Power grid icing prediction initial field modeling method and device based on multi-source data, computer equipment, medium and product

Abstract

The invention relates to a power grid icing prediction initial field modeling method and device based on multi-source data, computer equipment, a medium and a product, and relates to the technical field of icing prediction. The method comprises the following steps: acquiring multi-source environment data of a target power grid region; performing feature extraction on the multi-source environment data to obtain first environment feature data influencing icing in a spatial dimension and second environment feature data influencing icing in a time dimension; performing correlation analysis on the first environment feature data and the second environment feature data to obtain a feature correlation degree between the first environment feature data and the second environment feature data; under the condition that the feature association degree meets a feature association condition, constructing a three-dimensional environment model of the target power grid region based on the first environment feature data and the second environment feature data; the three-dimensional environment model is used for performing icing prediction on the target power grid area. By adopting the method, the icing prediction accuracy can be improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of ice prediction, in particular to an initial field modeling method and device for power grid ice prediction based on multi-source data, computer equipment, medium and product. BACKGROUND

[0002] Line icing has become one of the important meteorological disasters threatening the safe operation of power systems. Icing can cause serious accidents such as conductor jump, conductor breakage, tower collapse, etc. Therefore, it is urgent to realize accurate perception and risk warning of the icing state of the transmission line.

[0003] In the traditional technology, the icing prediction is usually realized by analyzing multi-source environmental data. However, since the multi-source environmental data is often random or selected according to experience, the reliability is not high, so the accuracy of the current icing prediction is also relatively low. SUMMARY

[0004] Therefore, it is necessary to provide an initial field modeling method and device for power grid ice prediction based on multi-source data, computer equipment, computer readable storage medium and computer program product, which can improve the accuracy of ice prediction.

[0005] In a first aspect, the present application provides an initial field modeling method for power grid ice prediction based on multi-source data, comprising: obtaining multi-source environmental data of a target power grid area; performing feature extraction on the multi-source environmental data to obtain first environmental feature data affecting icing in the spatial dimension and second environmental feature data affecting icing in the time dimension; performing correlation analysis on the first environmental feature data and the second environmental feature data to obtain a feature correlation degree between the first environmental feature data and the second environmental feature data; in the case that the feature correlation degree meets a feature correlation condition, constructing a three-dimensional environmental model of the target power grid area based on the first environmental feature data and the second environmental feature data; and the three-dimensional environmental model is used for icing prediction of the target power grid area.

[0006] In one of the embodiments, obtaining the multi-source environmental data of the target power grid area comprises: obtaining initial environmental data of the target power grid area under different data sources; performing data format conversion on each initial environmental data to obtain intermediate environmental data conforming to a target data format; and performing spatio-temporal alignment on each intermediate environmental data to obtain the multi-source environmental data after spatio-temporal alignment.

[0007] In one of the embodiments, performing spatio-temporal alignment on each intermediate environmental data to obtain the multi-source environmental data after spatio-temporal alignment comprises: obtaining the time stamp of each intermediate environmental data; determining the time alignment weight of each intermediate environmental data based on the time difference between the time stamp and the preset alignment time; and performing time alignment on each intermediate environmental data according to the time alignment weight to obtain the multi-source environmental data.

[0008] In one of the embodiments, the spatio-temporal alignment of the intermediate environment data to obtain the multi-source environment data after spatio-temporal alignment further comprises: spatial position alignment of the intermediate environment data to obtain the target environment data after spatial position alignment; and interpolation of the target environment data to obtain the multi-source environment data.

[0009] In one of the embodiments, the initial environment data of the target power grid region under different data sources is obtained, comprising: obtaining a plurality of candidate environment data of the target power grid region under different data sources; based on the icing prediction requirement of the target power grid region, screening the target environment data matched with the icing prediction requirement from the plurality of candidate environment data; and taking the target environment data and the historical icing data of the target power grid region as the initial environment data of the target power grid region.

[0010] In one of the embodiments, the initial field modeling method for power grid icing prediction based on multi-source data further comprises: obtaining the feature constraint conditions for the first environment feature data and the second environment feature data respectively; and in the case that the feature correlation degree meets the feature correlation condition, constructing the three-dimensional environment model of the target power grid region based on the first environment feature data and the second environment feature data, comprising: in the case that the feature correlation degree meets the feature correlation condition, and the first environment feature data and the second environment feature data meet the respective feature constraint conditions, constructing the three-dimensional environment model of the target power grid region based on the first environment feature data and the second environment feature data.

[0011] In a second aspect, the application further provides an initial field modeling device for power grid icing prediction based on multi-source data, comprising: a multi-source data acquisition module for acquiring multi-source environment data of a target power grid region; a feature data extraction module for feature extraction of the multi-source environment data to obtain first environment feature data affecting icing in spatial dimension and second environment feature data affecting icing in time dimension; a feature data correlation analysis module for correlation analysis of the first environment feature data and the second environment feature data to obtain the feature correlation degree between the first environment feature data and the second environment feature data; and a three-dimensional environment modeling module for constructing a three-dimensional environment model of the target power grid region based on the first environment feature data and the second environment feature data in the case that the feature correlation degree meets the feature correlation condition; and the three-dimensional environment model is used for icing prediction of the target power grid region.

