A power distribution cabinet remote operation and maintenance monitoring method and system based on an internet of things

Abstract

This invention relates to the field of IoT privacy technology, specifically an IoT-based remote operation and maintenance monitoring method and system for power distribution cabinets. The method includes: collecting operational data from the power distribution cabinet under monitoring using multiple IoT devices to obtain multiple raw operational time-series data; performing preliminary anomaly pattern identification based on the raw operational time-series data to obtain an initial identification result; if the initial identification result indicates a suspected anomaly in the power distribution cabinet, then performing node security authentication on the power distribution cabinet under monitoring to obtain an authenticated IoT device set; and using the authenticated IoT device set, the operation and maintenance monitoring platform, and the multiple raw operational time-series data for remote anomaly identification to obtain an anomaly value for the power distribution cabinet. This invention can improve the accuracy and security of power distribution cabinet monitoring, and reduce false alarm rates and resource consumption.

Description

Technical Field

[0001] This invention relates to the field of Internet of Things (IoT) privacy technology, and in particular to a method and system for remote operation and maintenance monitoring of power distribution cabinets based on IoT. Background Technology

[0002] In modern power systems, distribution cabinets are key nodes in power distribution, and the stability of their operation is directly related to the reliability of power supply and the safety of electricity use. With the maturity of Internet of Things (IoT) technology, deploying a large number of sensor devices in distribution cabinets to collect multi-dimensional status data such as current, voltage, and temperature in real time has become the core foundation for predictive maintenance and intelligent operation and maintenance.

[0003] Traditional methods for remote monitoring of power distribution cabinets typically involve simply uploading raw data collected by all IoT devices directly to a remote central platform for unified analysis and anomaly detection. This method involves massive data transmission of raw data, placing enormous pressure on network bandwidth and cloud platform computing resources. Furthermore, abnormal data generated by individual sensor malfunctions can easily trigger false alarms, interfering with operational decisions. Additionally, the lack of continuous security authentication for terminal devices poses a risk of malicious node access. Summary of the Invention

[0004] This invention provides a remote operation and maintenance monitoring method for power distribution cabinets based on the Internet of Things and a computer-readable storage medium. Its main purpose is to improve the accuracy and security of power distribution cabinet monitoring and reduce false alarm rate and resource consumption.

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

[0006] Receive the operation and maintenance instructions for the power distribution cabinet, and identify the power distribution cabinet to be monitored based on the operation and maintenance instructions. The power distribution cabinet to be monitored includes multiple IoT devices, and each IoT device contains an abnormal pattern recognition unit.

[0007] Based on the collection of working data from multiple IoT devices on the power distribution cabinet to be monitored, multiple raw working time series data are obtained, in which each raw working time series data corresponds one-to-one with an IoT device.

[0008] Anomaly patterns were initially identified based on multiple raw working time series data to obtain initial identification results, in which the initial identification results indicated that the power distribution cabinet was suspected of being abnormal or that the power distribution cabinet was not abnormal.

[0009] If the initial identification result is that the power distribution cabinet is normal, then return to the step of collecting working data of the power distribution cabinet to be monitored based on multiple IoT devices, until the initial identification result is that the power distribution cabinet is suspected of being abnormal.

[0010] If the initial identification result is that the power distribution cabinet is suspected to be abnormal, then the node security authentication of the power distribution cabinet to be monitored is performed to obtain the certified IoT device set, which includes multiple certified IoT devices.

[0011] Remote anomaly identification is performed using a set of certified IoT devices, a pre-built operation and maintenance monitoring platform, and multiple raw working time-series data to obtain abnormal values ​​for the power distribution cabinet.

[0012] Optionally, the preliminary identification of abnormal patterns based on multiple original working time-series data to obtain initial identification results includes:

[0013] Perform the following operation on each of the multiple raw working time series data sets:

[0014] Identify the original IoT device corresponding to the original working time sequence data;

[0015] The abnormal pattern recognition unit in the original IoT device is used to identify single data anomalies in the original working time series data to obtain single data anomaly values.

[0016] If a single data anomaly value exceeds the preset standard data anomaly value, a local data anomaly command will be generated.

[0017] Summarize the local data anomaly commands to obtain the local data anomaly command set;

[0018] Count the number of abnormal commands in the local data abnormal command set;

[0019] If the number of abnormal commands is greater than the preset value of 1, then the suspected abnormality of the power distribution cabinet is recorded as the initial identification result; otherwise, the absence of abnormality in the power distribution cabinet is recorded as the initial identification result.

[0020] Optionally, the step of using the abnormal pattern recognition unit in the original IoT device to perform single data anomaly identification on the original working time-series data to obtain a single data anomaly value includes:

[0021] The storage operation time series dataset is extracted based on the abnormal pattern recognition unit. The storage operation time series dataset includes multiple storage operation time series data, and each storage operation time series data contains a normal behavior label.

[0022] Storage time series data are extracted sequentially from the storage time series dataset. Time series features are extracted from the extracted storage time series data and the original time series data to obtain storage time series feature vectors and original time series feature vectors.

[0023] Calculate the normal mode deviation value based on the stored time series feature vector and the original time series feature vector;

[0024] Summarize the normal pattern deviation values ​​corresponding to each storage time-series data to obtain the normal pattern deviation value set;

[0025] The mean is calculated based on the deviation set of normal patterns to obtain a single data outlier.

[0026] Optionally, the node security authentication of the power distribution cabinet to be monitored, to obtain an authenticated IoT device set, includes:

[0027] For each IoT device in the monitored power distribution cabinet, perform the following operations:

[0028] Obtain the current behavior data of the IoT device, wherein the current behavior data includes: data packet reception rate, data packet transmission rate, and historical trust value;

[0029] The trust value of IoT devices is calculated based on current behavioral data to obtain the current trust value.

[0030] If the current trust value is greater than the preset standard trust value, a multi-dimensional authentication request is generated based on the IoT device. The multi-dimensional authentication request includes: node ID, request timestamp, and node MAC address.

[0031] The multidimensional authentication request is used to perform security authentication using a pre-built authentication server to obtain an authentication result, wherein the authentication result is authentication passed or authentication failed.

[0032] If the authentication result is successful, the IoT device will be marked as authenticated, thus obtaining an authenticated IoT device.

[0033] The certified IoT devices are aggregated to obtain a set of certified IoT devices.

[0034] Optionally, the remote anomaly identification using a set of certified IoT devices, a pre-built operation and maintenance monitoring platform, and multiple sets of raw operational time-series data to obtain abnormal values ​​for the power distribution cabinet includes:

[0035] Multiple raw working time-series data are encrypted using modular exponentiation based on a preset data encryption public key to obtain multiple IoT data to be transmitted.

[0036] Based on the certified IoT device set, multiple IoT data to be transmitted are transmitted to the operation and maintenance monitoring platform to obtain multiple IoT data to be decrypted.

[0037] An anomaly detection model for power distribution cabinets in the operation and maintenance monitoring platform is identified. The power distribution cabinet anomaly detection model includes: a data input layer, a local feature convolutional layer, a temporal feature extraction layer, an attention feature enhancement layer, a variational autoencoder, and a data output layer.

[0038] An anomaly detection model for power distribution cabinets and multiple undecrypted IoT data are used to detect anomalies in power distribution cabinets and obtain abnormal values ​​for the power distribution cabinets.

[0039] Optionally, the certified IoT device set transmits multiple IoT data to be transmitted to the operation and maintenance monitoring platform to obtain multiple IoT data to be decrypted, including:

[0040] Obtain the location of the certified IoT device for each certified IoT device in the certified IoT device set to obtain the certified device location set;

[0041] Identify multiple IoT devices to be relayed in the power distribution cabinet to be monitored based on the certified IoT device set;

[0042] Based on the location set of the certified device, multiple IoT devices to be relayed are assigned relay clusters to obtain multiple IoT device sets in the same cluster. Each IoT device set in the same cluster includes multiple IoT devices in the same cluster, and the location of each IoT device in the same cluster corresponds one-to-one with the location of the certified device.

