Monitoring method and system for preventing and controlling leakage of important information of centralized monitoring of plant station

Abstract

The invention relates to the technical field of power system monitoring, in particular to a monitoring method and system for preventing and treating important information leakage of centralized monitoring of a plant station. The method comprises the following steps: extracting four-remote-point table data from a master station OCS system database, and preprocessing the data; splitting an equipment alarm signal from the preprocessed data; the method comprises the following steps: describing a relationship among alarm signals, equipment and a monitoring responsibility area, creating a standard alarm signal model, initializing node attributes, and automatically updating alarm signal node tags through a signal classification algorithm; and an equipment monitoring alarm signal analysis method is adopted, and corresponding response measures are taken according to a classification result. According to the invention, the phenomena of false alarm and missing alarm are effectively reduced, the operation safety and stability of a plant station are improved, and the overall operation and maintenance efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power system monitoring, and in particular to a method and system for preventing leakage of important information in centralized monitoring of power plants and stations. Background Art

[0002] As the scale and complexity of modern power plants continue to increase, traditional monitoring systems face huge challenges, including a surge in data volume, a wide variety of equipment, and complex monitoring requirements. These systems often rely on a large number of sensors and monitoring equipment, generating massive amounts of operating data that need to be collected, processed, and analyzed in a timely manner in order to make accurate alarm and response decisions.

[0003] The main problem of traditional plant monitoring systems is the high incidence of missed monitoring and false alarms, which usually stems from insufficient data processing capabilities, inaccurate alarm signal recognition, and untimely system response. Due to the complexity and dynamic nature of data, it is difficult for existing systems to analyze and classify alarm signals in a timely and accurate manner when facing emergencies, which may lead to omissions of key information and erroneous responses. The operating status of the equipment is difficult to be reliably monitored, which may affect the stability and safety of the entire power system. Due to long-term operation, equipment may have short-term abnormalities. These frequently triggered alarms can easily paralyze monitoring personnel and make it impossible to conduct closed-loop control in a timely manner. Frequent missed monitoring events occur because the system cannot upload important monitoring information in a timely manner, or the interface alarm fails to clearly identify the importance of the information. The frequent missed monitoring events in regional power grids in recent years show that this has become one of the most serious risk points in power grid regulation. An effective method is urgently needed to improve the intelligence and reliability of the monitoring system in order to reduce the occurrence of missed monitoring events and ensure the stability and safety of the power system. Summary of the invention

[0004] In view of the above problems existing in the prior art, the present invention is proposed.

[0005] Therefore, the present invention provides a method for preventing and controlling the leakage of important information in centralized monitoring of power plants, which can solve the problems of missed monitoring, false alarms and untimely response often faced by traditional power plant monitoring systems due to large data volumes and complex equipment.

[0006] To solve the above technical problems, the present invention provides the following technical solutions, a method for preventing and controlling the omission of important information in centralized monitoring of power plants, comprising: extracting four-remote point table data from the master station OCS system database and preprocessing the data; splitting the equipment alarm signal from the preprocessed data; describing the relationship between the alarm signal, equipment and monitoring responsibility area, creating a standard alarm signal model, initializing node attributes, and automatically updating the alarm signal node label through a signal classification algorithm; adopting an equipment monitoring alarm signal analysis method, and taking corresponding response measures according to the classification results.

[0007] As a preferred solution of the method for preventing leakage of important information in centralized monitoring of power plants described in the present invention, the four-remote point table data extracted includes operation data, equipment data and monitoring responsibility area data;

[0008] The operating data includes voltage, current, active power, reactive power, energy consumption, temperature, humidity and wind speed;

[0009] The device data includes device category, device identifier, location information, operating data and status information;

[0010] The surveillance area of ​​responsibility data includes an area identifier, area location, area function, associated equipment, and security and access rights;

[0011] The preprocessing includes using interpolation methods to fill missing values, ensuring the continuity and integrity of data, converting string dates into date objects, ensuring that data types meet expectations, detecting and processing outliers through statistical methods, removing duplicate records in the data set, removing data that does not meet the conditions, and performing data verification.

[0012] As a preferred solution of the method for preventing leakage of important information in centralized monitoring of power plants described in the present invention, the splitting device alarm signal includes filtering noise signals, comparing the collected voltage data with the rated voltage, and if the fluctuation is less than 10%, judging the signal as a normal signal, and if the fluctuation is greater than 10%, judging the signal as a noise signal for filtering, analyzing the change trend of the filtered signal, and if the signal value has a significant change trend within a continuous time window, marking the signal as an alarm signal, and when the signal value does not show a significant change trend within the continuous time window, marking the signal as a noise signal and performing a secondary judgment;

[0013] Compare the collected temperature data with the preset temperature threshold. If the temperature of the device does not exceed the temperature threshold, the signal is judged to be a normal signal. If the temperature of the device exceeds the temperature threshold, the signal is judged to be a noise signal and filtered. For the filtered signal, analyze the change trend. If the signal value has an upward trend in the continuous time window, the signal is marked as an alarm signal. If the signal value has a downward trend in the continuous time window, the signal is marked as a noise signal and a secondary judgment is performed.