[0012] In a third aspect, the present application also provides a computer device comprising a memory and a processor, the memory storing a computer program, and the processor implementing the following steps when executing the computer program: obtaining multi-source environment data of a target power grid region; performing feature extraction on the multi-source environment data to obtain first environment feature data affecting icing in a spatial dimension and second environment feature data affecting icing in a time dimension; performing correlation analysis on the first environment feature data and the second environment feature data to obtain a feature correlation degree between the first environment feature data and the second environment feature data; in a case where the feature correlation degree meets a feature correlation condition, constructing a three-dimensional environment model of the target power grid region based on the first environment feature data and the second environment feature data; and the three-dimensional environment model is used for icing prediction of the target power grid region.

[0013] In a fourth aspect, the present application also provides a computer-readable storage medium having a computer program stored thereon, and the computer program is executed by a processor to implement the following steps: obtaining multi-source environment data of a target power grid region; performing feature extraction on the multi-source environment data to obtain first environment feature data affecting icing in a spatial dimension and second environment feature data affecting icing in a time dimension; performing correlation analysis on the first environment feature data and the second environment feature data to obtain a feature correlation degree between the first environment feature data and the second environment feature data; in a case where the feature correlation degree meets a feature correlation condition, constructing a three-dimensional environment model of the target power grid region based on the first environment feature data and the second environment feature data; and the three-dimensional environment model is used for icing prediction of the target power grid region.

[0014] In a fifth aspect, the present application also provides a computer program product comprising a computer program, and the computer program is executed by a processor to implement the following steps: obtaining multi-source environment data of a target power grid region; performing feature extraction on the multi-source environment data to obtain first environment feature data affecting icing in a spatial dimension and second environment feature data affecting icing in a time dimension; performing correlation analysis on the first environment feature data and the second environment feature data to obtain a feature correlation degree between the first environment feature data and the second environment feature data; in a case where the feature correlation degree meets a feature correlation condition, constructing a three-dimensional environment model of the target power grid region based on the first environment feature data and the second environment feature data; and the three-dimensional environment model is used for icing prediction of the target power grid region.

[0015] The power grid icing prediction initial field modeling method, device, computer equipment, computer readable storage medium and computer program product based on multi-source data have the advantages that first, multi-source environmental data of a target power grid region are acquired, and feature extraction is performed on the multi-source environmental data to obtain first environmental feature data that affects icing in a spatial dimension and second environmental feature data that affects icing in a time dimension. Then, correlation analysis is performed on the first environmental feature data and the second environmental feature data to obtain a feature correlation degree between the first environmental feature data and the second environmental feature data. In a case where the feature correlation degree meets a feature correlation condition, a three-dimensional environmental model of the target power grid region is constructed based on the first environmental feature data and the second environmental feature data, and the three-dimensional environmental model is used for icing prediction of the target power grid region. In this way, on the one hand, environmental feature data that affects icing is extracted from two dimensions of time and space, ensuring the accuracy and reliability of the environmental feature data and improving the adaptability and generalization capability of the subsequent three-dimensional environmental model to the spatiotemporal distribution of icing. On the other hand, by analyzing the correlation degree between the first environmental feature data and the second environmental feature data, the synergistic effect between features in the spatial dimension and the time dimension can be effectively identified, and the icing prediction deviation caused by feature conflicts can be avoided. Therefore, the present application can effectively improve the accuracy of icing prediction. BRIEF DESCRIPTION OF DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the related art, the drawings needed to be used in the description of the embodiments of the present application or the related art will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other related drawings can be obtained by those skilled in the art without any creative effort.

[0017] Figure 1 An application environment diagram of the power grid icing prediction initial field modeling method based on multi-source data in an embodiment;

[0018] Figure 2 A flowchart of the power grid icing prediction initial field modeling method based on multi-source data in an embodiment;

[0019] Figure 3 A flowchart of data format conversion in an embodiment;

[0020] Figure 4 A flowchart of time alignment in an embodiment;

[0021] Figure 5 A flowchart of spatial alignment in an embodiment;

[0022] Figure 6 A flowchart of obtaining initial environmental data in an embodiment;

[0023] Figure 7 1 is a flow chart of an initial field modeling method for power grid icing prediction based on multi-source data in a specific embodiment;

[0024] Figure 8 A structural block diagram of an initial field modeling device for power grid icing prediction based on multi-source data in one embodiment;

[0025] Figure 9 FIG. 1 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0026] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0027] It should be noted that the terms "first", "second", etc. used in this application may be used to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish the first element from the second element. The terms "including" and "having" used in this application and any variations thereof are intended to cover non-exclusive inclusions. The term "plurality" used in this application refers to two or more. The term "and / or" used in this application refers to one of the solutions or any combination of multiple solutions.

