Substation power system remote monitoring and fault early warning method based on Internet of Things

Abstract

The invention discloses a transformer substation power system remote monitoring and fault early warning method based on the Internet of Things, and relates to the technical field of power equipment monitoring, and the method comprises the steps: laying a monitoring terminal, carrying out the abnormality analysis, determining the laying position of a central gateway, constructing a monitoring terminal communication topology network through employing a star topology structure, and carrying out the fault early warning of the monitoring terminal; the network is used for carrying out asynchronous sensing communication on the monitoring terminal set, the collected monitoring data set is transmitted to the fault detection module for fault prototype matching, and if matching succeeds, fault early warning information is generated according to the matched fault prototype. The technical problems that in the prior art, traditional transformer substation monitoring depends on manual routing inspection and local automatic monitoring, real-time and comprehensive monitoring and early warning means are lacked, fault response lags behind, and precision is insufficient are solved, real-time and comprehensive equipment state monitoring and intelligent fault early warning are carried out based on the Internet of Things, and the fault early warning efficiency is improved. And the fault detection precision and the response speed are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power equipment monitoring, and in particular to a remote monitoring and fault early warning method for a substation power system based on the Internet of Things. Background Art

[0002] With the expansion of power systems and the increasing complexity of substation equipment, traditional equipment monitoring methods are no longer able to meet the demands for real-time, comprehensive, and intelligent monitoring. Existing substation monitoring systems typically rely on manual inspections or localized automated monitoring, failing to provide comprehensive, real-time monitoring of equipment status. When faults occur, they often lack timely warnings and accurate fault diagnosis. Furthermore, the use of single sensors and data transmission delays make traditional systems unable to meet the demands for rapid response and efficient management. Summary of the Invention

[0003] This application provides a remote monitoring and fault warning method for substation power systems based on the Internet of Things, which is used to solve the technical problems in the existing technology that traditional substation monitoring relies on manual inspections and local automatic monitoring, lacks real-time and comprehensive monitoring and warning means, and leads to delayed fault response and insufficient accuracy.

[0004] The present application provides a remote monitoring and fault warning method for a substation power system based on the Internet of Things, the method comprising: traversing the equipment of the substation power system to deploy monitoring terminals and obtain a monitoring terminal set; traversing the monitoring terminal set to perform monitoring anomaly analysis, determining the deployment position of the central gateway based on the analysis results, connecting them using a star topology structure, and constructing a monitoring terminal communication topology network; utilizing the monitoring terminal communication topology network to perform asynchronous perception communication on the monitoring terminal set, transmitting the collected monitoring data set to a fault detection module for fault prototype matching, and if the match is successful, generating fault warning information based on the matched fault prototype.

[0005] One or more technical solutions provided in this application have at least the following technical effects or advantages: The present application provides a remote monitoring and fault warning method for a substation power system based on the Internet of Things, which relates to the technical field of power equipment monitoring. By deploying monitoring terminals and performing anomaly analysis, the location of the central gateway is determined, a star topology is used to build a communication network, and monitoring data is transmitted to a fault detection module through asynchronous sensing communication for fault prototype matching. If the match is successful, fault warning information is generated. This solves the technical problem in the prior art that traditional substation monitoring relies on manual inspections and local automatic monitoring, lacks real-time and comprehensive monitoring and warning means, and results in delayed fault response and insufficient accuracy. This method realizes real-time and comprehensive equipment status monitoring and intelligent fault warning based on the Internet of Things, and improves the accuracy of fault detection and the response speed. BRIEF DESCRIPTION OF THE DRAWINGS

[0006] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0007] Figure 1 A flowchart of a remote monitoring and fault warning method for a substation power system based on the Internet of Things provided in an embodiment of the present application; Figure 2 A schematic diagram of the process of determining the location of the central gateway in the remote monitoring and fault warning method of the substation power system based on the Internet of Things provided in an embodiment of the present application. DETAILED DESCRIPTION

[0008] This application provides a remote monitoring and fault warning method for substation power systems based on the Internet of Things, which is used to solve the technical problems in the existing technology that traditional substation monitoring relies on manual inspections and local automatic monitoring, lacks real-time and comprehensive monitoring and warning means, and leads to delayed fault response and insufficient accuracy.

[0009] The following will be combined with the accompanying drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only some of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0010] It should be noted that the terms "first", "second", etc. in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products or devices.

[0011] Example 1, as Figure 1 As shown, the present application provides a remote monitoring and fault warning method for a substation power system based on the Internet of Things, the method comprising: P10: Traverse the equipment of the substation power system to deploy monitoring terminals and obtain a monitoring terminal set.

