Load balancing calculation method and system for electric power internet of things

Abstract

The invention relates to the technical field of edge computing, in particular to a load balancing computing method and system for an electric power internet of things. The invention discloses a load balancing calculation system for an electric power internet of things. The load balancing calculation system comprises a data reissuing operation triggering module, a data reissuing operation module and an updating module. According to the invention, the data reissuing priority of the local acquisition data and the power monitoring terminal equipment is determined through the stability, predictability and proximity correlation of the local acquisition data, and the data reissuing operation is executed in a polling manner according to the data reissuing priority of the local acquisition data and the power monitoring terminal equipment. In this way, only one power monitoring terminal device executes the data reissuing operation in each scheduling period, the situation of network congestion caused by the fact that multiple devices execute data reissuing at the same time is avoided, and the load situation of the power Internet of Things is balanced.

Description

Technical Field

[0001] The present invention relates to the field of edge computing technology, and in particular to a load balancing calculation method and system for an electric power Internet of Things. Background Art

[0002] With the widespread application of the Power Internet of Things (PoT) in smart grids, smart meters, and environmental monitoring, an increasing number of terminal devices are being deployed at various monitoring points, generating massive amounts of data. These devices continuously collect power data and upload it to centralized processing systems for analysis and monitoring. However, in real-world applications, device network connectivity is often affected by various factors, such as signal interference, network congestion, and device failures, leading to data upload interruptions or loss. To ensure the integrity and accuracy of system data, a data retransmission mechanism must be implemented to re-upload these lost data segments. A major challenge with data retransmission is that when multiple devices need to retransmit data after a disconnection or signal restoration, upload requests from all devices simultaneously flood in, creating data retransmission congestion. This puts severe pressure on communication networks, easily leading to bandwidth contention and overload, increased data upload delays, and exacerbated packet loss. Especially when communication channels are limited, network resources cannot be effectively allocated, resulting in some devices failing to upload data in a timely manner. Furthermore, traditional retransmission mechanisms often fail to consider device retransmission priority. Devices may rely on the same transmission channel without clear priority control, resulting in the delay of important data upload, impacting the system's real-time monitoring and decision-making. Summary of the Invention

[0003] The present invention determines the data retransmission priority of local collected data and power monitoring terminal equipment through the stability, predictability and proximity correlation of local collected data, and performs data retransmission operations in a polling manner according to the data retransmission priority of local collected data and power monitoring terminal equipment, so that only one power monitoring terminal equipment performs data retransmission operations in each scheduling cycle, avoiding network congestion caused by multiple devices performing data retransmission at the same time, so as to balance the load of the power Internet of Things.

[0004] The present invention provides a load balancing calculation method for electric power Internet of Things, comprising: The edge gateway sends a data retransmission polling beacon to all power monitoring terminal devices in the communication domain at each interval scheduling period. When the power monitoring terminal device receives the data retransmission polling beacon, it performs the following operations: the power monitoring terminal device traverses the locally stored retransmission scheduling table, which includes retransmission intention request data arranged from large to small according to the device retransmission priority value. The retransmission intention request data includes a device ID and a device retransmission priority value, and the device retransmission priority value is determined by the data segment confidence map inside the power monitoring terminal device. The data segment confidence map includes the local collected data of all time periods that the power monitoring terminal device has not uploaded during the offline period and the corresponding data retransmission priority value, and the data retransmission priority value is determined based on the stability, predictability and proximity correlation of the local collected data. If the device ID corresponding to the retransmission intention request data at the first position in the retransmission scheduling table is the power monitoring terminal device itself, the power monitoring terminal device performs the data retransmission operation. If the device ID corresponding to the retransmission intention request data at the first position in the retransmission scheduling table is not the power monitoring terminal device itself, no operation is performed, and the terminal device waits for the next data retransmission polling beacon to be received. Whenever a data resending operation is completed, the data segment confidence map in the power monitoring terminal device that has performed the data resending operation will be updated, and the resending scheduling table will be updated.

[0005] As a preferred aspect, the data retransmission operation specifically includes: the power monitoring terminal device selects the local collected data with the largest data retransmission priority value in the data segment confidence map, records it as the target retransmission data, and combines the target retransmission data and the corresponding device ID and timestamp into a target retransmission data segment, and transmits the target retransmission data segment through the data retransmission channel to complete the data retransmission operation.