[0028] The grid icing prediction initial field modeling method based on multi-source data provided in the embodiment of the present application can be applied to Figure 1 In the application environment shown, terminal 102 communicates with server 104 via a network. A data storage system can store data that server 104 needs to process. The data storage system can be integrated with server 104 or placed in the cloud or on other network servers. Terminal 102 can be, but is not limited to, various personal computers, laptops, smartphones, tablets, and IoT devices. Server 104 can be a standalone physical server, a server cluster or distributed system consisting of multiple physical servers, or a cloud server providing cloud computing services.

[0029] Specifically, the server 104 first acquires multi-source environmental data of a target power grid region, and performs feature extraction on the multi-source environmental data to obtain first environmental feature data in a spatial dimension affecting icing and second environmental feature data in a time dimension affecting icing. Then, the first environmental feature data and the second environmental feature data are subjected to correlation analysis to obtain a feature correlation degree between the first environmental feature data and the second environmental feature data. In a case where the feature correlation degree meets a feature correlation condition, a three-dimensional environmental model of the target power grid region is constructed based on the first environmental feature data and the second environmental feature data, and the three-dimensional environmental model is used for icing prediction of the target power grid region. The server 104 can send the icing prediction result to the terminal 102, so that the user can obtain the icing prediction result through the terminal 102. Of course, the user can also initiate an icing prediction instruction through the terminal 102, which is not limited in the embodiment.

[0030] In one exemplary embodiment, as shown in Figure 2 , a multi-source data-based power grid icing prediction initial field modeling method is provided. Taking the server 104 in Figure 1 as an example, the method includes the following steps:

[0031] Step S202, acquiring multi-source environmental data of a target power grid region.

[0032] The target power grid region can refer to a power grid region that needs to be subjected to line icing prediction. The multi-source environmental data refers to a data set from multiple different sources, such as different sensors, databases or systems. These data can be heterogeneous or homogeneous. The multi-source environmental data can include at least one of meteorological data, terrain data, power grid equipment state data, historical icing data, etc. The meteorological data can include at least one of temperature, humidity, wind speed, precipitation, etc. The terrain data can include at least one of elevation, slope, topography, etc. The power grid equipment state data can include at least one of conductor current, voltage, tension, etc. The historical icing data refers to icing occurrence records of the target power grid region in a historical time period, which can include at least one of icing grade, icing occurrence start and end time, icing thickness, etc.

[0033] In the embodiment, the server can access multiple data source interfaces in advance, and each data source interface can correspond to a sensor, a database or a system. The server can acquire multi-source data related to power grid icing prediction collected by the sensor, the database or the system through the data source interfaces.

[0034] Step S204, performing feature extraction on the multi-source environmental data to obtain first environmental feature data in a spatial dimension affecting icing and second environmental feature data in a time dimension affecting icing.

[0035] The feature data extraction in the spatial dimension refers to a process of quantifying spatial distribution rules related to ice formation from static geographic environment attributes of the target power grid region. The first environment feature data is related data that affects ice formation in the spatial dimension, including but not limited to at least one of a height change rate, i.e., a height change amount per unit distance, a terrain slope, a line direction azimuth, a humidity distribution gradient, and the like. The feature data extraction in the time dimension refers to a process of capturing process indicators related to ice formation from dynamic time series data, including but not limited to at least one of a temperature / humidity change rate, a wind speed extreme value frequency, i.e., a number of times that a wind speed reaches an extreme value per unit time, a conductor tension fluctuation amplitude in a sliding window, an ice duration, i.e., a cumulative duration of continuous icing, and the like.

[0036] In some embodiments, a calculation expression of the height change rate is as follows:

[0037]

[0038] wherein, is a height value, is a height change rate of a grid point .

[0039] A calculation expression of the terrain slope is as follows:

[0040]

[0041] The line direction azimuth can be calculated by a difference between longitude and latitude of adjacent line nodes.

[0042] A calculation expression of the humidity distribution gradient is as follows:

[0043]

[0044] wherein, is a relative humidity field, is a spatial gradient operator.

[0045] In some embodiments, a calculation expression of the temperature / humidity change rate is as follows:

[0046]

[0047] wherein, is a temperature / humidity value at a time point , and is a time interval.

[0048] A calculation expression of the wind speed extreme value frequency is as follows:

[0049]

[0050] wherein, is the first environment feature data, is the second environment feature data, is the wind speed value at the time t, is a wind speed threshold value, which can be determined according to actual conditions, for example may be 10 or other values. is an indicator function, outputting 1 when the condition is true, otherwise outputting 0.

[0051] The calculation expression of the conductor tension fluctuation amplitude is:

[0052]

[0053] wherein, represents the actual tension value of the conductor at time t, in units of kN (kiloNewtons). represents the time step, [ t - Δt, t] represents the statistical time window, represents the instantaneous value of the maximum tension in the statistical time window, represents the instantaneous value of the minimum tension in the statistical time window.