[0043] Multiple IoT devices in the same cluster and an operation and maintenance monitoring platform are used to remotely transmit multiple IoT data to be transmitted, resulting in multiple IoT data to be decrypted.

[0044] Optionally, the step of using a power distribution cabinet anomaly detection model and multiple undecrypted IoT data to perform power distribution cabinet anomaly detection and obtain power distribution cabinet anomaly values ​​includes:

[0045] The target encrypted private key is obtained by reconstructing the private key based on multiple pre-defined fragmented IoT devices, where each fragmented IoT device contains a fragment value.

[0046] Multiple IoT data to be decrypted are decrypted using the target encryption private key to obtain multiple IoT input data;

[0047] Multiple IoT input data are merged to obtain fused input data;

[0048] The fused input data is fed into the distribution cabinet anomaly detection model to obtain the distribution cabinet anomaly value.

[0049] Optionally, before reconstructing the private key based on a preset set of multiple segmented IoT devices to obtain the target encrypted private key, the method further includes:

[0050] Based on a preset number of shards, select multiple trusted IoT devices from multiple IoT devices, wherein the number of trusted IoT devices is the same as the number of shards.

[0051] A random encrypted prime number is generated based on the preset data encryption private key, wherein the random encrypted prime number is greater than the data encryption private key;

[0052] Extract trusted IoT devices sequentially from multiple trusted IoT devices, and record the extracted trusted IoT devices as IoT devices to be fragmented;

[0053] Obtain the arrangement identifier of the IoT device to be fragmented among multiple trusted IoT devices;

[0054] A slice polynomial is generated based on a preset threshold value, a data encryption private key, and randomly encrypted prime numbers. The slice polynomial is: polynomial of degree. Indicates the threshold value;

[0055] The data is fragmented using permutation identifiers and fragmentation polynomials to obtain fragmentation values. Based on these fragmentation values, the fragmented IoT devices are allocated to obtain the fragmented IoT devices.

[0056] By aggregating the fragmented IoT devices, multiple fragmented IoT devices are obtained.

[0057] Optionally, the step of reconstructing the private key based on multiple pre-defined segmented IoT devices to obtain the target encrypted private key includes:

[0058] Based on the threshold value, multiple responding IoT devices are randomly selected from multiple segmented IoT devices to obtain multiple responding IoT devices, wherein the number of responding IoT devices is the same as the threshold value;

[0059] The target encrypted private key is calculated using multiple responding IoT devices and the aforementioned random encrypted prime number, wherein the formula for calculating the target encrypted private key is:

[0060] ;

[0061] in, Indicates the target encryption private key. Indicates the index number. Indicates and Different index numbers, This indicates the number of responding IoT devices out of a total of multiple responding IoT devices. Indicates the first of multiple responding IoT devices Each response IoT device corresponds to a fragment value. Represents the modulo function. This represents a randomly encrypted prime number.

[0062] To achieve the above objectives, the present invention also provides a remote operation and maintenance monitoring system for power distribution cabinets based on the Internet of Things, comprising:

[0063] The operation and maintenance instruction receiving module is used to receive operation and maintenance instructions for the power distribution cabinet and identify the power distribution cabinet to be monitored based on the operation and maintenance instructions. The power distribution cabinet to be monitored includes multiple IoT devices, and each IoT device contains an abnormal pattern recognition unit.

[0064] The working data acquisition module is used to collect working data from the power distribution cabinet under monitoring based on multiple IoT devices, and obtain multiple raw working time series data. The raw working time series data corresponds one-to-one with the IoT devices.

[0065] An anomaly pattern recognition module is used to perform preliminary anomaly pattern recognition based on multiple raw working time-series data to obtain an initial recognition result. The initial recognition result is either a suspected anomaly in the power distribution cabinet or no anomaly in the power distribution cabinet. If the initial recognition result is no anomaly in the power distribution cabinet, the module returns to the step of collecting working data from the power distribution cabinet to be monitored based on multiple IoT devices until the initial recognition result is a suspected anomaly in the power distribution cabinet. If the initial recognition result is a suspected anomaly in the power distribution cabinet, the module performs node security authentication on the power distribution cabinet to be monitored to obtain an authenticated IoT device set, which includes multiple authenticated IoT devices.

[0066] The remote data transmission module is used to remotely identify anomalies by utilizing a set of certified IoT devices, a pre-built operation and maintenance monitoring platform, and multiple raw working time-series data to obtain abnormal values ​​of the power distribution cabinet.

[0067] To address the above problems, the present invention also provides an electronic device, the electronic device comprising:

[0068] Memory, storing at least one instruction;

[0069] The processor executes the instructions stored in the memory to implement the IoT-based remote operation and maintenance monitoring method for power distribution cabinets described above.

[0070] To address the aforementioned issues, the present invention also provides a computer-readable storage medium storing at least one instruction, which is executed by a processor in an electronic device to implement the aforementioned IoT-based remote operation and maintenance monitoring method for power distribution cabinets.

[0071] To address the problems described in the background art, this invention first performs preliminary anomaly pattern identification based on multiple raw working time-series data to obtain initial identification results. This step involves identifying single data anomalies locally on each IoT device, and only determining a suspected anomaly in the power distribution cabinet when the total number of aggregated anomaly commands is greater than one. Compared to the existing technology where an alarm or upload is triggered by a single sensor data anomaly, this method effectively avoids false alarms caused by the failure of a single IoT device, reducing unnecessary communication resource consumption and platform computing load caused by false alarms. If the initial identification result indicates a suspected anomaly in the power distribution cabinet, node security authentication is performed on the power distribution cabinet to be monitored, resulting in an authenticated set of IoT devices. This step, after triggering a suspected anomaly, does not directly upload the data of all IoT devices, but first performs multi-layer node security authentication including dynamic trust assessment and static identity verification. Compared to the existing technology that typically only performs simple identity authentication or completely trusts all nodes, this method can... This invention identifies and isolates malicious nodes exhibiting abnormal behavior or potentially tampered with, ensuring the high credibility of the node set participating in subsequent data aggregation and transmission. After authentication, data is encrypted using modular exponentiation to guarantee transmission confidentiality. Through spatial location-based clustering, a multi-hop relay transmission network with authenticated IoT devices as cluster heads is formed, optimizing the network topology and resolving the issue of edge devices being unable to directly connect to remote platforms due to signal coverage or power consumption limitations, thus improving data transmission success rate. Furthermore, a secret sharing technology is employed to store the decryption private key in fragments across multiple trusted devices. The number of fragments required for private key reconstruction reaches a threshold, avoiding the single-point leakage risk of centralized private key storage and enhancing key security. Finally, on the operation and maintenance monitoring platform, a dedicated model integrating convolution, temporal feature extraction, attention mechanisms, and variational autoencoders is used to perform deep analysis and reconstruction error calculation on the decrypted multi-source time-series data, demonstrating higher accuracy and adaptability compared to traditional anomaly detection methods based on fixed thresholds. Therefore, this invention can improve the accuracy and security of power distribution cabinet monitoring, and reduce false alarm rates and resource consumption. Attached Figure Description

[0072] Figure 1 This is a flowchart illustrating a remote operation and maintenance monitoring method for power distribution cabinets based on the Internet of Things, provided in an embodiment of the present invention.

[0073] Figure 2 A functional block diagram of a remote operation and maintenance monitoring system for power distribution cabinets based on the Internet of Things provided in an embodiment of the present invention;

[0074] Figure 3 This is a schematic diagram of the structure of an electronic device for implementing the IoT-based remote operation and maintenance monitoring method for power distribution cabinets, as provided in an embodiment of the present invention.

[0075] Explanation of reference numerals in the attached figures:

[0076] 10. Electronic device; 11. Processor; 12. Memory; 13. Bus.

