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

Abstract

The application relates to the technical field of equipment monitoring, and discloses a power distribution equipment remote monitoring method and system based on the Internet of Things. The method comprises the following steps: comparing multi-source heterogeneous monitoring data with normal operation parameters to obtain a unified state image; performing characteristic analysis on historical fault types to obtain characteristic parameters, and selecting a fault state characteristic sequence which is strongly related to the characteristic parameters from the unified state image; independently analyzing the unified state image, mapping the obtained analysis result to a preset multi-dimensional evaluation space, constructing a state evaluation matrix and mapping the state evaluation matrix to the fault state characteristic sequence to obtain a matching degree, and generating a comprehensive evaluation report according to the matching degree and an operation state development trend; analyzing the influence intensity of multi-dimensional characteristic parameters on the operation state to generate a regulation and control instruction of the power distribution equipment; and delivering the regulation and control instruction to the power distribution equipment to complete closed-loop verification. The application can improve the efficiency of the power distribution equipment remote monitoring based on the Internet of Things.

Description

Technical Field

[0001] This invention relates to the field of equipment monitoring technology, and in particular to a method and system for remote monitoring of power distribution equipment based on the Internet of Things. Background Technology

[0002] In remote monitoring scenarios of power distribution equipment, multi-source heterogeneous monitoring data have problems such as format differences, abnormal interference, and inconsistencies in spatiotemporal dimensions. Existing technologies lack efficient preprocessing solutions, making it difficult to accurately remove invalid and abnormal data and achieve standardized data alignment. This leads to deviations in the representation of equipment operating status, which cannot provide a reliable data foundation for subsequent analysis and directly affects the accuracy of monitoring results.

[0003] Existing monitoring methods mostly adopt a centralized data processing model, which is difficult to adapt to the needs of efficient analysis of massive monitoring data. Furthermore, the depth of historical fault feature mining is insufficient, and it is impossible to establish an accurate correlation between the current operating status and historical fault cases. This results in delayed fault prediction and response, and an incomplete comprehensive assessment. At the same time, the generation of control commands does not fully consider the influence intensity of multi-dimensional feature parameters, lacks closed-loop verification and optimization mechanisms, and the control measures are not targeted and effective enough to meet the monitoring requirements for stable and reliable operation of power distribution equipment. Therefore, how to improve the efficiency of remote monitoring of power distribution equipment based on the Internet of Things has become an urgent problem to be solved. Summary of the Invention

[0004] This invention provides a method and system for remote monitoring of power distribution equipment based on the Internet of Things (IoT) to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides a method for remote monitoring of power distribution equipment based on the Internet of Things, comprising:

[0006] S1. Compare the multi-source heterogeneous monitoring data of the power distribution equipment with the normal operating parameters of the power distribution equipment, remove abnormal data in the multi-source heterogeneous monitoring data, and perform spatiotemporal alignment on the multi-source heterogeneous monitoring data to obtain a unified state image of the power distribution equipment.

[0007] S2. Perform feature analysis on the historical fault types of the power distribution equipment to obtain the feature parameters of the historical fault types, and select fault state feature sequences that are strongly correlated with the feature parameters from the unified state image;

[0008] S3. Distribute the unified state image to the analysis nodes in the Internet of Things, and perform independent analysis on the unified state image to obtain the analysis results of the analysis nodes. Map the analysis results to a preset multi-dimensional evaluation space to construct the state evaluation matrix of the power distribution equipment.

[0009] S4. Map the state evaluation matrix to the fault state feature sequence to obtain the matching degree between the current operating state and historical fault state cases in the power distribution equipment, and generate a comprehensive evaluation report of the power distribution equipment based on the matching degree and the operating state development trend of the power distribution equipment.

[0010] S5. Extract the multi-dimensional feature parameters from the comprehensive evaluation report, analyze the influence intensity of the multi-dimensional feature parameters on the operating status of the power distribution equipment, and generate control instructions for the power distribution equipment based on the influence intensity.

[0011] S6. The control command is sent to the power distribution equipment through the network layer of the Internet of Things, the execution effect of the control command is monitored in real time, and closed-loop verification is completed based on the feedback data of the power distribution equipment.

[0012] In a preferred embodiment, the step of comparing the multi-source heterogeneous monitoring data of the power distribution equipment with the normal operating parameters of the power distribution equipment, removing abnormal data from the multi-source heterogeneous monitoring data, and performing spatiotemporal alignment of the multi-source heterogeneous monitoring data to obtain a unified state image of the power distribution equipment includes:

[0013] The multi-source heterogeneous monitoring data is processed to unify the format, resulting in a preliminary standardized dataset of the power distribution equipment.

[0014] Within a sliding time window, the parameter fluctuation thresholds during normal operation of the power distribution equipment are analyzed, the preliminary standardized dataset is mapped to the parameter fluctuation thresholds, and abnormal data sequences exceeding the parameter fluctuation thresholds are identified.

[0015] The contextual consistency of the abnormal data sequence is verified, and the abnormal data sequence is corrected according to the normal operating parameters of the power distribution equipment to obtain a clean data sequence of the power distribution equipment.

[0016] The timestamps of the clean data sequence are synchronized, the missing data points of the clean data sequence are filled by time interpolation, and the spatial coordinates of the clean data sequence are normalized and mapped to obtain a unified state image of the power distribution equipment.

[0017] In a preferred embodiment, the step of performing feature analysis on the historical fault types of the power distribution equipment to obtain feature parameters of the historical fault types, and selecting fault state feature sequences strongly correlated with the feature parameters from the unified state image, includes:

[0018] Extract multi-source data fragments of the historical fault types and standardize the multi-source data fragments to obtain the standard fault dataset of the power distribution equipment;

[0019] The standard fault dataset is divided into time windows, and the multidimensional features of the standard fault dataset are quantified to obtain the multidimensional feature vector of the standard fault dataset.

[0020] Cluster analysis is performed on the multidimensional feature vectors to obtain the distinguishability of the multidimensional feature vectors, and feature parameters of the multidimensional feature vectors are selected based on the distinguishability.

[0021] In a preferred embodiment, the step of performing feature analysis on the historical fault types of the power distribution equipment to obtain feature parameters of the historical fault types, and selecting fault state feature sequences strongly correlated with the feature parameters from the unified state image, includes:

[0022] The feature parameters are mapped to the data dimension of the unified state image to obtain the potential associated data of the power distribution equipment.

[0023] The potentially related data is segmented to obtain data segments of the potentially related data;

[0024] By integrating the similarity between the data segments and the feature parameters, a similarity sequence of the power distribution equipment is obtained.

[0025] Peak detection is performed on the similarity sequence to mark the high similarity data interval of the power distribution equipment;

[0026] By fusing the high-similarity data intervals and verifying the temporal continuity of the high-similarity data intervals, the fault state feature sequence of the power distribution equipment is obtained.

[0027] In a preferred embodiment, distributing the unified state image to analysis nodes in the Internet of Things (IoT), independently analyzing the unified state image to obtain the analysis results of the analysis nodes, and mapping the analysis results to a preset multi-dimensional evaluation space to construct the state evaluation matrix of the power distribution equipment includes:

[0028] Based on the computing power and current load of the analysis node in the Internet of Things, the data segments of the unified state image are transmitted in parallel to the analysis node;

[0029] In the analysis node, multi-dimensional feature extraction is performed on the data segment to obtain the state feature set of the unified state image;

[0030] The multi-source feature values ​​in the state feature set are weighted and fused to obtain the local evaluation value of the analysis node;

[0031] The local evaluation value is quantified to obtain the local evaluation vector of the analysis node, and the local evaluation vector is spatially aggregated to obtain the state evaluation matrix of the power distribution equipment.

[0032] In a preferred embodiment, quantifying the local evaluation value to obtain the local evaluation vector of the analysis node, and spatially aggregating the local evaluation vector to obtain the state evaluation matrix of the power distribution equipment, includes:

[0033] Obtain the local evaluation vector returned by the analysis node, and calculate the comprehensive status index of the power distribution equipment based on the preset baseline state vector. The calculation formula for the comprehensive status index is as follows:

[0034] ;

[0035] in, This indicates the comprehensive status index. Indicates the first The local evaluation vectors mentioned above. This represents the preset baseline state vector. Indicates the first The magnitude of each of the local evaluation vectors, This represents the magnitude of the reference state vector. This indicates the number of analysis nodes;

[0036] Based on the comprehensive status index, a status assessment matrix for the power distribution equipment is constructed.

[0037] In a preferred embodiment, mapping the state evaluation matrix to the fault state feature sequence to obtain the matching degree between the current operating state and historical fault state cases of the power distribution equipment, and generating a comprehensive evaluation report of the power distribution equipment based on the matching degree and the operating state development trend of the power distribution equipment, includes:

[0038] The evaluation dimensions of the state evaluation matrix are compared with the corresponding features of the fault state feature sequence in a multi-dimensional pattern comparison to obtain the similar areas of the current operating state of the power distribution equipment and the historical fault state cases in terms of feature morphology and change trend.

[0039] By analyzing the evolution of the state assessment matrix over time, the change direction and intensity characteristics of the assessment parameters in the state assessment matrix are extracted to obtain a description of the operating state development trend of the power distribution equipment.

[0040] The identification results of the similar regions are fused with the description of the operating status development trend to obtain a comprehensive evaluation report of the power distribution equipment.