[0054] The icing duration is the cumulative duration of the ice thickness in the continuous period, wherein, is the lower limit of the ice thickness, in units of millimeters, in an example, may be 3, of course other values can also be set according to actual conditions, and the present embodiment does not limit this.

[0055] In the present embodiment, after obtaining the multi-source environment data of the target power grid area, feature extraction can be performed from the spatial dimension and the time dimension respectively, to obtain the first environment feature data affecting the formation of icing in the spatial dimension, and the second environment feature data affecting the formation of icing in the time dimension.

[0056] In step S206, the first environment feature data and the second environment feature data are analyzed for relevance, to obtain the feature correlation degree between the first environment feature data and the second environment feature data.

[0057] wherein, the feature correlation degree refers to the correlation between the first environment feature data and the second environment feature data, such as the Pearson correlation coefficient between the first environment feature data and the second environment feature data.

[0058] In this embodiment, after the first environmental feature data and the second environmental feature data are extracted, the first environmental feature data and the second environmental feature data can be further analyzed for correlation, so as to obtain the feature correlation degree between the first environmental feature data and the second environmental feature data. In this embodiment, the Pearson correlation coefficient is preferably used as a statistical index for measuring the correlation degree between the first environmental feature data and the second environmental feature data. In an example, the expression of the feature data correlation analysis is as follows:

[0059]

[0060] wherein, is the Pearson correlation coefficient between the first environmental feature data x and the second environmental feature data y, and the value range is [-1, 1], greater than zero means that the two feature data are positively correlated, less than zero means that the two feature data are negatively correlated. represents the ith first environmental feature data, represents the ith second environmental feature data, represents the mean value of each first environmental feature data, represents the mean value of each second environmental feature data, is the total number of data points.

[0061] In step S208, when the feature correlation degree meets the feature correlation condition, a three-dimensional environmental model of the target power grid region is constructed based on the first environmental feature data and the second environmental feature data. The three-dimensional environmental model is used for ice prediction of the target power grid region.

[0062] The feature correlation condition refers to a prerequisite condition that the feature correlation degree between the first environmental feature data and the second environmental feature data needs to meet. It can be a feature correlation degree threshold or other conditions, which can be set according to actual conditions. If the feature correlation degree meets the condition, for example, the feature correlation degree is greater than or equal to the feature correlation degree threshold, it means that the first environmental feature data and the second environmental feature data have strong correlation and can be used as the input data set for subsequent three-dimensional environmental modeling. If the feature correlation degree does not meet the feature correlation condition, for example, the feature correlation degree is less than the feature correlation degree threshold, it means that the correlation between the first environmental feature data and the second environmental feature data is weak and is not used as the input data set for subsequent three-dimensional environmental modeling. The three-dimensional environmental model can refer to the initial environmental field model of the target power grid region. The initial field is the initial condition used to start the calculation in numerical simulation or experimental analysis. In this embodiment, the initial environmental field can include the initial physical state distribution of the target power grid region, such as the temperature, humidity, wind speed, etc. distribution of the target power grid region.

[0063] In this embodiment, after obtaining the feature correlation degree between the first environmental feature data and the second environmental feature data, the feature correlation degree can be compared with a preset feature correlation condition. If the feature correlation degree meets the feature correlation condition, the first environmental feature data and the second environmental feature data are taken as input data sets for subsequent three-dimensional environmental modeling. If the feature correlation degree does not meet the feature correlation condition, the first environmental feature data and the second environmental feature data can be discarded or not processed.

[0064] In this embodiment, first, multi-source environmental data of a target power grid region is obtained, and feature extraction is performed on the multi-source environmental data to obtain first environmental feature data affecting icing in the spatial dimension and second environmental feature data affecting icing in the time dimension. Then, correlation analysis is performed on the first environmental feature data and the second environmental feature data to obtain a feature correlation degree between the first environmental feature data and the second environmental feature data. In the case where the feature correlation degree meets a feature correlation condition, a three-dimensional environmental model of the target power grid region is constructed based on the first environmental feature data and the second environmental feature data, and the three-dimensional environmental model is used for icing prediction of the target power grid region. In this way, on the one hand, environmental feature data affecting icing is extracted from two dimensions of time and space, ensuring the accuracy and reliability of the environmental feature data and improving the adaptability and generalization ability of the subsequent three-dimensional environmental model to the spatio-temporal distribution of icing. On the other hand, by analyzing the correlation degree between the first environmental feature data and the second environmental feature data, the synergistic effect between features in the spatial dimension and the time dimension can be effectively identified, and the deviation of icing prediction caused by feature conflict can be avoided. Therefore, the scheme can effectively improve the accuracy of icing prediction.

[0065] In one exemplary embodiment, as shown in Figure 3 the multi-source environmental data of the target power grid region is obtained, including:

[0066] In step S302, initial environmental data of a target power grid region under different data sources is obtained.

[0067] In step S304, data format conversion is performed on each initial environmental data to obtain intermediate environmental data conforming to a target data format.

