A power distribution monitoring management method and system based on the Internet of Things

Abstract

The present invention provides a power distribution monitoring management method and system based on the Internet of Things, which relates to the technical field of the Internet of Things. The method comprises: determining the number of categories into which the pre-processed data is to be divided. k , randomly determined k data points as the initial centroid, which represents the center of each cluster; for each point in the preprocessed data, calculate its k The distance between the centroids is calculated and assigned to the cluster represented by the corresponding centroid; for each cluster, the average value of all data points is recalculated and the average value is set as the new centroid. The steps are repeated until the preset number of iterations is reached to obtain the clustering result. The preprocessed data is divided into k Clusters are formed, and the data points in each cluster have similar characteristics in voltage, current, power factor, temperature and humidity; dynamic adjustment factors are calculated based on historical data and real-time data. The present invention can realize comprehensive monitoring and early warning of the power distribution system, thereby ensuring the stable operation of the power system.

Description

Technical Field

[0001] The present invention relates to the technical field of Internet of Things, and in particular to a power distribution monitoring management method and system based on the Internet of Things. Background Art

[0002] The distribution system monitoring and management methods commonly used in the past, such as regular manual inspections and alarms based on fixed thresholds, may have certain limitations in efficiency and response speed, and sometimes it may be difficult to capture and respond to potential problems or anomalies in a timely manner.

[0003] In previous power distribution monitoring practices, due to certain limitations of technology and equipment, there may still be some blind spots in the continuous and comprehensive monitoring of key parameters of the power distribution system. At the same time, relying on fixed thresholds and simple logic to evaluate the operating status of the power distribution system may not always make accurate and timely abnormal identification when facing changeable and complex actual situations. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide a power distribution monitoring management method and system based on the Internet of Things, which can realize comprehensive monitoring and early warning of the power distribution system, thereby ensuring the stable operation of the power system.

[0005] In order to solve the above technical problems, the technical solution of the present invention is as follows:

[0006] In a first aspect, a power distribution monitoring and management method based on the Internet of Things, the method comprising:

[0007] Obtain various raw data of the power distribution system in real time, including voltage, current, power factor, temperature and humidity;

[0008] Preprocessing the original data to obtain preprocessed data;

[0009] Determine the number of categories to split the preprocessed data into , randomly determined data points as the initial centroid, which represents the center of each cluster; for each point in the preprocessed data, calculate its The distance between the centroids is calculated and assigned to the cluster represented by the corresponding centroid; for each cluster, the average value of all data points is recalculated and the average value is set as the new centroid. The steps are repeated until the preset number of iterations is reached to obtain the clustering result. The preprocessed data is divided into clusters, and the data points in each cluster have similarities in voltage, current, power factor, temperature, and humidity characteristics;

[0010] Calculate dynamic adjustment factors based on historical data and real-time data;

[0011] By comparing the current operating status data of the power distribution system with the centroid of each cluster, a comparison result is obtained;

[0012] Analyze the comparison results and the dynamic adjustment factors to obtain analysis results;

[0013] Based on the analysis results, it is determined whether the power distribution system is abnormal. If it is abnormal, the alarm information will be sent to the management personnel in real time, so that the management personnel can remotely access the power distribution system through the Internet of Things after receiving the alarm information and adjust the power distribution parameters.

[0014] Furthermore, for each point in the preprocessed data, calculate its The distance between the centroids is calculated and assigned to the clusters represented by the corresponding centroids, including:

[0015] For each point in the preprocessed data, Calculate its The distance between the centroids, where Represents data points and centroid The distance between Represents data points In the The value of a feature; Represents the centroid In the The value of a feature; and is the coefficient; Represents the dimensions of the data points and centroids on each feature; Is a summation symbol, which means the sum of All The values ​​are summed;

[0016] According to the distance, each point in the preprocessed data is assigned to the cluster represented by the corresponding centroid.

[0017] Furthermore, the dynamic adjustment factor is calculated based on historical data and real-time data, including:

[0018] Based on historical data and real-time data, Calculating Dynamic Adjustment Factors ;

[0019] in, Indicates The weight of each data point; Indicates The value of a real-time data point represents the actual observation value in the current environment; Representation and The corresponding The values ​​of historical data points represent the values ​​observed in the past under similar conditions; Represents the total number of data points; Represents the summation symbol, which indicates the sum from 1 to All The values ​​are summed.

[0020] Furthermore, by comparing the current operating status data of the power distribution system with the centroids of each cluster, a comparison result is obtained, including:

[0021] For the current operating status data of the distribution system, calculate the Euclidean distance between it and the centroid of each cluster;

[0022] Compare all calculated Euclidean distances to obtain the centroid corresponding to the current running status data;

[0023] According to the distance to the centroid corresponding to the current running status data, the cluster corresponding to the current running status data and the comparison result are determined.

