A distributed photovoltaic data acquisition method based on 5G communication

Abstract

The application provides a kind of distributed photovoltaic data acquisition method based on 5G communication, it is related to data acquisition technical field, step 1: corresponding acquisition equipment is configured to different working objects in photovoltaic area, and the working parameters of corresponding working object are acquired based on the acquisition equipment;Step 2: the acquisition data is transmitted to local database based on 5G base station and edge node, and the acquisition data is preliminarily processed based on the local database;Step 3: the data security analysis of the data after preliminary processing is carried out based on monitoring host, and the working state of each working object and data security coefficient are determined;Step 4: the early warning instruction related to each working object and the area instruction related to each working area are generated based on working state and data security coefficient, and are issued to dispatching center to display and early warning.A layer by layer realization of data simplification and data acquisition under channel capacity sharing, improve data acquisition efficiency.

Description

Technical Field

[0001] This invention relates to the field of data acquisition technology, and in particular to a distributed photovoltaic data acquisition method based on 5G communication. Background Technology

[0002] Photovoltaic power distribution areas involve a wide variety of equipment, including core equipment such as photovoltaic modules and inverters, supporting and transmission equipment such as support systems and cables, and auxiliary equipment such as monitoring systems and lightning protection and grounding systems. To ensure the normal operation of photovoltaic power distribution areas, it is necessary to start from multiple aspects such as equipment maintenance, environmental management, and system monitoring. Among these, equipment maintenance is extremely important, and the prerequisite for equipment maintenance is the collection of equipment operating data. Generally, data collection is carried out through fixed channels to achieve the collection of data from the previous level equipment by the next level equipment. The channel capacity is limited, and there are generally only two devices between the previous and next level equipment: the transmission end and the destination end. No other levels are involved in between. When the data volume is too large, data collection not only wastes business resources, but also limits the data collection efficiency of the next level equipment from the previous level equipment due to capacity limitations.

[0003] Therefore, this invention proposes a distributed photovoltaic data acquisition method based on 5G communication. Summary of the Invention

[0004] This invention provides a distributed photovoltaic data acquisition method based on 5G communication. It simplifies the data and distributes the channel capacity during data acquisition by using 5G base stations, edge nodes, local databases, and monitoring hosts in a layered manner, thereby improving the data acquisition efficiency. Furthermore, by performing status and safety factor analysis on the work object, it is convenient to carry out early warning and timely handling of anomalies, thus indirectly improving the acquisition efficiency.

[0005] This invention provides a distributed photovoltaic data acquisition method based on 5G communication, comprising:

[0006] Step 1: Configure corresponding data acquisition devices for different work objects in the photovoltaic area, and collect the working parameters of the corresponding work objects based on the data acquisition devices;

[0007] Step 2: Transmit the collected data to a local database via 5G base stations and edge nodes, and perform preliminary processing on the collected data based on the local database;

[0008] Step 3: Based on the monitoring host, perform data security analysis on the preliminarily processed data to determine the working status and data security level of each work object;

[0009] Step 4: Based on the working status and data security factor, generate early warning instructions related to each work object and area instructions related to each work area, and send them to the dispatch center for display and early warning.

[0010] Preferably, the work objects include photovoltaic-related work equipment and equipment for monitoring the operational behavior of workers.

[0011] Preferably, the process of transmitting the collected data to the local database via 5G base stations and edge nodes includes:

[0012] Based on the settings of each work object, implement functions and obtain network requirements by matching from the function-network database;

[0013] Cluster analysis is performed on all network requirements to obtain the clusters of each cluster result, and the first position of the cluster is locked to the second position of the first requirement that is farthest from the cluster and the third position of the second requirement that is second farthest from the cluster.

[0014] Based on the distance between any two positions among the first, second, and third positions, set the required range value for the corresponding clustering results. ;

[0015] ;

[0016] in, These represent the distances between the first and second positions, the first and third positions, and the second and third positions, respectively; max indicates the maximum value. This represents the radius of the specified region where the corresponding clustering result is placed, and it is greater than... ; This indicates that, except for the corresponding clustering results The average distance between any two positions other than 3; This represents the variance of all distance lengths involved in the corresponding clustering results; This indicates that, except for the corresponding clustering results The variance of the remaining distance length beyond;

[0017] The shortest distance is selected from the three distance lengths, and a circular region is divided with the first position of the cluster as the center, and the required cluster density of the circular region is determined.