[0014] For the secondary judged signal, the Pearson correlation coefficient is used to quantify the correlation between the signal and the key signal. If the signal is irrelevant to the key signal or the correlation is weak, it is determined to be an irrelevant interference signal and filtered. If the signal is strongly correlated with the key signal, it is determined to be an alarm signal and marked. The Pearson correlation coefficient formula is expressed as:

[0015]

[0016] Among them, R is the Pearson correlation coefficient, i is the sampling point index value, n is the total number of sampling points, X i , Y i are the values ​​of the original signal and the key signal at the i-th sampling point, respectively. are the average values ​​of the original signal and the key signal respectively;

[0017] If R = 0, the judgment signal is irrelevant to the key signal;

[0018] If R ≥ θ, the correlation between the judgment signal and the key signal is strong, and if R < θ, the correlation between the judgment signal and the key signal is weak;

[0019] Among them, θ is the correlation threshold.

[0020] As a preferred solution of the method for preventing leakage of important information in centralized monitoring of power plants described in the present invention, wherein: the creation of the standard alarm signal model includes describing the alarm signal and related attributes based on graph theory, setting model elements, the node entities include alarm signal nodes, equipment nodes, monitoring responsibility area nodes, the edges include signal-equipment edges, equipment-monitoring responsibility area edges, the process of constructing the graph represents all nodes and node relationships as a graph structure, and creates nodes for each signal, equipment and monitoring responsibility area;

[0021] Initialize node attributes. Alarm signal node attributes include alarm category label, occurrence time, duration, signal source and alarm content. Alarm signals of known categories are directly assigned corresponding alarm category labels. For new or unclassified alarm signals, default labels are assigned and iteratively updated through signal classification algorithms. Device node attributes include device category, device identifier, location information, operating data and status information. Monitoring responsibility area node attributes include area identifier, area location, area function and associated equipment. Directed edges are established based on the alarm signals generated by the equipment and the subordinate relationship between the equipment and the monitoring responsibility area node, and directed edge weights are assigned.

[0022] Use a graph search algorithm to find all paths from the alarm signal to the monitoring responsibility area, apply a centrality analysis algorithm to identify the devices or monitoring responsibility areas that are most critical to alarm delivery, and use a community detection algorithm to find highly correlated device groups that jointly affect specific monitoring responsibility areas or alarm types.

[0023] As a preferred solution of the method for preventing leakage of important information in centralized monitoring of a plant and station described in the present invention, the signal classification algorithm includes a label propagation algorithm based on which the label information of known categories is used to infer data of unknown labels, and a directed graph G = (V, E) is set, wherein V is a node set, E is an edge set, and each alarm signal node v∈V has an initial label l v, the label set is given by L, each label corresponds to an alarm category, and the label initialization formula is expressed as,

[0024]

[0025] in, is the label of node v at the beginning of the iteration;

[0026] In each iteration m, the weight w of the edge (u,v) is introduced uv , using weighted majority voting, the new label of the node is determined by the labels of neighboring nodes and the corresponding weights. The label update rule formula is expressed as,

[0027]

[0028] Where N(v) is the set of neighbor nodes u of node v, and mode is a function that returns the most frequently appearing label;

[0029] The Euclidean distance is used to determine the similarity between nodes. The formula is:

[0030]

[0031] Among them, d uv is the Euclidean distance between node v and node u, i is the feature vector index, n is the dimension of the feature vector, x u,i 、x v,i are the i-th component of the eigenvectors of node u and node v respectively;

[0032] Considering that the weight in the label propagation algorithm is similarity, the calculated distance is converted into a weight. The smaller the distance, the higher the similarity between nodes, and the larger the weight. An exponential decay function is used for conversion, and the formula is expressed as:

[0033]

[0034] The iteration stops when the labels of all nodes no longer change or the preset maximum number of iterations is reached.

[0035] As a preferred solution of the method for preventing leakage of important information in centralized monitoring of power plants described in the present invention, the equipment monitoring alarm signal analysis method includes outputting the alarm signal analysis result according to the alarm signal attribute through the alarm signal analysis algorithm, and the alarm signal analysis algorithm formula is expressed as follows:

[0036]

[0037] Among them, G(x) is the alarm signal score, t0 and t are the start and end time of the analysis time window, k is the normalization factor, γ is the adjustment coefficient, and g vis the alarm content and occurrence time of node v, ψ(v,t) is the difference between node v and the historical state, and σ(v,t) is the dynamic adjustment function;

[0038] According to the output results of the alarm signal analysis algorithm, the type and severity of the alarm signal are classified as follows:

[0039] When G(v)<α, the alarm signal is judged to be a Class I emergency alarm signal;

[0040] When α≤G(v)<β, the alarm signal is judged to be a Class II abnormal alarm signal;

[0041] When G(v)≥β, the signal is judged to be a potential alarm signal of Class III.