[0024] Further, the comparison results and the dynamic adjustment factors are analyzed to obtain analysis results, including:

[0025] pass Calculate the correlation between the comparison result and the dynamic adjustment factor, where is the correlation coefficient; and They are Comparison results and dynamic adjustment factors for each data point; and They are The comparison results of the data points and the weights of the dynamic adjustment factors; and is the weighted average; is the total number of data points;

[0026] The correlation is analyzed to obtain the analysis results.

[0027] Furthermore, the alarm information is sent to the management personnel in real time, including:

[0028] When the power distribution system detects an abnormality, it generates corresponding alarm information according to the type of abnormality;

[0029] Classify the generated alarm information and set an initial urgency score for each type of alarm information;

[0030] Initialize a population, where each individual in the population represents a sequence of sending alarm information;

[0031] Define the fitness function and iteratively optimize the population to find the final order of sending alarm information;

[0032] Sort the alarm information according to the final alarm information sending order;

[0033] Store the sorted alarm information in a priority queue and create an alarm information sending task;

[0034] The sending task periodically checks the priority queue and takes out the alarm information in the order in the queue;

[0035] According to the type and urgency score of the retrieved alarm information, the communication method is determined and the alarm information is sent to the management personnel in real time.

[0036] Furthermore, the calculation formula of the fitness function is:

[0037] ;

[0038] in, is the value of the fitness function, which indicates the fitness of the order in which the alarms are sent; is the total number of alarm messages; , and is the weight coefficient; Indicates The time interval from when an alarm message is generated to when it is sent; , and is the weight coefficient; Indicates Duration of the alert message; Indicates The priority coefficient of the alarm information; is the sum of the number of alarm types; It is Preset severity factors for each alarm type; is a binary variable indicating the Is the alarm of type, if yes, then ,otherwise, ; It is the The number of devices affected by the alert; and are index variables, which are used to traverse the alarm information and alarm type respectively.

[0039] In the second aspect, a power distribution monitoring and management system based on the Internet of Things includes:

[0040] The acquisition module is used to acquire various raw data of the power distribution system in real time, including voltage, current, power factor, temperature and humidity; preprocess the raw data to obtain preprocessed data; determine the number of categories K into which the preprocessed data is to be divided, and randomly determine K data points as initial centroids, where the centroids represent the centers of each cluster; for each point in the preprocessed data, calculate its distance from the K centroids, and assign it to the cluster represented by the corresponding centroid; for each cluster, recalculate the average value of all its data points, set the average value as the new centroid, and repeat the steps until a preset number of iterations is reached to obtain a clustering result, wherein the preprocessed data is divided into K clusters, and the data points in each cluster have similarities in voltage, current, power factor, temperature and humidity characteristics;

[0041] The processing module is used to calculate the dynamic adjustment factor based on historical data and real-time data; obtain the comparison result by comparing the current operating status data of the distribution system with the centroid of each cluster; analyze the comparison result and the dynamic adjustment factor to obtain the analysis result; and determine whether the distribution system is abnormal based on the analysis result. If it is abnormal, the alarm information is sent to the management personnel in real time, so that the management personnel can remotely access the distribution system through the Internet of Things after receiving the alarm information and adjust the distribution parameters.

[0042] According to a third aspect, a computing device includes:

[0043] one or more processors;

[0044] The storage device is used to store one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors implement the method described.

[0045] In a fourth aspect, a computer-readable storage medium stores a program, and when the program is executed by a processor, the method described is implemented.

[0046] The above solution of the present invention includes at least the following beneficial effects:

[0047] This method can continuously monitor the operating status of the system by acquiring various raw data of the distribution system in real time. Once an abnormality is found, the system can immediately issue an alarm to ensure timely discovery and handling of the problem. By preprocessing and clustering the raw data, this method can help managers better understand the operating status of the distribution system and classify the system status according to data characteristics. By calculating the dynamic adjustment factor and combining historical data with real-time data, this method can adapt to the dynamic changes of the distribution system and provide managers with more accurate adjustment suggestions to optimize system performance. The application of IoT technology enables managers to remotely access the distribution system and adjust the distribution parameters. This not only improves the convenience of operation, but also reduces the need for on-site intervention and reduces operation and maintenance costs.

[0048] Through real-time monitoring and early warning mechanisms, as well as data-driven decision support, this method helps to discover and resolve potential safety hazards in advance, thereby significantly improving the safety and stability of the distribution system. The entire monitoring and management method embodies a high degree of intelligence, can automatically process large amounts of data, provide intelligent analysis and early warning, simplify traditional manual inspections and data processing processes, and improve the overall management efficiency and intelligence level of the distribution system. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 It is a flow chart of a power distribution monitoring and management method based on the Internet of Things provided by an embodiment of the present invention.