[0041] In a preferred embodiment, the step of extracting multidimensional feature parameters from the comprehensive evaluation report, analyzing the influence intensity of the multidimensional feature parameters on the operating status of the power distribution equipment, and generating control instructions for the power distribution equipment based on the influence intensity includes:

[0042] The evaluation indicators that change drastically or have a significant impact in the comprehensive evaluation report are selected as the key status indicators of the comprehensive evaluation report, and trend analysis is performed on the key status indicators to obtain the changing trend characteristics of the comprehensive evaluation report.

[0043] By integrating the key status indicators and the changing trend characteristics, the multidimensional feature parameters of the comprehensive evaluation report are obtained.

[0044] Analyze the correlation between the multidimensional feature parameters and the historical operating status changes of the power distribution equipment, and evaluate the degree of influence and response priority of the multidimensional feature parameters on the operating status of the power distribution equipment based on the correlation;

[0045] Based on the degree of impact and the response priority, a mapping relationship for the control strategy of the power distribution equipment is established, and control instructions adapted to the operating state of the power distribution equipment are generated.

[0046] In a preferred embodiment, the step of sending the control command to the power distribution equipment through the network layer of the Internet of Things, monitoring the execution effect of the control command in real time, and completing closed-loop verification based on the feedback data from the power distribution equipment includes:

[0047] The control command and the identification information of the power distribution equipment are protocol-encapsulated to obtain a standard command data packet of the power distribution equipment, and the standard command data packet is distributed and transmitted through the secure channel of the network layer.

[0048] After the standard instruction data packet is sent, the status confirmation information and real-time operation data stream returned by the power distribution equipment are received to obtain the instruction response dataset of the power distribution equipment;

[0049] The command response dataset is parsed to obtain the key state parameters of the command response dataset, and the key state parameters are compared with the expected effect target of the control command in multiple dimensions to obtain the comparative analysis results of the power distribution equipment.

[0050] The degree of achievement of the control command is determined based on the comparative analysis results, and the control strategy of the power distribution equipment is updated based on the degree of achievement, thus completing the closed-loop verification.

[0051] To address the above problems, the present invention also provides a remote monitoring system for power distribution equipment based on the Internet of Things, the system comprising:

[0052] The data processing module is used to compare the multi-source heterogeneous monitoring data of the power distribution equipment with the normal operating parameters of the power distribution equipment, remove abnormal data in the multi-source heterogeneous monitoring data, and perform spatiotemporal alignment on the multi-source heterogeneous monitoring data to obtain a unified state image of the power distribution equipment.

[0053] The fault state analysis module is used to perform feature analysis on the historical fault types of the power distribution equipment, obtain the feature parameters of the historical fault types, and select fault state feature sequences that are strongly correlated with the feature parameters from the unified state image.

[0054] The current state analysis module is used to distribute the unified state image to the analysis nodes in the Internet of Things, and to perform independent analysis on the unified state image to obtain the analysis results of the analysis nodes. The analysis results are then mapped to a preset multi-dimensional evaluation space to construct the state evaluation matrix of the power distribution equipment.

[0055] The operation status assessment module is used to map the status assessment matrix to the fault status feature sequence, obtain the matching degree between the current operation status and historical fault status cases of the power distribution equipment, and generate a comprehensive assessment report of the power distribution equipment based on the matching degree and the development trend of the operation status of the power distribution equipment.

[0056] The control instruction generation module is used to extract multi-dimensional feature parameters from the comprehensive evaluation report, analyze the influence intensity of the multi-dimensional feature parameters on the operating status of the power distribution equipment, and generate control instructions for the power distribution equipment based on the influence intensity.

[0057] The closed-loop verification module is used to send the control command to the power distribution equipment through the network layer of the Internet of Things, monitor the execution effect of the control command in real time, and complete the closed-loop verification based on the feedback data of the power distribution equipment.

[0058] Compared with the prior art, the present invention has the following beneficial effects:

[0059] This invention forms a unified state image by removing anomalies and aligning spatiotemporally multi-source heterogeneous monitoring data. At the same time, relying on the parallel processing capabilities of distributed analysis nodes in the Internet of Things, it efficiently extracts multi-dimensional state features and constructs an accurate state assessment matrix, which significantly improves the efficiency of power distribution equipment operation status monitoring and the accuracy of data processing, making equipment status feedback more timely and assessment more comprehensive.

[0060] This invention deeply mines the correlation between historical fault characteristics and current operating status, accurately predicts equipment operating trends and generates targeted comprehensive assessment reports. It combines the influence intensity analysis of multi-dimensional characteristic parameters to formulate appropriate control instructions, and then optimizes the control strategy in real time through a closed-loop verification mechanism. This effectively enhances the ability to predict equipment operating risks and the effectiveness of control measures, providing strong support for the stable and reliable operation of power distribution equipment. Attached Figure Description

[0061] Figure 1 This is a flowchart illustrating a method for remote monitoring of power distribution equipment based on the Internet of Things, according to an embodiment of the present invention.

[0062] Figure 2 A functional block diagram of a remote monitoring system for power distribution equipment based on the Internet of Things (IoT) is provided in an embodiment of the present invention.

[0063] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0064] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0065] This application provides a method for remote monitoring of power distribution equipment based on the Internet of Things (IoT). The executing entity of this IoT-based remote monitoring method includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the method provided in this application: a server, a terminal, etc. In other words, the IoT-based remote monitoring method for power distribution equipment can be executed by software or hardware installed on a terminal device or a server device. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster. The server can be an independent server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0066] Reference Figure 1 The diagram shown is a flowchart illustrating a remote monitoring method for power distribution equipment based on the Internet of Things (IoT) according to an embodiment of the present invention. In this embodiment, the remote monitoring method for power distribution equipment based on the IoT includes:

[0067] S1. Compare the multi-source heterogeneous monitoring data of the power distribution equipment with the normal operating parameters of the power distribution equipment, remove abnormal data in the multi-source heterogeneous monitoring data, and perform spatiotemporal alignment on the multi-source heterogeneous monitoring data to obtain a unified state image of the power distribution equipment.

[0068] In this embodiment of the invention, the step of comparing the multi-source heterogeneous monitoring data of the power distribution equipment with the normal operating parameters of the power distribution equipment, removing abnormal data from the multi-source heterogeneous monitoring data, and performing spatiotemporal alignment of the multi-source heterogeneous monitoring data to obtain a unified state image of the power distribution equipment includes:

[0069] The multi-source heterogeneous monitoring data is processed to unify the format, resulting in a preliminary standardized dataset of the power distribution equipment.

[0070] Within a sliding time window, the parameter fluctuation thresholds during normal operation of the power distribution equipment are analyzed, the preliminary standardized dataset is mapped to the parameter fluctuation thresholds, and abnormal data sequences exceeding the parameter fluctuation thresholds are identified.

[0071] The contextual consistency of the abnormal data sequence is verified, and the abnormal data sequence is corrected according to the normal operating parameters of the power distribution equipment to obtain a clean data sequence of the power distribution equipment.

[0072] The timestamps of the clean data sequence are synchronized, the missing data points of the clean data sequence are filled by time interpolation, and the spatial coordinates of the clean data sequence are normalized and mapped to obtain a unified state image of the power distribution equipment.

[0073] The multi-source heterogeneous monitoring data of power distribution equipment is processed to unify the format. This data may come from different types of monitoring sensors, covering various parameters such as voltage, current, temperature, and humidity, and the original formats are different. Some data are in hexadecimal, some are in decimal, and the units are not uniform. Some voltage data are in kilovolts, some are in volts, and some current data are in kiloamperes, some are in amperes. All hexadecimal data are converted to decimal numbers, and parameters with different units are converted to consistent units according to industry standards. Voltage is uniformly converted to volts, current is uniformly converted to amperes, temperature is uniformly converted to degrees Celsius, and humidity is uniformly converted to percentage. This ensures that all monitoring data are completely consistent in format and units, and finally, a preliminary standardized dataset of power distribution equipment is obtained.

[0074] A fixed-duration sliding time window is set, with the window duration set to 1 minute based on the operating characteristics of the power distribution equipment. The historical monitoring data of the power distribution equipment during normal operation is traversed segment by segment using this window as the unit. For each operating parameter within each sliding window, the maximum and minimum values ​​are calculated to determine the reasonable fluctuation range of the parameter under normal operating conditions. This range is the parameter fluctuation threshold. Each data point in the preliminary standardized dataset is compared with the fluctuation threshold of the corresponding parameter to determine whether the data is within the threshold range. If a data point exceeds the fluctuation threshold of its corresponding parameter, the data is marked as abnormal data. All marked abnormal data are arranged in the order of their generation to form the abnormal data sequence of the power distribution equipment.

[0075] Examine each normal data point adjacent to the abnormal data point in the abnormal data sequence one by one, analyze the numerical change trend and fluctuation pattern of the normal data before and after, and determine whether the abnormal data is consistent with the change logic of the normal data before and after. If the change trend of the abnormal data is consistent with the change trend of the normal data before and after, and the only reason for the abnormal data is that the instantaneous fluctuation exceeds the threshold, it is judged as a false abnormal data. If the change pattern of the abnormal data contradicts that of the normal data before and after, and deviates significantly from the normal range, it is confirmed as real abnormal data. For real abnormal data, refer to the standard range of normal operating parameters of the power distribution equipment and the values ​​of the adjacent normal data before and after, and adjust the abnormal data according to the principle of gradual change of values ​​to correct it to a reasonable normal range. After context consistency verification and abnormal data correction, a clean data sequence with no abnormalities and accurate data is obtained.