[0068] The initial environmental data refers to multi-source environmental data in an initial state, and the initial environmental data has not been subjected to any data processing. The data format conversion is a process of converting the format of the initial environmental data. The target data format is a preset expected data format.

[0069] In this embodiment, first, the initial environment data of the target power grid region under different data sources is obtained, and the initial environment data is standardized, that is, data format conversion, the purpose is to unify the initial environment data according to the field structure to standard key-value pairs or table structure, and ensure that the same data has consistent field name, data type and data unit, etc. Specifically, the initial environment data is uniformly named, and based on the pre-established field mapping table, the fields with the same meaning but different names are uniformly mapped to the standard field name. The fields with physical quantities are converted, and the same physical meaning data is ensured to have a unified unit. Based on the results of field standardization and unit unification, at least one of the hierarchical labels, data type label, space-time label, and data source label, is added to each initial environment data. Among them, the data type label can include but is not limited to meteorological data label, terrain data label, and power grid equipment data label. The space-time label can include time stamp and grid position. The data source label is used to identify the data source of the initial environment data. It can be understood that after the subsequent space-time alignment is completed, the data structure of each multi-source environment data can be updated again, such as updating the space-time label.

[0070] In some embodiments, a standardized time stamp can also be added to each initial environment data obtained, in the coordinated universal time format, with a uniform granularity of one hour. For example, if the original time stamp is , the standardized time stamp is , wherein is the standardized time stamp.

[0071] In some embodiments, in addition to time stamp standardization, data source identification coding can also be performed, that is, a unique data source identification code is added to each initial environment data, and the coding rule is , wherein is the data type code, is the administrative region or site code, is the device number or measurement point code. The specific coding rule can be determined according to the actual situation, and this embodiment does not limit it.

[0072] Step S306, space-time alignment of each intermediate environment data is performed to obtain multi-source environment data after space-time alignment.

[0073] Among them, space-time alignment refers to the process of integrating multi-source environment data after data format conversion to a unified space-time reference, so that the data has a basic fusion. Space-time alignment can include time alignment and space alignment.

[0074] In this embodiment, after the data format conversion of each initial environment data is completed, the converted intermediate environment data is subjected to time alignment and space alignment, so as to obtain multi-source environment data.

[0075] In this embodiment, the data format conversion is performed on the initial environment data, so as to improve the quality of the multi-source environment data, lay a data foundation for subsequent three-dimensional environment modeling, avoid the icing prediction deviation caused by the difference between data sources, and thus ensure the accuracy of the icing prediction.

[0076] In one exemplary embodiment, as shown in Figure 4 The time and space alignment of each intermediate environment data is performed to obtain multi-source environment data after time and space alignment, which includes:

[0077] In step S402, the time stamp of each intermediate environment data is obtained.

[0078] In step S404, the time alignment weight of each intermediate environment data is determined based on the time difference between the time stamp and the preset alignment time.

[0079] The intermediate environment data refers to the data in the initial environment data that has been preliminarily processed but has not been subjected to time and space alignment. The time stamp is the data collection time after standardization processing. The preset alignment time refers to a standard time point that is set in advance, such as an hour time, a half hour time, etc., which can be set according to actual conditions. The time difference refers to the difference between the time stamp of the intermediate environment data and the preset alignment time. The time alignment weight is a contribution weight determined according to the time difference, which satisfies the principle that the closer to the preset alignment time, the higher the weight.

[0080] In this embodiment, for each intermediate environment data, further time and space alignment is required, and the process of time alignment is described in this embodiment. First, the time stamp of each intermediate environment data is obtained, and the difference between the time stamp and the preset standard time point is calculated, so as to determine the time alignment weight matched with the difference.

[0081] In step S406, the time alignment of each intermediate environment data is performed according to the time alignment weight, so as to obtain multi-source environment data.

[0082] In this embodiment, the time window aggregation method is used to align the intermediate environment data with different time efficiencies to the preset standard time point, and the expression is as follows:

[0083]

[0084] wherein, is the standard value of the intermediate environment data after time alignment at the preset alignment time point , is the a data value of the intermediate environment data falling into the time window, for the first time alignment weight of the intermediate environment data, a number of valid data falling into the time window. It can be understood that the time window is centered on the preset alignment time point , and the time difference allowed to float is obtained.

[0085] In the embodiment, by time aligning the multi-source environment data, the time dislocation conflict of the multi-source environment data can be eliminated, and the multi-source environment data can be fused and analyzed under a unified time reference, so as to improve the accuracy of the three-dimensional environment model, and further improve the accuracy of the icing prediction.

[0086] In an exemplary embodiment, as Figure 5 shown, the spatio-temporal alignment of the intermediate environment data is performed to obtain the multi-source environment data after spatio-temporal alignment, and the method further comprises:

[0087] In step S502, the spatial position alignment of the intermediate environment data is performed to obtain the target environment data after spatial position alignment.

[0088] The spatial position alignment, i.e., the spatial alignment, refers to mapping the intermediate environment data with different spatial resolutions to a unified grid structure. The unified grid structure can be a grid structure with a fixed spatial scale generated based on the path center line of the target power grid region, and can be expressed as , where m represents the number of grid points. The target environment data is data after spatial alignment.