[0018] Based on the aforementioned demand range value, and combined with the number of network demands involved in the corresponding clustering results and the demand clustering density, determine the maximum boundary value of the number of additional slices that can be added based on the corresponding clustering results.

[0019] The data volume of the photovoltaic work park at each historical time point for each work object was randomly obtained from the historical database for a time period of length T, and a data volume matrix was constructed. ,in, Let represent the data volume vectors at the 1st and Nth historical time points, respectively. , This represents the data volume vector at the j-th historical time point; This represents the amount of data collected for the i-th working object based on the j-th historical time point; m represents the number of working objects.

[0020] The column vectors of the working objects involved in the same clustering result are extracted sequentially from the data volume matrix, and a new clustering matrix is ​​constructed.

[0021] The number of network slices required for normal operation is determined based on the new clustering matrix;

[0022] The network constructed based on the sum of the number of network slices and the maximum boundary value of the number of additional slices is used as the transmission slice network for all working objects involved in the corresponding clustering results.

[0023] The collected data is transmitted based on the transmission slice network using 5G base stations.

[0024] Preferably, the maximum boundary value for the number of additional slices that can be added based on the corresponding clustering results is determined. ,include:

[0025] ;

[0026] in, This indicates the number of network requirements involved in the corresponding clustering results; This indicates the number of network requirements involved in the corresponding circular area; This indicates the required cluster density for the corresponding circular region; Indicates the preset cluster density; This represents the average of the range of demand values ​​based on all clustering results; This indicates the floor function.

[0027] Preferably, determining the required number of network slices based on the new clustering matrix includes:

[0028] ;

[0029] in, , , Let each represent a conditional function, and , , , This indicates the number of column vectors present in the corresponding new clustering matrix; This represents the average of all values ​​in the k-th column vector; This represents the coefficient of the midpoint on the fitted line obtained after plotting the curve and performing linear fitting on the k-th column vector; The variance of all values ​​in the k-th column vector; This indicates the set transmission volume for each network slice; This indicates the number of network slices typically required.

[0030] Preferably, preliminary processing refers to data cleaning and filling in missing data values.

[0031] Preferably, the working status and data security factor of each working object are determined, including:

[0032] Extract the first processed data for each work object from the pre-processed data, and compare and analyze the first processed data with the standard data of the work object to identify abnormal data.

[0033] Retrieve a state analysis model that matches the work object from the state analysis database;

[0034] The abnormal data is input into the state analysis model to obtain the working status of the corresponding work object and the data security coefficient.

[0035] Preferably, the system generates warning instructions related to each work object and area instructions related to each work area, including:

[0036] The warning instructions for the corresponding work object are obtained by matching the object-status-coefficient-instruction lookup table;

[0037] Each work object is categorized according to the regional division results of the photovoltaic area, and a list is generated from all early warning instructions under the same category to obtain the regional instructions.

[0038] Compared with the prior art, the beneficial effects of this application are as follows:

[0039] By using 5G base stations, edge nodes, local databases, and monitoring hosts in a layered manner, data simplification and channel capacity sharing during data collection are achieved, improving data collection efficiency. Furthermore, subsequent analysis of the status and security of the work objects facilitates early warning and timely handling of anomalies, thereby indirectly improving collection efficiency. Attached Figure Description

[0040] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0041] Figure 1 This is a flowchart illustrating a distributed photovoltaic data acquisition method based on 5G communication provided in an embodiment of the present invention.

[0042] Figure 2 It is a photovoltaic distribution network structure based on 5G communication;

[0043] Figure 3 This is a structural diagram of a distributed photovoltaic local communication scheme;

[0044] Figure 4 This is a transmission diagram based on a 5G slicing private network;

[0045] Figure 5 This is a call structure diagram of the monitoring host;

[0046] Figure 6 This is a structural diagram of the distributed photovoltaic peak-shaving command model. Detailed Implementation

[0047] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0048] This invention provides a distributed photovoltaic data acquisition method based on 5G communication, such as... Figure 1 As shown, it includes:

[0049] Step 1: Configure corresponding data acquisition devices for different work objects in the photovoltaic area, and collect the working parameters of the corresponding work objects based on the data acquisition devices;

[0050] Step 2: Transmit the collected data to a local database via 5G base stations and edge nodes, and perform preliminary processing on the collected data based on the local database;

[0051] Step 3: Based on the monitoring host, perform data security analysis on the preliminarily processed data to determine the working status and data security level of each work object;

[0052] Step 4: Based on the working status and data security factor, generate early warning instructions related to each work object and area instructions related to each work area, and send them to the dispatch center for display and early warning.

