An electric meter collection task intelligent scheduling method and system

Abstract

The application provides an electric meter collection task intelligent scheduling method and system, the method comprises the following steps: obtaining the maximum task number of each base station and the electric meter list; wherein the maximum task number is dynamically adjusted based on the collection success rate of each base station in each time interval and the maximum connection number corresponding to the collection success rate of each base station in the past 24 hours; according to the obtained maximum task number and the electric meter list, the electric meter collection task is assigned to each base station; based on the electric meter collection task, the collection is carried out, and the base station outputs the current collection task quantity and the collection task result in real time; based on the quantity of the current collection task and the execution result of the electric meter collection task, the electric meter collection task issued by the base station to the electric meter is adjusted. The application can dynamically adjust the number of electric meters concurrently communicated by a single base station, prevent the problem of channel quality decline caused by the congestion of the NB-IOT network due to a large number of tasks, and improve the data reading and collection success rate of the electric meter.

Description

Technical Field

[0001] This invention relates to an intelligent scheduling method and system for electricity meter data collection tasks, belonging to the fields of power metering automation and electricity consumption information collection technology. Background Technology

[0002] With the full launch of the electricity information collection system, the era of manual meter reading has been completely over for electricity data collection. The remote automatic meter reading method of the electricity information collection system has greatly improved meter reading efficiency.

[0003] As the construction of smart cities becomes increasingly important in government digitalization, smart cities require the Internet of Things (IoT) to be interconnected and a unified IoT network to be built. The development of information technology is also driving mobile communication to move from connecting people to connecting people and things, as well as things to things. The Internet of Things is an inevitable trend, and the connectivity requirements of IoT are very different from those of traditional wireless cellular networks.

[0004] As a fundamental infrastructure for smart city construction, the traditional remote automatic meter reading based on cellular mobile networks cannot meet the data collection and communication needs of smart grids. NB-IoT, an emerging technology in the IoT field, has emerged to meet the demands of smart IoT. Compared to traditional cellular networks, NB-IoT networks are particularly suitable for data collection from various IoT devices in smart city scenarios. Power data collection is an indispensable part of smart city construction.

[0005] The business intelligence of smart grids is built on rapid data acquisition and processing. Smart grid power data has high requirements in terms of data volume and acquisition frequency. In the process of power data acquisition based on NB-IoT network, when the power information acquisition system needs to actively collect data from the meter, such as for business needs like fee control, parameter setting, and firmware upgrade, it initiates a network connection to the corresponding NB-IoT power data acquisition device. The NB-IoT network has a limit on the number of downlink connection requests that a single base station can make at the same time. Too many concurrent requests will cause a large number of requests to fail to be established. The requests from the master station will be directly rejected or the requests will exceed the timeout time specified by the business, which will greatly limit the data reading efficiency and data reading success rate.

[0006] To achieve the advantages of low power consumption and wide coverage in NB-IoT networks, some network characteristics are sacrificed, such as reducing the bandwidth of a single base station and adopting a half-duplex communication mode. In smart power metering systems, due to the high coverage density of electricity meters, a single NB-IoT base station can connect to a large number of meters. For example, in a typical mid-rise residential community with 20 buildings, the number of electricity meters connected to a single base station can reach over 2,000. Considering more complex environments, the number of electricity meters connected to a single base station far exceeds this number. If all electricity meters communicate simultaneously, it will inevitably cause congestion in the NB-IoT network. Summary of the Invention

[0007] The purpose of this invention is to overcome the shortcomings of existing technologies and provide an intelligent scheduling method and system for electricity meter data collection tasks. This method dynamically adjusts the number of electricity meters communicating concurrently under a single base station, preventing channel quality degradation caused by NB-IoT network congestion due to a large number of tasks, and improving the success rate of electricity meter data reading and collection. To achieve the above objective, this invention employs the following technical solution:

[0008] In a first aspect, the present invention provides an intelligent scheduling method for electricity meter data collection tasks, comprising:

[0009] Obtain the maximum number of tasks and the electricity meter list for each base station; wherein, the maximum number of tasks is dynamically adjusted based on the data acquisition success rate of each base station in each time interval and the maximum number of connections corresponding to the data acquisition success rate of each base station in the past 24 hours;

[0010] The maximum number of tasks and the list of electricity meters are used to determine the electricity meter collection tasks to be assigned to each base station.