[0089] In the embodiment, for each intermediate environment data, the original coordinates of the intermediate environment data can be converted to a standard coordinate system, and each data is labeled with a grid number, so as to output the target environment data.

[0090] In step S504, the interpolation of the target environment data is performed to obtain the multi-source environment data.

[0091] The interpolation refers to the operation of completing the data of the grid missing area.

[0092] In the embodiment, the data value of each grid point can be interpolated by using a bilinear interpolation method, and the interpolation expression is as follows:

[0093]

[0094] wherein, is the value of the grid point after resampling, , , , are four adjacent point values around , , are normalized positions relative to the lower left corner point in the original lattice.

[0095] Further, the spatio-temporal joint interpolation strategy is adopted to complete the multi-source environmental data with missing values after time synchronization and resampling, and the expression is as follows:

[0096]

[0097] wherein, is the completed multi-source environmental data value, is the effective value at the same spatial position at the adjacent time point, and there are a total of a, is the effective value of the adjacent spatial position at the same time point, and there are a total of b. After time synchronization, spatial resampling and missing value completion, the standardized multi-source environmental data can be generated.

[0098] In some embodiments, the execution order of time alignment and spatial alignment can be determined according to specific application scenarios. In this embodiment, the order of time alignment first and then spatial alignment is preferred.

[0099] In this embodiment, the intermediate environmental data is mapped to a unified spatial grid by bilinear difference, the spatial position is aligned, and the missing values in the uncovered area are completed by joint spatio-temporal difference, so as to finally generate standardized multi-source environmental data, improve the accuracy of multi-source environmental data, and improve the accuracy of ice prediction.

[0100] In an exemplary embodiment, as shown in Figure 6 , the initial environmental data of the target power grid area under different data sources is obtained, including:

[0101] Step S602, obtaining a plurality of candidate environmental data of the target power grid area under different data sources.

[0102] Step S604, based on the ice prediction demand of the target power grid area, screening the target environmental data matched with the ice prediction demand from the plurality of candidate environmental data.

[0103] Wherein, the candidate environmental data can be the original environmental data set of the target power grid area under different data sources. The ice prediction demand refers to the related demand when the ice prediction is performed, such as accuracy requirement, timeliness requirement, etc. The target environmental data is the environmental data matched with the ice prediction demand.

[0104] ​In this embodiment, multiple candidate environmental data from different sources for the target power grid area are first acquired through multiple data source interfaces. Data is then filtered based on the current icing prediction requirements of the target power grid area to select the environmental data that meets these requirements. This ensures that the multi-source environmental data matches actual business needs, further improving the accuracy of icing predictions.

[0105] Step S606: The target environmental data and the historical ice coverage data of the target power grid area are used together as the initial environmental data of the target power grid area.

[0106] In this embodiment, the target environmental data and historical ice coverage data for the target grid area are combined as the initial environmental data for the target grid area. By incorporating historical ice coverage data, the physical laws of ice coverage in the target grid area can be analyzed, allowing for deviation correction of multi-source environmental data, further improving the accuracy of the multi-source environmental data.

[0107] In an exemplary embodiment, the method for initial field modeling of power grid icing prediction based on multi-source data further includes: obtaining characteristic constraint conditions for the first environmental characteristic data and the second environmental characteristic data respectively.

[0108] Among them, the characteristic constraint conditions refer to the constraint conditions for the environmental characteristic data, which may include the constraint conditions for the first environmental characteristic data and the constraint conditions for the second environmental characteristic data. For example, the constraint conditions for the second environmental characteristic data may include temperature constraints, which eliminate or set to zero all environmental characteristic data with temperatures above the freezing point, and only retain environmental characteristic data that meet the freezing point temperature conditions. It may also include wind speed constraints, which impose safety upper limit constraints on wind speed values ​​to avoid constructing data in wind-induced ice melting or aerodynamic de-icing risk areas. Data that exceeds the wind speed upper limit is considered to have an unstable flow field and needs to be eliminated. The constraint conditions for the first environmental characteristic data may include joint constraints on humidity and slope, combined with the humidity gradient. and terrain slope , a joint threshold rule is applied to determine whether the point has high ice cover potential. The judgment condition is and ,in, Humidity at point The spatial gradient of is the terrain slope, is the lower limit threshold of humidity gradient, is the upper slope threshold.

[0109] In a case where the feature correlation degree meets the feature correlation condition, a three-dimensional environment model of the target power grid region is constructed based on the first environment feature data and the second environment feature data, including: in a case where the feature correlation degree meets the feature correlation condition, and the first environment feature data and the second environment feature data respectively meet the respective corresponding feature constraint conditions, a three-dimensional environment model of the target power grid region is constructed based on the first environment feature data and the second environment feature data.