[0042] As a preferred solution of the method for preventing leakage of important information in centralized monitoring of power plants described in the present invention, the response measures include: if the signal is a Class I emergency alarm signal, using a graph search algorithm to find the affected equipment and key nodes and give priority to processing, the system automatically triggers the emergency response program, starts safety lock measures, shutdown operations or activates emergency plans, the system immediately notifies relevant personnel and emergency services, including the safety management team and maintenance personnel, and the system records detailed information of all relevant events and responses;

[0043] If the signal is a Class II abnormal alarm signal, apply centrality analysis to identify key devices or nodes in the network, predict and prevent the spread of faults, use community detection methods to identify the affected device groups, optimize monitoring and response strategies, conduct detailed system inspections, determine the specific cause of the abnormality, repair or adjust the fault point, and strengthen monitoring of key components;

[0044] If the signal is a Class III potential alarm signal, use the label propagation algorithm to continue observing and classifying the signal, ensure that all relevant information is processed, update the attributes and status of the nodes in the graph model, reflect the latest monitoring data and prediction information, adjust the operating parameters or perform scheduled maintenance work based on the analysis results, analyze the signal regularly, and evaluate the impact on system performance.

[0045] Another object of the present invention is to provide a system for preventing leakage of important information in centralized monitoring of prevention and control plants, which ensures the integrity and accuracy of the data by extracting the four-remote point table data from the main station OCS system database and performing systematic preprocessing on these data, such as filling in missing values, format conversion and outlier processing. By creating a standard alarm signal model based on graph theory, describing in detail the relationship between the alarm signal and the equipment and monitoring area, and automatically updating the alarm signal node label through the signal classification algorithm, accurate classification and processing of the alarm signal can be achieved. With the help of the equipment monitoring alarm signal analysis method, the system can quickly identify the type and severity of the signal and take corresponding response measures, thereby reducing the occurrence of missed monitoring and improving the overall efficiency and safety of the monitoring system.

[0046] As a preferred solution of the centralized monitoring system for important information leakage of a plant and station described in the present invention, it includes: a data acquisition and preprocessing module, an alarm signal splitting module, a signal classification and analysis module and a response and control module;

[0047] The data acquisition and preprocessing module extracts four-telemetry table data from the master station OCS system database, including operation data, equipment data and monitoring responsibility area data, and ensures the integrity and accuracy of the data through preprocessing, filling missing values, data type conversion and abnormal value detection;

[0048] The alarm signal splitting module analyzes the collected and preprocessed data, compares the voltage and temperature fluctuations, filters the noise signal, uses the Pearson correlation coefficient to determine the correlation between the signal and the key signal, and splits the device alarm signal;

[0049] The signal classification and analysis module establishes a standard alarm signal model based on graph theory, assigns initial labels to signals by initializing node attributes, and updates signal labels using a label propagation algorithm;

[0050] The response and control module calculates the alarm signal scores and classifies the signals into three alarm types. According to the alarm type, the system automatically takes corresponding response measures.

[0051] A computer device includes a memory and a processor, wherein the memory stores a computer program, and is characterized in that when the processor executes the computer program, the steps of any one of the methods for preventing and controlling the leakage of important information in centralized monitoring of power plants are implemented.

[0052] A computer-readable storage medium having a computer program stored thereon, characterized in that when the computer program is executed by a processor, the steps of any one of the methods for preventing and controlling leakage of important information in centralized monitoring of power plants are implemented.

[0053] Beneficial effects of the present invention: The present invention improves the efficiency and reliability of the plant monitoring system through data analysis technology and graph theory models. Through the automated data preprocessing process, the system can ensure the accuracy and completeness of the monitoring data. The alarm signal model constructed based on graph theory can accurately describe the relationship between signals, equipment and monitoring areas, and enhance the accuracy of signal classification and analysis. Through the label propagation algorithm, the system can automatically update and classify alarm signals to ensure rapid identification and response. The alarm signals are scored through the signal analysis algorithm, and the signals are divided into emergency alarms, abnormal alarms and potential alarm categories. The system can automatically adjust the response strategy according to the category, thereby improving the response speed and processing efficiency to various alarm events. The present invention effectively reduces false alarms and missed alarms, improves the safety and stability of plant operation, and improves the overall operation and maintenance efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0055] Figure 1 A flowchart of a method for preventing leakage of important information in centralized monitoring of power plants and stations is provided as an embodiment of the present invention.

[0056] Figure 2 A module diagram of a monitoring system for preventing and controlling leakage of important information in centralized monitoring of power plants provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0057] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0058] Example 1, reference Figure 1 , which is the first embodiment of the present invention, provides a method for preventing and controlling leakage of important information in centralized monitoring of power plants, including:

[0059] S1: Extract the four-telemetry table data from the master station OCS system database and pre-process the data.