[0011] The system controls the base station to collect data according to the meter data collection tasks, and obtains the current number of data collection tasks and the results of the data collection tasks output by the base station in real time. Based on the current number of data collection tasks and the execution results of the meter data collection tasks, the system adjusts the meter data collection tasks sent by the base station to the meters.

[0012] In conjunction with the first aspect, the adjustment base station's electricity meter data collection task sent to the electricity meter further includes:

[0013] Based on the current number of data collection tasks at the base station, if the current number of data collection tasks at the base station has not reached the maximum number of tasks, a data collection task is sent to the electricity meter; if the current number of data collection tasks at the base station reaches the maximum number of tasks, the maximum number of tasks reached by the base station is recorded, the data collection task sending process is paused, and the system enters a waiting-to-wake-up state.

[0014] Based on the execution result of the current data collection task of the electricity meter, if the execution result is successful, continue to send data collection tasks to the electricity meter's subordinate units; if the execution result is unsuccessful, reduce the number of data collection tasks under the base station corresponding to the electricity meter. If the number of data collection tasks of the base station has reached the maximum number of tasks before reducing the number of data collection tasks, then wake up the data collection task distribution process and continue to send data collection tasks to the electricity meter's subordinate units.

[0015] In conjunction with the first aspect, further, the success / failure result of the current data collection task of the electricity meter is determined by the following conditions:

[0016] Condition 1: When the current data collection task of the electricity meter is the execution result of the electricity meter data collection task dispatched to each base station,

[0017] Condition 2: The task status indicator in the execution result is a network connection related error.

[0018] The execution result of the current data collection task of the electricity meter is failure if and only if both conditions 1 and 2 are met; otherwise, the execution result of the current data collection task of the electricity meter is success.

[0019] In conjunction with the first aspect, the maximum number of tasks further includes an initial value, which is set according to the network operator's recommendations.

[0020] In conjunction with the first aspect, preferably, the maximum number of tasks is set to no more than 100, and the default value of the maximum number of tasks is 10.

[0021] In conjunction with the first aspect, furthermore, the maximum number of tasks is dynamically adjusted based on the data acquisition success rate of each base station within each time interval, including:

[0022] Initialize the first threshold for execution success rate p1 and the first threshold for execution success rate p2;

[0023] The time interval t is used to calculate the success rate p of the electricity meter collection task of each base station within the time period from the last statistical point to the current point.

[0024] If p < p1, the maximum number of tasks for the base station is too large. Then reduce the maximum number of connections for the base station by M, while the maximum number of tasks for the base station N is greater than or equal to 2.

[0025] If p > p2, and the maximum number of tasks of the base station reaches its maximum value within 10 minutes, the maximum number of tasks supported by the base station is relatively high, then the maximum number of tasks N of the base station is increased by 2.

[0026] If p1≤p≤p2, then the maximum number of tasks N for this base station remains unchanged.

[0027] In conjunction with the first aspect, preferably, the default value of the first threshold p1 for execution success rate is 80%.

[0028] In conjunction with the first aspect, preferably, the default value of the first threshold p2 for the execution success rate is 95%.

[0029] In conjunction with the first aspect, preferably, the default value of the time interval t is 10 minutes.

[0030] In conjunction with the first aspect, further, the execution success rate p within the execution time interval t is the percentage of tasks that the base station successfully returns within the time interval t out of the total number of tasks, expressed by the following formula:

[0031]

[0032] In equation (1), N s N represents the number of tasks that successfully returned within time interval t. t This represents the total number of tasks within a time interval t.

[0033] In conjunction with the first aspect, further, the maximum number of tasks is dynamically adjusted based on the maximum number of connections corresponding to the data collection success rate of each base station over the past 24 hours, including:

[0034] At 0:00 every day, the success rate p of the electricity meter collection task of each base station and the corresponding maximum number of connections are calculated for each time interval t in the past 24 hours.

[0035] The execution success rate p and the corresponding maximum number of connections are preprocessed and then input into a pre-built LSTM model for prediction to obtain the predicted value.