[0110] In a specific embodiment, as shown in FIG. 1, the steps of the multi-source data-based power grid icing prediction initial field modeling method include: Figure 7

[0111] S1, obtaining initial environment data of a target power grid region, including meteorological data, terrain data, power grid equipment state data and historical icing records;

[0112] S2, performing data format conversion on each initial environment data, including timestamp labeling, data source identification and formatting processing, to obtain each intermediate environment data conforming to a target data format;

[0113] S3, performing time alignment, spatial resampling and missing completion on each intermediate environment data to obtain standardized multi-source environment data;

[0114] S4, performing feature extraction on the multi-source environment data to obtain first environment feature data such as elevation change rate, terrain slope, line direction azimuth angle and humidity distribution gradient in the spatial dimension, and second environment feature data such as humidity change rate, wind speed extreme value frequency, conductor tension fluctuation amplitude and icing duration in the time dimension;

[0115] S5, performing correlation analysis on the first environment feature data and the second environment feature data to obtain a Pearson correlation coefficient between the first environment feature data and the second environment feature data;

[0116] S6, in a case where the Pearson correlation coefficient meets the feature correlation condition, and the first environment feature data and the second environment feature data respectively meet the respective corresponding feature constraint conditions, a three-dimensional environment model of the target power grid region is constructed based on the first environment feature data and the second environment feature data;

[0117] S7, performing icing prediction on the target power grid region through the three-dimensional environment model.

[0118] ​In this embodiment, first, multi-source environmental data of a target power grid region is acquired, and feature extraction is performed on the multi-source environmental data to obtain first environmental feature data affecting icing in the spatial dimension and second environmental feature data affecting icing in the time dimension. Then, correlation analysis is performed on the first environmental feature data and the second environmental feature data to obtain a feature correlation degree between the first environmental feature data and the second environmental feature data. In the case where the feature correlation degree meets a feature correlation condition, a three-dimensional environmental model of the target power grid region is constructed based on the first environmental feature data and the second environmental feature data, and the three-dimensional environmental model is used for icing prediction of the target power grid region. In this way, on the one hand, environmental feature data affecting icing is extracted from the time and space dimensions, ensuring the accuracy and reliability of the environmental feature data and improving the adaptability and generalization capability of the subsequent three-dimensional environmental model for the spatio-temporal distribution of icing. On the other hand, by analyzing the correlation degree between the first environmental feature data and the second environmental feature data, the synergistic effect between the features in the spatial dimension and the time dimension can be effectively identified, and the icing prediction deviation caused by feature conflicts can be avoided. Therefore, the scheme can effectively improve the accuracy of icing prediction.

[0119] It should be understood that, although each step in the flowchart involved in each embodiment as described above is shown in sequence according to the arrow, these steps are not necessarily executed in sequence according to the arrow. Unless otherwise specified herein, the execution of these steps is not strictly limited in sequence, and these steps can be executed in other sequences. Moreover, at least part of the steps in the flowchart involved in each embodiment as described above can include multiple steps or stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution sequence of these steps or stages is not necessarily sequential, but can be executed alternately or alternately with at least part of other steps or steps or stages in other steps. It can be understood that the steps in different embodiments can be freely combined as needed, and various non-contradictory schemes formed by the combination are within the scope of protection of the present application.

[0120] Based on the same inventive concept, the embodiments of the present application also provide a multi-source data-based power grid icing prediction initial field modeling device for implementing the above-mentioned multi-source data-based power grid icing prediction initial field modeling method. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme described in the above method, so the specific limitations in one or more multi-source data-based power grid icing prediction initial field modeling device embodiments provided below can be referred to the limitations of the multi-source data-based power grid icing prediction initial field modeling method in the above, which will not be repeated here.

[0121] In one exemplary embodiment, as Figure 8As shown, an initial field modeling device for power grid icing prediction based on multi-source data is provided, comprising: a multi-source data acquisition module 802 configured to acquire multi-source environmental data of a target power grid region; a feature data extraction module 804 configured to perform feature extraction on the multi-source environmental data to obtain first environmental feature data affecting icing in a spatial dimension and second environmental feature data affecting icing in a time dimension; a feature data correlation analysis module 806 configured to perform correlation analysis on the first environmental feature data and the second environmental feature data to obtain a feature correlation degree between the first environmental feature data and the second environmental feature data; and a three-dimensional environmental modeling module 808 configured to, when the feature correlation degree meets a feature correlation condition, construct a three-dimensional environmental model of the target power grid region based on the first environmental feature data and the second environmental feature data; and the three-dimensional environmental model is used for icing prediction of the target power grid region.

[0122] In one embodiment, the multi-source data acquisition module 802 is further configured to: acquire initial environmental data of the target power grid region under different data sources; perform data format conversion on each initial environmental data respectively to obtain intermediate environmental data conforming to a target data format; and perform spatio-temporal alignment on each intermediate environmental data to obtain multi-source environmental data after spatio-temporal alignment.

[0123] In one embodiment, the multi-source data acquisition module 802 is further configured to: acquire timestamps of each intermediate environmental data; determine a time alignment weight of each intermediate environmental data based on a time difference between the timestamp and a preset alignment time; and perform time alignment on each intermediate environmental data according to the time alignment weight to obtain the multi-source environmental data.