[0076] The process involves collecting the original timestamps of all data in the clean data sequence, converting timestamps from different time zones and formats to the UTC standard time format to ensure that all data records remain synchronized. For any missing data points in the clean data sequence, a reasonable value is calculated based on the time interval and numerical changes of two adjacent valid data points before and after the missing point, and this value is then used to fill in the missing position to ensure the continuity of the data sequence. For spatial location-related data in the clean data sequence, such as the installation location information of different monitoring sensors, a unified spatial coordinate origin and three-dimensional coordinate axes are set, and the original location data of all sensors are converted into coordinate values ​​based on this unified coordinate system to achieve spatial coordinate normalization mapping. After timestamp synchronization, data missing point filling, and spatial coordinate normalization mapping, a unified state image that comprehensively and coherently reflects the operating status of the power distribution equipment is obtained.

[0077] The beneficial effects are that by unifying the format of multi-source heterogeneous monitoring data and identifying and correcting abnormal data, invalid and erroneous data are effectively eliminated, ensuring the accuracy and reliability of the data. After timestamp synchronization, missing point filling and spatial coordinate normalization mapping, spatiotemporal alignment of multi-source data is achieved. The resulting unified state image can comprehensively, accurately and coherently present the operating status of power distribution equipment, providing high-quality and standardized data support for subsequent historical fault feature analysis, state assessment matrix construction and other links, and greatly improving the data processing efficiency of remote monitoring of power distribution equipment and the accuracy of subsequent analysis results.

[0078] S2. Perform feature analysis on the historical fault types of the power distribution equipment to obtain the feature parameters of the historical fault types, and select fault state feature sequences that are strongly correlated with the feature parameters from the unified state image;

[0079] In this embodiment of the invention, the step of performing feature analysis on the historical fault types of the power distribution equipment to obtain feature parameters of the historical fault types, and selecting fault state feature sequences strongly correlated with the feature parameters from the unified state image, includes:

[0080] Extract multi-source data fragments of the historical fault types and standardize the multi-source data fragments to obtain the standard fault dataset of the power distribution equipment;

[0081] The standard fault dataset is divided into time windows, and the multidimensional features of the standard fault dataset are quantified to obtain the multidimensional feature vector of the standard fault dataset.

[0082] Cluster analysis is performed on the multidimensional feature vectors to obtain the distinguishability of the multidimensional feature vectors, and feature parameters of the multidimensional feature vectors are selected based on the distinguishability.

[0083] The step involves performing feature analysis on the historical fault types of the power distribution equipment to obtain feature parameters for the historical fault types, and selecting fault state feature sequences strongly correlated with the feature parameters from the unified state image, including:

[0084] The feature parameters are mapped to the data dimension of the unified state image to obtain the potential associated data of the power distribution equipment.

[0085] The potentially related data is segmented to obtain data segments of the potentially related data;

[0086] By integrating the similarity between the data segments and the feature parameters, a similarity sequence of the power distribution equipment is obtained.

[0087] Peak detection is performed on the similarity sequence to mark the high similarity data interval of the power distribution equipment;

[0088] By fusing the high-similarity data intervals and verifying the temporal continuity of the high-similarity data intervals, the fault state feature sequence of the power distribution equipment is obtained.

[0089] Multi-source data segments of historical fault types were extracted from the power distribution equipment. These fault types included short-circuit faults, overload faults, insulation aging faults, and poor contact faults. For each fault type, all relevant monitoring data from one hour before the fault occurred to 30 minutes after the fault was resolved were retrieved from the equipment operation database. This data included multi-dimensional data such as voltage, current, temperature, equipment vibration frequency, insulation resistance, and number of switch actions, forming independent multi-source data segments corresponding to each fault type. The extracted multi-source data segments were standardized. First, the data format was unified by converting all data into decimal values. If there was text-based status description data, it was converted into the corresponding coded values. Then, the parameter units were unified: voltage data was uniformly converted to volts, current data to amperes, temperature data to degrees Celsius, vibration frequency to hertz, and insulation resistance to megohms. Finally, the numerical range of all data was mapped to the 0-1 interval. This was achieved by calculating (actual data value - historical minimum value of the parameter) / (historical maximum value of the parameter - historical minimum value of the parameter), ensuring the comparability of fault data of different types and units, and finally obtaining a standard fault dataset for the power distribution equipment.

[0090] The standard fault dataset is divided into time windows. Based on the typical cycle of power distribution equipment fault development, the time window length is set to 30 seconds. Using this time window as a fixed interval, data of corresponding durations are sequentially extracted starting from the start time of the standard fault dataset. Each time window contains complete data for all monitored parameters within that 30 seconds. The entire standard fault dataset is divided into multiple continuous and non-overlapping time window data blocks. Multidimensional feature quantization is performed on the standard fault data in each time window data block. For voltage parameters, the average, maximum, and minimum voltage values ​​within the window, as well as the voltage variation amplitude between adjacent data points, are calculated. For current parameters, the average, maximum, and minimum voltage values, and the current variation amplitude are calculated. For temperature parameters, the average, maximum, and minimum temperature values, as well as the temperature rise rate per unit time, are calculated. For the vibration frequency parameter, calculate the average, maximum, and minimum values, as well as the number of frequency fluctuations. For the insulation resistance parameter, calculate the average, minimum values, and the resistance decrease rate per unit time. Arrange the quantitative characteristic values ​​of voltage, current, temperature, vibration frequency, and insulation resistance calculated within each time window in a fixed order: "average voltage - maximum voltage - minimum voltage - voltage change amplitude - average current - maximum current - minimum current - current change amplitude - average temperature - maximum temperature - minimum temperature - temperature rise rate - average vibration frequency - maximum vibration frequency - minimum vibration frequency - number of vibration frequency fluctuations - average insulation resistance - minimum insulation resistance - insulation resistance decrease rate". This forms a multidimensional feature vector of the standard fault dataset corresponding to that time window.

[0091] Cluster analysis is performed on the multidimensional feature vectors corresponding to all time windows. Each multidimensional feature vector is treated as a data point. The differences in the corresponding feature values ​​between different data points are compared. If the voltage, current, temperature, and other feature values ​​of two data points are relatively close, it indicates that their corresponding fault states are similar, and they are classified into the same category. If the feature values ​​differ significantly, they are classified into different categories. For example, multidimensional feature vectors with obvious characteristics of sudden current increase and voltage drop are classified into the short-circuit fault category, and those with obvious characteristics of persistently high current and slow temperature rise are classified into the overload fault category. The average value of the multidimensional feature vectors between different categories is calculated. The degree of difference is determined by randomly selecting 100 multidimensional feature vectors from each of the two different categories, calculating the absolute difference between the feature values ​​of each pair of vectors and summing them, and then averaging the sum of the differences of all 100 pairs of vectors. This average value is the degree of distinction between the two categories. The larger the degree of distinction value, the more significant the feature difference between the two types of faults. The features corresponding to the feature values ​​with the highest degree of distinction between different categories are selected. For example, compared with overload faults, the current change amplitude and the minimum voltage value have the highest degree of distinction. These features are the feature parameters that can effectively distinguish different types of historical faults.

[0092] The unified state image is defined to include all data dimensions, covering the dimensions corresponding to various parameters such as real-time voltage, current, temperature, vibration frequency, insulation resistance, and humidity monitoring values ​​of the power distribution equipment. The selected feature parameters are matched with the data dimensions of the unified state image. If the feature parameters are "current change amplitude," "minimum voltage value," or "temperature rise rate," the dimensions corresponding to these parameters are found in the unified state image. This is because the current change amplitude can be calculated from continuous real-time current monitoring values, the minimum voltage value can be extracted from real-time voltage monitoring values, and the temperature rise rate can be calculated from real-time temperature monitoring values. All real-time monitoring data under these matching dimensions are extracted; this data represents the potential correlation data that is potentially related to the feature parameters.

[0093] Based on the frequency of monitoring data recording and the daily operating status changes of the power distribution equipment, the data segment duration is set to 5 minutes. This duration ensures that each data segment contains enough data to reflect the equipment status, while avoiding excessively long segments that could lead to blurred features. Starting from the starting time of the potentially related data, data is extracted sequentially at fixed durations of 5 minutes. The first data segment contains the potentially related data from the starting time to 5 minutes after the starting time, the second data segment contains the potentially related data from 5 minutes to 10 minutes after the starting time, and so on, thus completely dividing the entire potentially related data into multiple consecutive data segments of equal duration.

[0094] Each data segment is processed one by one. For each data item in the data segment, it is compared with the corresponding feature parameter. For example, if the average current change amplitude in a data segment over a 5-minute period is 20 amperes / second, and the standard value of the current change amplitude in the corresponding feature parameter is 18 amperes / second, the similarity between the two is calculated. The similarity is obtained by (1 - |data segment value - feature parameter value| / feature parameter value). Here, the calculation result is (1 - |20 - 18| / 18) ≈ 0.89, which is a high similarity. If the difference between the data segment value and the feature parameter value is large, the calculation result will be close to 0. A comprehensive similarity score is calculated for each data segment, which is the arithmetic mean of the similarity between all data in the data segment and the corresponding feature parameter. The comprehensive similarity scores of each data segment are arranged in the order of time of the data segments to form a similarity sequence of the power distribution equipment.