[0053] Preferably, the work objects include photovoltaic-related work equipment and equipment for monitoring the operational behavior of workers.

[0054] Preferably, the preliminary processing refers to data cleaning and missing value filling, which are existing technologies.

[0055] In this embodiment, the equipment involved may include transformer boxes, inverters, distribution cabinets, combiner boxes, photovoltaic arrays, environmental monitoring instruments, drones, cameras, inspection robots, etc., which means that the collected data involves power grid data, power plant internal equipment data, and power plant internal environmental data.

[0056] In this embodiment, the operator's operation, such as opening or closing a power distribution cabinet, is captured by a camera.

[0057] In this embodiment, data security analysis is implemented based on relevant analysis models, and the working states include: normal operation state and abnormal operation state.

[0058] In this embodiment, the data security factor ranges from 0 to 1.

[0059] In this embodiment, the regional command is implemented based on the warning command.

[0060] In this embodiment, the photovoltaic power station network structure based on 5G communication is as follows: Figure 2 As shown, in this framework, SA networking provides the power station with the ability to provide customized 5G services; edge computing alleviates the pressure on the monitoring center, processes data locally, and feeds the processed data back to the monitoring center; 5G network slicing divides the network according to the application needs of the power station; and UPF sinking shortens the physical distance of data forwarding and protects data security.

[0061] In this embodiment, the local communication scheme is designed as follows:

[0062] Embedded system development: High-performance processors and embedded operating systems are selected to achieve rapid system response and stability. The main control chip used in the intelligent terminal is the SCM701D. A highly reliable, fast, large-capacity, and open intelligent terminal platform is built through a platform-based hardware solution and an independently controllable operating system. The intelligent terminal platform's external interfaces include a cross-connection board interface, a MINIUSB interface (mirror loading interface), a debug serial port (TYPE-C interface RS232 protocol), a local communication module socket (low voltage end), a local communication module socket (high voltage end), a remote signaling interface (4-channel switch input interface), RS485 / 232 and pulse interfaces, 2 Ethernet interfaces, a reserved module socket, a remote communication module socket, a high-voltage connection socket, and a wireless Bluetooth module, etc. These external interfaces enable communication with a mobile app, connection to a local database, and communication with 5G networks.

[0063] Communication Method: The distributed photovoltaic (PV) smart terminal uses HPLC technology for downlink communication. Users with PV inverters already configured to support HPLC can directly interact with the smart terminal in their distribution area. Users with PV inverters that do not support HPLC can add a PV controller on the PV distribution area side, using RS485 to communicate with the PV inverter and grid-connected switch, responsible for data acquisition, encryption, and uploading of voltage, current, active power, reactive power, energy consumption, and switch status. The local communication scheme for distributed PV is shown in Figure 3.

[0064] The specific design of the remote communication solution is as follows:

[0065] A dedicated 5G slicing network is used for uploading photovoltaic (PV) data and issuing control commands. The 5G slicing network features ultra-large-scale connectivity, ultra-low latency, and ultra-high bandwidth, significantly improving network performance. Multi-access edge computing (MEC) and user plane functions (UPF) are deployed at the power communication station. PV data is uploaded to the power communication station via the dedicated 5G slicing network, and then connected to the secure access zone via a dedicated power optical transmission network line or dispatch data network channel, enabling the uploading of collected information and the issuance of control commands. Figure 4 As shown.

[0066] In this embodiment, the monitoring host also includes data retrieval, as detailed below:

[0067] The monitoring host (data center) is located at the center of information interaction between the cloud, edge, and endpoints. Data collection apps are responsible for collecting data from endpoint devices and writing it to the data center. Analysis apps acquire data from the data center, perform secondary processing and analysis to achieve edge computing. Main station apps acquire data from the data center and upload it to the cloud main station. All apps share data through the data center. Figure 5 As shown.