[0036] The weighted average of the number of connections is calculated using the success rate and the number of predicted tasks in each time interval, and then rounded down. The weighted average under the corresponding base station is taken as the maximum number of tasks for the base station when tasks are started on that day, and the intelligent scheduling of the daily meter data collection tasks begins.

[0037] In conjunction with the first aspect, preferably, the preprocessing of the execution success rate p and the corresponding maximum number of connections includes extracting feature vectors after data standardization.

[0038] In conjunction with the first aspect, preferably, the LSTM model makes predictions through forward propagation to obtain predicted values; and it is trained through backpropagation to update the model parameters.

[0039] In conjunction with the first aspect, preferably, the forward propagation formula of the LSTM model is expressed by the following equation:

[0040]

[0041] In equation (2), Wf W i W c W o These represent the forget gate, input gate, output gate, and memory gate of an LSTM neural network, respectively; b f b i b c b o These represent the bias terms for the forget gate, input gate, output gate, and memory gate of the LSTM neural network, respectively; σ(·) represents the sigmoid function; tanh(·) represents the hyperbolic tangent function; […,…] represents the concatenation of vectors; The symbols... It represents the Hadama accumulation.

[0042] In conjunction with the first aspect, preferably, the forward propagation calculation process of the LSTM model includes:

[0043] According to equation (2), the input time series x = (x1, x2, ..., x...) is... n The output h of the neuron at each time interval t is calculated sequentially. t Finally, the output result h of the last time interval n is obtained. n The prediction result is obtained after activation by the sigmoid function.

[0044] In conjunction with the first aspect, preferably, the backpropagation calculation process of the LSTM neural network includes:

[0045] The loss is calculated using the following formula:

[0046]

[0047] In equation (3), MAE represents the mean absolute error. Represents the predicted value, y i Represents the actual value.

[0048] By using stochastic gradient descent to optimize the bias terms of the forget gate, input gate, output gate, and memory gate of the LSTM neural network with the goal of minimizing loss, the parameters of the LSTM model are updated iteratively.

[0049] In addition to the first aspect, this also includes the correspondence between the monitored electricity meter and its corresponding base station:

[0050] When the correspondence changes, update the correspondence between the electricity meter and its base station, delete the electricity meter that has changed from the electricity meter list of the original base station, and add it to the electricity meter list of the latest base station.

[0051] If the correspondence remains unchanged, continue monitoring the correspondence between the electricity meter and its corresponding base station.

[0052] Secondly, the present invention provides an intelligent scheduling system for electricity meter data collection tasks, comprising:

[0053] Acquisition module: used to acquire the maximum number of tasks and the list of electricity meters for each base station; wherein, the maximum number of tasks is dynamically adjusted based on the acquisition success rate of each base station in each time interval and the maximum number of connections corresponding to the acquisition success rate of each base station in the past 24 hours;

[0054] Dispatch module: used to determine the dispatch of meter collection tasks for each base station based on the maximum number of tasks obtained and the list of meters;

[0055] Adjustment module: Used to control the base station to collect data according to the meter collection task, obtain the current number of collection tasks and the collection task results output by the base station in real time; and adjust the meter collection tasks sent by the base station to the meter based on the current number of collection tasks and the execution results of the meter collection tasks.

[0056] Thirdly, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described in the first aspect.

[0057] Compared with the prior art, the intelligent scheduling method and system for electricity meter data collection provided in this embodiment of the invention achieves the following beneficial effects:

[0058] This invention obtains the maximum number of tasks and the meter list for each base station. The maximum number of tasks is dynamically adjusted based on the data collection success rate of each base station within each time interval and the maximum number of connections corresponding to the data collection success rate of each base station over the past 24 hours. Based on the obtained maximum number of tasks and meter list, meter collection tasks are assigned to each base station. Data collection is performed based on these meter collection tasks, and the base station outputs the current number of collection tasks and the results in real time. Based on the current number of collection tasks and the execution results of the meter collection tasks, the number of meter collection tasks issued by the base station to the meters is adjusted. This invention can dynamically adjust the number of meters communicating concurrently under a single base station, preventing channel quality degradation caused by NB-IoT network congestion due to a large number of tasks, and improving the data reading and collection success rate of the meters.