[0012] Specifically, during the deployment of monitoring terminals in a substation power system, a comprehensive and detailed traversal of all equipment within the substation power system is first required. This operation ensures that every device is included in the monitoring scope, laying the foundation for subsequent monitoring and fault warning functions. Traversal involves accessing each device in the system one by one according to a specific sequence or rule, ensuring that monitoring terminals are deployed across the entire substation power system without omission. Substation power systems typically include a variety of critical equipment, such as transformers, circuit breakers, disconnectors, busbars, and transmission lines.

[0013] During the traversal process, the appropriate type of monitoring terminal must be selected based on factors such as the equipment type, function, operating environment, and importance. The monitoring terminal is the core component for achieving device status awareness. Its functions include, but are not limited to, real-time collection of electrical parameters (such as voltage, current, and power), as well as monitoring of equipment operating environment parameters (such as temperature, humidity, and partial discharge). For example, for critical equipment such as transformers, a composite monitoring terminal capable of simultaneously monitoring both its electrical performance and internal insulation status is typically required. For outdoor transmission lines, a weather-resistant monitoring terminal capable of monitoring parameters such as the contamination level of line insulators and conductor temperature is required. Furthermore, the selection of the monitoring terminal must consider technical indicators such as accuracy, stability, anti-interference capabilities, and compatibility with existing equipment to ensure long-term stable operation in complex substation environments.

[0014] When deploying monitoring terminals, the location and method of terminal installation must also be considered. The installation location should be selected to ensure that the monitoring terminal can accurately and stably collect the required data while avoiding external interference. For example, the monitoring terminal for a current transformer should be installed on the secondary side of the transformer to ensure accurate collection of the current signal; while the monitoring terminal for high-voltage equipment should be installed in a well-insulated location to prevent damage to the terminal from high-voltage electrical breakdown. In addition, the installation method of the monitoring terminal must be designed based on the specific structure and operating conditions of the equipment to ensure its long-term stable operation. For example, for monitoring terminals installed outdoors, protective measures such as waterproofing, dustproofing, and lightning protection must be taken; while for monitoring terminals installed inside equipment, the impact on the normal operation of the equipment must be considered to ensure that the installation process does not damage the performance and safety of the equipment.

[0015] After deploying the monitoring terminals, each terminal needs to be debugged and calibrated to ensure proper operation and accurate data collection. During debugging, the monitoring terminal's communication functionality needs to be tested to ensure stable data exchange with the subsequent central gateway or other data transmission equipment. Furthermore, the data collected by the monitoring terminal needs to be verified for accuracy and reliability. For example, the accuracy of the monitoring terminal can be verified by comparing the voltage and current data collected by the monitoring terminal with the readings of the device's built-in meter. Simulating equipment failures can also verify the monitoring terminal's ability to detect abnormal conditions.

[0016] Ultimately, through the above process, a complete set of monitoring terminals is obtained. Each monitoring terminal in this set has a unique identifier and is associated with a corresponding device, capable of collecting and transmitting the device's operating status data in real time. This set of monitoring terminals forms the foundation for subsequent monitoring and fault warning functions, providing the necessary technical support for comprehensive, real-time monitoring of substation power system equipment.

[0017] P20: traverse the monitoring terminal set to perform monitoring anomaly analysis, determine the central gateway layout location based on the analysis results, use a star topology structure to connect, and build a monitoring terminal communication topology network.

[0018] Further, such as Figure 2 As shown, step P20 in this embodiment of the application also includes: P21: Perform time series extraction on the monitoring terminal set according to preset monitoring anomaly indicators to obtain a monitoring terminal time series indicator sequence set; P22: Perform trend feature identification on the monitoring terminal time series indicator sequence set to obtain an indicator trend feature set; P23: Perform monitoring anomaly analysis on the indicator trend feature set and the monitoring terminal time series indicator sequence set to obtain an analysis result; P24: Perform center identification on the analysis result, and use the location corresponding to the center identification result as the central gateway deployment location. The preset monitoring anomaly indicators include data fluctuation gradient, data transmission duration error, and data deviation.

[0019] It should be understood that the construction process of the monitoring terminal communication topology network can be further refined, especially in determining the layout location of the central gateway, and a monitoring anomaly analysis mechanism is introduced to ensure the efficiency and reliability of the communication topology network.

[0020] First, the set of monitoring terminals is subjected to time series extraction of indicators according to the preset monitoring anomaly indicators. The preset monitoring anomaly indicators include data fluctuation gradient, data transmission time error, and data deviation. These indicators are key parameters for measuring the operating status of the monitoring terminal and the quality of data transmission. For each monitoring terminal, the corresponding indicator sequence is extracted by performing time series analysis on the data collected by it. Specifically, the data fluctuation gradient reflects the rate of change of the monitoring data in the time series, the data transmission time error measures the time deviation of the data transmitted from the monitoring terminal to the central node, and the data deviation indicates the degree of deviation between the monitoring data and the standard value under normal operating conditions. Through the time series extraction of these indicators, the operating status characteristics of each monitoring terminal at different time points can be obtained to form a set of monitoring terminal time series indicator sequences, with each monitoring terminal corresponding to an indicator sequence.

