data model and conversion rules, heterogeneous data can be converted into a unified format. During the conversion process, the original meaning and association of the data remain unchanged.

[0091] In a further embodiment, data from different data sources are fused and a comprehensive analysis result is obtained based on the fused data, including:

[0092] S11: Determine the current operating state of the power grid using Bayesian reasoning based on information from the current data source.

[0093] It should be noted that in practical applications, Bayesian reasoning is used for fusion analysis, which comprehensively considers the information of multiple data sources and obtains a comprehensive evaluation result by calculating the correlation weight, as shown in formula (7):

[0094]

[0095] In the formula, A represents a certain operating state of the power grid, B 1 ,B 2 ,...,B n Represents information from different data sources. When multiple data source information B is known 1 ,B 2 ,...,B n Under the condition of , the probability of the power grid operating state A is calculated, and the most likely operating state evaluation result is obtained by comparing the probabilities of different operating states.

[0096] S12: Calculate the correlation weight between each current data source and the current operating state according to the relationship between different data sources and different operating states.

[0097] It should be noted that in the fusion analysis process, it is necessary to calculate the correlation weights between different data sources to reflect their influence on the final evaluation results, as shown in formula (8):

[0098]

[0099] In the formula, w i represents the relevance weight of data source i, Cov(X i ,Y) represents the covariance between data source i and the power grid operation status Y, n represents the total number of data sources. The covariance between each data source and the power grid operation status is calculated and normalized to obtain the correlation weight of each data source. The larger the weight, the greater the influence of the data source on the final evaluation result.

[0100] S13: The evaluation results of all current data sources are integrated with the relevance weights to obtain a comprehensive evaluation result.

[0101] It should be noted that the comprehensive evaluation result is calculated based on the relevance weight and the information of each data source, as shown in formula (9):

[0102]

[0103] In the formula, R represents the comprehensive evaluation result, w i represents the relevance weight of data source i, r i Represents the evaluation result provided by data source i. The evaluation result of each data source is multiplied by its correlation weight, and all the results are added together to obtain a comprehensive evaluation result, which can fully reflect the operation status of the power grid. Finally, the comprehensive evaluation result is compared with the preset threshold. If it exceeds the threshold range, the early warning mechanism is triggered to monitor the changing trend of the comprehensive evaluation result in real time, so as to promptly discover potential problems in the operation of the power grid and take corresponding measures to deal with them.

[0104] Through the above steps, the correlation weights between different data sources are calculated, and the information from multiple data sources is fused into a comprehensive evaluation result. The heterogeneous data fusion analysis module can effectively standardize and transform and fuse heterogeneous data from different manufacturers and different types of equipment, providing a comprehensive understanding of the operating status of the power grid.

[0105] In one embodiment, an edge intelligence deployment management module is also included, and the edge intelligence deployment management module includes: an intelligent algorithm deployment unit and an edge intelligence management unit;

[0106] The intelligent algorithm deployment unit is used to deploy intelligent algorithms on edge devices to achieve localized computing. Intelligent algorithms are efficiently deployed on edge devices using containerization technology and edge computing platforms. Containerization technology enables algorithms to run in a consistent manner on different edge devices, achieving localized computing and reducing pressure on the cloud.

[0107] The edge intelligence management unit is used to monitor and manage the intelligent algorithms on edge devices. The execution strategy of the algorithm is dynamically adjusted through a lightweight management framework, which monitors and manages the intelligent algorithms on edge devices, including load balancing and energy consumption management, to ensure that the algorithm runs efficiently on resource-constrained edge devices.

[0108] In one embodiment, the real-time warning response module includes: a real-time warning unit and an automatic response unit;

[0109] The real-time warning unit is used to predict the power grid status based on the real-time sensor data, compare the predicted status results with the comprehensive evaluation results, and issue a warning signal based on the comparison results.