[0077] 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

[0078] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0079] This application provides a remote operation and maintenance monitoring method for power distribution cabinets based on the Internet of Things (IoT). The executing entity of this IoT-based remote operation and maintenance 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 operation and maintenance monitoring method for power distribution cabinets can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster.

[0080] Reference Figure 1 The diagram shown is a flowchart illustrating a remote operation and maintenance monitoring method for power distribution cabinets based on the Internet of Things (IoT) according to an embodiment of the present invention. In this embodiment, the remote operation and maintenance monitoring method for power distribution cabinets based on the IoT includes:

[0081] S1. Receive the power distribution cabinet operation and maintenance instruction, and confirm the power distribution cabinet to be monitored based on the power distribution cabinet operation and maintenance instruction. The power distribution cabinet to be monitored includes multiple IoT devices, and each IoT device contains an abnormal pattern recognition unit.

[0082] It is clear that the aforementioned power distribution cabinet maintenance instruction refers to a manually initiated instruction to monitor a specific power distribution cabinet unit. The power distribution cabinet to be monitored refers to the power distribution cabinet unit that needs to be monitored, as specified in the power distribution cabinet maintenance instruction. The aforementioned IoT device refers to a device deployed on the power distribution cabinet to be monitored, used to collect operational status data of the cabinet. This operational status data includes, for example, operating parameters such as current, voltage, temperature, and vibration. The IoT device includes, but is not limited to, current sensors, voltage monitoring modules, and temperature and humidity sensors. This IoT device can be installed at key monitoring points of the power distribution cabinet, such as busbars, circuit breakers, and cable joints. The IoT device also includes a communication module used to upload its own collected operational status data or receive operational status data sent by other IoT devices. The aforementioned abnormal pattern recognition unit refers to a unit integrated into a specific IoT device used to identify anomalies in single data points collected by that IoT device.

[0083] S2. Based on multiple IoT devices, the working data of the power distribution cabinet to be monitored is collected to obtain multiple raw working time series data, wherein the raw working time series data corresponds one-to-one with the IoT devices.

[0084] Understandably, the raw operating time sequence data refers to the operating data of the power distribution cabinet to be monitored collected by a certain Internet of Things device over a period of time, such as current value sequence, voltage value sequence, temperature value sequence, active power sequence, etc.

[0085] S3. Based on multiple original working time sequence data, perform preliminary identification of abnormal patterns to obtain initial identification results, where the initial identification results are either suspected abnormality of the power distribution cabinet or no abnormality of the power distribution cabinet.

[0086] It is clear that the initial identification result refers to the result obtained after preliminary identification of abnormal patterns, indicating whether the monitored distribution cabinet is operating abnormally. "Suspected abnormality" means that the monitored distribution cabinet may have entered an abnormal operating state, such as an abnormal voltage increase. "No abnormality" means that the monitored distribution cabinet has not entered an abnormal operating state.

[0087] In detail, the preliminary identification of abnormal patterns based on multiple original working time-series data to obtain initial identification results includes:

[0088] Perform the following operation on each of the multiple raw working time series data sets:

[0089] Identify the original IoT device corresponding to the original working time sequence data;

[0090] The abnormal pattern recognition unit in the original IoT device is used to identify single data anomalies in the original working time series data to obtain single data anomaly values.

[0091] If a single data anomaly value exceeds the preset standard data anomaly value, a local data anomaly command will be generated.

[0092] Summarize the local data anomaly commands to obtain the local data anomaly command set;

[0093] Count the number of abnormal commands in the local data abnormal command set;

[0094] If the number of abnormal commands is greater than the preset value of 1, then the suspected abnormality of the power distribution cabinet is recorded as the initial identification result; otherwise, the absence of abnormality in the power distribution cabinet is recorded as the initial identification result.

[0095] It should be explained that the "original IoT device" refers to the IoT device corresponding to the original working time-series data. The "single data anomaly value" refers to a numerical value representing the degree of anomaly in the original working time-series data; the larger the single data anomaly value, the higher the probability that the original working time-series data is anomalous. The "standard data anomaly value" refers to a manually set threshold used to determine whether the original working time-series data is anomalous; this standard data anomaly value can be set based on historical experience, for example, 0.8. When the single data anomaly value is greater than the standard data anomaly value, it indicates that the original working time-series data is anomalous compared to the previous normal working time-series data of the original IoT device. In this case, a local data anomaly instruction should be generated. This local data anomaly instruction is a command generated by the original IoT device to mark the anomaly in its collected original working time-series data. The "number of anomaly instructions" refers to the number of local data anomaly instructions in the local data anomaly instruction set. If no abnormalities are found in any of the original working time sequence data, it indicates that the entire monitored power distribution cabinet is operating normally without any abnormalities. If only one piece of original working time sequence data is abnormal, it may be that the original IoT device for that original working time sequence data is faulty, rather than the entire monitored power distribution cabinet being abnormal. In this case, in order to reduce the consumption of computing and communication resources, it is not necessary to transmit all the original working time sequence data to the operation and maintenance monitoring platform. It is only necessary to notify the relevant maintenance personnel to conduct further inspection of the original IoT device.

[0096] In detail, the step of using the abnormal pattern recognition unit in the original IoT device to perform single data anomaly identification on the original working time-series data to obtain a single data anomaly value includes:

[0097] The storage operation time series dataset is extracted based on the abnormal pattern recognition unit. The storage operation time series dataset includes multiple storage operation time series data, and each storage operation time series data contains a normal behavior label.

[0098] Storage time series data are extracted sequentially from the storage time series dataset. Time series features are extracted from the extracted storage time series data and the original time series data to obtain storage time series feature vectors and original time series feature vectors.

[0099] Calculate the normal mode deviation value based on the stored time series feature vector and the original time series feature vector;

[0100] Summarize the normal pattern deviation values ​​corresponding to each storage time-series data to obtain the normal pattern deviation value set;

[0101] The mean is calculated based on the deviation set of normal patterns to obtain a single data outlier.

[0102] It should be explained that the stored working time series dataset refers to a collection of multiple stored working time series data. The method for obtaining this stored working time series dataset is as follows: Raw working time series data collected at a certain moment is recorded as historical working time series data, and the collection time of this historical working time series data is recorded. If the historical working time series data and the collection time meet the following storage conditions: the difference between the collection time and the current time is less than a manually set dynamic change duration, and the historical working time series data is not marked as abnormal data, then the historical working time series data is stored in the storage module of the original IoT device. Simultaneously, data in the storage module that does not meet the above conditions is stored elsewhere. Other historical working time-series data of the storage conditions are deleted to ensure that the historical working time-series data in the storage module can reflect the normal data pattern of the original IoT device in the recent period. The historical working time-series data stored in the storage module constitutes the storage working time-series dataset. The aforementioned dynamic change duration refers to a predefined time window length. The dynamic change duration can be determined based on the change characteristics of the monitored physical quantity, the device operating cycle, or the frequency of change of environmental factors. For example, if the monitored physical quantity is temperature, it can be determined based on the temperature change cycle in a day, such as using 1 hour as the dynamic change duration.

[0103] Furthermore, the stored time-series feature vector refers to a numerical vector representing the time-series characteristics of the stored working time-series data. This stored time-series feature vector includes the average, minimum, maximum, and standard deviation of the stored working time-series data. The original time-series feature vector refers to a numerical vector representing the time-series characteristics of the original working time-series data. This original time-series feature vector includes the average, minimum, maximum, and standard deviation of the original working time-series data. The normal mode deviation value refers to a numerical value representing the degree of deviation between the original time-series feature vector and the stored time-series feature vector. This normal mode deviation value can be expressed as: ,in, This indicates the deviation value from the normal mode. Represents the cosine of a vector. and These represent the stored time-series feature vector and the original time-series feature vector, respectively.