[0006] As a preferred aspect, determining the data retransmission priority value corresponding to the locally collected data based on the stability, predictability, and proximity correlation of the locally collected data specifically includes the following steps: The power monitoring terminal device corresponding to the local collected data of the data reissue priority value to be calculated is recorded as the target power monitoring terminal device; The first data set is composed of the locally collected data from all time periods that the target power monitoring terminal device did not upload during the offline period. The second data set is then selected from all the locally collected data from the target power monitoring terminal device during normal use using a sliding window of size C, where C is the same as the number of locally collected data in the first data set. The standard deviation of all the locally collected data in the first and second data sets is calculated and recorded as the stability coefficient. Taking the current local collected data as the end, a time series analysis set of collected data is selected from the historical records through a sliding window of window size N. The first N-1 local collected data in the time series analysis set are then fed into the corresponding time series prediction model for processing, and the predicted data is output. The absolute value of the difference between the predicted data and the current local collected data is used as the predictability coefficient; Obtaining a set of neighboring power monitoring terminal devices corresponding to the target power monitoring terminal device, the set of neighboring power monitoring terminal devices including power monitoring terminal devices that are in a neighboring relationship with the target power monitoring terminal device, obtaining, for each power monitoring terminal device in the set of neighboring power monitoring terminal devices, a third data set based on a timestamp corresponding to the locally collected data in the first data set, i.e., the timestamps of the locally collected data in the first data set and the third data set correspond one to one, calculating a correlation coefficient between the first data set and the third data set using a Pearson correlation coefficient, and taking the reciprocal of the average of all correlation coefficients as a neighboring correlation coefficient; A weighted sum operation is performed on the stability coefficient, the predictability coefficient, and the proximity correlation coefficient to obtain a data retransmission priority value corresponding to the locally collected data.

[0007] As a preferred aspect, the device reissue priority value is determined by the data segment confidence map inside the power monitoring terminal device, and a reissue scheduling table is constructed, which specifically includes the following steps: Select the average of the K lowest data resending priority values ​​in the data segment confidence map as the device resending priority value; Whenever the power monitoring terminal device is reconnected, the power monitoring terminal device will combine the device ID and the device reissue priority value into reissue intention request data, and broadcast the reissue intention request data to all power monitoring terminal devices in the communication domain. After receiving the reissue intention request data, the power monitoring terminal device will sort all the received reissue intention request data from the largest to the smallest according to the device reissue priority value to build a reissue scheduling table; After the data reissue operation is completed, the target reissue data and the corresponding data reissue priority value that have performed the data reissue operation in the data segment confidence map are deleted, and the device reissue priority value is recalculated. Whenever the device reissue priority value is recalculated, the reissue intention request data is updated, and the corresponding power monitoring terminal device broadcasts the updated reissue intention request data to all power monitoring terminal devices in the communication domain, and updates the reissue scheduling table.

[0008] As a preferred aspect, it also includes a synchronization operation on the reissue schedule, the specific contents are as follows: Each time the reissue schedule is constructed or updated, the reissue schedule version number is updated. It should be noted that each time the reissue schedule version number is updated, the update of the reissue schedule version number is represented by a numerical increase; Each power monitoring terminal device sends a reissue schedule and a corresponding reissue schedule version number to the other power monitoring terminal devices in the communication domain. Each power monitoring terminal device only stores the reissue schedule corresponding to the latest reissue schedule version number.

[0009] As a preferred aspect, it also includes the self-sensing data retransmission operation of the power monitoring terminal equipment, the specific content is as follows: the power monitoring terminal equipment at the first position in the retransmission scheduling table continuously monitors the status of the data retransmission channel. If the status of the monitored data retransmission channel is idle, the power monitoring terminal equipment performs the data retransmission operation.

[0010] As a preferred aspect, the time series prediction model is established based on the LSTM model, and each power monitoring terminal device will train the corresponding time series prediction model based on the local collected data in the historical records.