[0050] Figure 2 It is a schematic diagram of a power distribution monitoring and management system based on the Internet of Things provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0051] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0052] like Figure 1 As shown, an embodiment of the present invention proposes a power distribution monitoring and management method based on the Internet of Things, and the method includes the following steps:

[0053] Step 11, obtaining various raw data of the power distribution system in real time, including voltage, current, power factor, temperature and humidity;

[0054] Step 12, preprocessing the original data to obtain preprocessed data;

[0055] Step 13: Determine the number of categories into which the preprocessed data should be divided , randomly determined data points as the initial centroid, which represents the center of each cluster; for each point in the preprocessed data, calculate its The distance between the centroids is calculated and assigned to the cluster represented by the corresponding centroid; for each cluster, the average value of all data points is recalculated and the average value is set as the new centroid. The steps are repeated until the preset number of iterations is reached to obtain the clustering result. The preprocessed data is divided into clusters, and the data points in each cluster have similarities in voltage, current, power factor, temperature, and humidity characteristics;

[0056] Step 14, calculating a dynamic adjustment factor based on historical data and real-time data;

[0057] Step 15, comparing the current operating status data of the power distribution system with the centroids of each cluster to obtain a comparison result;

[0058] Step 16, analyzing the comparison result and the dynamic adjustment factor to obtain an analysis result;

[0059] Step 17, based on the analysis results, determine whether the power distribution system is abnormal. If it is abnormal, send the alarm information to the management personnel in real time, so that the management personnel can remotely access the power distribution system through the Internet of Things after receiving the alarm information and adjust the power distribution parameters.

[0060] In the embodiment of the present invention, step 11, through real-time data collection, it can ensure that the monitoring and management system always grasps the latest status of the distribution system. Step 12, data preprocessing can clean and organize the original data, eliminate outliers, noise and redundant information, and improve data quality and availability. Step 13, through cluster analysis, data points with similar characteristics can be classified into one category, which is helpful to identify different operating modes or states in the distribution system. Step 14, the introduction of dynamic adjustment factors makes the monitoring and management method more flexible and adaptive, and can be dynamically adjusted according to the historical performance and current state of the distribution system, improving the accuracy and sensitivity of anomaly detection. Step 15, by comparing with the historical clustering results, it can quickly identify whether the current state of the distribution system deviates from the normal operating range, providing key information for timely anomaly detection and early warning. Step 16, comprehensive comparison results and analysis of dynamic adjustment factors can provide managers with a more comprehensive and in-depth evaluation of system operation status. Step 17, real-time anomaly detection and alarm mechanism can ensure that managers are informed of abnormal conditions of the distribution system at the first time, so as to quickly take measures to intervene, prevent the problem from expanding, and ensure the safe and stable operation of the distribution system. At the same time, the remote access and adjustment functions realized through Internet of Things technology have greatly improved management efficiency and response speed.

[0061] In another preferred embodiment of the present invention, the above step 11 may include:

[0062] Install corresponding sensors and data acquisition devices in the power distribution system, which can measure and record key parameters such as voltage, current, power factor, temperature and humidity in real time; the data collected by the sensors are transmitted to the central data processing unit by wired or wireless means, for example, using Internet of Things (IoT) technology such as MQTT (Message Queue Telemetry Transport) protocol to send data to cloud servers or local servers. The raw data is stored in the database for subsequent processing and analysis. The choice of database can be determined according to actual needs and system scale. For example, you can choose a relational database (such as MySQL, PostgreSQL) or a non-relational database (such as MongoDB, Cassandra).

[0063] The above step 12 may include:

[0064] Calculate the Z-score for each data point. The Z-score is a statistical measurement that indicates the distance of a value relative to the mean and standard deviation of the entire data set. For each feature (such as voltage, current, etc.), calculate its mean and standard deviation. For each data point, calculate its Z-score. If the Z-score of a data point exceeds a preset threshold (such as ±3), the data point can be considered an outlier and considered for removal or replacement. By comparing each row or record in the data set, check whether there are exactly the same records. If so, keep one record and delete the remaining duplicate records. Since different features (such as voltage, current, temperature, etc.) may have different dimensions and ranges, normalization is required so that the metrics can be unified in subsequent data processing and analysis, for example, scaling the data to the range of [0, 1]. Extract the peak, valley or average value of power from the original data as a new feature.