[0110] In the real-time warning unit, a real-time warning response mechanism is established:

[0111] Construct an LSTM network model, select appropriate network structure and parameters, such as the number of hidden layers and the number of neurons, and use historical power grid data to train the LSTM network so that it can learn the changing laws of the power grid state. Use a loss function that combines mean square error (MSE) and cross entropy to optimize the model performance, as shown in formula (10):

[0112]

[0113] In the formula, α is the weight coefficient, which is used to balance the influence of the two loss functions. By combining MSE and Cross-Entropy, the accuracy of power grid state prediction and the sensitivity of anomaly detection can be considered at the same time, thereby improving the accuracy of real-time warning. The real-time power grid data is input into the trained LSTM network for state prediction. The prediction result is compared with the preset threshold. If the predicted value exceeds the threshold range, a warning signal is issued.

[0114] The automatic response unit is used to execute corresponding emergency response measures according to predefined rules after receiving the early warning signal.

[0115] In the automatic response unit, configure the IFTTT engine and define the "if-then" rules, that is, when specific conditions (such as power grid abnormality warning) are met, automatically perform corresponding operations (such as automatic power outage or load adjustment), and connect the IFTTT engine to the power grid monitoring system so that it can receive warning signals in real time and trigger emergency response measures;

[0116] When the IFTTT engine receives a warning signal, it automatically executes the corresponding emergency response measures according to predefined rules. For example, if it is predicted that the grid load is too high and may cause a failure, the IFTTT engine can automatically trigger load adjustment operations, such as shutting down some non-critical equipment or adjusting the equipment operation mode.

[0117] Evaluate the effectiveness of automatic response measures, such as measuring changes in grid status after load adjustment, and feed the evaluation results back to the LSTM network and IFTTT engine to optimize the configuration of models and rules and improve the accuracy and efficiency of future warnings and responses.

[0118] See also Figure 2 The embodiment of the present invention also proposes a method for constructing a real-time monitoring and early warning system for a power grid, which is used to construct a real-time monitoring and early warning system for a power grid, and the process is as follows:

[0119] S1. System integration and architecture design:

[0120] Adopt distributed microservice architecture to ensure the modularity and scalability of the system, use Docker containerization technology for deployment, and support cross-platform operation and elastic expansion of the system;

[0121] S2. Perform data collection and edge computing:

[0122] Implement lightweight data collection and preprocessing functions on intelligent edge devices, and deploy and run intelligent algorithms through embedded Linux and EdgeX Foundry edge computing platform;

[0123] S3. Build a cloud data analysis and early warning system:

[0124] Build the Apache Spark big data analysis framework on the cloud computing platform, and combine it with the TensorFlow artificial intelligence tool to perform in-depth data analysis and anomaly detection to achieve real-time early warning response;

[0125] S4. Establish security and privacy protection mechanisms:

[0126] Implement data encryption and access control at all levels in the system, intelligent edge devices, cloud computing platforms, and network communication facilities, and ensure data security and privacy protection through role-based access control and end-to-end encryption technology;

[0127] S5. System monitoring and adaptive optimization:

[0128] Combined with AIOps technology, the system operation status is monitored in real time and dynamically adjusted, and the reinforcement learning algorithm is used for adaptive optimization. The present invention proposes a real-time monitoring and early warning system for power grids using cloud technology. The system is implemented based on intelligent edge devices, cloud computing platforms, and network communication facilities;

[0129] Based on the above method, the real-time monitoring and early warning system of the power grid consists of an edge data acquisition preprocessing module, a cloud-edge data synchronization consistency module, a heterogeneous data fusion analysis module, an edge intelligent deployment management module, a real-time early warning response module, and an environmental perception module;

[0130] Among them, it should be noted that this system uses the edge data collection and preprocessing module for edge device data collection and data preprocessing, this system uses the cloud-edge data synchronization consistency module to control data synchronization and provide data buffering and retransmission mechanism, this system uses the heterogeneous data fusion analysis module to standardize heterogeneous data and perform fusion analysis, this system uses the edge intelligent deployment management module to deploy intelligent algorithms and realize localized computing and edge intelligent management, this system uses the real-time early warning response module to perform real-time early warning and response operations, and this system uses the environmental perception module to perceive and adapt to the environment.