[0095] A high similarity threshold of 0.8 was set. This threshold was determined by analyzing the similarity between a large amount of historical fault data and feature parameters, which can effectively distinguish between high and low similarity data. Each score in the similarity sequence was iterated. If the score was ≥0.8, it was marked as a peak point. The scores before and after each peak point were examined. If the adjacent scores were between 0.75 and 0.8, it meant that the similarity of these time periods was close to that of the peak point. The time period from the first score that continuously met the condition to the last score that continuously met the condition was determined as the high similarity data interval. The start time, end time, and score range of the interval were marked in the sequence.

[0096] Examine all high-similarity data intervals. If the time of two intervals overlaps, or the time interval between the end time and the start time is less than 1 minute, it indicates that they reflect continuous high-similarity states, and they are merged into a complete interval. Verify the time continuity of the merged interval and check whether there is any missing data in the interval. If there is corresponding data at all time points and the data recording interval is consistent with the monitoring frequency, it is determined that the time is continuous. If there is a breakpoint, the interval is removed. Integrate the potential related data in all time-continuous high-similarity data intervals in chronological order to form an operating status data sequence that can completely reflect the association between the equipment and historical fault characteristics, that is, the fault status characteristic sequence of the power distribution equipment.

[0097] The beneficial effects are as follows: By standardizing the multi-source data fragments of historical faults, interference caused by differences in data format and units is eliminated, laying a unified foundation for subsequent analysis; by dividing the time window and quantifying multi-dimensional features, fault data is transformed into analyzable vectors, and cluster analysis is used to select high-discrimination feature parameters, ensuring that the feature parameters can accurately represent the core features of the fault; and through feature parameter mapping, data segmentation, similarity calculation, peak detection, and time continuity verification, the final fault state feature sequence can accurately associate historical faults with the current equipment state, providing accurate and reliable data support for subsequent equipment state assessment and fault prediction, effectively improving the accuracy and efficiency of fault feature identification in remote monitoring of power distribution equipment.

[0098] S3. Distribute the unified state image to the analysis nodes in the Internet of Things, and perform independent analysis on the unified state image to obtain the analysis results of the analysis nodes. Map the analysis results to a preset multi-dimensional evaluation space to construct the state evaluation matrix of the power distribution equipment.

[0099] In this embodiment of the invention, the step of distributing the unified state image to analysis nodes in the Internet of Things, independently analyzing the unified state image to obtain the analysis results of the analysis nodes, and mapping the analysis results to a preset multi-dimensional evaluation space to construct the state evaluation matrix of the power distribution equipment includes:

[0100] Based on the computing power and current load of the analysis node in the Internet of Things, the data segments of the unified state image are transmitted in parallel to the analysis node;

[0101] In the analysis node, multi-dimensional feature extraction is performed on the data segment to obtain the state feature set of the unified state image;

[0102] The multi-source feature values ​​in the state feature set are weighted and fused to obtain the local evaluation value of the analysis node;

[0103] The local evaluation value is quantified to obtain the local evaluation vector of the analysis node, and the local evaluation vector is spatially aggregated to obtain the state evaluation matrix of the power distribution equipment.

[0104] The process of quantifying the local evaluation value to obtain the local evaluation vector of the analysis node, and spatially aggregating the local evaluation vector to obtain the state evaluation matrix of the power distribution equipment includes:

[0105] Obtain the local evaluation vector returned by the analysis node, and calculate the comprehensive status index of the power distribution equipment based on the preset baseline state vector. The calculation formula for the comprehensive status index is as follows:

[0106] ;

[0107] in, This indicates the comprehensive status index. Indicates the first The local evaluation vectors mentioned above. This represents the preset baseline state vector. Indicates the first The magnitude of each of the local evaluation vectors, This represents the magnitude of the reference state vector. This indicates the number of analysis nodes;

[0108] Based on the comprehensive status index, a status assessment matrix for the power distribution equipment is constructed.

[0109] First, the computing power and current load of each analysis node in the IoT are assessed. Computing power is determined by the number of CPU cores and memory capacity of the node. For example, a node with 4 CPU cores and 8GB of memory is defined as having medium computing power, while a node with 2 CPU cores and 4GB of memory is defined as having low computing power. The current load is determined by the node's real-time CPU utilization and memory usage. A CPU utilization of less than 50% and a memory usage of less than 40% indicates a low load, while a CPU utilization of more than 70% and a memory usage of more than 60% indicates a high load. The unified state image is divided into several data segments of equal duration according to the time sequence of data generation. Each data segment contains 10 minutes of device operation data. Data segments are allocated according to the node assessment results. Nodes with medium computing power and low load are allocated 2 data segments, while nodes with low computing power or high load are allocated 1 data segment, ensuring that the computing resources of each node are used reasonably and are not overloaded. The allocated data segments are simultaneously transmitted to the corresponding analysis nodes through the IoT communication link to achieve parallel data distribution.

[0110] In each analysis node, multi-dimensional feature extraction is performed on the received data segment. First, the core operating parameters contained in the data segment are identified, including voltage, current, temperature, and vibration frequency. For voltage parameters, the average, maximum, and minimum values ​​of all voltage values ​​in the data segment, as well as the maximum absolute value of the difference between two adjacent voltage values, are calculated. For current parameters, the average, maximum, and minimum values, as well as the current fluctuation amplitude, are calculated in the same way. For temperature parameters, the average, maximum, and minimum values, as well as the maximum value of the ratio of the difference between two adjacent temperature values ​​to the time interval, are calculated. For vibration frequency parameters, the average, maximum, and minimum values, as well as the number of times the frequency value exceeds the normal range (48-52Hz), are calculated. All these extracted voltage, current, temperature, and vibration frequency-related feature values ​​are collected and organized to form the feature set corresponding to the data segment. After summing the feature sets corresponding to the data segments processed by all analysis nodes, the state feature set of the unified state image is obtained.

[0111] First, weights are assigned based on the degree of influence of each feature value on the operating status of the power distribution equipment. Among them, voltage fluctuation amplitude and current fluctuation amplitude have the greatest impact on the safe operation of the equipment, and each is assigned a weight of 0.25; temperature change rate has a relatively large impact, and is assigned a weight of 0.2; vibration frequency fluctuation frequency has a moderate impact, and is assigned a weight of 0.15; the average voltage, average current, average temperature, and average vibration frequency have relatively small impacts, and are each assigned a weight of 0.0375. Each feature value in the state feature set is converted into a score of 0-100, based on the value range of the feature when the equipment is operating normally. For example, the normal range of voltage fluctuation amplitude is 50-150V. When a feature value is 100V, the score is calculated as (100-50) / (150-50)×100=50 points. The score of each feature value is multiplied by its corresponding weight to obtain the weighted score of that feature. For example, if the voltage fluctuation amplitude score is 50 points, the weighted score is 50×0.25=12.5 points. The weighted scores of all features are added together, and the sum is the local evaluation value of the analysis node.

[0112] First, determine the dimensions of the local evaluation vector. These dimensions correspond to the core operating index categories of the power distribution equipment, including four dimensions: voltage stability, current stability, temperature stability, and vibration stability. Decompose the local evaluation values ​​into these four dimensions: the voltage stability score is obtained by adding the weighted score of voltage fluctuation amplitude to the weighted score of the average voltage; the current stability score is obtained by adding the weighted score of current fluctuation amplitude to the weighted score of the average current; the temperature stability score is obtained by adding the weighted score of temperature change rate to the weighted score of the average temperature; and the vibration stability score is obtained by adding the weighted score of vibration frequency fluctuation number to the weighted score of vibration frequency. Arrange the scores of these four dimensions in a fixed order of "voltage stability – current stability – temperature stability – vibration stability," and the resulting numerical sequence is the local evaluation vector for that analysis node. Collect the local evaluation vectors returned by all analysis nodes. For each dimension, calculate the average score of that dimension across all vectors, and record the deviation of each dimension's score from the average value for each vector, thus completing the spatial aggregation of the local evaluation vectors.

[0113] First, the system receives local evaluation vectors from all analysis nodes via the IoT communication link, ensuring that each vector has complete dimensions and valid values. The preset baseline state vector is determined based on historical data of the power distribution equipment's long-term normal operation, reflecting the equipment's optimal operating state. For example, the baseline state vector is [15, 15, 12, 10], corresponding to the standard scores of voltage stability, current stability, temperature stability, and vibration stability dimensions, respectively. The system calculates the similarity coefficient between each local evaluation vector and the baseline state vector. First, it calculates the sum of the products of the corresponding dimension scores of the two vectors, then calculates the square root of the sum of the squares of the scores of each dimension of each vector, and divides the sum of the products by the product of the square roots of the two vectors to obtain the similarity coefficient of a single node. The system calculates the sum of the deviations between all local evaluation vectors and the baseline state vector, the sum of the absolute values ​​of the differences between the scores of each dimension of each node and the corresponding dimension scores of the baseline vector, and then calculates the average of the sum of the deviations of all nodes. The average is substituted into the log(1 + average deviation) formula to calculate the logarithmic term. The system adds the similarity coefficients of all nodes together with the calculated logarithmic term, and the final result is the comprehensive state index of the power distribution equipment.

[0114] A status assessment matrix is ​​constructed based on comprehensive status indicators. The rows of the matrix represent each IoT analysis node involved in the analysis, and the columns represent the core operating indicator dimensions of the power distribution equipment, the comprehensive status indicators, and the deviation values ​​of each node's score from the average value. In each cell of the matrix, the score of the corresponding dimension of the corresponding node, the comprehensive status indicator value, or the deviation value is filled in to ensure that each data is accurately matched, forming a power distribution equipment status assessment matrix with a clear structure and complete data.