[0065] In a preferred embodiment of the present invention, the above step 13 may include:

[0066] Determine the number of categories , randomly selected from the preprocessed data data points as the initial centroid, which is the point representing the center of each cluster; for each point in the preprocessed data, calculate its The distance between the centroids is calculated; each data point is assigned to the cluster represented by the nearest centroid; for each cluster, the average value of all data points in each feature (voltage, current, power factor, temperature and humidity) is calculated, and the calculated average value is set as the new centroid position; the steps are repeated, that is, the data points are reallocated to the nearest centroid and the centroid position is updated until the stop condition is met (the preset number of iterations is reached). After multiple iterations, the data will be divided into The data points in each cluster have similar features.

[0067] For example, suppose there is a data set about a power distribution system, which contains 1000 data points, each of which has 5 features: voltage, current, power factor, temperature, and humidity. It is hoped that these data points can be divided into 3 clusters ( =3), set =3, which means that we want to divide the data into 3 clusters; randomly select 3 points from 1000 data points as the initial centroids; for each point in the data set, calculate its distance from the 3 centroids; according to the principle of the nearest distance, assign the data points to the corresponding clusters; for each cluster, calculate the average values ​​of voltage, current, power factor, temperature and humidity. Repeat the above assignment and centroid update process until the preset number of iterations (for example, 100 times) is reached; finally, 3 clusters are obtained, and the data points in each cluster have high similarity in voltage, current, power factor, temperature and humidity. These clusters may represent the data distribution of the distribution system under different operating conditions, which helps to better understand and monitor the behavior of the system.

[0068] In a preferred embodiment of the present invention, for each point in the preprocessed data, its The distance between the centroids is calculated and assigned to the clusters represented by the corresponding centroids, including:

[0069] For each point in the preprocessed data, Calculate its The distance between the centroids, where Represents data points and centroid The distance between Represents data points In the The value of a feature; Represents the centroid In the The value of a feature; and is the coefficient; Represents the dimensions of the data points and centroids on each feature; Is a summation symbol, which means the sum of All The values ​​are summed;

[0070] According to the distance, each point in the preprocessed data is assigned to the cluster represented by the corresponding centroid, specifically including: after calculating the distance between the current data point and all the centroids, find the centroid with the smallest distance, which is the centroid that is "most similar" or "closest" to the current data point; assign the current data point to the cluster represented by the nearest centroid found. In actual operation, this usually means adding the index or reference of the data point to the list associated with the centroid, and repeating the steps until all preprocessed data points are assigned to the corresponding clusters. When all data points are processed, each cluster will contain a group of data points that are closest to its centroid.

[0071] In the embodiment of the present invention, the algorithm can adapt to different data types and feature spaces through the defined distance function. By calculating the distance between each data point and all centroids and assigning them based on the minimum distance, the process ensures that each data point is accurately assigned to the cluster represented by the centroid that is most similar to it, which helps to form compact and distinct clusters in cluster analysis. Each data point is assigned to a cluster represented by a centroid, which helps to understand the grouping of data points and the similarity within the group, and can be easily extended to a large number of data points and multiple feature dimensions.

[0072] In a preferred embodiment of the present invention, the above step 14, calculating the dynamic adjustment factor according to historical data and real-time data, may include:

[0073] Step 141, based on historical data and real-time data, Calculating Dynamic Adjustment Factors ;

[0074] in, Indicates The weight of each data point; Indicates The value of a real-time data point represents the actual observation value in the current environment; Representation and The corresponding The values ​​of historical data points represent the values ​​observed in the past under similar conditions; Represents the total number of data points; Represents the summation symbol, which indicates the sum from 1 to All The values ​​are summed.

[0075] In an embodiment of the present invention, the method can adapt to changing environmental conditions by dynamically adjusting the factor. Since real-time data is included in the calculation, the adjustment factor can be dynamically updated as the actual situation changes, thereby improving flexibility and adaptability. By comprehensively considering historical data and real-time data, as well as the weight of each data point, the method can more accurately reflect the actual situation in the current environment. This accuracy not only helps to make more accurate decisions, but also improves the performance and efficiency of the overall system. Using the ratio of historical data to real-time data to calculate the dynamic adjustment factor can reduce the impact of outliers or noise data. When the real-time data is abnormal, the historical data can provide a benchmark so that the adjustment factor will not change dramatically due to the fluctuation of a single data, thereby enhancing the stability of the system. The method calculates the dynamic adjustment factor through a clear mathematical formula, which increases the interpretability of the model. Users can clearly understand how the adjustment factor is calculated based on historical data and real-time data, which helps to build trust and promote the acceptance of the model in practical applications.

[0076] In a preferred embodiment of the present invention, the above step 15, by comparing the current operating status data of the power distribution system with the centroids of each cluster to obtain a comparison result, may include:

[0077] Step 151, for the current operating status data of the power distribution system, calculate the Euclidean distance between it and the centroid of each cluster;

[0078] Step 152, comparing all calculated Euclidean distances to obtain the centroid corresponding to the current running state data;

[0079] Step 153: Determine the cluster corresponding to the current running status data and the comparison result according to the distance to the centroid corresponding to the current running status data.