[0059] This invention addresses the problem of low meter reading success rate in high-frequency (hourly) and high-concurrency data acquisition under NB-IoT networks during advanced metering systems in smart power systems, due to limitations in NB-IoT network base station communication channels and low downlink connection success rates. The invention's electricity information acquisition master station system supports analysis and processing of the number of concurrent communication connections for a single base station, dynamically scheduling the number of concurrent communication connections under a single base station based on the acquisition success rate, thereby improving the meter reading success rate. Attached Figure Description

[0060] Figure 1 This is a flowchart of an intelligent scheduling method for electricity meter data collection tasks provided in Embodiment 1 of the present invention;

[0061] Figure 2 This is the prediction process of the LSTM model in the intelligent scheduling method for electricity meter data collection tasks provided in Embodiment 1 of the present invention;

[0062] Figure 3 This is a network structure diagram of the LSTM neural network in an intelligent scheduling method for electricity meter data collection tasks provided in Embodiment 1 of the present invention;

[0063] Figure 4 This is a flowchart of the prediction results obtained by the LSTM neural network in an intelligent scheduling method for electricity meter data collection tasks provided in Embodiment 1 of the present invention. Detailed Implementation

[0064] The present invention will be further described below with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and should not be used to limit the scope of protection of the present invention.

[0065] Example 1:

[0066] like Figure 1 As shown, this embodiment of the invention provides an intelligent scheduling method for electricity meter data collection tasks, including:

[0067] Obtain the maximum number of tasks and the electricity meter list for each base station; wherein, the maximum number of tasks is dynamically adjusted based on the data acquisition success rate of each base station in each time interval and the maximum number of connections corresponding to the data acquisition success rate of each base station in the past 24 hours;

[0068] The maximum number of tasks and the list of electricity meters are used to determine the electricity meter collection tasks to be assigned to each base station.

[0069] The system controls the base station to collect data according to the meter data collection tasks, and obtains the current number of data collection tasks and the results of the data collection tasks output by the base station in real time. Based on the current number of data collection tasks and the execution results of the meter data collection tasks, the system adjusts the meter data collection tasks sent by the base station to the meters.

[0070] The main station stores the relationship between the electricity meter and its corresponding base station, including the base station number corresponding to the electricity meter, the IP address of the electricity meter, and the communication port of the electricity meter.

[0071] like Figure 1 As shown, adjusting the meter data collection tasks sent from the base station to the meter includes:

[0072] Based on the current number of data collection tasks at the base station, if the current number of data collection tasks at the base station has not reached the maximum number of tasks, a data collection task is sent to the electricity meter; if the current number of data collection tasks at the base station reaches the maximum number of tasks, the maximum number of tasks reached by the base station is recorded, the data collection task sending process is paused, and the system enters a waiting-to-wake-up state.

[0073] Based on the execution result of the current data collection task of the electricity meter, if the execution result is successful, continue to send data collection tasks to the electricity meter's subordinate units; if the execution result is unsuccessful, reduce the number of data collection tasks under the base station corresponding to the electricity meter. If the number of data collection tasks of the base station has reached the maximum number of tasks before reducing the number of data collection tasks, then wake up the data collection task distribution process and continue to send data collection tasks to the electricity meter's subordinate units.

[0074] Specifically, the execution result of the current data collection task of the electricity meter is determined as success / failure based on the following conditions:

[0075] Condition 1: When the current data collection task of the electricity meter is the execution result of the electricity meter data collection task dispatched to each base station,

[0076] Condition 2: The task status indicator in the execution result is a network connection related error.

[0077] The execution result of the current data collection task of the electricity meter is failure if and only if both conditions 1 and 2 are met; otherwise, the execution result of the current data collection task of the electricity meter is success.

[0078] The maximum number of tasks also includes an initial value, which is set according to the network operator's recommendations. The maximum number of tasks can be set to no more than 100, and the default value is 10.

[0079] like Figure 1 As shown, the maximum number of tasks is dynamically adjusted based on the success rate of data collection at each base station within each time interval, including:

[0080] Initialize the first threshold for execution success rate p1 and the first threshold for execution success rate p2;

[0081] The time interval t is used to calculate the success rate p of the electricity meter collection task of each base station within the time period from the last statistical point to the current point.

[0082] If p < p1, the maximum number of tasks for the base station is too large. Then reduce the maximum number of connections for the base station by M, while the maximum number of tasks for the base station N is greater than or equal to 2.