[0104] S4. If the initial identification result is that the power distribution cabinet is normal, return to the step of collecting working data of the power distribution cabinet to be monitored based on multiple IoT devices until the initial identification result is that the power distribution cabinet is suspected of being abnormal.

[0105] It is clear that if the initial identification result shows that there is no abnormality in the power distribution cabinet, then it is not necessary to upload the multiple original working time sequence data at this time to the operation and maintenance monitoring platform, and the working data collection of the power distribution cabinet to be monitored can continue.

[0106] S5. If the initial identification result is that the power distribution cabinet is suspected to be abnormal, then the node security authentication of the power distribution cabinet to be monitored is performed to obtain the certified IoT device set, which includes multiple certified IoT devices.

[0107] Understandably, the certified IoT device set refers to a collection of multiple certified IoT devices, wherein certified IoT devices refer to IoT devices that have undergone node security authentication and possess high security.

[0108] Specifically, the node security authentication of the power distribution cabinet to be monitored, resulting in a set of authenticated IoT devices, includes:

[0109] For each IoT device in the monitored power distribution cabinet, perform the following operations:

[0110] Obtain the current behavior data of the IoT device, wherein the current behavior data includes: data packet reception rate, data packet transmission rate, and historical trust value;

[0111] The trust value of IoT devices is calculated based on current behavioral data to obtain the current trust value.

[0112] If the current trust value is greater than the preset standard trust value, a multi-dimensional authentication request is generated based on the IoT device. The multi-dimensional authentication request includes: node ID, request timestamp, and node MAC address.

[0113] The multidimensional authentication request is used to perform security authentication using a pre-built authentication server to obtain an authentication result, wherein the authentication result is authentication passed or authentication failed.

[0114] If the authentication result is successful, the IoT device will be marked as authenticated, thus obtaining an authenticated IoT device.

[0115] The certified IoT devices are aggregated to obtain a set of certified IoT devices.

[0116] It should be explained that the current behavior data refers to data representing the data communication status of the IoT device within a recent period. The data packet reception rate refers to the number of data packets successfully received by the IoT device per unit time, the data packet transmission rate refers to the number of data packets actively sent by the IoT device per unit time, and the historical trust value refers to the most recently calculated current trust value at the most recent moment. The current trust value represents the degree of trustworthiness of the IoT device at the current moment. The higher the current trust value, the more trustworthy the IoT device's behavior, and the lower the probability that it is a malicious or faulty node. The current trust value is calculated as follows: ,in, Indicates the current trust value. Indicates historical trust value, Indicates the data packet receiving rate. This represents the data packet sending rate. A higher data packet receiving rate or sending rate indicates more active network communication between the IoT device and other IoT devices or the operation and maintenance monitoring platform, meaning a higher current trust value. The standard trust value refers to a threshold used to determine the trustworthiness of an IoT device. This standard trust value is set by obtaining the current trust values ​​of all IoT devices to obtain the current trust value set, which is then expressed as: ,in, Indicates the standard trust value. and These represent the average and standard deviation of the current trust value set, respectively. The multidimensional authentication request refers to the data packet used to request identity legitimacy from the authentication server, where the node ID refers to the ID corresponding to the IoT device, the request timestamp refers to the timestamp at which the multidimensional authentication request was generated, and the node MAC address refers to the physical address of the IoT device's network interface.

[0117] Furthermore, the authentication server refers to a dedicated server located on the operation and maintenance monitoring platform, responsible for executing device authentication and authorization decisions, such as a server based on RADIUS or TLS protocols. The authentication result refers to the result obtained after security authentication, where authentication success means that the authentication server has verified the legality of the multidimensional authentication request and generated the corresponding digital signature, and authentication failure means that the verification failed. The specific process of the above security authentication is as follows: After receiving the multidimensional authentication request, the authentication server first verifies whether the request timestamp in the multidimensional authentication request is within a reasonable time window, thereby ensuring the timeliness of the multidimensional authentication request. Then, it verifies whether the binding relationship between the node ID and the node MAC address in the multidimensional authentication request is already in the list of registered IoT devices, thereby preventing unauthorized devices from impersonating and accessing the device. When all the above verifications pass, it is considered that the authentication is successful; otherwise, it is considered that the authentication is unsuccessful. The authentication server digitally signs the multidimensional authentication request and returns the digital signature to the corresponding IoT device, thus completing the authentication of the IoT device. The authenticated IoT device refers to the IoT device that has undergone security authentication, where the authentication mark refers to the digital signature returned by the authentication server to the IoT device.

[0118] S6. Remote anomaly identification is performed using a set of certified IoT devices, a pre-built operation and maintenance monitoring platform, and multiple original working time-series data to obtain abnormal values ​​of the power distribution cabinet.

[0119] It is clear that the aforementioned operation and maintenance monitoring platform refers to a centralized monitoring system located in a remote data center or cloud, integrating secure communication, data decryption, and intelligent analysis capabilities. The aforementioned abnormal value of the power distribution cabinet refers to the probability value identified by the operation and maintenance monitoring platform, indicating that the power distribution cabinet under monitoring is in an abnormal state.

[0120] In detail, the method of remotely identifying anomalies using a set of certified IoT devices, a pre-built operation and maintenance monitoring platform, and multiple sets of raw operational time-series data to obtain abnormal values ​​for the power distribution cabinet includes:

[0121] Multiple raw working time-series data are encrypted using modular exponentiation based on a preset data encryption public key to obtain multiple IoT data to be transmitted.

[0122] Based on the certified IoT device set, multiple IoT data to be transmitted are transmitted to the operation and maintenance monitoring platform to obtain multiple IoT data to be decrypted.

[0123] An anomaly detection model for power distribution cabinets in the operation and maintenance monitoring platform is identified. The power distribution cabinet anomaly detection model includes: a data input layer, a local feature convolutional layer, a temporal feature extraction layer, an attention feature enhancement layer, a variational autoencoder, and a data output layer.

[0124] An anomaly detection model for power distribution cabinets and multiple undecrypted IoT data are used to detect anomalies in power distribution cabinets and obtain abnormal values ​​for the power distribution cabinets.

[0125] It should be explained that the data encryption public key refers to the public key used to encrypt the original working time-series data. The IoT data to be transmitted refers to the original working time-series data encrypted by modular exponentiation. The specific calculation method of modular exponentiation encryption is existing technology and will not be elaborated here. The multiple IoT data to be decrypted refers to multiple IoT data to be transmitted to the operation and maintenance monitoring platform. The power distribution cabinet anomaly detection model refers to the mathematical model for anomaly identification of the power distribution cabinet to be monitored. The data input layer refers to the first layer of the power distribution cabinet anomaly detection model. This data input layer can perform preprocessing such as normalization, interpolation, and time alignment on the input data (such as subsequent fused input data). The local feature convolutional layer refers to the network layer used to extract local correlations and spatial features. This local feature convolutional layer can receive the data output from the data input layer and use a one-dimensional convolutional network to perform local convolution on the data. The temporal feature extraction layer refers to the network layer used to model the long-term dependencies and temporal dynamics in the input data. Optionally, this temporal feature extraction layer is a Long Short-Term Memory (LSTM) network. The attention feature enhancement layer refers to a network layer that can adaptively assign different weights to different time points or different feature channels. Optionally, this attention feature enhancement layer is a self-attention mechanism or a channel attention (SENet) module. The variational autoencoder refers to a generative model containing an encoder and a decoder, used to receive the data output by the attention feature enhancement layer (denoted as the original input data), compress and encode the original input data to learn its latent distribution, reconstruct the original input data based on the learned latent distribution, and then compare the difference between the reconstructed data and the original input data. This difference is the reconstruction error. The larger the reconstruction error, the more likely the original input data at the corresponding time point is to be abnormal.