[0080] In the embodiment of the present invention, step 151, obtain the current operating status data of the power distribution system, including a series of characteristic values, such as voltage, current, power factor, etc.; extract the centroid data of each cluster from the previous K-means clustering results, these centroid data represent the center point of each cluster, and are also composed of a series of characteristic values; for the current operating status data and the centroid of each cluster, calculate the Euclidean distance between them. Step 152, list all the Euclidean distances calculated in step 151, and find the smallest one from the listed Euclidean distances. This minimum distance represents the closest degree of proximity between the current operating status data and the centroid of a cluster, which may specifically include:

[0081] The current operating status data is obtained from the power distribution system. These data include key parameters such as voltage, current, and power factor. These data are represented as a vector, denoted as X current, where each element corresponds to a eigenvalue; from the previous K-means clustering results, obtain the centroid data of all clusters. Each centroid is also a vector composed of the same features, recorded as C i ( i Represents the index of the cluster); create an array or list to store the Euclidean distance between the current running status data and each cluster centroid; for each cluster centroid C i , perform the following steps to calculate the current running status data X current The Euclidean distance between:

[0082] Initialize an accumulator variable sum of is 0; traverse X current and C i Each corresponding element (eigenvalue) in is denoted as x j and c ij ( j represents the index of the feature); for each pair of corresponding elements, calculate the square of the difference, that is, ( x j - c ij ) 2 , and add to sum of When all eigenvalues ​​have been traversed, calculate sum of The square root of the current running status data and cluster centroid is obtained C i The Euclidean distance between the clusters is calculated; the calculated Euclidean distance is stored in the distance array initialized previously. When the Euclidean distances between all cluster centroids and the current running status data are calculated, the distance array is output. Each element in this array represents the degree of closeness between the current running status data and the corresponding cluster centroid.

[0083] Step 153, based on the minimum Euclidean distance found in step 152, determine the cluster corresponding to the current operating status data. This cluster is the cluster that is most similar to the current operating status data. The determined cluster is output as the comparison result, which can be used for subsequent tasks such as distribution system status monitoring and fault warning.

[0084] In a preferred embodiment of the present invention, the above step 16, analyzing the comparison result and the dynamic adjustment factor to obtain the analysis result, includes:

[0085] Step 161, pass Calculate the correlation between the comparison result and the dynamic adjustment factor, where is the correlation coefficient; and They are Comparison results and dynamic adjustment factors for each data point; and They are The comparison results of the data points and the weights of the dynamic adjustment factors; and is the weighted average; is the total number of data points;

[0086] Step 162, analyze the correlation to obtain the analysis result; the value range of the correlation coefficient is between -1 and 1. If the correlation coefficient is close to 1, it means that there is a strong positive correlation between the comparison result and the dynamic adjustment factor; if it is close to -1, it means a strong negative correlation; if it is close to 0, it means that the relationship between the two is weak or there is no linear relationship; according to the calculated correlation coefficient value, the strength and direction of the correlation between the comparison result and the dynamic adjustment factor are determined. For example, if the correlation coefficient is 0.8, it means that there is a strong positive correlation between the two.

[0087] In an embodiment of the present invention, by calculating the correlation coefficient, the method can quantify the strength and direction of the relationship between the comparison result and the dynamic adjustment factor, which helps to understand the mutual influence between the two more accurately. Correlation analysis based on actual data can improve the accuracy and effectiveness of decision-making. For example, if the dynamic adjustment factor has a significant impact on the comparison result, then resources can be invested in optimizing these factors as a priority. By analyzing the correlation, possible risk factors can be predicted and adjusted. If a dynamic adjustment factor is highly correlated with the comparison result, then any change to this factor may have a significant impact on the final result, so that risk management can be carried out in advance.

[0088] The above step 17 may include:

[0089] Analyze the correlation analysis results calculated in step 16. This includes understanding the correlation coefficient value between the comparison result and the dynamic adjustment factor, as well as the correlation strength and direction represented by this value. Before performing the analysis, a threshold or normal range is set based on historical data. The threshold can be a specific value of the correlation coefficient, which is used to define what kind of correlation strength is considered normal or abnormal; compare the calculated correlation coefficient with the preset threshold. If the correlation coefficient exceeds the preset normal range, this may mean that there has been an abnormal change in the relationship between the comparison result and the dynamic adjustment factor. Based on the above comparison, if the analysis result exceeds the preset normal range or threshold, and this change cannot be explained by normal fluctuations or known factors, then it can be determined that there is an abnormality in the distribution system.