[0011] The present invention also provides a load balancing computing system for the electric power Internet of Things, comprising: The data reissue operation trigger module is used for the edge gateway to send data reissue polling beacons to all power monitoring terminal devices in the communication domain at each interval scheduling period. When the power monitoring terminal device receives the data reissue polling beacon, it performs the following operations: the power monitoring terminal device traverses the locally stored reissue scheduling table, which includes reissue intention request data arranged from large to small according to the device reissue priority value. The reissue intention request data includes the device ID and the device reissue priority value, and the device reissue priority value is determined by the data segment confidence map inside the power monitoring terminal device. The data segment confidence map includes Including the local collected data of all time periods that the power monitoring terminal device has not uploaded during the offline period and the corresponding data retransmission priority value, and the data retransmission priority value is determined based on the stability, predictability and proximity correlation of the local collected data. If the device ID corresponding to the retransmission intention request data at the first position in the retransmission schedule is the power monitoring terminal device itself, the power monitoring terminal device performs the data retransmission operation. If the device ID corresponding to the retransmission intention request data at the first position in the retransmission schedule is not the power monitoring terminal device itself, no operation is performed, and the terminal device waits for the next data retransmission polling beacon to be received; The data resending operation module is used to select the local collected data with the largest data resending priority value in the data segment confidence map of the force monitoring terminal device, record it as the target resending data, and combine the target resending data with the corresponding device ID and timestamp to form a target resending data segment, and transmit the target resending data segment through the data resending channel to complete the data resending operation; The updating module is used to update the data segment confidence map in the power monitoring terminal device that has performed the data reissue operation and to update the reissue scheduling table.

[0012] The present invention has the following advantages: The present invention determines the data retransmission priority of local collected data and power monitoring terminal equipment through the stability, predictability and proximity correlation of local collected data, and performs data retransmission operations in a polling manner according to the data retransmission priority of local collected data and power monitoring terminal equipment, so that only one power monitoring terminal equipment performs data retransmission operations in each scheduling cycle, avoiding network congestion caused by multiple devices performing data retransmission at the same time, so as to balance the load of the power Internet of Things. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 This is a schematic diagram of the structure of the load balancing computing system for the power Internet of Things adopted in an embodiment of the present invention. DETAILED DESCRIPTION

[0014] In order to enable those skilled in the art to better understand the technical solutions in the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0015] Example 1, a load balancing calculation method for the power Internet of Things, comprising: The edge gateway sends a data retransmission polling beacon to all power monitoring terminal devices in the communication domain at each interval scheduling period. The power monitoring terminal devices here include smart meters and distribution sensors, etc. The data retransmission polling beacon is used to remind the power monitoring terminal devices to start the next round of data retransmission process. When the power monitoring terminal device receives the data retransmission polling beacon, it performs the following operations: the power monitoring terminal device traverses the locally stored retransmission scheduling table, which includes retransmission intention request data arranged from large to small according to the device retransmission priority value. The retransmission intention request data includes the device ID and the device retransmission priority value. The device retransmission priority value here represents the urgency of the current power monitoring terminal device needing to retransmit data, and the device retransmission priority value is determined by the data segment confidence map inside the power monitoring terminal device. The data segment confidence map includes the local collected data of all time periods that the power monitoring terminal device has not uploaded during the offline period and the corresponding data retransmission priority value, and the data retransmission priority value is determined based on the stability, predictability and proximity correlation of the local collected data. The stability of the locally collected data refers to the degree of change of the locally collected data in all time periods that were not uploaded during the offline period. If the degree of change is small, it means that linear prediction can be performed and the urgency of data reissue is low. The predictability of the locally collected data refers to the accuracy of prediction based on historical data. If the prediction accuracy is high, it means that the master data station can also make predictions based on historical data and the urgency of data reissue is low. The proximity correlation of the locally collected data refers to the correlation between the data collected by the neighboring power monitoring terminal devices. If there are multiple highly correlated local collected data in the neighboring devices, it means that the correlation can be easily reconstructed at the master data station and the urgency of data reissue is low. If the device ID corresponding to the reissue intention request data at the first position in the reissue schedule is the power monitoring terminal device itself, the power monitoring terminal device performs the data reissue operation. If the device ID corresponding to the reissue intention request data at the first position in the reissue schedule is not the power monitoring terminal device itself, no operation is performed and the next data reissue polling beacon is waited for to be received. The data retransmission operation specifically includes: the power monitoring terminal device selects the locally collected data with the highest data retransmission priority value in the data segment confidence map, records it as the target retransmission data, and combines the target retransmission data with the corresponding device ID and timestamp to form a target retransmission data segment. The target retransmission data segment is transmitted through the data retransmission channel to complete the data retransmission operation. The timestamp is the collection time corresponding to the locally collected data. The data retransmission channel here is generally built based on NB-IoT. Whenever a data resending operation is completed, the data segment confidence map in the power monitoring terminal device that performed the data resending operation will be updated, and the resending schedule will be updated; It should be noted that during the period when the power monitoring terminal device is offline, the power monitoring terminal device will continue to collect data according to the pre-set collection frequency, and store the collected local data (such as voltage, current and power, etc.) locally on the power monitoring terminal device. For example, if a smart meter is offline from 08:00 to 08:15, according to the pre-set sampling once per minute, the smart meter will store 15 local collected data and the corresponding data re-issuance priority value locally. The data re-issuance priority value here represents the urgency of the corresponding local collected data re-issuance. The communication domain refers to the communication network environment composed of various power monitoring terminal devices (such as smart meters and distribution sensors), edge gateways, and relay nodes. The data retransmission polling beacon is a lightweight control frame broadcast by the gateway. It is generally a short packet of dozens of bytes. The data structure includes an identification header, the number of the current scheduling cycle, and a check bit. The identification header is used to identify the data retransmission polling beacon, and the check bit is used for CRC verification. This application determines the data retransmission priority of local collected data and power monitoring terminal devices through the stability, predictability and proximity correlation of local collected data, and performs data retransmission operations in a polling manner according to the data retransmission priority of local collected data and power monitoring terminal devices, so that only one power monitoring terminal device performs data retransmission operations in each scheduling cycle, avoiding network congestion caused by multiple devices performing data retransmission at the same time, so as to balance the load of the power Internet of Things.