[0021] Next, trend feature identification is performed on the set of monitoring terminal time series indicator sequences to obtain a set of indicator trend features. This process aims to identify characteristics such as rising, falling, or stable trends in the data through trend analysis of time series data. For example, if the data fluctuation gradient of a monitoring terminal shows a continuous upward trend, it may indicate that there are potential instability factors in the equipment monitored by the terminal; if the data transmission duration error shows periodic changes, it may indicate that there is interference in the communication link. By identifying trend features of the time series indicator sequence of each monitoring terminal, we can have a more comprehensive understanding of the operating status of the monitoring terminal and its changing trends, thereby providing richer information for subsequent monitoring anomaly analysis.

[0022] Subsequently, a monitoring anomaly analysis is performed on the indicator trend feature set and the monitoring terminal time series indicator sequence set to obtain the analysis results. The monitoring anomaly analysis is based on the extracted indicator sequence and identified trend features, and comprehensively evaluates whether the operating status of each monitoring terminal is abnormal. For example, if the data fluctuation gradient of a monitoring terminal is far higher than the normal range and the error in its data transmission duration is also large, then the monitoring anomaly of this terminal is high, indicating that it may have a communication failure or the operating status of the monitored equipment is unstable. By analyzing the monitoring anomaly of all monitoring terminals, it is possible to identify which terminals have abnormalities during data collection or transmission, thereby providing a basis for further optimizing the layout of the communication topology network.

[0023] Finally, the results of the monitoring anomaly analysis are used for center identification, and the location corresponding to the center identification result is used as the layout location of the central gateway. The purpose of center identification is to find a relatively optimal location in the set of monitoring terminals to deploy the central gateway, so that the performance of the entire communication topology network is optimized. Generally, the location of the central gateway should be selected in an area with low monitoring anomalies and high data transmission stability to ensure that the central gateway can efficiently aggregate and process data from various monitoring terminals. For example, if the monitoring terminals in a certain area generally have low data fluctuation gradients and data transmission duration errors, then this area can be considered as the preferred layout location for the central gateway.

[0024] After determining the location of the central gateway, a star topology is used to connect the monitoring terminals to the central gateway, creating a monitoring terminal communication topology network. The advantages of a star topology lie in its simplicity, ease of management and expansion, and the central gateway's ability to centrally manage and process data from each monitoring terminal. In this topology, each monitoring terminal is connected to the central gateway via an independent communication link, providing a clear data transmission path and facilitating fault location and troubleshooting. Furthermore, a star topology offers high fault tolerance; even if a monitoring terminal fails, it will not affect communication between other terminals and the central gateway.

[0025] Through the above steps, not only can the layout location of the central gateway be determined scientifically and reasonably, but also an efficient and stable monitoring terminal communication topology network can be constructed, providing a solid foundation for subsequent fault detection and early warning functions.

[0026] Furthermore, step P22 of the embodiment of the present application further includes: P22-1: Sample the monitoring terminal timing indicator sequence set according to the first identification scale and the second identification scale respectively to obtain a first monitoring terminal timing indicator sampling sequence set and a second monitoring terminal timing indicator sampling sequence set.

[0027] Furthermore, before sampling the monitoring terminal timing indicator sequence set according to the first identification scale and the second identification scale respectively, step P22-1 of the embodiment of the present application further includes: P22-11: Traverse the monitoring terminal set to extract abnormal interval durations and obtain abnormal interval duration clusters; P22-12: Use the maximum abnormal interval duration in the abnormal interval duration cluster as the first identification scale; P22-13: Use the minimum abnormal interval duration in the abnormal interval duration cluster as the second identification scale.

[0028] P22-2: Traverse the first monitoring terminal time series indicator sampling sequence set and the second monitoring terminal time series indicator sampling sequence set to identify trend features and determine the first trend feature set and the second trend feature set; P22-3: Perform one-to-one mapping interaction enhancement on the first trend feature set and the second trend feature set to obtain the indicator trend feature set.

[0029] Optionally, the trend feature identification process of the set of monitoring terminal time series indicator sequences can be further refined. In order to improve the accuracy and reliability of trend feature identification, multi-scale sampling and abnormal interval duration analysis are introduced. Specifically, it is first necessary to traverse the monitoring terminal set and extract the abnormal interval duration of the time series indicator sequence of each monitoring terminal. The abnormal interval duration refers to the time interval when the monitoring data fluctuates significantly or deviates from the normal range in the time series. By analyzing these abnormal interval durations, the instability that may occur in the monitoring terminal during the data collection process can be obtained, and the abnormal interval durations of all monitoring terminals can be summarized to form an abnormal interval duration cluster.