[0060] It should be noted that the four remote point table data include telemetry data, telesignaling data, remote control data and remote adjustment data, and the extracted four remote point table data include operation data, equipment data and monitoring responsibility area data. The operation data includes the voltage, current, active power, reactive power, energy consumption, temperature, humidity and wind speed of the telemetry data; the equipment data includes the equipment category, equipment identifier, location information, operation data and status information of the telesignaling data; the monitoring responsibility area data includes the area identifier, area location, area function, associated equipment and security and access rights. Real-time monitoring of these data is crucial for timely detection of anomalies and response to faults, and helps to improve the reliability and safety of the power grid.

[0061] The preprocessing includes using interpolation methods to fill missing values, ensuring the continuity and integrity of data, converting string dates into date objects, ensuring that data types meet expectations, detecting and processing outliers through statistical methods, removing duplicate records in the data set, removing data that does not meet the conditions, and performing data verification.

[0062] S2: Split the device alarm signal from the pre-processed data.

[0063] It should be noted that the split device alarm signal includes filtering the noise signal, comparing the collected voltage data with the rated voltage, if the fluctuation is less than 10%, the signal is judged to be a normal signal, if the fluctuation is greater than 10%, the signal is judged to be a noise signal for filtering, and the filtered signal is analyzed for the change trend. If the signal value has a significant change trend within the continuous time window, the signal is marked as an alarm signal. When the signal value does not show a significant change trend within the continuous time window, the signal is marked as a noise signal and a secondary judgment is performed;

[0064] Compare the collected temperature data with the preset temperature threshold. If the temperature of the device does not exceed the temperature threshold, the signal is judged to be a normal signal. If the temperature of the device exceeds the temperature threshold, the signal is judged to be a noise signal and filtered. For the filtered signal, analyze the change trend. If the signal value has an upward trend in the continuous time window, the signal is marked as an alarm signal. If the signal value has a downward trend in the continuous time window, the signal is marked as a noise signal and a secondary judgment is performed.

[0065] Furthermore, key signals refer to alarm signals that are closely related to the safety and efficiency of system operation. Key signals should be selected based on their direct impact on plant safety, equipment health and operating efficiency, including alarm signals that can directly reflect equipment failures, abnormal operations or environmental risks. For example, for power systems, key signals may include circuit breaker action signals, protection device trigger signals, etc.

[0066] Furthermore, for the secondary judged signal, the Pearson correlation coefficient is used to quantify the correlation between the signal and the key signal. If the signal is irrelevant to the key signal or the correlation is weak, it is determined to be an irrelevant interference signal and filtered. If the signal is strongly correlated with the key signal, it is determined to be an alarm signal and marked. The Pearson correlation coefficient formula is expressed as,

[0067]

[0068] Among them, R is the Pearson correlation coefficient, i is the sampling point index value, n is the total number of sampling points, X i , Y i are the values ​​of the original signal and the key signal at the i-th sampling point, respectively. are the average values ​​of the original signal and the key signal respectively;

[0069] If R = 0, the judgment signal is irrelevant to the key signal;

[0070] If R ≥ θ, the correlation between the judgment signal and the key signal is strong, and if R < θ, the correlation between the judgment signal and the key signal is weak;

[0071] Among them, θ is the correlation threshold.

[0072] S3: Describe the relationship between alarm signals, equipment, and monitoring responsibility areas, create a standard alarm signal model, initialize node attributes, and automatically update alarm signal node labels through signal classification algorithms.

[0073] It should be noted that the creation of the standard alarm signal model includes describing the alarm signal and related attributes based on graph theory, setting model elements, the node entities include alarm signal nodes, device nodes, monitoring responsibility area nodes, the edges include signal-device edges, device-monitoring responsibility area edges, and the process of constructing the graph involves representing all nodes and node relationship relationships as a graph structure, creating nodes for each signal, device, and monitoring responsibility area;

[0074] Initialize node attributes. Alarm signal node attributes include alarm category label, occurrence time, duration, signal source and alarm content. Alarm signals of known categories are directly assigned corresponding alarm category labels. For new or unclassified alarm signals, default labels are assigned and iteratively updated through signal classification algorithms. Device node attributes include device category, device identifier, location information, operating data and status information. Monitoring responsibility area node attributes include area identifier, area location, area function and associated equipment. Directed edges are established based on the alarm signals generated by the equipment and the subordinate relationship between the equipment and the monitoring responsibility area node, and directed edge weights are assigned.

[0075] The signal-device directed edge connects the alarm signal node and the device node, indicating which specific device the alarm signal is sent from. This edge setting helps to track the origin of the alarm and clarify which devices the alarm information is sent from, so as to facilitate targeted fault diagnosis and processing.