[0016] The data re-issuance priority value corresponding to the locally collected data is determined based on the stability, predictability, and proximity correlation of the locally collected data, specifically including the following steps: The power monitoring terminal device corresponding to the local collected data of the data reissue priority value to be calculated is recorded as the target power monitoring terminal device; The local collected data of all time periods that the target power monitoring terminal device has not uploaded during the offline period are combined into a first data set, and then a second data set is selected from all the local collected data when the target power monitoring terminal device is in normal use through a sliding window with a window size of C. Here, C is consistent with the number of locally collected data in the first data set. It should be noted that the normal use of the target power monitoring terminal device refers to the situation where it is not offline. The target power monitoring terminal device will also perform data collection during normal use and store the collected data locally. The standard deviation of all locally collected data in the first data set and the second data set is calculated and recorded as the stability coefficient. The standard deviation can reflect the degree of change of the collected data. If the degree of change is small, then this collected data can be roughly estimated based on the linear model, so the urgency of data reissue is not high; Taking the current locally collected data as the end, a time series analysis set of collected data is selected from the historical records using a sliding window of window size N. The first N-1 locally collected data in the time series analysis set are then fed into the corresponding time series prediction model for processing, and the predicted data is output. The absolute value of the difference between the predicted data and the current locally collected data is used as the predictability coefficient to measure the accuracy of the prediction based on historical data. The time series prediction model is built based on the LSTM model. Each power monitoring terminal device trains the corresponding time series prediction model based on the local collected data in the historical records. Obtaining a set of neighboring power monitoring terminal devices corresponding to the target power monitoring terminal device, the set of neighboring power monitoring terminal devices including power monitoring terminal devices in a neighboring relationship with the target power monitoring terminal device. The neighboring relationship here generally refers to proximity in spatial position, and the corresponding neighboring relationship can be configured by an operator in the background. For each power monitoring terminal device in the set of neighboring power monitoring terminal devices, obtaining a third data set based on the timestamp corresponding to the locally collected data in the first data set, that is, the timestamps of the locally collected data in the first data set and the third data set correspond one to one, calculating the correlation coefficient between the first data set and the third data set using the Pearson correlation coefficient, and taking the reciprocal of the average value of all correlation coefficients as the neighboring correlation coefficient; Perform a weighted sum operation on the stability coefficient, the predictability coefficient, and the proximity correlation coefficient to obtain the data re-issuance priority value corresponding to the locally collected data. It should be noted that the weights used in the weighted summation process are set by the operator. The device reissue priority value is determined by the data segment confidence map inside the power monitoring terminal device, and a reissue scheduling table is constructed, which specifically includes the following steps: The average of the lowest K data re-send priority values ​​in the data segment confidence map is selected as the device re-send priority value. K is configured by the operator and is generally 3. Whenever the power monitoring terminal device is reconnected, the power monitoring terminal device will combine the device ID and the device reissue priority value into reissue intention request data, and broadcast the reissue intention request data to all power monitoring terminal devices in the communication domain. After receiving the reissue intention request data, the power monitoring terminal device will sort all the received reissue intention request data from the largest to the smallest according to the device reissue priority value to build a reissue scheduling table; When the data reissue operation is completed, the target reissue data and the corresponding data reissue priority value for which the data reissue operation has been performed in the data segment confidence map are deleted, and the device reissue priority value is recalculated. Whenever the device reissue priority value is recalculated, the reissue intention request data is updated, and the corresponding power monitoring terminal device broadcasts the updated reissue intention request data to all power monitoring terminal devices in the communication domain, and updates the reissue scheduling table. Specifically, the reissue intention request data is marked with a version number. Each time the reissue intention request data is updated, the version number is modified. When the reissue scheduling table is updated, the device ID is used as an index to ensure that the reissue intention request data is the latest version number. During the scheduling cycle, before the edge gateway sends the data resending polling beacon, in order to ensure the consistency of the resending schedule in all power monitoring terminal devices, it also includes the synchronization operation of the resending schedule. The specific contents are as follows: Each time the reissue schedule is constructed or updated, the reissue schedule version number is updated. It should be noted that each time the reissue schedule version number is updated, the update of the reissue schedule version number is represented by a numerical increase; Each power monitoring terminal device sends a reissue schedule and a corresponding reissue schedule version number to the other power monitoring terminal devices in the communication domain. Each power monitoring terminal device only stores the reissue schedule corresponding to the latest reissue schedule version number.