[0126] Furthermore, the various model parameters in the aforementioned distribution cabinet anomaly detection model can be determined through unsupervised training. Specifically, the following method is used: First, the time-series data of the distribution cabinet under monitoring during its historical normal operation is obtained (referred to as normal time-series data). During the training phase, the distribution cabinet anomaly detection model attempts to encode and reconstruct these normal time-series data and calculates the reconstruction error after encoding and reconstruction. By minimizing this reconstruction error, the various model parameters are optimized. After the distribution cabinet anomaly detection model is trained, when new time-series data (such as subsequent fused input data) is input during the application phase, the magnitude of the reconstruction error output by the distribution cabinet anomaly detection model can be used as a quantitative basis for judging whether the input time-series data is abnormal. The larger the reconstruction error, the further the input time-series data deviates from the normal mode, and the higher the probability of anomaly. The reconstruction error output by the distribution cabinet anomaly detection model is recorded as the distribution cabinet anomaly value.

[0127] In detail, the certified IoT device set transmits multiple IoT data to be transmitted to the operation and maintenance monitoring platform, obtaining multiple IoT data to be decrypted, including:

[0128] Obtain the location of the certified IoT device for each certified IoT device in the certified IoT device set to obtain the certified device location set;

[0129] Identify multiple IoT devices to be relayed in the power distribution cabinet to be monitored based on the certified IoT device set;

[0130] Based on the location set of the certified device, multiple IoT devices to be relayed are assigned relay clusters to obtain multiple IoT device sets in the same cluster. Each IoT device set in the same cluster includes multiple IoT devices in the same cluster, and the location of each IoT device in the same cluster corresponds one-to-one with the location of the certified device.

[0131] Multiple IoT devices in the same cluster and an operation and maintenance monitoring platform are used to remotely transmit multiple IoT data to be transmitted, resulting in multiple IoT data to be decrypted.

[0132] It should be explained that the "certified device location set" refers to a collection of multiple certified device locations, where each certified device location refers to the actual location of a specific certified IoT device within the certified IoT device set. The "IoT devices to be relayed" refers to IoT devices in the monitored distribution cabinet other than those in the certified IoT device set. Because these devices have limitations in communication module power consumption, poor signal coverage, or have not been granted security credentials for direct communication with the remote platform, their IoT data needs to be transmitted to a certified IoT device. The certified IoT device then transmits the received IoT data and its own IoT data to the operation and maintenance monitoring platform. The "clustered IoT device set" refers to a collection of multiple IoT devices obtained after relay cluster allocation, where each clustered IoT device set contains a certified IoT device corresponding to a certified device location.

[0133] Furthermore, the above method of allocating relay clusters for multiple IoT devices to be relayed based on the location set of certified devices is as follows: The number of certified device locations in the location set is determined and used as the clustering number. The relay location of each IoT device to be relayed is obtained, resulting in multiple relay locations. These relay locations refer to the actual locations of the IoT devices to be relayed. Then, based on the spatial distance between these relay locations and the locations of each certified device, the K-means clustering algorithm is used to divide all IoT devices to be relayed into the cluster containing the nearest certified IoT device. After clustering, multiple sets of IoT devices in the same cluster are obtained. The above method of remotely transmitting multiple IoT data to be transmitted using multiple sets of IoT devices in the same cluster and an operation and maintenance monitoring platform refers to the following: For each set of IoT devices in the same cluster: First, all IoT devices to be relayed within the set transmit their own IoT data to the corresponding certified IoT device. Then, the certified IoT device uploads all the IoT data to be transmitted to the operation and maintenance monitoring platform, completing the data upload for the set of IoT devices in the same cluster. Once all sets of IoT devices in the same cluster have completed the data upload, the remote transmission step is complete.

[0134] In detail, the method of using a power distribution cabinet anomaly detection model and multiple undecrypted IoT data to perform power distribution cabinet anomaly detection and obtain power distribution cabinet anomaly values ​​includes:

[0135] The target encrypted private key is obtained by reconstructing the private key based on multiple pre-defined fragmented IoT devices, where each fragmented IoT device contains a fragment value.

[0136] Multiple IoT data to be decrypted are decrypted using the target encryption private key to obtain multiple IoT input data;

[0137] Multiple IoT input data are merged to obtain fused input data;

[0138] The fused input data is fed into the distribution cabinet anomaly detection model to obtain the distribution cabinet anomaly value.

[0139] It should be explained that the "sharded IoT device" refers to an IoT device containing shard values. The method for obtaining this sharded IoT device will be given in subsequent embodiments. This sharded IoT device is used to securely store shard values ​​and provides them to participate in reconstructing the complete target encrypted private key when data decryption is required. The shard value refers to a partial value representing the complete target encrypted private key, obtained through subsequent sharding polynomial calculations. The target encrypted private key refers to the private key obtained after reconstruction. The multiple IoT input data refer to multiple IoT data to be transmitted after decryption. The decryption steps are existing technology and will not be described further here. The fused input data refers to the data set of multiple IoT input data after data merging.

[0140] Specifically, before reconstructing the private key based on multiple pre-defined segmented IoT devices to obtain the target encrypted private key, the method further includes:

[0141] Based on a preset number of shards, select multiple trusted IoT devices from multiple IoT devices, wherein the number of trusted IoT devices is the same as the number of shards.

[0142] A random encrypted prime number is generated based on the preset data encryption private key, wherein the random encrypted prime number is greater than the data encryption private key;

[0143] Extract trusted IoT devices sequentially from multiple trusted IoT devices, and record the extracted trusted IoT devices as IoT devices to be fragmented;

[0144] Obtain the arrangement identifier of the IoT device to be fragmented among multiple trusted IoT devices;

[0145] A slice polynomial is generated based on a preset threshold value, a data encryption private key, and randomly encrypted prime numbers. The slice polynomial is: polynomial of degree. Indicates the threshold value;

[0146] The data is fragmented using permutation identifiers and fragmentation polynomials to obtain fragmentation values. Based on these fragmentation values, the fragmented IoT devices are allocated to obtain the fragmented IoT devices.

[0147] By aggregating the fragmented IoT devices, multiple fragmented IoT devices are obtained.

[0148] It should be explained that the number of shards refers to the total number of shard values ​​to be generated. This number of shards can be set according to the scale of IoT devices, for example, 5% of the total number of IoT devices. The trusted IoT device refers to the IoT device selected before monitoring the distribution cabinet to be monitored. This trusted IoT device has a high degree of trustworthiness and can be used as a candidate for sharding IoT devices. The selection method for this trusted IoT device is: obtaining the average trust value of each IoT device in the previous monitoring process, and selecting the one with the largest value. The average confidence value corresponds to One IoT device acts as multiple trusted IoT devices, among which... This indicates the number of fragments. The data encryption private key refers to the private key corresponding to the aforementioned data encryption public key. To ensure that the data encryption private key is not at risk of being stolen or destroyed due to storage on a single node, it needs to be split into multiple fragments through a secret sharing scheme and distributed across multiple trusted IoT devices. When multiple IoT data to be transmitted need to be decrypted, a sufficient number of fragments must be collected from the multiple fragmented IoT devices to reconstruct the complete data encryption private key. The sufficient number is the subsequent threshold value. The random encryption prime number refers to a random prime number greater than the data encryption private key. For example, if the data encryption private key is 1234, then the random encryption prime number can be 1999. The permutation identifier refers to the permutation identifier of the IoT devices to be fragmented among the multiple trusted IoT devices. The threshold value refers to the minimum number of trusted IoT devices required to reconstruct the target encryption private key. This threshold value is less than the number of fragments but greater than half the number of fragments. It can be set according to the number of fragments. For example, if the number of fragments is 5, then the threshold value can be set to 3.