[0076] The device-monitoring responsibility area edge connects the device node and the monitoring responsibility area node, indicating which monitoring area a device is located in, or which devices a monitoring area contains. This helps to quickly locate the responsibility area when an alarm or fault occurs, so as to more effectively organize and dispatch maintenance resources and improve response efficiency.

[0077] Use a graph search algorithm to find all paths from the alarm signal to the monitoring responsibility area, apply a centrality analysis algorithm to identify the devices or monitoring responsibility areas that are most critical to alarm delivery, and use a community detection algorithm to find highly correlated device groups that jointly affect specific monitoring responsibility areas or alarm types.

[0078] Further, graph search algorithms include breadth-first search (BFS) and depth-first search (DFS). Considering that DFS deeply explores each branch until it reaches the end or finds the target, it is suitable for exploring all possible paths. The algorithm steps are as follows:

[0079] The node that initiates the alarm signal is used as the starting point of the DFS search. A stack is used to support the last-in, first-out (LIFO) feature of DFS to ensure that the algorithm can explore each branch in depth. A visit mark is set for each node to ensure that each node is only visited once during the search process to prevent loops. Recursive function: Implement the recursive function of DFS, starting from the starting point, and explore all unvisited adjacent nodes.

[0080] During the recursive process, the path from the starting point to the current node is recorded. Whenever a monitoring responsibility area node is reached, the current complete path is saved. Whenever a node is returned from, backtracking is performed, that is, the last node is removed from the current path to explore other possible paths.

[0081] All found paths are stored for subsequent analysis and decision-making. These paths are analyzed to identify critical paths and critical equipment, and the impact range of alarm signals is evaluated to provide a basis for developing response measures. Based on the found critical paths, the alarms with the greatest impact are prioritized, and situations that may affect critical equipment or areas of responsibility are quickly responded to. Through regular analysis of alarm paths, system configuration and monitoring strategies are optimized to prevent possible problems in the future.

[0082] Furthermore, the centrality analysis algorithm is applied to identify the equipment or monitoring responsibility areas that are most critical to alarm transmission. The algorithm aims to evaluate the importance of each node in the network. In the monitoring and alarm management of the power system, identifying these key nodes can help optimize resource allocation and response strategies. The algorithm steps are as follows:

[0083] Use graph theory software or libraries (such as NetworkX) to calculate the betweenness centrality of each node, identify the shortest paths between all pairs of nodes, and calculate the frequency with which each node acts as an intermediary in these paths. Nodes with high betweenness centrality may be critical equipment or monitoring areas. Analyze the calculated centrality results and identify the nodes with the highest betweenness centrality. These nodes are the priority focus of monitoring and response strategies. By optimizing the monitoring and maintenance of these key nodes, the response speed and efficiency of the system can be improved and the risk of missed monitoring can be reduced.

[0084] Furthermore, identifying highly associated device groups through community detection algorithms can significantly improve the management and response capabilities of alarm signals. Select a suitable community detection algorithm, such as the Louvain algorithm or the Girvan-Newman algorithm. The following uses the Louvain algorithm to describe the steps of the community detection algorithm:

[0085] The Louvain algorithm is applied to the constructed network graph. The Louvain algorithm forms communities by gradually merging nodes to improve modularity and eventually identify node groups (communities). Each device node is assigned to a community, which represents a group of devices that are closely interconnected in the network. Through the community detection results, highly associated device groups are identified. The devices in these groups are usually closely related in function or physical location, and may jointly affect a certain monitoring responsibility area or have similar alarm types. Modularity is analyzed to evaluate the quality of community division. Higher modularity usually means a more significant community structure. Based on the detected device groups, the allocation of monitoring resources is optimized to ensure that devices in the same community can be coordinated and effectively monitored.

[0086] It should be noted that the Euclidean distance is used to determine the similarity between nodes, and the formula is expressed as:

[0087]

[0088] Among them, d uv is the Euclidean distance between node v and node u, i is the feature vector index, n is the dimension of the feature vector, x u,i 、x v,i are the i-th component of the eigenvectors of node u and node v respectively;

[0089] Considering that the weight in the label propagation algorithm is similarity, the calculated distance is converted into a weight. The smaller the distance, the higher the similarity between nodes, and the larger the weight. An exponential decay function is used for conversion, and the formula is expressed as:

[0090]

[0091] The signal classification algorithm includes a label propagation algorithm based on which the label information of known categories is used to infer the data of unknown labels. A directed graph G = (V, E) is set, where V is a node set, E is an edge set, and each alarm signal node v∈V has an initial label l v , the label set is given by L, each label corresponds to an alarm category, and the label initialization formula is expressed as,

[0092]

[0093] in, is the label of node v at the beginning of the iteration;

[0094] In each iteration m, the weight w of the edge (u,v) is introduced uv , using weighted majority voting, the new label of the node is determined by the labels of neighboring nodes and the corresponding weights. The label update rule formula is expressed as,

[0095]

[0096] Where N(v) is the set of neighbor nodes u of node v, and mode is a function that returns the most frequently appearing label;

[0097] Iterations continue until either of the following conditions is met:

[0098] Condition 1: When the labels of all nodes do not change, that is, for all v

[0099] Condition 2: Reach the preset maximum number of iterations.