[0115] The local evaluation vector comes from the analysis node in the Internet of Things. The analysis node first receives the data segment of the unified state image, performs multi-dimensional feature extraction on the data segment to obtain the state feature set, then weights and fuses the multi-source feature values ​​in the state feature set to obtain the local evaluation value, and finally quantizes the local evaluation value to form the local evaluation vector.

[0116] The baseline state vector is a pre-defined standard vector used to measure the normal operating status of power distribution equipment and serves as a reference for constructing the state assessment matrix.

[0117] The number of analysis nodes is the total number of analysis nodes in the Internet of Things (IoT) participating in the independent analysis of the unified state image data segment. It is determined based on the computing power of the analysis nodes in the IoT and the data segment allocation according to the current load.

[0118] The modulus of a local evaluation vector is the result of calculating the modulus of each local evaluation vector. The calculation process involves squaring the values ​​of each dimension in the local evaluation vector, summing them, and then taking the square root of the sum.

[0119] The magnitude of the reference state vector is the result of calculating the magnitude of the preset reference state vector. The calculation process involves squaring the values ​​of each dimension in the reference state vector, summing them, and then taking the square root of the summation result.

[0120] The dot product of the local evaluation vector and the baseline state vector is obtained by multiplying the values ​​of the corresponding dimensions of the two vectors separately, and then summing all the results.

[0121] The magnitude of the difference between the local evaluation vector and the baseline state vector is obtained by first calculating the difference between the corresponding dimension values ​​of the two vectors, then summing the squares of all the differences, and finally taking the square root of the summation result.

[0122] The core of this calculation process is to comprehensively reflect the degree of fit between the current operating status of the power distribution equipment and the preset benchmark status, and to obtain a comprehensive status index by superimposing the results of the two calculations.

[0123] The first part calculates the sum of the similarity values ​​between all local evaluation vectors and the baseline state vector. The higher the similarity, the larger this part is, and the better it reflects that the current running state is close to the baseline state.

[0124] The second part calculates the logarithmic correlation value of the average deviation between all local evaluation vectors and the baseline state vector. The smaller the average deviation, the smaller this value, which better reflects the high consistency of the evaluation results of each analysis node.

[0125] The comprehensive status index obtained by adding the results of the two parts can fully integrate the evaluation results of each analysis node, providing key data support for constructing the status evaluation matrix of power distribution equipment.

[0126] When the similarity between the local evaluation vector and the baseline state vector of each analysis node increases as a whole, the summation result of the first part will increase, and the comprehensive state index will rise accordingly, indicating that the current operating state of the power distribution equipment is closer to the preset baseline state.

[0127] When the average deviation between the local evaluation vector and the baseline state vector of each analysis node decreases as a whole, the logarithmic calculation result of the second part will decrease, and the comprehensive state index will decrease accordingly, indicating that the evaluation of the equipment state by each analysis node is more consistent and closer to the baseline state.

[0128] When the similarity between the local evaluation vector of each analysis node and the baseline state vector decreases as a whole, the summation result of the first part will decrease, and the comprehensive state index will decrease accordingly, indicating that the current operating state of the power distribution equipment deviates from the preset baseline state.

[0129] When the average deviation between the local evaluation vector and the baseline state vector of each analysis node increases as a whole, the logarithmic calculation result of the second part will increase, and the comprehensive state index will rise accordingly, indicating that the evaluation differences of the equipment state by each analysis node have increased and deviated from the baseline state.

[0130] The beneficial effects include: by analyzing the computing power of nodes and the data segments allocated to load, parallel transmission and processing of a unified state image are achieved, avoiding node resource overload or idleness and improving data processing efficiency; multi-dimensional feature extraction comprehensively covers the key features of the core operating parameters of power distribution equipment, providing rich data support for subsequent evaluation; the weighted fusion process highlights the impact of key features on equipment status through differentiated weights, making local evaluation values ​​more in line with actual operating needs; the quantification and spatial aggregation of local evaluation vectors, combined with the benchmark state vector to calculate comprehensive state indicators and construct a matrix, allows the operating status of power distribution equipment to be presented in an intuitive and systematic form, facilitating accurate understanding of equipment status and providing a reliable basis for subsequent fault prediction and control, thus improving the scientificity and effectiveness of remote monitoring of power distribution equipment.

[0131] S4. Map the state evaluation matrix to the fault state feature sequence to obtain the matching degree between the current operating state and historical fault state cases in the power distribution equipment, and generate a comprehensive evaluation report of the power distribution equipment based on the matching degree and the operating state development trend of the power distribution equipment.

[0132] In this embodiment of the invention, mapping the state evaluation matrix to the fault state feature sequence to obtain the matching degree between the current operating state and historical fault state cases of the power distribution equipment, and generating a comprehensive evaluation report of the power distribution equipment based on the matching degree and the operating state development trend of the power distribution equipment, includes:

[0133] The evaluation dimensions of the state evaluation matrix are compared with the corresponding features of the fault state feature sequence in a multi-dimensional pattern comparison to obtain the similar areas of the current operating state of the power distribution equipment and the historical fault state cases in terms of feature morphology and change trend.

[0134] By analyzing the evolution of the state assessment matrix over time, the change direction and intensity characteristics of the assessment parameters in the state assessment matrix are extracted to obtain a description of the operating state development trend of the power distribution equipment.

[0135] The identification results of the similar regions are fused with the description of the operating status development trend to obtain a comprehensive evaluation report of the power distribution equipment.

[0136] First, define the evaluation dimensions of the state assessment matrix, specifically covering four core dimensions: voltage stability, current stability, temperature stability, and vibration stability. Each dimension corresponds to specific operating parameter characteristics. For example, the voltage stability dimension is associated with characteristics such as voltage fluctuation amplitude and average voltage; the current stability dimension is associated with characteristics such as current fluctuation amplitude and average current; the temperature stability dimension is associated with characteristics such as temperature change rate and average temperature; and the vibration stability dimension is associated with characteristics such as the number of vibration frequency fluctuations and average vibration frequency. Simultaneously, determine the characteristics in the fault state feature sequence corresponding to each evaluation dimension. For instance, in the fault state feature sequence of historical short-circuit faults, the characteristic corresponding to the voltage stability dimension is "a sudden increase in voltage fluctuation amplitude and a sudden decrease in average voltage," while the characteristic corresponding to the current stability dimension is "a sudden increase in current fluctuation amplitude and a sudden increase in average current." In the fault state feature sequence of historical overload faults, the characteristic corresponding to the temperature stability dimension is "a continuous increase in temperature change rate." "Rising, with the average temperature gradually increasing," corresponds to the current stability dimension's characteristics of "a slight increase in current fluctuation amplitude and a persistently high average current." The real-time numerical changes of each dimension in the state assessment matrix are compared dimension-by-dimensionally and time-by-time with the corresponding numerical changes in the fault state feature sequence. For example, comparing the real-time value change curve of the voltage stability dimension with the voltage fluctuation characteristic change curve in historical short-circuit faults, we observe whether the shapes of the two curves are consistent. If both show a shape of "first remaining stable within the normal range, then rapidly deviating from the normal range within 5 minutes," and the change trend is "from stable to sudden change," then this time period is marked as a similar area between the current operating state and historical fault state cases. The start time, end time, corresponding fault type, and specific details of feature matching of the similar area are recorded, ultimately obtaining the similar area between the current operating state of the power distribution equipment and historical fault state cases in terms of feature shape and change trend.

[0137] State assessment matrices were collected sequentially at different time points, with a 10-minute interval between each point, forming a continuous sequence of state assessment matrices. For example, state assessment matrices were collected for four time points: 13:00, 13:10, 13:20, and 13:30. For each assessment dimension, the dimension values ​​were extracted sequentially from the matrices at each time point, and the direction of change in the values ​​was analyzed. For example, if the voltage stability dimension value decreased from 15 at 13:00 to 14.2 at 13:10, and then further to 13.5 at 13:20... If the value finally drops to 12.8 at 13:30, then the direction of change for this dimension is determined to be "continuous decline"; if the current stability dimension value is 14.5 at 13:00, 14.6 at 13:10, 14.4 at 13:20, and 14.5 at 13:30, then the direction of change is determined to be "small fluctuations, overall stability"; at the same time, the intensity of change of each dimension value is calculated, and the absolute value of the difference between values ​​at adjacent time points is used as the indicator of the intensity of change, for example, the absolute value of the difference between the voltage stability dimension values ​​at 13:00 and 13:10. The values ​​for 13:10-13:20 and 13:20-13:30 are 0.7, indicating that the decreasing intensity of this dimension remains stable and at a moderate level. If the absolute values ​​for the temperature stability dimension are 0.3 for 13:00-13:10, 0.6 for 13:10-13:20, and 1.1 for 13:20-13:30, it indicates that the increasing intensity of this dimension is gradually strengthening. Integrating the direction and intensity characteristics of each dimension forms a complete description of the operating status of the power distribution equipment, for example, " The voltage stability dimension has shown a continuous downward trend over the past 40 minutes, with the decrease intensity remaining between 0.7 and 0.8, which is at a moderate level; the current stability dimension shows a slight fluctuation and an overall stable trend, with a weak change intensity; the temperature stability dimension shows a gradual upward trend, with the increase intensity gradually increasing from 0.3 to 1.1, and the intensity continuously strengthening; the vibration stability dimension value remains between 12.0 and 12.2, with no obvious direction of change and an extremely weak change intensity. This description is the development trend of the operating status of the power distribution equipment.