[0017] In actual use, the edge gateway may experience disconnection or failure, resulting in the inability to send data retransmission polling beacons. In order to ensure the normal retransmission of data, the self-sensing data retransmission operation of the power monitoring terminal equipment is also included. The specific contents are as follows: The power monitoring terminal device at the first position in the retransmission scheduling table continuously monitors the status of the data retransmission channel. If the status of the monitored data retransmission channel is idle, that is, no data retransmission operation is in progress, the power monitoring terminal device performs the data retransmission operation.

[0018] Example 2, a load balancing computing system for the power Internet of Things, see Figure 1 ,include: The data reissue operation trigger module is used for the power monitoring terminal device to traverse the locally stored reissue scheduling table. The reissue scheduling table includes reissue intention request data arranged from large to small according to the device reissue priority value. The reissue intention request data includes the device ID and the device reissue priority value. The device reissue priority value here represents the urgency of the current power monitoring terminal device in needing to reissue data, and the device reissue priority value is determined by the data segment confidence map inside the power monitoring terminal device. The data segment confidence map includes the local collected data of all time periods that the power monitoring terminal device has not uploaded during the offline period and the corresponding data reissue priority value, and the data reissue priority value is determined based on the stability, predictability and proximity correlation of the local collected data. The stability of the local collected data refers to the degree of change of the local collected data of all time periods that have not been uploaded during the offline period. If the degree of change is small, it means that linear The urgency of data reissue is low. The predictability of locally collected data refers to the accuracy of prediction based on historical data. If the prediction accuracy is high, it means that the master data station can also make predictions based on historical data, and the urgency of data reissue is low. The proximity correlation of locally collected data refers to the correlation between the data collected by the neighboring power monitoring terminal equipment. If there are multiple locally collected data with high correlation in the neighboring equipment, it means that the correlation can be easily reconstructed at the master data station, and the urgency of data reissue is low. If the device ID corresponding to the reissue intention request data at the first position in the reissue schedule is the power monitoring terminal equipment itself, the power monitoring terminal equipment performs the data reissue operation. If the device ID corresponding to the reissue intention request data at the first position in the reissue schedule is not the power monitoring terminal equipment itself, no operation is performed, and the terminal equipment waits for the next data reissue polling beacon to be received. The data retransmission operation module is used by the power monitoring terminal device to select the locally collected data with the highest data retransmission priority value in the data segment confidence map, record it as the target retransmission data, and combine the target retransmission data with the corresponding device ID and timestamp to form a target retransmission data segment. The target retransmission data segment is transmitted through the data retransmission channel to complete the data retransmission operation. The timestamp is the collection time corresponding to the locally collected data. The data retransmission channel here is generally built based on NB-IoT. The updating module is used to update the data segment confidence map in the power monitoring terminal device that has performed the data reissue operation and to update the reissue scheduling table.