[0149] Furthermore, the piecewise polynomial refers to the polynomial used to generate the piecewise values, and the piecewise polynomial is expressed as: ,in, Represents a piecewise polynomial. This represents the private key used for data encryption. , … Both indicate that the interval is within the range Integer coefficients are randomly selected within the range, for example: if Then select randomly. , Placeholders indicating permutation identifiers, used to substitute subsequent permutation identifiers. This represents a randomly encrypted prime number. The above sharding using permutation identifiers and sharding polynomials refers to substituting the permutation identifier into the placeholder in the sharding polynomial; the output is the sharding value of the trusted IoT device corresponding to that permutation identifier. Upon receiving a private key reconstruction request, the sharded IoT device will provide its own permutation identifier and sharding value to participate in the private key reconstruction.

[0150] In detail, the process of reconstructing the private key based on multiple pre-defined segmented IoT devices to obtain the target encrypted private key includes:

[0151] Based on the threshold value, multiple responding IoT devices are randomly selected from multiple segmented IoT devices to obtain multiple responding IoT devices, wherein the number of responding IoT devices is the same as the threshold value;

[0152] The target encrypted private key is calculated using multiple responding IoT devices and the aforementioned random encrypted prime number, wherein the formula for calculating the target encrypted private key is:

[0153] ;

[0154] in, Indicates the target encryption private key. Indicates the index number. Indicates and Different index numbers, This indicates the number of responding IoT devices out of a total of multiple responding IoT devices. Indicates the first of multiple responding IoT devices Each response IoT device corresponds to a fragment value. Represents the modulo function. This represents a randomly encrypted prime number.

[0155] It is clear that the responding IoT device refers to a segmented IoT device randomly selected from multiple segmented IoT devices that participates in the reconstruction of the target encrypted private key.

[0156] To address the problems described in the background art, this invention first performs preliminary anomaly pattern identification based on multiple raw working time-series data to obtain initial identification results. This step involves identifying single data anomalies locally on each IoT device, and only determining a suspected anomaly in the power distribution cabinet when the total number of aggregated anomaly commands is greater than one. Compared to the existing technology where an alarm or upload is triggered by a single sensor data anomaly, this method effectively avoids false alarms caused by the failure of a single IoT device, reducing unnecessary communication resource consumption and platform computing load caused by false alarms. If the initial identification result indicates a suspected anomaly in the power distribution cabinet, node security authentication is performed on the power distribution cabinet to be monitored, resulting in an authenticated set of IoT devices. This step, after triggering a suspected anomaly, does not directly upload the data of all IoT devices, but first performs multi-layer node security authentication including dynamic trust assessment and static identity verification. Compared to the existing technology that typically only performs simple identity authentication or completely trusts all nodes, this method can... This invention identifies and isolates malicious nodes exhibiting abnormal behavior or potentially tampered with, ensuring the high credibility of the node set participating in subsequent data aggregation and transmission. After authentication, data is encrypted using modular exponentiation to guarantee transmission confidentiality. Through spatial location-based clustering, a multi-hop relay transmission network with authenticated IoT devices as cluster heads is formed, optimizing the network topology and resolving the issue of edge devices being unable to directly connect to remote platforms due to signal coverage or power consumption limitations, thus improving data transmission success rate. Furthermore, a secret sharing technology is employed to store the decryption private key in fragments across multiple trusted devices. The number of fragments required for private key reconstruction reaches a threshold, avoiding the single-point leakage risk of centralized private key storage and enhancing key security. Finally, on the operation and maintenance monitoring platform, a dedicated model integrating convolution, temporal feature extraction, attention mechanisms, and variational autoencoders is used to perform deep analysis and reconstruction error calculation on the decrypted multi-source time-series data, demonstrating higher accuracy and adaptability compared to traditional anomaly detection methods based on fixed thresholds. Therefore, this invention can improve the accuracy and security of power distribution cabinet monitoring, and reduce false alarm rates and resource consumption.

[0157] like Figure 2 The diagram shown is a functional block diagram of a remote operation and maintenance monitoring system for power distribution cabinets based on the Internet of Things, provided in an embodiment of the present invention.

[0158] The IoT-based remote operation and maintenance monitoring system 100 for power distribution cabinets described in this invention can be installed in electronic devices. Depending on the functions implemented, the IoT-based remote operation and maintenance monitoring system 100 may include an operation and maintenance instruction receiving module 101, a work data acquisition module 102, an anomaly pattern recognition module 103, and a remote data transmission module 104. The module described in this invention can also be called a unit, which refers to a series of computer program segments that can be executed by the processor of an electronic device and can perform a fixed function, and which are stored in the memory of the electronic device.

[0159] The operation and maintenance instruction receiving module 101 is used to receive the operation and maintenance instructions of the power distribution cabinet and identify the power distribution cabinet to be monitored based on the operation and maintenance instructions. The power distribution cabinet to be monitored includes multiple Internet of Things (IoT) devices, and each IoT device includes an abnormal pattern recognition unit.

[0160] The working data acquisition module 102 is used to acquire working data of the power distribution cabinet to be monitored based on multiple Internet of Things (IoT) devices to obtain multiple raw working time series data, wherein the raw working time series data corresponds one-to-one with the IoT devices.

[0161] The anomaly pattern recognition module 103 is used to perform preliminary anomaly pattern recognition based on multiple original working time series data to obtain an initial recognition result. The initial recognition result is either that the power distribution cabinet is suspected of being abnormal or that the power distribution cabinet is not abnormal. If the initial recognition result is that the power distribution cabinet is not abnormal, the step of collecting working data of the power distribution cabinet to be monitored based on multiple IoT devices is returned until the initial recognition result is that the power distribution cabinet is suspected of being abnormal. If the initial recognition result is that the power distribution cabinet is suspected of being abnormal, node security authentication is performed on the power distribution cabinet to be monitored to obtain a set of certified IoT devices. The set of certified IoT devices includes multiple certified IoT devices.

[0162] The remote data transmission module 104 is used to remotely identify anomalies by utilizing a set of certified IoT devices, a pre-built operation and maintenance monitoring platform, and multiple original working time-series data to obtain abnormal values ​​of the power distribution cabinet.

[0163] In detail, the modules in the IoT-based remote operation and maintenance monitoring system 100 for power distribution cabinets described in this embodiment of the invention employ the same methods as described above. Figure 1 The method used is the same as the IoT-based remote operation and maintenance monitoring method for power distribution cabinets described above, and it can produce the same technical effects, so it will not be repeated here.

[0164] like Figure 3 The diagram shown is a structural schematic of an electronic device for implementing a remote operation and maintenance monitoring method for power distribution cabinets based on the Internet of Things, according to an embodiment of the present invention.

[0165] The electronic device 1 may include a processor 10, a memory 11 and a bus 12, and may also include a computer program stored in the memory 11 and capable of running on the processor 10, such as a remote operation and maintenance monitoring method program for power distribution cabinets based on the Internet of Things.

[0166] The memory 11 includes at least one type of readable storage medium, such as flash memory, portable hard drive, multimedia card, card-type memory (e.g., SD or DX memory), magnetic memory, magnetic disk, optical disk, etc. In some embodiments, the memory 11 can be an internal storage unit of the electronic device 1, such as the portable hard drive of the electronic device 1. In other embodiments, the memory 11 can be an external storage device of the electronic device 1, such as a plug-in portable hard drive, smart media card (SMC), secure digital card (SD), flash card, etc., equipped on the electronic device 1. Furthermore, the memory 11 includes both internal storage units and external storage devices of the electronic device 1. The memory 11 can be used not only to store application software and various types of data installed on the electronic device 1, such as the code of a remote operation and maintenance monitoring method program for a power distribution cabinet based on the Internet of Things, but also to temporarily store data that has been output or will be output.

[0167] In some embodiments, the processor 10 may be composed of integrated circuits, such as a single packaged integrated circuit or multiple integrated circuits with the same or different functions, including combinations of one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and various control chips. The processor 10 is the control unit of the electronic device, connecting various components of the entire electronic device through various interfaces and lines. It executes programs or modules stored in the memory 11 (e.g., a remote operation and maintenance monitoring method program for power distribution cabinets based on the Internet of Things), and calls data stored in the memory 11 to perform various functions of the electronic device 1 and process data.