[0138] The identification results of similar regions are systematically integrated with the description of the operational status development trend. First, the comprehensive evaluation report clearly presents the details of the similar regions, including: "During the period from 13:20 to 13:30, the current change in the temperature stability dimension overlaps with the similar regions of temperature characteristics in the historical overload fault state characteristic sequence, with a similarity of 82%; during the same period, the change in the voltage stability dimension partially overlaps with the similar regions of the historical short-circuit fault characteristic sequence, with a similarity of 65%." Then, the description of the operational status development trend is incorporated, explaining the impact of the current direction and intensity of change in each dimension on the similar regions. For example, "Based on trend analysis, the increasing intensity of the temperature stability dimension will further enhance the similarity with historical overload fault characteristics; although the voltage stability dimension shows a decreasing trend, the decreasing intensity is stable, and..." The report states that "the similarity of historical short-circuit fault characteristics has not yet shown a significant increase." It then supplements this with an assessment of the overall operating status of the power distribution equipment, based on whether the values ​​of each dimension are within the normal range, the similarity of similar areas, and trend changes, to determine whether the equipment currently faces fault risks. For example, "the current voltage stability dimension value of 12.8 is still within the normal range, while the temperature stability dimension value of 15.2 is close to the upper limit of the normal range, indicating a slight overload risk." Finally, the report clarifies the key areas for future attention, such as "it is recommended to focus on monitoring changes in the temperature stability dimension, recording temperature-related parameters every 5 minutes, while continuously tracking the downward trend of the voltage stability dimension to prevent further deviation from the normal range." Through this integration process, a comprehensive assessment report of the power distribution equipment with complete content and clear logic is formed.

[0139] The beneficial effects include: accurately locating similar areas between the current operating status and historical fault status through multi-dimensional pattern comparison, providing a direct basis for fault risk identification; clearly grasping the direction and intensity of changes in equipment operating status by analyzing the evolution of the status assessment matrix over time, enabling accurate prediction of status trends; and generating a comprehensive assessment report by integrating similar area identification results with trend descriptions, which includes specific similarity feature information, dynamic trend analysis, and overall status judgment and attention suggestions. This report can comprehensively and accurately reflect the operating status of power distribution equipment, providing scientific and detailed decision support for subsequent generation of control commands and ensuring stable equipment operation, effectively improving the accuracy and practicality of remote monitoring of power distribution equipment.

[0140] S5. Extract the multi-dimensional feature parameters from the comprehensive evaluation report, analyze the influence intensity of the multi-dimensional feature parameters on the operating status of the power distribution equipment, and generate control instructions for the power distribution equipment based on the influence intensity.

[0141] In this embodiment of the invention, the step of extracting multidimensional feature parameters from the comprehensive evaluation report, analyzing the influence intensity of the multidimensional feature parameters on the operating status of the power distribution equipment, and generating control instructions for the power distribution equipment based on the influence intensity includes:

[0142] The evaluation indicators that change drastically or have a significant impact in the comprehensive evaluation report are selected as the key status indicators of the comprehensive evaluation report, and trend analysis is performed on the key status indicators to obtain the changing trend characteristics of the comprehensive evaluation report.

[0143] By integrating the key status indicators and the changing trend characteristics, the multidimensional feature parameters of the comprehensive evaluation report are obtained.

[0144] Analyze the correlation between the multidimensional feature parameters and the historical operating status changes of the power distribution equipment, and evaluate the degree of influence and response priority of the multidimensional feature parameters on the operating status of the power distribution equipment based on the correlation;

[0145] Based on the degree of impact and the response priority, a mapping relationship for the control strategy of the power distribution equipment is established, and control instructions adapted to the operating state of the power distribution equipment are generated.

[0146] All assessment indicators were extracted from the comprehensive assessment report. These indicators specifically include voltage stability, current stability, temperature stability, and vibration stability. Each indicator includes its current value, the magnitude of change over the past 30 minutes, and the duration of the change. The assessment determined whether each indicator fell into the categories of "drastic change" or "significant impact." "Dramatic change" was defined as a change exceeding the indicator's normal fluctuation range over the past 30 minutes. For example, temperature stability rising from 12.0 to 15.5 in the past 30 minutes (a change of 3.5%) far exceeded the normal fluctuation range and thus constituted drastic change. "Significant impact" was defined as an indicator directly related to the core operational safety of the power distribution equipment; even if the magnitude of the change did not exceed the normal range, it still required close monitoring. For example, voltage stability dropped from 13.1 to 12.8 in the past 30 minutes. The change of 0.3 was within the normal range, but voltage directly determines whether the equipment circuit can conduct normally, which is a significant factor. After screening temperature stability and voltage stability as key status indicators, trend analysis was performed on them. By recording the indicator values ​​every 10 minutes and analyzing the pattern of value changes, it was found that temperature stability showed a continuous and rapid upward trend of "an average increase of 1.17 every 10 minutes, and the rate of increase showed no signs of slowing down", while voltage stability showed a slow and continuous downward trend of "an average decrease of 0.27 every 10 minutes, and the rate of decrease remained stable". These trend information, which includes the rate of change and the persistence of change, are the trend characteristics of the comprehensive evaluation report.

[0147] Collect complete information on key status indicators, including the current value of temperature stability (15.5, with a change of 3.5 in the past 30 minutes) and the current value of voltage stability (12.8, with a change of 0.3 in the past 30 minutes). Simultaneously, compile descriptions of the rate and duration of change for both indicators from the trend characteristics. Following a fixed structure of "Indicator Name - Current Value - Change in the Past 30 Minutes - Trend," integrate the key status indicators and trend characteristics to form structured data such as "Temperature Stability - Current Value 15.5 - Change in the Past 30 Minutes 3.5 - Continuously and Rapidly Rising" and "Voltage Stability - Current Value 12.8 - Change in the Past 30 Minutes 0.3 - Slowly and Continuously Decreasing." These structured data collectively constitute the multidimensional feature parameters of the comprehensive evaluation report.

[0148] We retrieved historical operating data from the power distribution equipment over the past five years and filtered out historical data segments with similar multidimensional characteristic parameters to the current data. For example, we identified historical cases where temperature stability showed a continuous and rapid increase, while voltage stability showed a slow and continuous decrease, finding 12 matching cases. We analyzed the correlation between the multidimensional characteristic parameters and the changes in equipment operating status in each case. For instance, in eight cases of rapid temperature increases, when the temperature stability value reached 15.0 and continued to rise for one hour, the equipment triggered an overload alarm, and the operating status changed from "normal" to "warning." In all cases of slow voltage decreases, when the voltage stability value dropped to 12.0 and continued to decrease... Three hours later, the equipment displayed a low voltage warning, and its operating status changed from "normal" to "attention." By statistically analyzing the number of related cases and the time elapsed, the degree of impact was assessed: multi-dimensional characteristic parameters related to temperature stability could cause equipment status deterioration within 1 hour, and the degree of impact was judged as "high"; multi-dimensional characteristic parameters related to voltage stability required 3 hours to affect equipment status, and the degree of impact was judged as "medium". The response priority was set according to the degree of impact and the time elapsed of deterioration. Temperature stability-related parameters with high impact and rapid deterioration had a response priority of "level 1", while voltage stability-related parameters with medium impact and slow deterioration had a response priority of "level 2".

[0149] The corresponding rules for "impact level - response priority - control strategy" are clearly defined: A "high" impact level and a "level 1" response priority correspond to the control strategy of "reducing equipment load rate to decrease heat generation and enhance heat dissipation capacity," with specific parameters such as "load rate reduced by 10%, cooling fan switched to high-speed operation mode." A "medium" impact level and a "level 2" response priority correspond to the control strategy of "fine-tuning input voltage to compensate for voltage drop," with specific parameters such as "input voltage increased by 0.5%." Based on the impact level and response priority corresponding to the current multi-dimensional characteristic parameters, the corresponding control strategy is matched and transformed into specific instructions that the equipment can execute. The instructions include the execution object, execution action, and specific parameters. For example, "Instruction 1: Power distribution equipment load control module, adjust the load rate from 90% to 80%; Instruction 2: Power distribution equipment heat dissipation module, switch the cooling fan operation mode to high speed; Instruction 3: Power distribution equipment voltage regulation module, increase the input voltage from 220V to 221.1V." These instructions are the control instructions adapted to the current operating state of the power distribution equipment.

[0150] The beneficial effects are as follows: By selecting indicators that change drastically or have a significant impact as key status indicators, ineffective analysis of non-key indicators is avoided, and the focus is on core influencing factors; the multi-dimensional feature parameters formed by integrating key indicators and trend characteristics comprehensively reflect the core status and dynamic changes of equipment operation, providing a complete data foundation for subsequent analysis; by assessing the degree of impact and response priority by correlating historical operating data, the judgment of the impact of each parameter is ensured to be objective and accurate, avoiding subjective decision-making bias; and control instructions are generated based on the mapping relationship between the degree of impact and priority, enabling the instructions to accurately match the current status of the equipment, prioritizing the resolution of urgent high-impact issues while also taking into account medium-impact issues, effectively improving the targeting and effectiveness of control, ensuring the stable operation of power distribution equipment and preventing failures.

[0151] S6. The control command is sent to the power distribution equipment through the network layer of the Internet of Things, the execution effect of the control command is monitored in real time, and closed-loop verification is completed based on the feedback data of the power distribution equipment.