[0083] If p > p2, and the maximum number of tasks of the base station reaches its maximum value within 10 minutes, the maximum number of tasks supported by the base station is relatively high, then the maximum number of tasks N of the base station is increased by 2.

[0084] If p1≤p≤p2, then the maximum number of tasks N for this base station remains unchanged.

[0085] In this embodiment, the first threshold for execution success rate p1 is set to a default value of 80%, the first threshold for execution success rate p2 is set to a default value of 95%, and the time interval t is set to a default value of 10 minutes.

[0086] Specifically, the execution success rate p within time interval t is the percentage of tasks that the base station successfully returns within time interval t out of the total number of tasks, expressed by the following formula:

[0087]

[0088] In equation (1), N s N represents the number of tasks that successfully returned within time interval t. t This represents the total number of tasks within a time interval t.

[0089] like Figure 2 As shown, the maximum number of tasks is dynamically adjusted based on the maximum number of connections corresponding to the data collection success rate of each base station over the past 24 hours, including:

[0090] At 0:00 every day, the success rate p of the electricity meter collection task of each base station and the corresponding maximum number of connections are calculated for each time interval t in the past 24 hours.

[0091] The execution success rate p and the corresponding maximum number of connections are preprocessed and then input into a pre-built LSTM model for prediction to obtain the predicted value.

[0092] The weighted average of the number of connections is calculated using the success rate and the number of predicted tasks in each time interval, and then rounded down. The weighted average under the corresponding base station is taken as the maximum number of tasks for the base station when tasks are started on that day, and the intelligent scheduling of the daily meter data collection tasks begins.

[0093] like Figure 2 As shown, the execution success rate p and the corresponding maximum number of connections are preprocessed, including standardizing the data and extracting feature vectors.

[0094] The pre-built LSTM model includes an LSTM neural network, the structure of which is as follows: Figure 3 As shown. The LSTM neural network controls the state C through three control switches: the forget gate (for the previous state C). t-1 How much is retained up to the current time C? t ), Input gate (current input x) t How much is retained to cell state C? t Output gate (current cell state C) t How much output to the current output h t ).

[0095] The LSTM model makes predictions through forward propagation and updates the model parameters through backpropagation.

[0096] The forward propagation formula for the LSTM model is expressed by the following equation:

[0097]

[0098] In equation (2), W f W i W c W o These represent the forget gate, input gate, output gate, and memory gate of an LSTM neural network, respectively; b f b i b c b o These represent the bias terms for the forget gate, input gate, output gate, and memory gate of the LSTM neural network, respectively; σ(·) represents the sigmoid function; tanh(·) represents the hyperbolic tangent function; […,…] represents the concatenation of vectors; The symbols... It represents the Hadama accumulation.

[0099] like Figure 4 As shown, the process by which the LSTM model obtains prediction results is the forward propagation calculation process, including:

[0100] According to equation (2), the input time series x = (x1, x2, ..., x...) is... n The output h of the neuron at each time interval t is calculated sequentially. t Finally, the output result h of the last time interval n is obtained. n The prediction result is obtained after activation by the sigmoid function.

[0101] The backpropagation calculation process of an LSTM neural network includes:

[0102] The loss is calculated using the following formula:

[0103]

[0104] In equation (3), MAE represents the mean absolute error. Represents the predicted value, y i Represents the actual value.

[0105] By using stochastic gradient descent to optimize the bias terms of the forget gate, input gate, output gate, and memory gate of the LSTM neural network with the goal of minimizing loss, the parameters of the LSTM model are updated iteratively.

[0106] like Figure 1 As shown, it also includes the correspondence between the monitored electricity meters and their respective base stations:

[0107] When the correspondence changes, update the correspondence between the electricity meter and its base station, delete the electricity meter that has changed from the electricity meter list of the original base station, and add it to the electricity meter list of the latest base station.

[0108] If the correspondence remains unchanged, continue monitoring the correspondence between the electricity meter and its corresponding base station.