[0168] The bus 12 can be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus 12 can be divided into an address bus, a data bus, a control bus, etc. The bus 12 is configured to realize the connection and communication between the memory 11 and at least one processor 10, etc.

[0169] Figure 3 Only electronic devices with components are shown; it will be understood by those skilled in the art that... Figure 3The structure shown does not constitute a limitation on the electronic device 1, and may include fewer or more components than shown, or combine certain components, or have different component arrangements.

[0170] For example, although not shown, the electronic device 1 may also include a power supply (such as a battery) to power the various components. Preferably, the power supply can be logically connected to the at least one processor 10 through a power management device, thereby enabling functions such as charging management, discharging management, and power consumption management. The power supply may also include one or more DC or AC power supplies, recharging devices, power fault detection circuits, power converters or inverters, power status indicators, and other arbitrary components. The electronic device 1 may also include various sensors, Bluetooth modules, Wi-Fi modules, etc., which will not be described in detail here.

[0171] Furthermore, the electronic device 1 may also include a network interface. Optionally, the network interface may include a wired interface and / or a wireless interface (such as a Wi-Fi interface, a Bluetooth interface, etc.), which is typically used to establish communication connections between the electronic device 1 and other electronic devices.

[0172] Optionally, the electronic device 1 may further include a user interface, which may be a display, an input unit (such as a keyboard), and optionally, a standard wired interface or a wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, or an OLED (Organic Light-Emitting Diode) touchscreen, etc. The display may also be appropriately referred to as a screen or display unit, used to display information processed in the electronic device 1 and to display a visual user interface.

[0173] The IoT-based remote operation and maintenance monitoring method program for power distribution cabinets stored in the memory 11 of the electronic device 1 is a combination of multiple instructions. When run in the processor 10, it can achieve the following:

[0174] Receive the operation and maintenance instructions for the power distribution cabinet, and identify the power distribution cabinet to be monitored based on the operation and maintenance instructions. The power distribution cabinet to be monitored includes multiple IoT devices, and each IoT device contains an abnormal pattern recognition unit.

[0175] Based on the collection of working data from multiple IoT devices on the power distribution cabinet to be monitored, multiple raw working time series data are obtained, in which each raw working time series data corresponds one-to-one with an IoT device.

[0176] Anomaly patterns were initially identified based on multiple raw working time series data to obtain initial identification results, in which the initial identification results indicated that the power distribution cabinet was suspected of being abnormal or that the power distribution cabinet was not abnormal.

[0177] If the initial identification result is that the power distribution cabinet is normal, then return to the step of collecting working data of the power distribution cabinet to be monitored based on multiple IoT devices, until the initial identification result is that the power distribution cabinet is suspected of being abnormal.

[0178] If the initial identification result is that the power distribution cabinet is suspected to be abnormal, then the node security authentication of the power distribution cabinet to be monitored is performed to obtain the certified IoT device set, which includes multiple certified IoT devices.

[0179] Remote anomaly identification is performed using a set of certified IoT devices, a pre-built operation and maintenance monitoring platform, and multiple raw working time-series data to obtain abnormal values ​​for the power distribution cabinet.

[0180] Specifically, the processor 10's implementation method for the above instructions can be found in [reference needed]. Figures 1 to 3 The descriptions of the relevant steps in the corresponding embodiments are not repeated here.

[0181] Furthermore, if the modules / units integrated in the electronic device 1 are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. The computer-readable storage medium can be volatile or non-volatile. For example, the computer-readable medium may include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, or a read-only memory (ROM).

[0182] The present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor of an electronic device, can perform the following:

[0183] Receive the operation and maintenance instructions for the power distribution cabinet, and identify the power distribution cabinet to be monitored based on the operation and maintenance instructions. The power distribution cabinet to be monitored includes multiple IoT devices, and each IoT device contains an abnormal pattern recognition unit.

[0184] Based on the collection of working data from multiple IoT devices on the power distribution cabinet to be monitored, multiple raw working time series data are obtained, in which each raw working time series data corresponds one-to-one with an IoT device.

[0185] Anomaly patterns were initially identified based on multiple raw working time series data to obtain initial identification results, in which the initial identification results indicated that the power distribution cabinet was suspected of being abnormal or that the power distribution cabinet was not abnormal.

[0186] If the initial identification result is that the power distribution cabinet is normal, then return to the step of collecting working data of the power distribution cabinet to be monitored based on multiple IoT devices, until the initial identification result is that the power distribution cabinet is suspected of being abnormal.

[0187] If the initial identification result is that the power distribution cabinet is suspected to be abnormal, then the node security authentication of the power distribution cabinet to be monitored is performed to obtain the certified IoT device set, which includes multiple certified IoT devices.

[0188] Remote anomaly identification is performed using a set of certified IoT devices, a pre-built operation and maintenance monitoring platform, and multiple raw working time-series data to obtain abnormal values ​​for the power distribution cabinet.

[0189] In the embodiments provided by this invention, it should be understood that the disclosed devices, systems, and methods can be implemented in other ways. For example, the system embodiments described above are merely illustrative, and actual implementations may have other classification methods.

[0190] 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.

[0191] 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.

[0192] 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.

[0193] 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 method for remote operation and maintenance monitoring of power distribution cabinets based on the Internet of Things, characterized in that, The method includes: Receive the operation and maintenance instructions for the power distribution cabinet, and identify the power distribution cabinet to be monitored based on the operation and maintenance instructions. The power distribution cabinet to be monitored includes multiple IoT devices, and each IoT device contains an abnormal pattern recognition unit. Based on the collection of working data from multiple IoT devices on the power distribution cabinet to be monitored, multiple raw working time series data are obtained, in which each raw working time series data corresponds one-to-one with an IoT device. Anomaly patterns were initially identified based on multiple raw working time series data to obtain initial identification results, in which the initial identification results indicated that the power distribution cabinet was suspected of being abnormal or that the power distribution cabinet was not abnormal. If the initial identification result is that the power distribution cabinet is normal, then return to the step of collecting working data of the power distribution cabinet to be monitored based on multiple IoT devices, until the initial identification result is that the power distribution cabinet is suspected of being abnormal. If the initial identification result is that the power distribution cabinet is suspected to be abnormal, then the node security authentication of the power distribution cabinet to be monitored is performed to obtain the certified IoT device set, which includes multiple certified IoT devices. Remote anomaly identification is performed using a set of certified IoT devices, a pre-built operation and maintenance monitoring platform, and multiple raw working time-series data to obtain abnormal values ​​for the power distribution cabinet.

2. The method for remote operation and maintenance monitoring of power distribution cabinets based on the Internet of Things as described in claim 1, characterized in that, The preliminary identification of abnormal patterns based on multiple original working time series data yields initial identification results, including: Perform the following operation on each of the multiple raw working time series data sets: Identify the original IoT device corresponding to the original working time sequence data; The abnormal pattern recognition unit in the original IoT device is used to identify single data anomalies in the original working time series data to obtain single data anomaly values. If a single data anomaly value exceeds the preset standard data anomaly value, a local data anomaly command will be generated. Summarize the local data anomaly commands to obtain the local data anomaly command set; Count the number of abnormal commands in the local data abnormal command set; If the number of abnormal commands is greater than the preset value of 1, then the suspected abnormality of the power distribution cabinet is recorded as the initial identification result; otherwise, the absence of abnormality in the power distribution cabinet is recorded as the initial identification result.