[0131] In this embodiment, it should also be noted that the edge data acquisition preprocessing module of this system also includes an edge device data acquisition unit and a data preprocessing unit:

[0132] The edge device data acquisition unit collects various sensor data from the power grid, including current, voltage, and temperature, through a dynamic sampling algorithm. The data preprocessing unit applies an adaptive noise suppression algorithm and the TinyML edge intelligent model for real-time processing, which is used for preliminary data processing on the edge device, including denoising, filtering, and normalization.

[0133] In this embodiment, it should also be noted that the cloud-edge data synchronization consistency module of this system also includes a data synchronization control unit and a data buffer retransmission mechanism unit:

[0134] The data synchronization control unit is used to ensure data synchronization between edge devices and cloud platforms. It avoids data inconsistency caused by network delays by adopting a distributed consistency algorithm. The data buffering and retransmission mechanism unit adopts an ACK-based transmission protocol to detect data loss or delay during data transmission and automatically trigger the data retransmission mechanism.

[0135] In this embodiment, it should also be noted that the heterogeneous data fusion analysis module of this system also includes a data standardization unit and a fusion analysis unit:

[0136] The data standardization unit uses standardized conversion tools based on XML or JSON to convert heterogeneous data from different manufacturers and different types of equipment into a unified format for subsequent analysis and processing. The fusion analysis unit uses a fusion method based on Bayesian reasoning to perform fusion analysis on standardized data and provide a comprehensive assessment of the operating status of the power grid.

[0137] In this embodiment, it should also be noted that the edge intelligent deployment management module of this system also includes an intelligent algorithm deployment unit and an edge intelligent management unit:

[0138] The intelligent algorithm deployment unit uses containerization technology and edge computing platform to efficiently deploy intelligent algorithms on edge devices, realize localized computing and reduce pressure on the cloud. The edge intelligent management unit dynamically adjusts the algorithm's execution strategy through a lightweight management framework to monitor and manage intelligent algorithms on edge devices, including load balancing and energy consumption management.

[0139] In this embodiment, it should also be noted that the real-time warning response module of this system also includes a real-time warning unit and an automatic response unit:

[0140] The real-time warning unit monitors data in real time through the LSTM network and issues real-time warnings for abnormal grid conditions. The automatic response unit automatically triggers predetermined emergency response measures, including automatic power outages or load adjustments, when an abnormal grid condition is detected through the IFTTT engine.

[0141] It should also be noted that the intelligent edge devices of this system use intelligent gateways, RTUs and PLCs to obtain device sensor data in real time and perform preliminary processing and local operation of intelligent algorithms. The intelligent edge devices are connected to the cloud computing platform via a bus or sensor network.

[0142] It should also be noted that the cloud computing platform of this system is a server group, storage cluster and network resources built on a private cloud based on a large-scale distributed cloud computing architecture. It supports modular functional deployment through a microservice architecture and is used for data storage, in-depth analysis, real-time processing and system coordination management.

[0143] It should also be noted that the network communication facilities of this system are based on a multi-layer network architecture of fiber optic networks, 5G base stations and satellite communication links. The network performance is optimized through SD-WAN technology and is used for data transmission between the smart edge devices and the cloud computing platform.

[0144] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A real-time monitoring and early warning system for power grids using cloud technology, characterized in that: include: The edge data collection and preprocessing module, cloud-edge data synchronization consistency module and heterogeneous data fusion analysis module deployed on the edge device side and the real-time warning response module deployed on the cloud side; The edge device end collects and preprocesses data from various sensors in the power grid through the edge data collection and preprocessing module; The edge device also replicates and synchronizes the state of the pre-processed data between various nodes through the cloud-edge data synchronization consistency module to ensure that all nodes between the edge device and the cloud platform have the same data view; The edge device also fuses data from different data sources through the heterogeneous data fusion analysis module; The cloud monitors and issues warnings on the fused data and real-time sensor data through the real-time warning response module.