[0100] S4: Use equipment monitoring alarm signal analysis method and take corresponding response measures according to the classification results.

[0101] It should be noted that the device monitoring alarm signal analysis method includes outputting an alarm signal analysis result according to the alarm signal attribute through an alarm signal analysis algorithm. The alarm signal analysis algorithm formula is expressed as:

[0102]

[0103] Among them, G(x) is the alarm signal score, t0 and t are the start and end time of the analysis time window, k is the normalization factor, γ is the adjustment coefficient, and gv is the alarm content and occurrence time of node v, ψ(v,t) is the difference between node v and the historical state, and σ(v,t) is the dynamic adjustment function;

[0104] According to the output results of the alarm signal analysis algorithm, the type and severity of the alarm signal are classified as follows:

[0105] When G(v)<α, the alarm signal is judged to be a Class I emergency alarm signal;

[0106] When α≤G9v)<β, the alarm signal is judged to be a Class II abnormal alarm signal;

[0107] When G9v)≥β, the signal is judged to be a potential alarm signal of Class III.

[0108] It should be noted that the response measures include, if the signal is a Class I emergency alarm signal, using a graph search algorithm to find the affected equipment and key nodes and prioritize processing, the system automatically triggers the emergency response procedure, initiates safety lock measures, shutdown operations or activates emergency plans, the system immediately notifies relevant personnel and emergency services, including the security management team and maintenance personnel, and the system records detailed information on all relevant events and responses;

[0109] If the signal is a Class II abnormal alarm signal, apply centrality analysis to identify key devices or nodes in the network, predict and prevent the spread of faults, use community detection methods to identify the affected device groups, optimize monitoring and response strategies, conduct detailed system inspections, determine the specific cause of the abnormality, repair or adjust the fault point, and strengthen monitoring of key components;

[0110] If the signal is a Class III potential alarm signal, use the label propagation algorithm to continue observing and classifying the signal, ensure that all relevant information is processed, update the attributes and status of the nodes in the graph model, reflect the latest monitoring data and prediction information, adjust the operating parameters or perform scheduled maintenance work based on the analysis results, analyze the signal regularly, and evaluate the impact on system performance.

[0111] The above is a schematic scheme of a method for preventing leakage of important information from centralized monitoring of plants and stations in this embodiment. It should be noted that the technical scheme of the system for preventing leakage of important information from centralized monitoring of plants and stations and the technical scheme of the method for preventing leakage of important information from centralized monitoring of plants and stations are of the same concept. For details not described in detail in the technical scheme of the system for preventing leakage of important information from centralized monitoring of plants and stations in this embodiment, please refer to the description of the technical scheme of the method for preventing leakage of important information from centralized monitoring of plants and stations.

[0112] The centralized monitoring system for important information leakage of the prevention and control plant in this embodiment includes: a data acquisition and preprocessing module, an alarm signal splitting module, a signal classification and analysis module, and a response and control module;

[0113] The data acquisition and preprocessing module extracts four-telemetry table data from the master station OCS system database, including operation data, equipment data and monitoring responsibility area data, and ensures the integrity and accuracy of the data through preprocessing, filling missing values, data type conversion and abnormal value detection;

[0114] The alarm signal splitting module analyzes the collected and preprocessed data, compares the voltage and temperature fluctuations, filters the noise signal, uses the Pearson correlation coefficient to determine the correlation between the signal and the key signal, and splits the device alarm signal;

[0115] The signal classification and analysis module establishes a standard alarm signal model based on graph theory, assigns initial labels to signals by initializing node attributes, and updates signal labels using a label propagation algorithm;

[0116] The response and control module calculates the alarm signal scores and classifies the signals into three alarm types. According to the alarm type, the system automatically takes corresponding response measures.

[0117] This embodiment also provides a computing device, which is suitable for preventing and controlling the leakage of important information in centralized monitoring of power plants, and includes:

[0118] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program codes.

[0119] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0120] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0121] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0122] Embodiment 2 is the second embodiment of the present invention, which provides a method for preventing and controlling leakage of important information in centralized monitoring of power plants. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through experiments.

[0123] Two similar plants were selected, one using the method proposed in this invention and the other using a traditional monitoring and alarm processing system. Four-telemeter data of the same time period were extracted from the master station OCS systems of the two plants, including operation data, equipment data, and monitoring responsibility area data. The response measures under the two methods were recorded and compared, including the rapidity of response, accuracy, and stability of the system. The experimental results are shown in Table 1.