[0109] Example 2:

[0110] This invention provides an intelligent scheduling system for electricity meter data collection tasks, comprising:

[0111] Acquisition module: used to acquire the maximum number of tasks and the list of electricity meters for each base station; wherein, the maximum number of tasks is dynamically adjusted based on the acquisition success rate of each base station in each time interval and the maximum number of connections corresponding to the acquisition success rate of each base station in the past 24 hours;

[0112] Dispatch module: used to determine the dispatch of meter collection tasks for each base station based on the maximum number of tasks obtained and the list of meters;

[0113] Adjustment module: Used to control the base station to collect data according to the meter collection task, obtain the current number of collection tasks and the collection task results output by the base station in real time; and adjust the meter collection tasks sent by the base station to the meter based on the current number of collection tasks and the execution results of the meter collection tasks.

[0114] Example 3:

[0115] This invention also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the method described in Embodiment 1.

[0116] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0117] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0118] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0119] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0120] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for intelligent scheduling of electricity meter data collection tasks, characterized in that, include: Obtain the maximum number of tasks and the electricity meter list for each base station; wherein, the maximum number of tasks is dynamically adjusted based on the data acquisition success rate of each base station in each time interval and the maximum number of connections corresponding to the data acquisition success rate of each base station in the past 24 hours; The maximum number of tasks and the list of electricity meters are used to determine the electricity meter collection tasks to be assigned to each base station. The system controls the base station to collect data according to the meter data collection tasks, and obtains the current number of data collection tasks and the results of the data collection tasks output by the base station in real time. Based on the current number of data collection tasks and the results of the meter data collection tasks, the system adjusts the meter data collection tasks sent by the base station to the meters. The meter data collection task sent from the base station to the meter includes: Based on the current number of data collection tasks at the base station, if the current number of data collection tasks at the base station has not reached the maximum number of tasks, a data collection task is sent to the electricity meter; if the current number of data collection tasks at the base station reaches the maximum number of tasks, the maximum number of tasks reached by the base station is recorded, the data collection task sending process is paused, and the system enters a waiting-to-wake-up state. Based on the execution result of the current data collection task of the electricity meter, if the execution result is successful, continue to send data collection tasks to the electricity meter; if the execution result is unsuccessful, reduce the number of data collection tasks of the electricity meter under the base station corresponding to the electricity meter. If the number of data collection tasks of the base station has reached the maximum number of tasks before reducing the number of data collection tasks, then wake up the data collection task sending process and continue to send data collection tasks to the electricity meter. The maximum number of tasks is dynamically adjusted based on the data acquisition success rate of each base station within each time interval, including: First threshold for initialization execution success rate p 1 and the first threshold for execution success rate p 2; Interval time interval t The success rate of meter data collection tasks at each base station was calculated from the last statistical point in time to the current point in time. p ; like p < p 1. If the maximum number of tasks for this base station is too high, then reduce the maximum number of connections for this base station. M Meanwhile, the maximum number of tasks for this base station N Greater than or equal to 2; like p > p 2. If the maximum number of tasks a base station can support reaches its maximum value within 10 minutes, and the maximum number of tasks a base station can support is relatively high, then the maximum number of tasks for that base station will be reduced. N Increase by 2; like p 1≤ p ≤ p 2, then the maximum number of tasks for this base station. N constant; The maximum number of tasks is dynamically adjusted based on the maximum number of connections corresponding to the data collection success rate of each base station over the past 24 hours, including: At 00:00 each day, statistics are collected for each base station at each time interval over the past 24 hours. t Success rate of electricity meter data collection tasks at each base station p and the corresponding maximum number of connections; Success rate p After preprocessing the corresponding maximum number of connections, the data is input into a pre-built LSTM model for prediction to obtain the predicted value. The weighted average of the number of connections is calculated using the success rate and the number of predicted tasks in each time interval, and then rounded down. The weighted average under the corresponding base station is taken as the maximum number of tasks for the base station when tasks are started on that day, and the intelligent scheduling of the daily meter data collection tasks begins.

2. The intelligent scheduling method for electricity meter data collection tasks according to claim 1, characterized in that, The result of the current data collection task of the electricity meter is determined as success / failure based on the following conditions: Condition 1: When the current data collection task of the electricity meter is the execution result of the electricity meter data collection task dispatched to each base station, Condition 2: The task status indicator in the execution result is a network connection related error. The execution result of the current data collection task of the electricity meter is failure if and only if both conditions 1 and 2 are met; otherwise, the execution result of the current data collection task of the electricity meter is success.