[0068] In this embodiment, the management based on the local database is as follows: The local communication management APP is responsible for acting as a proxy carrier module (CCO) and providing a calling interface to the business APP. The calling interface exists in two forms: JSON and ASN.1.

[0069] In this embodiment, during the data collection process, the collection tasks of lower-level data to higher-level data can be prioritized according to data usage, timeliness, importance, etc. For example, they can be sorted according to the priority of first-level events, 1-minute high-frequency collection, daily freeze, 15-minute, hourly high-frequency collection, second-level events, and third-level events to ensure reliable uploading of minute-level data.

[0070] In this embodiment, if there is a busy channel during the lower-level data acquisition process from the upper-level system, a peak-shaving command model can be used for adjustment, such as... Figure 6 As shown, the model consists of three parts: command type, command mode, and dispatch allocation mode. Peak shaving commands have two types: command by power and command by capacity.

[0071] The beneficial effects of the above technical solution are: by using 5G base stations, edge nodes, local databases, and monitoring hosts to simplify data and distribute channel capacity during data collection, the efficiency of data collection is improved. Furthermore, by performing status and security factor analysis on the work objects, it is convenient to conduct early warnings and enable timely handling of anomalies, thereby indirectly improving the collection efficiency.

[0072] This invention provides a distributed photovoltaic data acquisition method based on 5G communication. The process of transmitting the acquired data to a local database via a 5G base station and edge nodes includes:

[0073] Based on the settings of each work object, implement functions and obtain network requirements by matching from the function-network database;

[0074] Cluster analysis is performed on all network requirements to obtain the clusters of each cluster result, and the first position of the cluster is locked to the second position of the first requirement that is farthest from the cluster and the third position of the second requirement that is second farthest from the cluster.

[0075] Based on the distance between any two positions among the first, second, and third positions, set the required range value for the corresponding clustering results. ;

[0076] ;

[0077] in, These represent the distances between the first and second positions, the first and third positions, and the second and third positions, respectively; max indicates the maximum value. This represents the radius of the specified region where the corresponding clustering result is placed, and it is greater than... ; This indicates that, except for the corresponding clustering results The average distance between any two positions other than 3; This represents the variance of all distance lengths involved in the corresponding clustering results; This indicates that, except for the corresponding clustering results The variance of the remaining distance length beyond;

[0078] The shortest distance is selected from the three distance lengths, and a circular region is divided with the first position of the cluster as the center, and the required cluster density of the circular region is determined.

[0079] Based on the aforementioned demand range value, and combined with the number of network demands involved in the corresponding clustering results and the demand clustering density, determine the maximum boundary value of the number of additional slices that can be added based on the corresponding clustering results.

[0080] The data volume of the photovoltaic work park at each historical time point for each work object was randomly obtained from the historical database for a time period of length T, and a data volume matrix was constructed. ,in, Let represent the data volume vectors at the 1st and Nth historical time points, respectively. , This represents the data volume vector at the j-th historical time point; This represents the amount of data collected for the i-th working object based on the j-th historical time point; m represents the number of working objects.

[0081] The column vectors of the working objects involved in the same clustering result are extracted sequentially from the data volume matrix, and a new clustering matrix is ​​constructed.

[0082] The number of network slices required for normal operation is determined based on the new clustering matrix;

[0083] The network constructed based on the sum of the number of network slices and the maximum boundary value of the number of additional slices is used as the transmission slice network for all working objects involved in the corresponding clustering results.

[0084] The collected data is transmitted based on the transmission slice network using 5G base stations.

[0085] Preferably, the maximum boundary value for the number of additional slices that can be added based on the corresponding clustering results is determined. ,include:

[0086] ;

[0087] in, This indicates the number of network requirements involved in the corresponding clustering results; This indicates the number of network requirements involved in the corresponding circular area; This indicates the required cluster density for the corresponding circular region; Indicates the preset cluster density; This represents the average of the range of demand values ​​based on all clustering results; This indicates the floor function.

[0088] Preferably, determining the required number of network slices based on the new clustering matrix includes:

[0089] ;

[0090] in, , , Let each represent a conditional function, and , , , This indicates the number of column vectors present in the corresponding new clustering matrix; This represents the average of all values ​​in the k-th column vector; This represents the coefficient of the midpoint on the fitted line obtained after plotting the curve and performing linear fitting on the k-th column vector; The variance of all values ​​in the k-th column vector; This indicates the set transmission volume for each network slice; This indicates the number of network slices typically required.