2. The real-time monitoring and early warning system for power grids using cloud technology according to claim 1 is characterized in that: The edge data acquisition and preprocessing module includes: an edge device data acquisition unit and a data preprocessing unit; The edge device data acquisition unit is used to collect data from various sensors in the power grid using a dynamic sampling algorithm, and the dynamic sampling algorithm defines the relationship between different data change speeds and corresponding sampling rates; The data preprocessing unit is used to preprocess the collected data so as to improve the data quality.

3. The real-time monitoring and early warning system for power grids using cloud technology according to claim 2 is characterized in that: In the dynamic sampling algorithm, the data change speed and the sampling rate satisfy the following relationship: In the formula, f s is the adjusted sampling frequency, f base is the basic sampling frequency, α is the adjustment coefficient, x t and are the data values ​​of the current and previous sampling points, respectively, x max and x min are the maximum and minimum values ​​of the data.

4. The real-time monitoring and early warning system for power grids using cloud technology according to claim 1 is characterized in that: The cloud-edge data synchronization consistency module includes: a data synchronization control unit and a data buffer retransmission mechanism unit; The data synchronization control unit is used to replicate and synchronize data status between various nodes using a distributed consistency algorithm; The data buffer retransmission mechanism unit is used to trigger a retransmission mechanism when data is lost or delayed, so as to retransmit the corresponding data.

5. The real-time monitoring and early warning system for power grids using cloud technology according to claim 4 is characterized in that: The data state is determined according to the following formula: Where V(t) represents the data state vector of all nodes at time t, ΔV i (t) represents the amount of data state change of node i at time t, and n represents the total number of nodes.

6. The real-time monitoring and early warning system for power grids using cloud technology according to claim 4 is characterized in that: In the retransmission mechanism, the retransmission timeout threshold is determined according to the following formula: Where, T timeout is the timeout threshold, T base is the basic timeout, γ is the adjustment factor, RTT is the round-trip time, min is the minimum round trip time.

7. The real-time monitoring and early warning system for power grids using cloud technology according to claim 1 is characterized in that: The heterogeneous data fusion analysis module includes: a data standardization unit and a fusion analysis unit; The data standardization unit is used to perform standardization processing on the data; The fusion analysis unit is used to fuse data from different data sources and obtain a comprehensive analysis result based on the fused data.

8. The real-time monitoring and early warning system for power grids using cloud technology according to claim 7 is characterized in that: The data from different data sources are fused and processed, and comprehensive analysis results are obtained based on the fused data, including: Based on the information from the current data source, Bayesian reasoning is used to determine the current operating status of the power grid; Calculate the correlation weight between each current data source and the current operating state according to the relationship between different data sources and different operating states; The evaluation results of all current data sources are integrated with the relevance weights to obtain a comprehensive evaluation result.

9. The real-time monitoring and early warning system for power grids using cloud technology according to claim 8, characterized in that: The real-time warning response module includes: a real-time warning unit and an automatic response unit; The real-time warning unit is used to predict the state of the power grid according to the real-time sensor data, compare the predicted state result with the comprehensive evaluation result, and issue a warning signal according to the comparison result; The automatic response unit is used to execute corresponding emergency response measures according to predefined rules after receiving the early warning signal.

10. The real-time monitoring and early warning system for power grids using cloud technology according to claim 1, characterized in that: It also includes an edge intelligence deployment management module, which includes: an intelligent algorithm deployment unit and an edge intelligence management unit; The intelligent algorithm deployment unit is used to deploy the intelligent algorithm on the edge device to achieve localized computing; The edge intelligence management unit is used to monitor and manage the intelligent algorithm on the edge device.

Images