[0124] Table 1. Data comparison table

[0125] Parameters / Test Application of the present invention Traditional application factory Data preprocessing time (seconds) 120 180 Data preprocessing error rate (%) 2 5 Alarm signal recognition rate (%) 95 80 False alarm rate of alarm signal (%) 5 15 Response action execution time (seconds) 30 60 System stability evaluation high middle

[0126] As can be seen from Table 1, our invention application plant is superior to the traditional application plant in data preprocessing time and error rate, indicating that our invention has significant advantages in data processing efficiency and accuracy. In terms of the recognition rate and false alarm rate of alarm signals, our invention application also shows better performance, indicating that our invention can process and classify alarm signals more accurately. The execution time of response measures is shorter and the system stability is higher, showing the significant improvement of our invention in improving response efficiency and system stability.

[0127] Through comparative tests, we can fully demonstrate the advantages of our invention in improving the performance of plant monitoring systems and prove its innovation and practical value in practical applications.

[0128] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for preventing leakage of important information in centralized monitoring of power plants and stations, characterized in that: include, Extract the four-telemetry table data from the master station OCS system database and pre-process the data; Splitting device alarm signals from pre-processed data; Describe the relationship between alarm signals, equipment, and monitoring responsibility areas, create a standard alarm signal model, initialize node attributes, and automatically update alarm signal node labels through signal classification algorithms; Use equipment monitoring alarm signal analysis methods and take appropriate response measures based on the classification results.

2. A method for preventing leakage of important information in centralized monitoring of power plants as claimed in claim 1, characterized in that: The extracted four-remote point table data includes operation data, equipment data and monitoring responsibility area data; The operating data includes voltage, current, active power, reactive power, energy consumption, temperature, humidity and wind speed; The device data includes device category, device identifier, location information, operating data and status information; The surveillance area of ​​responsibility data includes an area identifier, area location, area function, associated equipment, and security and access rights; The preprocessing includes using interpolation methods to fill missing values, ensuring the continuity and integrity of data, converting string dates into date objects, ensuring that data types meet expectations, detecting and processing outliers through statistical methods, removing duplicate records in the data set, removing data that does not meet the conditions, and performing data verification.

3. A method for preventing leakage of important information in centralized monitoring of power plants as claimed in claim 2, characterized in that: The splitting device alarm signal includes filtering the noise signal, comparing the collected voltage data with the rated voltage, if the fluctuation is less than 10%, judging the signal as a normal signal, if the fluctuation is greater than 10%, judging the signal as a noise signal for filtering, analyzing the change trend of the filtered signal, if the signal value has a significant change trend within the continuous time window, marking the signal as an alarm signal, when the signal value does not show a significant change trend within the continuous time window, marking the signal as a noise signal and performing a secondary judgment; Compare the collected temperature data with the preset temperature threshold. If the temperature of the device does not exceed the temperature threshold, the signal is judged to be a normal signal. If the temperature of the device exceeds the temperature threshold, the signal is judged to be a noise signal and filtered. For the filtered signal, analyze the change trend. If the signal value has an upward trend in the continuous time window, the signal is marked as an alarm signal. If the signal value has a downward trend in the continuous time window, the signal is marked as a noise signal and a secondary judgment is performed. For the secondary judged signal, the Pearson correlation coefficient is used to quantify the correlation between the signal and the key signal. If the signal is irrelevant to the key signal or the correlation is weak, it is determined to be an irrelevant interference signal and filtered. If the signal is strongly correlated with the key signal, it is determined to be an alarm signal and marked. The Pearson correlation coefficient formula is expressed as: Among them, R is the Pearson correlation coefficient, i is the sampling point index value, n is the total number of sampling points, X i , Y i are the values ​​of the original signal and the key signal at the i-th sampling point, respectively. are the average values ​​of the original signal and the key signal respectively; If R = 0, the judgment signal is irrelevant to the key signal; If R≥θ, the correlation between the judgment signal and the key signal is strong, and if R<θ, the correlation between the judgment signal and the key signal is weak; Among them, θ is the correlation threshold.

4. A method for preventing leakage of important information in centralized monitoring of power plants as claimed in claim 3, characterized in that: The creation of the standard alarm signal model includes describing the alarm signal and related attributes based on graph theory, setting model elements, the node entities include alarm signal nodes, device nodes, monitoring responsibility area nodes, the edges include signal-device edges, device-monitoring responsibility area edges, the process of constructing the graph represents all nodes and node relationships as a graph structure, and creates nodes for each signal, device and monitoring responsibility area; Initialize node attributes. Alarm signal node attributes include alarm category label, occurrence time, duration, signal source and alarm content. Alarm signals of known categories are directly assigned corresponding alarm category labels. For new or unclassified alarm signals, default labels are assigned and iteratively updated through signal classification algorithms. Device node attributes include device category, device identifier, location information, operating data and status information. Monitoring responsibility area node attributes include area identifier, area location, area function and associated equipment. Directed edges are established based on the alarm signals generated by the equipment and the subordinate relationship between the equipment and the monitoring responsibility area node, and directed edge weights are assigned. Use a graph search algorithm to find all paths from the alarm signal to the monitoring responsibility area, apply a centrality analysis algorithm to identify the devices or monitoring responsibility areas that are most critical to alarm delivery, and use a community detection algorithm to find highly correlated device groups that jointly affect specific monitoring responsibility areas or alarm types.