[0152] In this embodiment of the invention, the step of sending the control command to the power distribution equipment through the network layer of the Internet of Things, monitoring the execution effect of the control command in real time, and completing closed-loop verification based on the feedback data from the power distribution equipment includes:

[0153] The control command and the identification information of the power distribution equipment are protocol-encapsulated to obtain a standard command data packet of the power distribution equipment, and the standard command data packet is distributed and transmitted through the secure channel of the network layer.

[0154] After the standard instruction data packet is sent, the status confirmation information and real-time operation data stream returned by the power distribution equipment are received to obtain the instruction response dataset of the power distribution equipment;

[0155] The command response dataset is parsed to obtain the key state parameters of the command response dataset, and the key state parameters are compared with the expected effect target of the control command in multiple dimensions to obtain the comparative analysis results of the power distribution equipment.

[0156] Based on the comparative analysis results, the target achievement rate of the control command is determined, and the control command and the identification information of the power distribution equipment are updated and protocol-encapsulated according to the target achievement rate. The control command includes specific execution requirements for the power distribution equipment load control module to adjust the load rate from 90% to 80%, the heat dissipation module to switch the cooling fan to high-speed operation mode, and the voltage regulation module to increase the input voltage from 220V to 221.1V. The equipment identification information includes the unique equipment code of the power distribution equipment, the substation number to which it belongs, and the installation location code. The MQTT Internet of Things communication protocol is used for encapsulation. During encapsulation, the data packet header is constructed first, and the protocol version number, equipment identification information, and data packet length are written in. The middle section contains the specific content of the control instructions, including the execution module, execution action, and target parameters. Finally, a checksum is generated at the end of the data packet. The checksum is calculated by using the CRC32 algorithm on the header and middle data to ensure that the data packet is not tampered with or damaged during transmission, thus obtaining the standard instruction data packet for the power distribution equipment. This standard instruction data packet is distributed and transmitted through the TLS encrypted secure channel of the Internet of Things network layer. Before transmission, the device and the network layer server first authenticate each other, and the two parties exchange digital certificates to verify the legitimacy. After successful authentication, they negotiate to use the AES-256 encryption algorithm to encrypt the standard instruction data packet, and then transmit the encrypted data packet to the target power distribution equipment through the network layer communication link.

[0157] Within 5 minutes of the standard command data packet being sent, data returned by the power distribution equipment is received in real time. First, the status confirmation information sent by the equipment is received. This information includes the command reception time, the corresponding standard command data packet number, the command parsing result, and the reception status. If the equipment successfully parses the command, the reception status is marked as "successful"; if the command format is abnormal, it is marked as "failure" with an error reason. At the same time, real-time operation data streams sent by the equipment at a frequency of 1 second are received. The data streams contain real-time monitoring values ​​of load rate, cooling fan speed, input voltage, temperature stability, and voltage stability. The status confirmation information received within 5 minutes and all real-time operation data streams are sorted and summarized in chronological order to ensure that there is no missing or duplicate data, and finally the command response dataset of the power distribution equipment is obtained.

[0158] The command response dataset of the power distribution equipment is analyzed. First, the command reception status and parsing results are extracted from the status confirmation information to confirm that the control command has been successfully received and parsed by the equipment. Then, key status parameters are extracted from the real-time operation data stream, and the average load rate, continuous monitoring value of the cooling fan speed, average input voltage, average temperature stability, and average voltage stability over a 5-minute period are calculated. The load rate is calculated from data converted from the data of the equipment's built-in current and power sensors. The cooling fan speed is read from the feedback signal of the fan drive module. The input voltage is directly collected by the voltage sensor. Temperature stability and voltage stability are calculated according to the real-time monitoring values ​​of the corresponding dimensions according to the previously set calculation rules. The expected effect target of the control command is defined, namely, the average load rate should be within the range of 80% ± 2%, and the cooling fan speed should be maintained within the range of 2000 rpm ± 100 rpm. The following parameters are required for the power distribution equipment: average input voltage must be within 221.1V ± 0.2V; average temperature stability must drop below 14.0V; and average voltage stability must be maintained within 12.8 ± 0.1V. The extracted key state parameters are compared with the expected results across multiple dimensions. For example, if the actual average load rate is 81%, within the expected range, this dimension is considered compliant; the actual cooling fan speed is 1950 rpm, within the expected range, and is considered compliant; the actual average input voltage is 221.0V, within the expected range, and is considered compliant; the actual average temperature stability is 14.2V, exceeding the expected range, and is considered non-compliant; and the actual average voltage stability is 12.7V, within the expected range, and is considered compliant. Based on this, a comparative analysis of the power distribution equipment is compiled, clearly recording the actual values, expected values, and compliance conclusions for each key state parameter.

[0159] Based on the comparative analysis results, the target achievement rate of the control commands was determined. The number of dimensions that met the key state parameters was counted. In this case, 4 out of 5 dimensions met the target and 1 did not. The target achievement rate was calculated as the ratio of the number of met dimensions to the total number of dimensions, i.e., 80%. A target achievement rate judgment standard was set. If the achievement rate is ≥90%, the current control strategy does not need to be adjusted. If 60% ≤ achievement rate < 90%, the control strategy needs to be optimized based on the dimension that did not meet the target. If the achievement rate is <60%, the multi-dimensional characteristic parameters need to be re-evaluated and a new control strategy needs to be formulated. In this case, the temperature stability dimension did not meet the target. The original control strategy only controlled the temperature by reducing the load rate by 10%. The updated control strategy is to further reduce the load rate to 75%. At the same time, in addition to the continuous high-speed operation of the cooling fan, a powerful cooling mode is added every 30 minutes for 5 minutes each time to enhance the cooling effect and promote the decrease of temperature stability. The updated control strategy is stored in the monitoring system database of the power distribution equipment for subsequent control command generation, completing the closed-loop verification of the control of the power distribution equipment.

[0160] The beneficial effects are as follows: by encapsulating control commands and equipment identification information protocols and transmitting them through a secure channel, it ensures that the commands accurately match the target power distribution equipment and prevents the commands from being leaked or tampered with during transmission, thus guaranteeing the accuracy and security of command transmission; real-time reception of equipment feedback status confirmation information and operating data streams allows for timely monitoring of command reception and equipment operating dynamics, providing complete data support for subsequent effect evaluation; parsing and multi-dimensional comparison of the response dataset can accurately determine whether the actual effect of each control dimension has met expectations, avoiding ambiguous judgments on the control effect; updating the control strategy based on the target achievement level forms a complete control closed loop, which can continuously optimize control measures, improve the accuracy and effectiveness of controlling the operating status of power distribution equipment, and further ensure the stable and reliable operation of power distribution equipment; the control strategy of the power distribution equipment completes closed-loop verification.

[0161] like Figure 2 The diagram shown is a functional block diagram of a remote monitoring system for power distribution equipment based on the Internet of Things provided in an embodiment of the present invention.

[0162] The IoT-based remote monitoring system 100 for power distribution equipment described in this invention can be installed in an electronic device. Depending on the functions implemented, the IoT-based remote monitoring system 100 may include a data processing module 101, a fault status analysis module 102, a current status analysis module 103, an operating status evaluation module 104, a control command generation module 105, and a closed-loop verification module 106. The modules described in this invention can also be referred to as units, which are a series of computer program segments that can be executed by the processor of an electronic device and can perform a fixed function, stored in the memory of the electronic device.

[0163] In this embodiment, the functions of each module / unit are as follows:

[0164] The data processing module 101 is used to compare the multi-source heterogeneous monitoring data of the power distribution equipment with the normal operating parameters of the power distribution equipment, remove abnormal data in the multi-source heterogeneous monitoring data, and perform spatiotemporal alignment on the multi-source heterogeneous monitoring data to obtain a unified state image of the power distribution equipment.

[0165] The fault state analysis module 102 is used to perform feature analysis on the historical fault types of the power distribution equipment, obtain the feature parameters of the historical fault types, and select fault state feature sequences that are strongly correlated with the feature parameters from the unified state image.

[0166] The current state analysis module 103 is used to distribute the unified state image to the analysis nodes in the Internet of Things, and to perform independent analysis on the unified state image to obtain the analysis results of the analysis nodes. The analysis results are then mapped to a preset multi-dimensional evaluation space to construct the state evaluation matrix of the power distribution equipment.

[0167] The operation status assessment module 104 is used to map the status assessment matrix to the fault status feature sequence to obtain the matching degree between the current operation status and historical fault status cases in the power distribution equipment, and generate a comprehensive assessment report of the power distribution equipment based on the matching degree and the development trend of the operation status of the power distribution equipment.

[0168] The control instruction generation module 105 is used to extract multi-dimensional feature parameters from the comprehensive evaluation report, analyze the influence intensity of the multi-dimensional feature parameters on the operating status of the power distribution equipment, and generate control instructions for the power distribution equipment based on the influence intensity.

[0169] The closed-loop verification module 106 is used to send the control command to the power distribution equipment through the network layer of the Internet of Things, monitor the execution effect of the control command in real time, and complete the closed-loop verification based on the feedback data of the power distribution equipment.

[0170] In the several embodiments provided by this invention, it should be understood that the disclosed methods and systems can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and other division methods may be used in actual implementation.