[0030] After obtaining anomaly interval duration clusters, the maximum anomaly interval duration is extracted as the first identification scale, used to capture long-term trend characteristics in the time series data, such as slow changes or periodic fluctuations in equipment operating status. Simultaneously, the minimum anomaly interval duration is extracted as the second identification scale, used to capture short-term trend characteristics in the time series data, such as sudden failures or rapidly changing anomalies. Determining these two identification scales provides a basis for subsequent multi-scale sampling.

[0031] Next, the monitoring terminal's time series indicator sequence set is sampled according to the first and second identification scales, respectively, to obtain two sampling sequence sets at different scales. The first monitoring terminal time series indicator sampling sequence set is sampled based on the first identification scale and is primarily used to capture long-term trend characteristics; the second monitoring terminal time series indicator sampling sequence set is sampled based on the second identification scale and is used to capture short-term trend characteristics. This multi-scale sampling approach can extract the operating status characteristics of the monitoring terminal from different time scales, providing richer and more comprehensive data support for subsequent trend feature identification.

[0032] After completing multi-scale sampling, trend feature identification is performed on both the first and second monitoring terminal time series indicator sampling sequence sets. The trend features identified from the first sampling sequence set primarily reflect long-term trends, such as slow upward or downward trends in data, or cyclical changes; whereas the trend features identified from the second sampling sequence set primarily reflect short-term trends, such as sudden fluctuations and rapid changes in data. By performing trend feature identification on both sampling sequence sets, we can obtain trend feature sets for each monitoring terminal at different time scales, namely the first and second trend feature sets.

[0033] In order to further improve the accuracy and completeness of trend features, it is necessary to perform one-to-one mapping and interactive enhancement on the first trend feature set and the second trend feature set. The specific operation is that, for each monitoring terminal, the long-term trend features in the first trend feature set are mapped one-to-one with the short-term trend features in the second trend feature set, and the two are fused through an interactive enhancement algorithm to generate a comprehensive trend feature. For example, if the long-term trend feature of a monitoring terminal shows that the data is in a slow upward trend, and the short-term trend feature shows that the data has sudden fluctuations, then the comprehensive trend feature can be described as "the data is in a slow upward trend, and there are sudden fluctuations." Through this one-to-one mapping and interactive enhancement method, the long-term trend features and the short-term trend features can be effectively fused, thereby obtaining a more comprehensive and accurate description of trend features.

[0034] Ultimately, the resulting indicator trend feature set encompasses the comprehensive trend characteristics of each monitoring terminal, more comprehensively reflecting the operating status and changing trends of the monitoring terminal. This indicator trend feature set will serve as an important basis for subsequent monitoring anomaly analysis, providing scientific and rational support for determining the deployment location of the central gateway and building an efficient monitoring terminal communication topology network, thereby effectively improving the accuracy and reliability of remote monitoring and fault warning of substation power systems.

[0035] Furthermore, step P22-3 of the embodiment of the present application also includes: P22-31: Perform mapping similarity analysis on the first trend feature set and the second trend feature set, and perform normalized matrix processing to obtain an interactive enhancement matrix set; P22-32: Perform convolution enhancement on the second trend feature set based on the interactive enhancement matrix set to obtain an indicator trend feature set.

[0036] Specifically, the process of interactively enhancing the first trend feature set and the second trend feature set can be further refined. Through mapping similarity analysis and convolution enhancement operations, the fusion effect of trend features can be further optimized to obtain a more accurate indicator trend feature set.

[0037] Specifically, after extracting the first trend feature set and the second trend feature set, a mapping similarity analysis is first performed on the two trend feature sets. Specifically, a similarity is calculated between each trend feature vector in the first trend feature set and the corresponding trend feature vector in the second trend feature set. Similarity calculation can be implemented using a variety of methods, such as cosine similarity and Euclidean distance, to quantify the degree of similarity between the two trend feature vectors. The calculated similarity values are then normalized to uniformly map the range to the [0, 1] interval to eliminate the impact of dimension and magnitude differences between the feature vectors and make the similarity values comparable. Next, the normalized similarity values are populated into an initially empty matrix to form an interaction enhancement matrix. Each row of this matrix corresponds to a trend feature vector in the first trend feature set, and each column corresponds to a trend feature vector in the second trend feature set. The matrix elements represent the normalized similarity between the two trend feature vectors. In this way, a complete interaction enhancement matrix is obtained, which contains similarity information between the trend features of all monitored terminals.

[0038] Subsequently, the second trend feature set is convolution-enhanced based on the interaction enhancement matrix set. The purpose of the convolution enhancement operation is to use the similarity information contained in the interaction enhancement matrix to perform weighted enhancement on the trend feature vectors in the second trend feature set. Specifically, for each trend feature vector in the second trend feature set, a weighted convolution operation is performed on other trend feature vectors according to their corresponding similarity values in the interaction enhancement matrix. The weighted convolution operation can be implemented by defining a convolution kernel, the weight of which is determined by the similarity value in the interaction enhancement matrix. In this way, the long-term trend feature information in the first trend feature set can be effectively integrated into the short-term trend features in the second trend feature set, thereby obtaining richer and more accurate comprehensive trend features. Finally, after the convolution enhancement operation, the obtained indicator trend feature set can more comprehensively reflect the operating status of the monitoring terminal and its changing trend, providing high-quality feature input for subsequent monitoring anomaly analysis.