[0019] It should be understood that those skilled in the art may make improvements or modifications based on the above description, and all such improvements and modifications shall fall within the scope of protection of the appended claims. Any portion of this specification not described in detail is prior art known to those skilled in the art.

Claims

1. A load balancing calculation method for power Internet of Things, characterized in that: include: The edge gateway sends a data retransmission polling beacon to all power monitoring terminal devices in the communication domain at each interval scheduling period. When the power monitoring terminal device receives the data retransmission polling beacon, it performs the following operations: the power monitoring terminal device traverses the locally stored retransmission scheduling table, which includes retransmission intention request data arranged from large to small according to the device retransmission priority value. The retransmission intention request data includes a device ID and a device retransmission priority value, and the device retransmission priority value is determined by the data segment confidence map inside the power monitoring terminal device. The data segment confidence map includes the local collected data of all time periods that the power monitoring terminal device has not uploaded during the offline period and the corresponding data retransmission priority value, and the data retransmission priority value is determined based on the stability, predictability and proximity correlation of the local collected data. If the device ID corresponding to the retransmission intention request data at the first position in the retransmission scheduling table is the power monitoring terminal device itself, the power monitoring terminal device performs the data retransmission operation. If the device ID corresponding to the retransmission intention request data at the first position in the retransmission scheduling table is not the power monitoring terminal device itself, no operation is performed, and the terminal device waits for the next data retransmission polling beacon to be received. Whenever a data resending operation is completed, the data segment confidence map in the power monitoring terminal device that has performed the data resending operation will be updated, and the resending scheduling table will be updated.

2. A load balancing calculation method for electric power Internet of Things according to claim 1, characterized in that: The data retransmission operation specifically includes: the power monitoring terminal device selects the local collected data with the largest data retransmission priority value in the data segment confidence map, records it as the target retransmission data, and combines the target retransmission data and the corresponding device ID and timestamp into a target retransmission data segment, and transmits the target retransmission data segment through the data retransmission channel to complete the data retransmission operation.

3. A load balancing calculation method for electric power Internet of Things according to claim 2, characterized in that: The data re-issuance priority value corresponding to the locally collected data is determined based on the stability, predictability, and proximity correlation of the locally collected data, specifically including the following steps: The power monitoring terminal device corresponding to the local collected data of the data reissue priority value to be calculated is recorded as the target power monitoring terminal device; The first data set is composed of the locally collected data from all time periods that the target power monitoring terminal device did not upload during the offline period. The second data set is then selected from all the locally collected data from the target power monitoring terminal device during normal use using a sliding window of size C, where C is the same as the number of locally collected data in the first data set. The standard deviation of all the locally collected data in the first and second data sets is calculated and recorded as the stability coefficient. Taking the current local collected data as the end, a time series analysis set of collected data is selected from the historical records through a sliding window of window size N. The first N-1 local collected data in the time series analysis set are then fed into the corresponding time series prediction model for processing, and the predicted data is output. The absolute value of the difference between the predicted data and the current local collected data is used as the predictability coefficient; Obtaining a set of neighboring power monitoring terminal devices corresponding to the target power monitoring terminal device, the set of neighboring power monitoring terminal devices including power monitoring terminal devices that are in a neighboring relationship with the target power monitoring terminal device, obtaining, for each power monitoring terminal device in the set of neighboring power monitoring terminal devices, a third data set based on a timestamp corresponding to the locally collected data in the first data set, i.e., the timestamps of the locally collected data in the first data set and the third data set correspond one to one, calculating a correlation coefficient between the first data set and the third data set using a Pearson correlation coefficient, and taking the reciprocal of the average of all correlation coefficients as a neighboring correlation coefficient; A weighted sum operation is performed on the stability coefficient, the predictability coefficient, and the proximity correlation coefficient to obtain a data retransmission priority value corresponding to the locally collected data.