3. The intelligent scheduling method for electricity meter data collection tasks according to claim 1, characterized in that, The maximum number of tasks also includes an initial value, which is set according to the network operator's recommendations.

4. The intelligent scheduling method for electricity meter data collection tasks according to claim 1, characterized in that, The execution success rate p The percentage of tasks successfully returned by the base station within time interval t is expressed by the following formula: （1） In equation (1), N s Indicates time interval t The number of tasks successfully returned. N t Indicates time interval t Total number of internal tasks.

5. The intelligent scheduling method for electricity meter data collection tasks according to claim 1, characterized in that, It also includes the mapping between the monitored electricity meters and their respective base stations: When the correspondence changes, update the correspondence between the electricity meter and its base station, delete the electricity meter that has changed from the electricity meter list of the original base station, and add it to the electricity meter list of the latest base station. If the correspondence remains unchanged, continue monitoring the correspondence between the electricity meter and its corresponding base station.

6. An intelligent scheduling system for electricity meter data collection tasks, characterized in that, include: Acquisition module: used to acquire the maximum number of tasks and the list of electricity meters for each base station; wherein, the maximum number of tasks is dynamically adjusted based on the acquisition success rate of each base station in each time interval and the maximum number of connections corresponding to the acquisition success rate of each base station in the past 24 hours; Dispatch module: used to determine the dispatch of meter collection tasks for each base station based on the maximum number of tasks obtained and the list of meters; Adjustment module: Used to control the base station to collect data according to the meter data collection task, obtain the current number of data collection tasks and the data collection task results output by the base station in real time; and adjust the meter data collection tasks sent by the base station to the meter based on the current number of data collection tasks and the execution results of the meter data collection tasks. The meter data collection task sent from the base station to the meter includes: Based on the current number of data collection tasks at the base station, if the current number of data collection tasks at the base station has not reached the maximum number of tasks, a data collection task is sent to the electricity meter; if the current number of data collection tasks at the base station reaches the maximum number of tasks, the maximum number of tasks reached by the base station is recorded, the data collection task sending process is paused, and the system enters a waiting-to-wake-up state. Based on the execution result of the current data collection task of the electricity meter, if the execution result is successful, continue to send data collection tasks to the electricity meter; if the execution result is unsuccessful, reduce the number of data collection tasks of the electricity meter under the base station corresponding to the electricity meter. If the number of data collection tasks of the base station has reached the maximum number of tasks before reducing the number of data collection tasks, then wake up the data collection task sending process and continue to send data collection tasks to the electricity meter. The maximum number of tasks is dynamically adjusted based on the data acquisition success rate of each base station within each time interval, including: First threshold for initialization execution success rate p 1 and the first threshold for execution success rate p 2; Interval time interval t The success rate of meter data collection tasks at each base station was calculated from the last statistical point in time to the current point in time. p ; like p < p 1. If the maximum number of tasks for this base station is too high, then reduce the maximum number of connections for this base station. M Meanwhile, the maximum number of tasks for this base station N Greater than or equal to 2; like p > p 2. If the maximum number of tasks a base station can support reaches its maximum value within 10 minutes, and the maximum number of tasks a base station can support is relatively high, then the maximum number of tasks for that base station will be reduced. N Increase by 2; like p 1≤ p ≤ p 2, then the maximum number of tasks for this base station. N constant; The maximum number of tasks is dynamically adjusted based on the maximum number of connections corresponding to the data collection success rate of each base station over the past 24 hours, including: At 00:00 each day, the success rate of meter data collection tasks for each base station in each time interval t over the past 24 hours is calculated. p and the corresponding maximum number of connections; Success rate p After preprocessing the corresponding maximum number of connections, the data is input into a pre-built LSTM model for prediction to obtain the predicted value. The weighted average of the number of connections is calculated using the success rate and the number of predicted tasks in each time interval, and then rounded down. The weighted average under the corresponding base station is taken as the maximum number of tasks for the base station when tasks are started on that day, and the intelligent scheduling of the daily meter data collection tasks begins.

7. A computer-readable storage medium, characterized in that: It stores a computer program that, when executed by a processor, implements the steps of the method according to any one of claims 1-5.

Images