[0039] Furthermore, step P24 of the embodiment of the present application further includes: P24-1: Extract the maximum value of the monitoring anomaly in the analysis results and use it as the initial center; P24-2: Construct the initial center neighborhood of the initial center according to the preset neighborhood radius; P24-3: From the two dimensions of monitoring anomaly consistency and position proximity, iterate the initial center in the initial center neighborhood to obtain the iterative center; P24-4: And so on, iterate the iterative center until the iteration stops to obtain the center recognition result.

[0040] It should be understood that the process of determining the central gateway placement location can be further refined by introducing an iterative optimization mechanism to gradually select the optimal central gateway placement location from the monitoring anomaly analysis results.

[0041] After completing the monitoring anomaly analysis, the maximum monitoring anomaly value is first extracted from the analysis results, and the corresponding monitoring terminal location is used as the initial center. The monitoring anomaly value reflects the stability and reliability of the monitoring terminal during data collection and transmission. The lower the anomaly value, the more stable the data transmission of the terminal and the less likely it is to be interfered with. Therefore, the terminal location corresponding to the maximum monitoring anomaly value is selected as the initial center based on its relative superiority in terms of data transmission and device operation status.

[0042] Next, an initial center neighborhood is constructed based on a preset neighborhood radius. The neighborhood radius can be determined based on factors such as substation layout, monitoring terminal density, and communication requirements. The goal of constructing the initial center neighborhood is to limit the search to a specific area around the initial center, enabling more efficient search for the central gateway's location. By limiting the neighborhood range, unnecessary computations can be reduced while ensuring the search process is targeted and effective.

[0043] The initial center is then iteratively optimized within its neighborhood. This iterative process is based on two dimensions: monitoring anomaly consistency and location proximity. Monitoring anomaly consistency involves searching for monitoring terminals within the neighborhood with similar monitoring anomaly levels as the initial center to ensure uniform data transmission stability around the central gateway. Location proximity considers the spatial distance between monitoring terminals, prioritizing those located close to the initial center to reduce communication link length, latency, and energy consumption. Through a comprehensive evaluation of these two dimensions, a new iteration center is selected from the neighborhood of the initial center.

[0044] The above iterative process is then repeated for the new iteration center. That is, based on the new iteration center, a new neighborhood is constructed according to the preset neighborhood radius. Within the new neighborhood, optimization is performed based on the two dimensions of monitoring anomaly consistency and location proximity to find the next more optimal iteration center. This process is repeated until the iteration stopping condition is met. The iteration stopping condition can be that the location of the iteration center no longer changes significantly, the monitoring anomaly reaches a preset optimal threshold, or the number of iterations reaches a preset maximum value. Ultimately, through this iterative optimization process, the center identification result is obtained, that is, the optimal layout location of the central gateway is determined.

[0045] Through the above steps, starting from the monitoring anomaly analysis results, a step-by-step iterative optimization method can be used to accurately determine the placement of the central gateway. This multi-dimensional iterative mechanism, based on monitoring anomaly and location proximity, not only considers data transmission stability, but also takes into account communication efficiency and system energy consumption. It can provide a scientific and reasonable layout plan for building an efficient and reliable monitoring terminal communication topology network.

[0046] Furthermore, step P24-3 of the embodiment of the present application further includes: P24-31: Traverse the initial center neighborhood and the initial center, use the two-dimensional analysis function to determine the center coefficient, and obtain the neighborhood center coefficient set and the initial center coefficient; P24-32: When there is a neighborhood center coefficient greater than or equal to the initial center coefficient in the neighborhood center coefficient set, the monitoring anomaly degree corresponding to the maximum value in the neighborhood center coefficient set is used as the iteration center.

[0047] Specifically, the process of iteratively optimizing the initial center within its neighborhood can be further refined. By introducing the concepts of a two-dimensional analysis function and a central coefficient, the initial center can be evaluated and optimized from two dimensions: consistency of monitored anomalies and proximity of locations, thereby more accurately determining the iterative center.

[0048] First, all monitoring terminals within the neighborhood of the initial center and the initial center itself are traversed. For each monitoring terminal, a two-dimensional analysis function is used to evaluate it from two dimensions: monitoring anomaly consistency and positional proximity. Specifically, monitoring anomaly consistency reflects the degree of similarity between the monitoring terminal and the initial center in terms of monitoring anomaly, while positional proximity measures the spatial distance between the monitoring terminal and the initial center. The two-dimensional analysis function combines the evaluation results of these two dimensions to calculate a centrality coefficient that is used to quantify the quality of each monitoring terminal as a potential center. In this way, a set of neighborhood centrality coefficients can be obtained, which includes the centrality coefficients of all monitoring terminals within the neighborhood of the initial center, as well as the centrality coefficient of the initial center itself.