[0171] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0172] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0173] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0174] This application embodiment can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0175] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A power distribution equipment remote monitoring method based on Internet of Things, characterized in that, The method comprises: S1, comparing the multi-source heterogeneous monitoring data of the power distribution equipment with the normal operation parameters of the power distribution equipment, removing the abnormal data in the multi-source heterogeneous monitoring data, and performing space-time alignment on the multi-source heterogeneous monitoring data to obtain a unified state image of the power distribution equipment; S2, performing feature analysis on the historical fault types of the power distribution equipment to obtain characteristic parameters of the historical fault types, and selecting fault state feature sequences strongly related to the characteristic parameters from the unified state image; S3, distributing the unified state image to analysis nodes in the Internet of Things, independently analyzing the unified state image to obtain analysis results of the analysis nodes, mapping the analysis results to a preset multi-dimensional evaluation space, and constructing a state evaluation matrix of the power distribution equipment, comprising: According to the computing capacity and current load of the analysis nodes in the Internet of Things, the data segments of the unified state image are transmitted to the analysis nodes in parallel; In the analysis nodes, multi-dimensional feature extraction is performed on the data segments to obtain a state feature set of the unified state image; Weighted fusion is performed on the multi-source feature values in the state feature set to obtain a local evaluation value of the analysis node; The local evaluation value is quantized to obtain a local evaluation vector of the analysis node, and the local evaluation vector is spatially aggregated to obtain a state evaluation matrix of the power distribution equipment, comprising: The local evaluation vector returned by the analysis node is obtained, and a comprehensive state index of the power distribution equipment is calculated according to a preset reference state vector, wherein the calculation formula of the comprehensive state index is: ； wherein, denotes the comprehensive status indicator, denotes the i-th local evaluation vector, denotes the i-th local evaluation vector, denotes a preset reference status vector, denotes the i-th local evaluation vector, denotes the i-th local evaluation vector, denotes the reference status vector, denotes the number of analysis nodes; According to the comprehensive state index, a state evaluation matrix of the power distribution equipment is constructed; S4, mapping the state evaluation matrix to the fault state feature sequence to obtain the matching degree of the current operating state and the historical fault state case in the power distribution equipment, and generating a comprehensive evaluation report of the power distribution equipment according to the matching degree and the operating state development trend of the power distribution equipment; S5, extracting multi-dimensional feature parameters in the comprehensive evaluation report, analyzing the influence intensity of the multi-dimensional feature parameters on the operating state of the power distribution equipment, and generating a control instruction of the power distribution equipment according to the influence intensity; S6, the network layer of the Internet of Things is used to issue the control instruction to the power distribution equipment, real-time monitor the execution effect of the control instruction, and complete closed-loop verification according to the feedback data of the power distribution equipment.

2. The power distribution equipment remote monitoring method based on the Internet of Things according to claim 1, characterized in that, The method comprises: Performing format unification processing on the multi-source heterogeneous monitoring data to obtain a preliminary standardized data set of the power distribution equipment; In the sliding time window, analyze the parameter fluctuation threshold when the power distribution equipment is in normal operation, map the preliminary standardized data set to the parameter fluctuation threshold, and identify abnormal data sequences that exceed the parameter fluctuation threshold; Check the context consistency of the abnormal data sequence, and correct the abnormal data sequence according to the normal operation parameters of the power distribution equipment to obtain a clean data sequence of the power distribution equipment; Synchronize the time stamp of the clean data sequence, fill in the data missing points of the clean data sequence through time interpolation, and normalize the spatial coordinates of the clean data sequence to obtain a unified state image of the power distribution equipment.

3. The power distribution equipment remote monitoring method based on the Internet of Things according to claim 1, characterized in that, The feature analysis of the historical fault type of the power distribution equipment obtains a feature parameter of the historical fault type, and a fault state feature sequence strongly related to the feature parameter is selected from the unified state image, including: Extracting a multi-source data segment of the historical fault type, and standardizing the multi-source data segment to obtain a standard fault data set of the power distribution equipment; Dividing the standard fault data set into time windows, and quantifying the multi-dimensional features of the standard fault data set to obtain a multi-dimensional feature vector of the standard fault data set; The clustering analysis of the multi-dimensional feature vector obtains the discrimination degree of the multi-dimensional feature vector, and the feature parameter of the multi-dimensional feature vector is selected according to the discrimination degree.

4. The power distribution equipment remote monitoring method based on the Internet of Things according to claim 1, characterized in that, The feature analysis of the historical fault type of the power distribution equipment obtains a feature parameter of the historical fault type, and a fault state feature sequence strongly related to the feature parameter is selected from the unified state image, including: Mapping the feature parameter to the data dimension of the unified state image to obtain potential associated data of the power distribution equipment; Segmenting the potential associated data to obtain a data segment of the potential associated data; Integrating the similarity of the data segment and the feature parameter to obtain a similarity sequence of the power distribution equipment; Performing peak value detection on the similarity sequence to label a high-similarity data interval of the power distribution equipment; Fusing the high-similarity data interval and verifying the time continuity of the high-similarity data interval to obtain a fault state feature sequence of the power distribution equipment.

5. The power distribution equipment remote monitoring method based on the Internet of Things according to claim 1, characterized in that, The mapping of the state evaluation matrix into the fault state feature sequence obtains the matching degree of the current running state and the historical fault state case in the power distribution equipment, and generates a comprehensive evaluation report of the power distribution equipment according to the matching degree and the running state development trend of the power distribution equipment, including: Multi-dimensional mode comparison of the evaluation dimension of the state evaluation matrix and the corresponding features of the fault state feature sequence to obtain a similar area of the current running state of the power distribution equipment and the historical fault state case in feature form and change trend; Analyzing the evolution law of the state evaluation matrix over time to extract the change direction and intensity features of the evaluation parameters in the state evaluation matrix to obtain a running state development trend description of the power distribution equipment; Fusing the recognition result of the similar area and the running state development trend description to obtain a comprehensive evaluation report of the power distribution equipment.

6. The power distribution equipment remote monitoring method based on the Internet of Things according to claim 1, characterized in that, The extracting the multi-dimensional characteristic parameters in the comprehensive evaluation report, analyzing the influence intensity of the multi-dimensional characteristic parameters on the operation state of the power distribution equipment, and generating the regulation and control instruction of the power distribution equipment according to the influence intensity comprises: Screening the evaluation indexes with sharp changes or great influence in the comprehensive evaluation report as key state indexes of the comprehensive evaluation report, and performing trend analysis on the key state indexes to obtain the change trend characteristics of the comprehensive evaluation report; Integrating the key state indexes and the change trend characteristics to obtain the multi-dimensional characteristic parameters of the comprehensive evaluation report; Analyzing the correlation between the multi-dimensional characteristic parameters and the historical operation state changes of the power distribution equipment, and evaluating the influence degree and response priority of the multi-dimensional characteristic parameters on the operation state of the power distribution equipment according to the correlation; Establishing the regulation and control strategy mapping relationship of the power distribution equipment according to the influence degree and the response priority, and generating the regulation and control instruction adapted to the operation state of the power distribution equipment.

7. The power distribution equipment remote monitoring method based on the Internet of Things according to claim 1, characterized in that, The regulation and control instruction is delivered to the power distribution equipment through the network layer of the Internet of Things, the execution effect of the regulation and control instruction is monitored in real time, and closed-loop verification is completed according to the feedback data of the power distribution equipment, which comprises: Protocol encapsulation is performed on the regulation and control instruction and the identification information of the power distribution equipment to obtain a standard instruction data packet of the power distribution equipment, and the standard instruction data packet is distributed and transmitted through a secure channel of the network layer; After the standard instruction data packet is delivered, state confirmation information and real-time operation data stream returned by the power distribution equipment are received to obtain an instruction response data set of the power distribution equipment; The instruction response data set is parsed to obtain key state parameters of the instruction response data set, and the key state parameters are compared with an expected effect target of the regulation and control instruction in multiple dimensions to obtain a comparative analysis result of the power distribution equipment; The target achievement degree of the regulation and control instruction is judged according to the comparative analysis result, and the regulation and control strategy of the power distribution equipment is updated according to the target achievement degree to complete closed-loop verification.

8. The power distribution equipment remote monitoring system based on the Internet of Things, used for realizing the power distribution equipment remote monitoring method based on the Internet of Things of claim 1, characterized in that, The system comprises: A data processing module is configured to compare multi-source heterogeneous monitoring data of a power distribution equipment with normal operation parameters of the power distribution equipment, remove abnormal data in the multi-source heterogeneous monitoring data, and perform spatio-temporal alignment on the multi-source heterogeneous monitoring data to obtain a unified state image of the power distribution equipment; A fault state analysis module is configured to perform feature analysis on historical fault types of the power distribution equipment to obtain feature parameters of the historical fault types, and select fault state feature sequences strongly related to the feature parameters from the unified state image; A current state analysis module is configured to distribute the unified state image to analysis nodes in an Internet of Things, independently analyze the unified state image to obtain analysis results of the analysis nodes, map the analysis results to a preset multi-dimensional evaluation space, and construct a state evaluation matrix of the power distribution equipment; The running state evaluation module is configured to map the state evaluation matrix to the fault state feature sequence to obtain a matching degree of a current running state and a historical fault state case of the power distribution equipment, and generate a comprehensive evaluation report of the power distribution equipment according to the matching degree and a running state development trend of the power distribution equipment. The regulation instruction generation module is configured to extract multi-dimensional feature parameters in the comprehensive evaluation report, analyze an influence intensity of the multi-dimensional feature parameters on the running state of the power distribution equipment, and generate a regulation instruction of the power distribution equipment according to the influence intensity. The closed-loop verification module is configured to issue the regulation instruction to the power distribution equipment through a network layer of the Internet of Things, monitor an execution effect of the regulation instruction in real time, and complete closed-loop verification according to feedback data of the power distribution equipment.

Images