4. A load balancing calculation method for electric power Internet of Things according to claim 3, characterized in that: The device reissue priority value is determined by the data segment confidence map inside the power monitoring terminal device, and a reissue scheduling table is constructed, which specifically includes the following steps: Select the average of the K lowest data resending priority values ​​in the data segment confidence map as the device resending priority value; Whenever the power monitoring terminal device is reconnected, the power monitoring terminal device will combine the device ID and the device reissue priority value into reissue intention request data, and broadcast the reissue intention request data to all power monitoring terminal devices in the communication domain. After receiving the reissue intention request data, the power monitoring terminal device will sort all the received reissue intention request data from the largest to the smallest according to the device reissue priority value to build a reissue scheduling table; After the data reissue operation is completed, the target reissue data and the corresponding data reissue priority value that have performed the data reissue operation in the data segment confidence map are deleted, and the device reissue priority value is recalculated. Whenever the device reissue priority value is recalculated, the reissue intention request data is updated, and the corresponding power monitoring terminal device broadcasts the updated reissue intention request data to all power monitoring terminal devices in the communication domain, and updates the reissue scheduling table.

5. A load balancing calculation method for electric power Internet of Things according to claim 4, characterized in that: It also includes the synchronization operation of the reissue schedule, the specific contents are as follows: Each time the reissue schedule is constructed or updated, the reissue schedule version number is updated for the reissue schedule; Each power monitoring terminal device sends a reissue schedule and a corresponding reissue schedule version number to the other power monitoring terminal devices in the communication domain. Each power monitoring terminal device only stores the reissue schedule corresponding to the latest reissue schedule version number.

6. A load balancing calculation method for electric power Internet of Things according to claim 5, characterized in that: It also includes the self-sensing data retransmission operation of the power monitoring terminal equipment. The specific contents are as follows: the power monitoring terminal equipment at the first position in the retransmission scheduling table continuously monitors the status of the data retransmission channel. If the status of the monitored data retransmission channel is idle, the power monitoring terminal equipment performs the data retransmission operation.

7. A load balancing calculation method for electric power Internet of Things according to claim 6, characterized in that: The time series prediction model is established based on the LSTM model. Each power monitoring terminal device will train the corresponding time series prediction model based on the local collected data in the historical records.

8. A load balancing computing system for electric power Internet of Things, characterized in that: The system applies a load balancing calculation method for the power Internet of Things according to any one of claims 1 to 7, including: The data reissue operation trigger module is used for the edge gateway to send data reissue polling beacons to all power monitoring terminal devices in the communication domain at each interval scheduling period. When the power monitoring terminal device receives the data reissue polling beacon, it performs the following operations: the power monitoring terminal device traverses the locally stored reissue scheduling table, which includes reissue intention request data arranged from large to small according to the device reissue priority value. The reissue intention request data includes the device ID and the device reissue priority value, and the device reissue priority value is determined by the data segment confidence map inside the power monitoring terminal device. The data segment confidence map includes Including the local collected data of all time periods that the power monitoring terminal device has not uploaded during the offline period and the corresponding data retransmission priority value, and the data retransmission priority value is determined based on the stability, predictability and proximity correlation of the local collected data. If the device ID corresponding to the retransmission intention request data at the first position in the retransmission schedule is the power monitoring terminal device itself, the power monitoring terminal device performs the data retransmission operation. If the device ID corresponding to the retransmission intention request data at the first position in the retransmission schedule is not the power monitoring terminal device itself, no operation is performed, and the terminal device waits for the next data retransmission polling beacon to be received; The data resending operation module is used to select the local collected data with the largest data resending priority value in the data segment confidence map of the force monitoring terminal device, record it as the target resending data, and combine the target resending data with the corresponding device ID and timestamp to form a target resending data segment, and transmit the target resending data segment through the data resending channel to complete the data resending operation; The updating module is used to update the data segment confidence map in the power monitoring terminal device that has performed the data reissue operation and to update the reissue scheduling table.

Images