[0049] Next, the neighborhood center coefficient set is analyzed. When there is a neighborhood center coefficient greater than or equal to the initial center coefficient in the neighborhood center coefficient set, it means that there are one or more monitoring terminals in the neighborhood of the initial center, and their comprehensive performance in the two dimensions of monitoring anomaly consistency and location proximity is better than or equal to the initial center. At this time, the monitoring terminal corresponding to the maximum value in the neighborhood center coefficient set is selected, and its monitoring anomaly is used as the new iterative center. This process is actually looking for a better center position in the neighborhood of the initial center to replace the initial center, thereby gradually optimizing the layout position of the central gateway.

[0050] This iterative optimization mechanism based on two-dimensional analysis and central coefficient not only takes into account the stability and reliability of data transmission, but also takes into account communication efficiency and system energy consumption. It can provide a more scientific, reasonable and optimized solution for the deployment of central gateways in the remote monitoring and fault warning system of substation power systems.

[0051] Furthermore, the embodiment of the present application further includes steps P24-32b: When there is no neighborhood center coefficient greater than or equal to the initial center coefficient in the neighborhood center coefficient set, the iteration is stopped and the initial center is used as the center recognition result.

[0052] In one possible embodiment of the present application, further analysis of the neighborhood center coefficient set is performed to determine whether to continue iterative optimization. If, during this analysis, no neighborhood center coefficient in the neighborhood center coefficient set is greater than or equal to the initial center coefficient, that is, no neighborhood center coefficient exceeds or equals the initial center coefficient, then iteration is terminated, and the initial center is used as the final center identification result.

[0053] This stopping condition is set to ensure the stability of the center position. In other words, if after a certain number of iterations, there is no room for further optimization of the center coefficient within the neighborhood, the initial center is considered to be the most suitable gateway deployment location and no further adjustments are required. At this point, the maximum monitoring anomaly corresponding to the initial center is considered to be the most representative of the most critical monitoring point in the substation power system and is therefore used as the final identification result.

[0054] The addition of this logical step helps prevent the system from over-optimizing during the iteration process, ensuring that the final selected center location has both a high degree of monitoring anomalies and is stable and reliable, and can effectively support subsequent monitoring and data transmission tasks.

[0055] P30: Use the monitoring terminal communication topology network to perform asynchronous sensing communication on the monitoring terminal set, and transmit the collected monitoring data set to the fault detection module for fault prototype matching. If the match is successful, generate fault warning information based on the matched fault prototype.

[0056] Specifically, by utilizing the established monitoring terminal communication topology network, all monitoring terminals can perform asynchronous sensing communication. The core of this process is to ensure that the system can accurately collect and transmit monitoring data in real time, and use this data for subsequent fault detection and early warning.

[0057] After completing the construction of the monitoring terminal communication topology network and the deployment of the central gateway, the real-time monitoring and fault warning stage begins. First, using the established monitoring terminal communication topology network, asynchronous perception communication is performed on the monitoring terminal set. Asynchronous perception communication means that after the monitoring terminal collects data, it independently transmits the data to the central gateway based on its own data generation rate and communication conditions, without waiting for other terminals or synchronization signals. This communication method can effectively avoid the delay and congestion problems caused by data synchronization, and is particularly suitable for complex scenarios in substation power systems with a large number of monitoring terminals and inconsistent data generation frequencies. Through asynchronous perception communication, each monitoring terminal can promptly transmit the collected monitoring data to the central gateway, ensuring the timeliness and integrity of the data.

[0058] After receiving data from each monitoring terminal, the central gateway aggregates and initially processes the data to form a complete monitoring data set. This data set contains real-time operating status information for all key equipment in the substation power system, such as voltage, current, temperature, humidity, and other parameters. This monitoring data set is then transmitted to the fault detection module. The fault detection module is the core component of the system. Its main function is to analyze the monitoring data and determine whether there are potential faults in the substation power system.

[0059] The fault detection module uses fault prototype matching technology to achieve rapid fault diagnosis. Fault prototypes refer to a collection of characteristic parameters of various known fault modes pre-stored in the system. These fault prototypes are derived by analyzing and modeling historical fault data and can reflect the typical manifestations of different fault types in the monitoring data. When the fault detection module receives the monitoring data set, it first preprocesses the data, such as filtering and normalization, to eliminate the effects of noise and data bias. Then, using specific matching algorithms (such as pattern recognition algorithms and machine learning algorithms), the monitoring data is matched one by one with the fault prototypes in the fault prototype library. The matching process is achieved by calculating the similarity between the monitoring data and the fault prototype. When the similarity exceeds a preset threshold, the match is considered successful, indicating that the monitoring data contains fault characteristics that match a certain fault prototype.