5. A method for preventing leakage of important information in centralized monitoring of power plants and stations as claimed in claim 4, characterized in that: The signal classification algorithm includes a label propagation algorithm based on which the label information of known categories is used to infer the data of unknown labels. A directed graph G = (V, E) is set, where V is a node set, E is an edge set, and each alarm signal node v∈V has an initial label l v , the label set is given by L, each label corresponds to an alarm category, and the label initialization formula is expressed as, in, is the label of node v at the beginning of the iteration; In each iteration m, the weight w of the edge (u,v) is introduced uv , using weighted majority voting, the new label of the node is determined by the labels of neighboring nodes and the corresponding weights. The label update rule formula is expressed as, Where N(v) is the set of neighbor nodes u of node v, and mode is a function that returns the most frequently appearing label; The Euclidean distance is used to determine the similarity between nodes. The formula is: Among them, d uv is the Euclidean distance between node v and node u, i is the feature vector index, n is the dimension of the feature vector, x u,i 、x v,i are the i-th component of the eigenvectors of node u and node v respectively; Considering that the weight in the label propagation algorithm is similarity, the calculated distance is converted into a weight. The smaller the distance, the higher the similarity between nodes, and the larger the weight. An exponential decay function is used for conversion, and the formula is expressed as: The iteration stops when the labels of all nodes no longer change or the preset maximum number of iterations is reached.

6. A method for preventing leakage of important information in centralized monitoring of power plants as claimed in claim 5, characterized in that: The device monitoring alarm signal analysis method includes outputting an alarm signal analysis result according to an alarm signal attribute through an alarm signal analysis algorithm. The alarm signal analysis algorithm formula is expressed as: Among them, G(x) is the alarm signal score, t0 and t are the start and end time of the analysis time window, k is the normalization factor, γ is the adjustment coefficient, and g v is the alarm content and occurrence time of node v, ψ(v,t) is the difference between node v and the historical state, and σ(v,t) is the dynamic adjustment function; According to the output results of the alarm signal analysis algorithm, the type and severity of the alarm signal are classified as follows: When G(v)<α, the alarm signal is judged to be a Class I emergency alarm signal; When α≤G(v)<β, the alarm signal is judged to be a Class II abnormal alarm signal; When G(v)≥β, the signal is judged to be a potential alarm signal of Class III.

7. A method for preventing leakage of important information in centralized monitoring of power plants and stations as claimed in claim 6, characterized in that: The response measures include: if the signal is a Class I emergency alarm signal, using a graph search algorithm to find the affected equipment and key nodes and prioritize processing, the system automatically triggers the emergency response procedure, initiates safety lock measures, shutdown operations or activates emergency plans, the system immediately notifies relevant personnel and emergency services, including the security management team and maintenance personnel, and the system records detailed information on all relevant events and responses; If the signal is a Class II abnormal alarm signal, apply centrality analysis to identify key devices or nodes in the network, predict and prevent the spread of faults, use community detection methods to identify the affected device groups, optimize monitoring and response strategies, conduct detailed system inspections, determine the specific cause of the abnormality, repair or adjust the fault point, and strengthen monitoring of key components; If the signal is a Class III potential alarm signal, use the label propagation algorithm to continue observing and classifying the signal, ensure that all relevant information is processed, update the attributes and status of the nodes in the graph model, reflect the latest monitoring data and prediction information, adjust the operating parameters or perform scheduled maintenance work based on the analysis results, analyze the signal regularly, and evaluate the impact on system performance.

8. A system based on the method for preventing leakage of important information in centralized monitoring of power plants according to any one of claims 1 to 7, characterized in that: It includes data acquisition and preprocessing module, alarm signal splitting module, signal classification and analysis module and response and control module; The data acquisition and preprocessing module extracts four-telemetry table data from the master station OCS system database, including operation data, equipment data and monitoring responsibility area data, and ensures the integrity and accuracy of the data through preprocessing, filling missing values, data type conversion and abnormal value detection; The alarm signal splitting module analyzes the collected and preprocessed data, compares the voltage and temperature fluctuations, filters the noise signal, uses the Pearson correlation coefficient to determine the correlation between the signal and the key signal, and splits the device alarm signal; The signal classification and analysis module establishes a standard alarm signal model based on graph theory, assigns initial labels to signals by initializing node attributes, and updates signal labels using a label propagation algorithm; The response and control module calculates the alarm signal scores and classifies the signals into three alarm types. According to the alarm type, the system automatically takes corresponding response measures.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images