[0090] In a preferred embodiment of the present invention, the alarm information is sent to the management personnel in real time, including:

[0091] When the power distribution system detects an abnormality, it generates corresponding alarm information according to the type of abnormality;

[0092] Classify the generated alarm information and set an initial urgency score for each type of alarm information;

[0093] Initialize a population, where each individual in the population represents a sequence of sending alarm information;

[0094] Define the fitness function and iteratively optimize the population to find the final order of sending alarm information;

[0095] Sort the alarm information according to the final alarm information sending order;

[0096] Store the sorted alarm information in a priority queue and create an alarm information sending task;

[0097] The sending task periodically checks the priority queue and takes out the alarm information in the order in the queue;

[0098] According to the type and urgency score of the retrieved alarm information, the communication method is determined and the alarm information is sent to the management personnel in real time.

[0099] In the embodiment of the present invention, it is possible to ensure that the alarm information can be quickly sent to the management personnel after it is generated, thereby reducing the time delay from the occurrence of the abnormality to the knowledge of the management personnel. By classifying the alarm information and scoring the urgency, the system can determine which alarms need to be handled first. This priority sorting ensures that the most important and urgent problems are handled first when resources are limited. The use of population optimization and fitness function to determine the order in which the alarm information is sent not only takes into account the urgency of the alarm, but also other factors (such as the availability of the management personnel, the reliability of the communication channel, etc.), thereby optimizing the alarm sending strategy more flexibly and comprehensively. By optimizing the order in which the alarm information is sent, the system can help the management personnel handle the alarm more efficiently, reduce unnecessary delays and confusion, and the management personnel can reasonably allocate time and resources according to the priority of the alarm information, thereby improving the overall response efficiency. The orderly and clear priority alarm sending method helps to reduce the information overload of the management personnel. The management personnel can focus more on handling high-priority alarms without having to find key information in a large number of chaotic alarms.

[0100] When specifically applied, this includes:

[0101] The power distribution system continuously monitors key indicators. Once it detects that the data exceeds the preset threshold or an abnormal pattern appears, it is judged as abnormal. The system automatically generates detailed alarm information based on the specific type of abnormality (such as voltage fluctuation, excessive temperature, equipment failure, etc.), including the type of abnormality, the time of occurrence, the specific value of the abnormal data, etc. The system classifies the generated alarm information, for example, the alarm is divided into categories such as "serious failure", "performance degradation", and "preventive maintenance reminder", and an initial urgency score is set for each type of alarm, such as from 1 (low urgency) to 5 (high urgency); a population is initialized, in which each individual represents a possible order for sending alarm information; a fitness function is defined number, iteratively optimize the population, and find the alarm sending order that can maximize the fitness function; sort the alarm information according to the optimized alarm information sending order; store the sorted alarm information in a priority queue, and the priority of the queue is determined by the urgency of the alarm and the sending order; create an alarm information sending task, which will periodically check the priority queue. When the task runs, it will take out the alarm information in the order of priority in the queue; determine the most appropriate communication method (such as SMS, email, application push, etc.) according to the type of alarm information taken out and the urgency score, and send the alarm information to managers in real time to ensure that they can respond in time.

[0102] For example, suppose that on a certain day, the power distribution system detects two anomalies: one is that the transformer temperature is too high, and the other is that a switch cabinet fails. The system detects that the transformer temperature exceeds the safety threshold and generates an alarm message: "Transformer temperature is too high, current temperature: XX℃, safety threshold: YY℃". At the same time, the system detects a switch cabinet failure and generates an alarm message: "Switch cabinet failure, fault code: ZZZ"; "Transformer temperature is too high" is classified as "performance degradation" and the urgency score is 4; "Switch cabinet failure" is classified as "serious failure" and the urgency score is 5; the sending order is determined by the optimization algorithm: "Switch cabinet failure" is sent first, and then "Transformer temperature is too high" is sent; according to the optimized order, the alarm information is stored in the priority queue; the alarm sending task is started, and the "switch cabinet failure" alarm is first taken out. Since "switch cabinet failure" is a serious fault with a high urgency, the system chooses to send the alarm to the management personnel through SMS and application push at the same time. Subsequently, the task takes out the "transformer temperature is too high" alarm and sends it to the management personnel by email because its urgency is slightly lower. In this way, managers can deal with the most serious faults first and then solve other less serious problems in a timely manner, thereby ensuring the stable operation of the distribution system.