[0060] Once the fault detection module successfully matches a fault prototype, it enters the fault warning phase. Based on the matched fault prototype, detailed fault warning information is generated. This fault warning information should include the following key information: fault type (e.g., short circuit, insulation aging, etc.), fault location (determined by the location information of the monitoring terminal), fault time, fault severity (assessed based on the matching similarity), and recommended action (e.g., immediate power outage for maintenance, enhanced monitoring, etc.). This information is promptly communicated to operations and maintenance personnel via the system interface, text messages, emails, or other communication methods, allowing them to quickly take appropriate measures to prevent further escalation of the fault and ensure the safe and stable operation of the substation power system.

[0061] Furthermore, the embodiment of the present application further includes the following steps: Acquire a historical monitoring fault log set; aggregate the historical monitoring fault log set by the same type to obtain an aggregated historical monitoring fault log cluster; traverse the aggregated historical monitoring fault log cluster to perform same type mean processing to obtain a fault prototype set, and store the fault prototype set in the fault detection module.

[0062] Optionally, in order to further optimize the fault prototype set in the fault detection module and improve the accuracy and reliability of fault diagnosis, the present application also generates and updates the fault prototype set by processing historical monitoring fault logs.

[0063] First, obtain a collection of historical monitoring fault logs. These logs record various fault events that occurred during the substation power system's past operations, including information such as the time, location, fault type, relevant monitoring data (such as abnormal values of parameters such as voltage, current, and temperature), and the results of fault handling. This collection of historical monitoring fault logs is the fundamental data source for generating fault prototypes, and its integrity and accuracy directly impact the quality of subsequent fault prototypes. Therefore, it is crucial to ensure that the log data is collected in a standardized process, stored securely, and easily accessible.

[0064] After obtaining the collection of historical monitoring fault logs, these logs are aggregated by similarity. The purpose of similarity aggregation is to group fault logs with similar characteristics into one category so that they can be processed uniformly later. The specific operation is to classify the collection of historical monitoring fault logs according to key information such as fault type, fault location, fault characteristic parameters, etc. For example, all faults caused by insulation aging are classified into one category, and all faults caused by short circuits are classified into another category. Through similarity aggregation, multiple clusters of aggregated historical monitoring fault logs can be formed, and each cluster contains a class of fault logs with similar characteristics. The process of similarity aggregation can be implemented through data mining algorithms (such as cluster analysis), or it can be manually classified according to pre-defined fault classification rules.

[0065] Subsequently, the aggregated historical monitoring fault log clusters are traversed, and the fault logs in each cluster are subjected to similar mean processing. The purpose of similar mean processing is to extract representative fault characteristic parameters, namely fault prototypes, from each fault cluster. Specifically, for each fault cluster, the mean of the monitoring data of all fault logs in it is calculated in each parameter dimension. For example, for parameters such as voltage, current, and temperature, their average values are calculated when the fault occurs. These average values can reflect the typical characteristics of this type of fault. In this way, the fault prototype of each fault cluster can be obtained, and all fault prototypes can be aggregated to form a fault prototype set.

[0066] Finally, the generated fault prototype set is stored in the fault detection module. This set serves as the core data for the fault detection module and is used in the subsequent fault prototype matching process. Regularly updating the fault prototype set ensures that the fault detection module can identify a wider range of fault types, effectively improving the performance and reliability of the substation power system remote monitoring and fault warning system.

[0067] In summary, the embodiments of the present application have at least the following technical effects: This application traverses the substation power system equipment, deploys monitoring terminals, performs anomaly analysis, determines the central gateway deployment location, and constructs a monitoring terminal communication topology network using a star topology. This network implements asynchronous sensing communication, transmits the collected monitoring data to the fault detection module for fault prototype matching, and generates fault warning information if a match is successful.

[0068] The technical effect of real-time and comprehensive equipment status monitoring and intelligent fault warning based on the Internet of Things has been achieved, thereby improving the accuracy of fault detection and the response speed.

[0069] It should be noted that the order in which the embodiments of the present application are presented is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. Furthermore, the foregoing descriptions of specific embodiments of this specification are provided. Furthermore, the processes depicted in the accompanying drawings do not necessarily require the specific order or sequential sequence shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0070] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application should be included in the scope of protection of the present application.

[0071] This specification and drawings are merely illustrative of the present application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Obviously, those skilled in the art may make various modifications and variations to this application without departing from the scope of this application. Thus, this application is intended to include such modifications and variations as fall within the scope of this application and its equivalents.