[0091] In this embodiment, the defined implementation function refers to the function performed by the working object. For example, the function of a photovoltaic array is to collect light energy and convert it into electrical energy.

[0092] In this embodiment, the function-network database includes different settings to implement functions and matching network requirements, because different objects require different network transmission channels, and the size of the network transmission channel is reflected in the number of network slices.

[0093] In this embodiment, cluster analysis is implemented based on the K-means algorithm, which is an existing technology. This allows for the direct determination of clusters, and the location of each requirement within the results can be effectively determined based on the clustering results.

[0094] In this embodiment, the shortest distance is based on It is achieved, and the division of the circular region is based on the first position and the shortest distance is used as the radius.

[0095] In this embodiment, the demand cluster density = the number of demands involved in the circular region / the area of ​​the circular region.

[0096] In this embodiment, the maximum boundary value is determined so that when the number of network slices required by the normal procedure is insufficient, the backup slices under the maximum boundary value can be selected to ensure that the lower level effectively collects data from the upper level.

[0097] In this embodiment, the historical database contains the amount of data generated by different working objects at different times.

[0098] In this embodiment, the transmission slice network is used to transmit data collected related to the work objects involved.

[0099] The beneficial effects of the above technical solution are: cluster analysis is performed according to network requirements to initially classify the collected data related to the device, so that different data can be transmitted using different slices to avoid data corruption and data congestion. Furthermore, the maximum boundary value is determined by the demand range value and the demand cluster density to set up backup slices to ensure that the lower level can effectively collect data from the upper level even when the data volume is too large.

[0100] This invention provides a distributed photovoltaic data acquisition method based on 5G communication, which determines the working status of each working object and the data security factor, including:

[0101] Extract the first processed data for each work object from the pre-processed data, and compare and analyze the first processed data with the standard data of the work object to identify abnormal data.

[0102] Retrieve a state analysis model that matches the work object from the state analysis database;

[0103] The abnormal data is input into the state analysis model to obtain the working status of the corresponding work object and the data security coefficient.

[0104] In this embodiment, each working object has its corresponding first data.

[0105] In this embodiment, the standard data is pre-set. For example, the inverter collects data such as DC side voltage, DC side current, DC side power, AC side voltage, AC side current, AC side power, AC side active power, AC side reactive power, and AC side power factor. The standard value for each parameter is pre-set as a reference standard.

[0106] In this embodiment, the state analysis database contains state analysis models for different work objects. The state analysis models are trained on neural network models based on parameter combinations of different work objects and the analysis results of the parameter combinations (state and data security coefficients). Therefore, the required results can be obtained directly based on the relevant models.

[0107] The beneficial effects of the above technical solution are: by comparing and analyzing the first processed data with standard data to identify abnormal data, and then by using a state analysis model to determine the working status and data security factor of the working object, a foundation is provided for subsequent early warning.

[0108] This invention provides a distributed photovoltaic data acquisition method based on 5G communication, which generates early warning instructions related to each work object and area instructions related to each work area, including:

[0109] The warning instructions for the corresponding work object are obtained by matching the object-status-coefficient-instruction lookup table;

[0110] Each work object is categorized according to the regional division results of the photovoltaic area, and a list is generated from all early warning instructions under the same category to obtain the regional instructions.

[0111] In this embodiment, the object-state-coefficient-instruction lookup table contains warning instructions for different working objects under different working states and data security coefficients, which can be one or more combinations of sound, light, text description, etc.

[0112] In this embodiment, the region division result can be the object result of the power grid end region, the power station internal equipment region, and the power station internal environment region.

[0113] In this embodiment, the list refers to the statistical analysis of all warning instructions under that category.

[0114] In this embodiment, "regional command" is a collective term for all warning commands involved in the list.

[0115] The beneficial effects of the above technical solution are: obtaining early warning instructions from the lookup table, and combining the regional division results to achieve reasonable classification of early warning instructions, and obtaining regional instructions, thus providing a basis for subsequent early warnings.