3. The method for remote operation and maintenance monitoring of power distribution cabinets based on the Internet of Things as described in claim 2, characterized in that, The step of using the abnormal pattern recognition unit in the original IoT device to perform single data anomaly identification on the original working time-series data to obtain a single data anomaly value includes: The storage operation time series dataset is extracted based on the abnormal pattern recognition unit. The storage operation time series dataset includes multiple storage operation time series data, and each storage operation time series data contains a normal behavior label. Storage time series data are extracted sequentially from the storage time series dataset. Time series features are extracted from the extracted storage time series data and the original time series data to obtain storage time series feature vectors and original time series feature vectors. Calculate the normal mode deviation value based on the stored time series feature vector and the original time series feature vector; Summarize the normal pattern deviation values ​​corresponding to each storage time-series data to obtain the normal pattern deviation value set; The mean is calculated based on the deviation set of normal patterns to obtain a single data outlier.

4. The method for remote operation and maintenance monitoring of power distribution cabinets based on the Internet of Things as described in claim 3, characterized in that, The node security authentication of the power distribution cabinet to be monitored yields a set of authenticated IoT devices, including: For each IoT device in the monitored power distribution cabinet, perform the following operations: Obtain the current behavior data of the IoT device, wherein the current behavior data includes: data packet reception rate, data packet transmission rate, and historical trust value; The trust value of IoT devices is calculated based on current behavioral data to obtain the current trust value. If the current trust value is greater than the preset standard trust value, a multi-dimensional authentication request is generated based on the IoT device. The multi-dimensional authentication request includes: node ID, request timestamp, and node MAC address. The multidimensional authentication request is used to perform security authentication using a pre-built authentication server to obtain an authentication result, wherein the authentication result is authentication passed or authentication failed. If the authentication result is successful, the IoT device will be marked as authenticated, thus obtaining an authenticated IoT device. The certified IoT devices are aggregated to obtain a set of certified IoT devices.

5. The method for remote operation and maintenance monitoring of power distribution cabinets based on the Internet of Things as described in claim 4, characterized in that, The method utilizes a set of certified IoT devices, a pre-built operation and maintenance monitoring platform, and multiple sets of raw operational time-series data for remote anomaly identification to obtain abnormal values ​​for the power distribution cabinet, including: Multiple raw working time-series data are encrypted using modular exponentiation based on a preset data encryption public key to obtain multiple IoT data to be transmitted. Based on the certified IoT device set, multiple IoT data to be transmitted are transmitted to the operation and maintenance monitoring platform to obtain multiple IoT data to be decrypted. An anomaly detection model for power distribution cabinets in the operation and maintenance monitoring platform is identified. The power distribution cabinet anomaly detection model includes: a data input layer, a local feature convolutional layer, a temporal feature extraction layer, an attention feature enhancement layer, a variational autoencoder, and a data output layer. An anomaly detection model for power distribution cabinets and multiple undecrypted IoT data are used to detect anomalies in power distribution cabinets and obtain abnormal values ​​for the power distribution cabinets.

6. The method for remote operation and maintenance monitoring of power distribution cabinets based on the Internet of Things as described in claim 5, characterized in that, The certified IoT device set transmits multiple IoT data to the operation and maintenance monitoring platform, resulting in multiple IoT data to be decrypted, including: Obtain the location of the certified IoT device for each certified IoT device in the certified IoT device set to obtain the certified device location set; Identify multiple IoT devices to be relayed in the power distribution cabinet to be monitored based on the certified IoT device set; Based on the location set of the certified device, multiple IoT devices to be relayed are assigned relay clusters to obtain multiple IoT device sets in the same cluster. Each IoT device set in the same cluster includes multiple IoT devices in the same cluster, and the location of each IoT device in the same cluster corresponds one-to-one with the location of the certified device. Multiple IoT devices in the same cluster and an operation and maintenance monitoring platform are used to remotely transmit multiple IoT data to be transmitted, resulting in multiple IoT data to be decrypted.

7. The method for remote operation and maintenance monitoring of power distribution cabinets based on the Internet of Things as described in claim 6, characterized in that, The method of using a power distribution cabinet anomaly detection model and multiple undecrypted IoT data to detect power distribution cabinet anomalies and obtain abnormal values ​​for the power distribution cabinet includes: The target encrypted private key is obtained by reconstructing the private key based on multiple pre-defined fragmented IoT devices, where each fragmented IoT device contains a fragment value. Multiple IoT data to be decrypted are decrypted using the target encryption private key to obtain multiple IoT input data; Multiple IoT input data are merged to obtain fused input data; The fused input data is fed into the distribution cabinet anomaly detection model to obtain the distribution cabinet anomaly value.

8. The method for remote operation and maintenance monitoring of power distribution cabinets based on the Internet of Things as described in claim 7, characterized in that, Before obtaining the target encrypted private key by reconstructing the private key based on multiple pre-defined segmented IoT devices, the method further includes: Based on a preset number of shards, select multiple trusted IoT devices from multiple IoT devices, wherein the number of trusted IoT devices is the same as the number of shards. A random encrypted prime number is generated based on the preset data encryption private key, wherein the random encrypted prime number is greater than the data encryption private key; Extract trusted IoT devices sequentially from multiple trusted IoT devices, and record the extracted trusted IoT devices as IoT devices to be fragmented; Obtain the arrangement identifier of the IoT device to be fragmented among multiple trusted IoT devices; A slice polynomial is generated based on a preset threshold value, a data encryption private key, and randomly encrypted prime numbers. The slice polynomial is: polynomial of degree. Indicates the threshold value; The data is fragmented using permutation identifiers and fragmentation polynomials to obtain fragmentation values. Based on these fragmentation values, the fragmented IoT devices are allocated to obtain the fragmented IoT devices. By aggregating the fragmented IoT devices, multiple fragmented IoT devices are obtained.

9. The method for remote operation and maintenance monitoring of power distribution cabinets based on the Internet of Things as described in claim 8, characterized in that, The process of reconstructing the private key based on multiple pre-defined segmented IoT devices to obtain the target encrypted private key includes: Based on the threshold value, multiple responding IoT devices are randomly selected from multiple segmented IoT devices to obtain multiple responding IoT devices, wherein the number of responding IoT devices is the same as the threshold value; The target encrypted private key is calculated using multiple responding IoT devices and the aforementioned random encrypted prime number, wherein the formula for calculating the target encrypted private key is: ； in, Indicates the target encryption private key. Indicates the index number. Indicates and Different index numbers, This indicates the number of responding IoT devices out of a total of multiple responding IoT devices. Indicates the first of multiple responding IoT devices Each response IoT device corresponds to a fragment value. Represents the modulo function. This represents a randomly encrypted prime number.

10. A remote operation and maintenance monitoring system for power distribution cabinets based on the Internet of Things, characterized in that, The system includes: The operation and maintenance instruction receiving module is used to receive operation and maintenance instructions for the power distribution cabinet and identify the power distribution cabinet to be monitored based on the operation and maintenance instructions. The power distribution cabinet to be monitored includes multiple IoT devices, and each IoT device contains an abnormal pattern recognition unit. The working data acquisition module is used to collect working data from the power distribution cabinet under monitoring based on multiple IoT devices, and obtain multiple raw working time series data. The raw working time series data corresponds one-to-one with the IoT devices. An anomaly pattern recognition module is used to perform preliminary anomaly pattern recognition based on multiple raw working time-series data to obtain an initial recognition result. The initial recognition result is either a suspected anomaly in the power distribution cabinet or no anomaly in the power distribution cabinet. If the initial recognition result is no anomaly in the power distribution cabinet, the module returns to the step of collecting working data from the power distribution cabinet to be monitored based on multiple IoT devices until the initial recognition result is a suspected anomaly in the power distribution cabinet. If the initial recognition result is a suspected anomaly in the power distribution cabinet, the module performs node security authentication on the power distribution cabinet to be monitored to obtain an authenticated IoT device set, which includes multiple authenticated IoT devices. The remote data transmission module is used to remotely identify anomalies by utilizing a set of certified IoT devices, a pre-built operation and maintenance monitoring platform, and multiple raw working time-series data to obtain abnormal values ​​of the power distribution cabinet.

Images