[0103] In a preferred embodiment of the present invention, the calculation formula of the fitness function is:

[0104] ;

[0105] in, is the value of the fitness function, which indicates the fitness of the order in which the alarms are sent; is the total number of alarm messages; , and is the weight coefficient; Indicates The time interval from when an alarm message is generated to when it is sent; , and is the weight coefficient; Indicates Duration of the alert message; Indicates The priority coefficient of the alarm information; is the sum of the number of alarm types; It is Preset severity factors for each alarm type; is a binary variable indicating the Is the alarm of type, if yes, then ,otherwise, ; It is the The number of devices affected by the alert; and are index variables, which are used to traverse the alarm information and alarm type respectively.

[0106] In an embodiment of the present invention, the function comprehensively considers multiple factors that affect the importance of alarm information, including the time interval, duration, priority, type severity, number of affected devices, etc. of the alarm generation. This comprehensive evaluation method can more comprehensively reflect the actual importance and urgency of the alarm, and help to formulate a more reasonable alarm sending order. By adjusting the weight coefficient, the importance of different factors in the fitness calculation can be flexibly set according to actual needs. Through the priority coefficient, the function can clearly reflect the priority of different alarms to ensure that high-priority alarms can be sent first. This helps managers to quickly respond to and handle critical alarms and reduce potential losses. Through the alarm type severity coefficient and binary variables, the function can distinguish different types of alarms and weight them according to their preset severity. This distinction helps managers better understand the nature and severity of the alarm and make a more appropriate response. By considering the number of devices affected by the alarm, the function can reflect the degree of impact of the alarm on the overall operation of the system, which helps managers to more comprehensively evaluate the urgency and importance of the alarm. By comparing the fitness values ​​of different alarm sending orders, the optimal sending strategy can be selected, thereby improving the decision-making efficiency and accuracy of managers.

[0107] like Figure 2 As shown, an embodiment of the present invention further provides a power distribution monitoring and management system 20 based on the Internet of Things, comprising:

[0108] The acquisition module 21 is used to acquire various raw data of the power distribution system in real time, including voltage, current, power factor, temperature and humidity; preprocess the raw data to obtain preprocessed data; determine the number of categories to be divided into for the preprocessed data; , randomly determined data points as the initial centroid, which represents the center of each cluster; for each point in the preprocessed data, calculate its The distance between the centroids is calculated and assigned to the cluster represented by the corresponding centroid; for each cluster, the average value of all data points is recalculated and the average value is set as the new centroid. The steps are repeated until the preset number of iterations is reached to obtain the clustering result. The preprocessed data is divided into clusters, and the data points in each cluster have similarities in voltage, current, power factor, temperature, and humidity characteristics;

[0109] The processing module 22 is used to calculate the dynamic adjustment factor based on the historical data and the real-time data; obtain the comparison result by comparing the current operating status data of the distribution system with the centroid of each cluster; analyze the comparison result and the dynamic adjustment factor to obtain the analysis result; and determine whether the distribution system is abnormal based on the analysis result. If it is abnormal, the alarm information is sent to the management personnel in real time, so that the management personnel can remotely access the distribution system through the Internet of Things after receiving the alarm information and adjust the distribution parameters.

[0110] It should be noted that the system is a system corresponding to the above method, and all implementation methods in the above method embodiment are applicable to this embodiment and can achieve the same technical effect.

[0111] The embodiment of the present invention further provides a computing device, comprising: a processor, a memory storing a computer program, wherein when the computer program is executed by the processor, the method described above is executed. All implementations in the above method embodiment are applicable to this embodiment and can achieve the same technical effect.

[0112] The embodiment of the present invention also provides a computer-readable storage medium storing instructions, which, when executed on a computer, enable the computer to execute the method described above. All implementations in the above method embodiment are applicable to this embodiment and can achieve the same technical effect.