[0116] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A distributed photovoltaic data acquisition method based on 5G communication, characterized in that, The method comprises the following steps: Step 1: configuring corresponding acquisition devices to different working objects in a photovoltaic area, and collecting working parameters of the corresponding working objects based on the acquisition devices; Step 2: the process of transmitting the collected data to a local database based on a 5G base station and an edge node comprises: based on the setting of each working object, realizing a function, and matching the network demand from a function-network database; performing cluster analysis on all network demands to obtain a cluster of each cluster result, and locking a first position of the cluster, a second position of a first demand farthest from the first position of the cluster, and a third position of a second demand farthest from the second position of the first demand; set a demand range value to the corresponding clustering result based on the distance length between any two of the first position, the second position, and the third position ; ； in, These represent the distances between the first and second positions, the first and third positions, and the second and third positions, respectively; max indicates the maximum value. This represents the radius of the specified region where the corresponding clustering result is placed, and it is greater than... ; This indicates that, except for the corresponding clustering results The average distance between any two positions other than 3; This represents the variance of all distance lengths involved in the corresponding clustering results; This indicates that, except for the corresponding clustering results The variance of the remaining distance length beyond; selecting the shortest distance from the three distance lengths, and dividing a circular area with the first position of the cluster as the center, and determining the demand cluster density of the circular area; determining the maximum boundary value of the number of additional slices based on the corresponding cluster result according to the demand range value, in combination with the number of network demands involved in the corresponding cluster result, and in combination with the demand cluster density; obtain the collected data amount of each work object of the photovoltaic station at each historical time point in a time period with a time length T from the historical database randomly, and construct a data amount matrix wherein, respectively represent the data amount vector at the 1st historical time point and the Nth historical time point, and , represents the data amount vector at the jth historical time point; represents the collected data amount of the ith work object based on the jth historical time point; m represents the number of work objects; extracting column vectors of the working objects involved in the same cluster result from the data amount matrix in sequence, and constructing a new cluster matrix; determining the number of network slices required in a regular manner based on the new cluster matrix; constructing a network composed of the sum of the number of network slices and the maximum boundary value of the number of additional slices as a transmission slice network for all working objects involved in the corresponding cluster result; transmitting the collected data according to the transmission slice network based on the 5G base station; and preliminarily processing the collected data based on the local database; Step 3: performing data security analysis on the preliminarily processed data based on a monitoring host to determine the working state and data security coefficient of each working object; Step 4: generating a warning instruction related to each working object and a region instruction related to each working region based on the working state and the data security coefficient, and delivering them to a dispatch center for display and warning.

2. The method of claim 1, wherein the method is a 5G communication based distributed photovoltaic data collection method. The working objects include working equipment related to photovoltaic and equipment for monitoring the operation behavior of working personnel.

3. The method of claim 1, wherein, Determining a maximum boundary value of an addable slice quantity based on a corresponding clustering result , comprising: ； wherein, represents the number of network demands involved in the corresponding cluster result; represents the number of network demands involved in the corresponding circular region; represents the demand cluster density of the corresponding circular region; represents the preset cluster density; represents the average value of the demand range values based on all the cluster results; represents a floor symbol.

4. The method of claim 1, wherein, Determining the number of network slices required in a regular manner based on the new cluster matrix comprises: ； wherein, , , respectively represent a decision function, and , , , represents the number of column vectors existing in the corresponding cluster new matrix; represents the average value of all values in the kth column vector; represents the coefficient of the middle position point on the fitting straight line obtained after curve plotting and linear fitting of the kth column vector; represents the variance of all values in the kth column vector; represents the set transmission amount of each network slice; represents the number of network slices required in general.

5. The method of claim 1, wherein, Preliminary processing refers to data cleaning and data missing value filling processing.

6. The method of claim 1, wherein, Determining the working state and data security coefficient of each working object comprises: extracting first processing data of each working object from the preliminarily processed data, and comparing and analyzing the first processing data with standard data of the working object to determine abnormal data; calling a state analysis model matched with the working object from a state analysis database; inputting the abnormal data into the state analysis model to obtain the working state and data security coefficient of the corresponding working object.

7. The method of claim 1, wherein, Generating a warning instruction related to each working object and a region instruction related to each working region comprises: matching the warning instruction of the corresponding working object from an object-state-coefficient-instruction table; classifying each working object according to the region division result of the photovoltaic area, and generating a list of all warning instructions under the same classification to obtain a region instruction.

Images