Claims

1. A remote monitoring and fault warning method for a substation power system based on the Internet of Things, characterized in that: The method comprises: Traversing the equipment of the substation power system to deploy monitoring terminals and obtain a monitoring terminal set; Traversing the monitoring terminal set to perform monitoring anomaly analysis, determining the central gateway layout location based on the analysis results, connecting them using a star topology structure, and building a monitoring terminal communication topology network; The monitoring terminal communication topology network is used to perform asynchronous sensing communication on the monitoring terminal set, and the collected monitoring data set is transmitted to the fault detection module for fault prototype matching. If the match is successful, fault warning information is generated according to the matched fault prototype.

2. The remote monitoring and fault warning method for substation power system based on Internet of Things according to claim 1, characterized in that: Traversing the monitoring terminal set to perform monitoring anomaly analysis, and determining the central gateway deployment location based on the analysis results, including: Performing time series extraction of indicators on the monitoring terminal set according to preset monitoring abnormality indicators to obtain a monitoring terminal time series indicator sequence set; Performing trend feature identification on the monitoring terminal time series indicator sequence set to obtain an indicator trend feature set; Perform monitoring anomaly analysis on the indicator trend feature set and the monitoring terminal time series indicator sequence set to obtain analysis results; Perform center identification on the analysis results, and use the location corresponding to the center identification result as the location for the central gateway layout.

3. The remote monitoring and fault warning method for substation power system based on Internet of Things according to claim 2, characterized in that: The preset monitoring abnormality indicators include data fluctuation gradient, data transmission time error and data deviation.

4. The remote monitoring and fault warning method for substation power system based on Internet of Things according to claim 2, characterized in that: Performing trend feature identification on the monitoring terminal time series indicator sequence set to obtain an indicator trend feature set includes: Sampling the monitoring terminal timing indicator sequence set according to the first identification scale and the second identification scale respectively to obtain a first monitoring terminal timing indicator sampling sequence set and a second monitoring terminal timing indicator sampling sequence set; Traversing the first monitoring terminal time series indicator sampling sequence set and the second monitoring terminal time series indicator sampling sequence set to perform trend feature identification, and determine a first trend feature set and a second trend feature set; One-to-one mapping interaction enhancement is performed on the first trend feature set and the second trend feature set to obtain an indicator trend feature set.

5. The remote monitoring and fault warning method for substation power system based on Internet of Things according to claim 4 is characterized in that: Perform one-to-one mapping interaction enhancement on the first trend feature set and the second trend feature set to obtain an indicator trend feature set, including: Performing mapping similarity analysis on the first trend feature set and the second trend feature set, and performing normalized matrix processing to obtain an interactive enhancement matrix set; The second trend feature set is subjected to convolution enhancement based on the interaction enhancement matrix set to obtain an indicator trend feature set.

6. The remote monitoring and fault warning method for substation power system based on Internet of Things according to claim 4, characterized in that: The monitoring terminal timing indicator sequence set is sampled according to the first identification scale and the second identification scale respectively, and the method further includes: Traverse the monitoring terminal set to extract abnormal interval duration and obtain abnormal interval duration clusters; Taking the maximum abnormal interval duration in the abnormal interval duration cluster as the first identification metric; The minimum abnormal interval duration in the abnormal interval duration cluster is used as the second identification metric.

7. The remote monitoring and fault warning method for substation power system based on Internet of Things according to claim 2, characterized in that: Perform center identification on the analysis results and use the location corresponding to the center identification result as the location for the central gateway deployment, including: Extract the maximum value of monitoring abnormality in the analysis results and use it as the initial center; Constructing an initial center neighborhood of the initial center according to a preset neighborhood radius; From the two dimensions of monitoring anomaly consistency and position proximity, the initial center is iterated in the neighborhood of the initial center to obtain an iterative center; Similarly, the iteration center is iterated until the iteration stops, and a center recognition result is obtained.

8. The remote monitoring and fault warning method for substation power system based on Internet of Things according to claim 7, characterized in that: From the two dimensions of monitoring anomaly consistency and position proximity, the initial center is iterated in the neighborhood of the initial center to obtain an iterative center, including: Traversing the initial center neighborhood and the initial center, determining the center coefficient using a two-dimensional analysis function, and obtaining a neighborhood center coefficient set and an initial center coefficient; When there is a neighborhood center coefficient greater than or equal to the initial center coefficient in the neighborhood center coefficient set, the monitoring abnormality corresponding to the maximum value in the neighborhood center coefficient set is used as the iteration center.

9. The remote monitoring and fault warning method for substation power system based on Internet of Things according to claim 8, characterized in that: When there is no neighborhood center coefficient greater than or equal to the initial center coefficient in the neighborhood center coefficient set, the iteration is stopped and the initial center is used as the center recognition result.

10. The remote monitoring and fault warning method for substation power system based on Internet of Things according to claim 1, characterized in that: include: Get the historical monitoring fault log collection; Aggregating the historical monitoring fault log set by the same type to obtain an aggregated historical monitoring fault log cluster; The aggregated historical monitoring fault log cluster is traversed to perform similar mean processing to obtain a fault prototype set, and the fault prototype set is stored in the fault detection module.

Images