Claims

1. A power distribution monitoring and management method based on the Internet of Things, characterized in that: The method comprises: Obtain various raw data of the power distribution system in real time, including voltage, current, power factor, temperature and humidity; Preprocessing the original data to obtain preprocessed data; Determine the number of categories to split the preprocessed data into , randomly determined data points as the initial centroid, which represents the center of each cluster; for each point in the preprocessed data, calculate its The distance between the centroids is calculated and assigned to the cluster represented by the corresponding centroid; for each cluster, the average value of all data points is recalculated and the average value is set as the new centroid. The steps are repeated until the preset number of iterations is reached to obtain the clustering result. The preprocessed data is divided into clusters, and the data points in each cluster have similarities in voltage, current, power factor, temperature, and humidity characteristics; Calculate dynamic adjustment factors based on historical data and real-time data; By comparing the current operating status data of the power distribution system with the centroids of each cluster, a comparison result is obtained; Analyze the comparison results and the dynamic adjustment factors to obtain analysis results; According to the analysis results, it is determined whether the power distribution system is abnormal. If it is abnormal, the alarm information is sent to the management personnel in real time, so that the management personnel can remotely access the power distribution system through the Internet of Things after receiving the alarm information and adjust the power distribution parameters; For each point in the preprocessed data, calculate its The distance between the centroids is calculated and assigned to the clusters represented by the corresponding centroids, including: For each point in the preprocessed data, Calculate its The distance between the centroids, where Represents data points and centroid The distance between Represents data points In the The value of a feature; Represents the centroid In the The value of a feature; and is the coefficient; Represents the dimensions of the data points and centroids on each feature; Is a summation symbol, which means the sum of All The values ​​are summed; According to the distance, each point in the preprocessed data is assigned to the cluster represented by the corresponding centroid; the dynamic adjustment factor is calculated based on the historical data and real-time data, including: Based on historical data and real-time data, Calculating Dynamic Adjustment Factors ; in, Indicates The weight of each data point; Indicates The value of a real-time data point represents the actual observation value in the current environment; Representation and The corresponding The values ​​of historical data points represent the values ​​observed in the past under similar conditions; Represents the total number of data points; Represents the summation symbol, which indicates the sum from 1 to All The values ​​are summed; By comparing the current operating status data of the power distribution system with the centroid of each cluster, the comparison results are obtained, including: For the current operating status data of the distribution system, calculate the Euclidean distance between it and the centroid of each cluster; Compare all calculated Euclidean distances to obtain the centroid corresponding to the current running status data; Determine the cluster corresponding to the current running status data and the comparison result according to the centroid corresponding to the current running status data; The comparison results and the dynamic adjustment factors are analyzed to obtain analysis results, including: pass Calculate the correlation between the comparison result and the dynamic adjustment factor, where is the correlation coefficient; and They are Comparison results and dynamic adjustment factors for each data point; and They are The comparison results of the data points and the weights of the dynamic adjustment factors; and is the weighted average; is the total number of data points; Analyze the correlation to obtain analysis results; Send alert information to managers in real time, including: When the power distribution system detects an abnormality, it generates corresponding alarm information according to the type of abnormality; Classify the generated alarm information and set an initial urgency score for each type of alarm information; Initialize a population, where each individual in the population represents a sequence of sending alarm information; Define the fitness function and iteratively optimize the population to find the final order of sending alarm information; Sort the alarm information according to the final alarm information sending order; Store the sorted alarm information in a priority queue and create an alarm information sending task; The sending task periodically checks the priority queue and takes out the alarm information in the order in the queue; Determine the communication method based on the type and urgency score of the retrieved alarm information, and send the alarm information to the management personnel in real time; The calculation formula of the fitness function is: ； in, is the value of the fitness function, which indicates the fitness of the order in which the alarms are sent; is the total number of alarm messages; , and is the weight coefficient; Indicates The time interval from when an alarm message is generated to when it is sent; , and is the weight coefficient; Indicates Duration of the alert message; Indicates The priority coefficient of the alarm information; is the sum of the number of alarm types; It is Preset severity factors for each alarm type; is a binary variable indicating the Is the alarm of type, if yes, then ,otherwise, ; It is subject to The number of devices affected by the alert; and are index variables, which are used to traverse the alarm information and alarm type respectively.

2. A power distribution monitoring and management system based on the Internet of Things, characterized in that: As used in claim 1, comprising: The acquisition module is used to obtain various raw data of the power distribution system in real time, including voltage, current, power factor, temperature and humidity; preprocess the raw data to obtain preprocessed data; determine the number of categories to be divided into for the preprocessed data; , randomly determined data points as the initial centroid, which represents the center of each cluster; for each point in the preprocessed data, calculate its The distance between the centroids is calculated and assigned to the cluster represented by the corresponding centroid; for each cluster, the average value of all data points is recalculated and the average value is set as the new centroid. The steps are repeated until the preset number of iterations is reached to obtain the clustering result. The preprocessed data is divided into clusters, and the data points in each cluster have similarities in voltage, current, power factor, temperature, and humidity characteristics; The processing module is used to calculate the dynamic adjustment factor based on historical data and real-time data; obtain the comparison result by comparing the current operating status data of the distribution system with the centroid of each cluster; analyze the comparison result and the dynamic adjustment factor to obtain the analysis result; and determine whether the distribution system is abnormal based on the analysis result. If it is abnormal, the alarm information is sent to the management personnel in real time, so that the management personnel can remotely access the distribution system through the Internet of Things after receiving the alarm information and adjust the distribution parameters.

3. A computing device, characterized in that: include: one or more processors; A storage device, used for storing one or more programs, when the one or more programs are executed by the one or more processors, the one or more processors implement the method as claimed in claim 1.

4. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a program, which implements the method according to claim 1 when executed by a processor.

Images