[0124] In one embodiment, the multi-source data acquisition module 802 is further configured to: perform spatial position alignment on each intermediate environmental data to obtain each target environmental data after spatial position alignment; and perform interpolation on each target environmental data to obtain the multi-source environmental data.

[0125] In one embodiment, the multi-source data acquisition module 802 is further configured to: acquire a plurality of candidate environmental data of the target power grid region under different data sources; select target environmental data matching an icing prediction requirement of the target power grid region from the plurality of candidate environmental data based on the icing prediction requirement; and use the target environmental data and historical icing data of the target power grid region together as initial environmental data of the target power grid region.

[0126] In one embodiment, the initial field modeling device for power grid icing prediction based on multi-source data is further configured to: acquire feature constraint conditions for the first environmental feature data and the second environmental feature data respectively.

[0127] In one of the embodiments, the three-dimensional environment modeling module 808 is further configured to: in a case where the feature correlation degree meets the feature correlation condition, and the first environment feature data and the second environment feature data respectively meet the respective corresponding feature constraint conditions, construct a three-dimensional environment model of the target power grid region based on the first environment feature data and the second environment feature data.

[0128] The modules in the above power grid icing prediction initial field modeling device based on multi-source data can be realized by software, hardware, and combinations thereof, in whole or in part. The modules can be embedded in or independent of a processor in a computer device in hardware form, or stored in a memory in a computer device in software form, so as to be called and executed by a processor to perform the operations corresponding to the modules.

[0129] In one exemplary embodiment, a computer device, which can be a server, is provided, and an internal structure diagram of the computer device can be as shown in Figure 9 The computer device includes a processor, a memory, an input / output interface (I / O), and a communication interface. The processor, the memory, and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operating system and the computer program in the non-volatile storage medium to run. The database of the computer device is configured to store power grid icing prediction initial field modeling data based on multi-source data. The input / output interface of the computer device is configured to exchange information between the processor and external devices. The communication interface of the computer device is configured to communicate with external terminals through network connection. The computer program is executed by the processor to implement a power grid icing prediction initial field modeling method based on multi-source data.

[0130] Those skilled in the art can understand that Figure 9 The structure shown in the above

[0131] In an example embodiment, a computer device is provided, comprising a memory and a processor, the memory storing a computer program, and the processor implementing the following steps when executing the computer program: obtaining multi-source environment data of a target power grid region; performing feature extraction on the multi-source environment data to obtain first environment feature data affecting icing in a spatial dimension and second environment feature data affecting icing in a time dimension; performing correlation analysis on the first environment feature data and the second environment feature data to obtain a feature correlation degree between the first environment feature data and the second environment feature data; in a case where the feature correlation degree meets a feature correlation condition, constructing a three-dimensional environment model of the target power grid region based on the first environment feature data and the second environment feature data; and the three-dimensional environment model is used for icing prediction of the target power grid region.

[0132] In an example embodiment, the processor further implements the following steps when executing the computer program: obtaining initial environment data of the target power grid region under different data sources; performing data format conversion on each initial environment data to obtain intermediate environment data conforming to a target data format; and performing space-time alignment on each intermediate environment data to obtain multi-source environment data after space-time alignment.

[0133] In an example embodiment, the processor further implements the following steps when executing the computer program: obtaining a timestamp of each intermediate environment data; determining a time alignment weight of each intermediate environment data based on a time difference between the timestamp and a preset alignment time; and performing time alignment on each intermediate environment data according to the time alignment weight to obtain the multi-source environment data.

[0134] In an example embodiment, the processor further implements the following steps when executing the computer program: performing spatial position alignment on each intermediate environment data to obtain each target environment data after spatial position alignment; and performing interpolation on each target environment data to obtain the multi-source environment data.

[0135] In an example embodiment, the processor further implements the following steps when executing the computer program: obtaining a plurality of candidate environment data of the target power grid region under different data sources; selecting target environment data matched with icing prediction requirements from the plurality of candidate environment data based on the icing prediction requirements of the target power grid region; and taking the target environment data and historical icing data of the target power grid region as initial environment data of the target power grid region.

[0136] In an example embodiment, the processor further implements the following steps when executing the computer program: obtaining feature constraint conditions for the first environment feature data and the second environment feature data, respectively.

[0137] In an embodiment, the processor, when executing the computer program, also implements the following steps: constructing, based on the first environment feature data and the second environment feature data, a three-dimensional environment model of the target power grid region, in a case where the feature correlation degree meets a feature correlation condition and the first environment feature data and the second environment feature data respectively meet respective corresponding feature constraint conditions.

[0138] In an embodiment, a computer readable storage medium is provided, and the computer readable storage medium stores a computer program. The computer program, when executed by a processor, implements the following steps: obtaining multi-source environment data of a target power grid region; performing feature extraction on the multi-source environment data to obtain first environment feature data affecting icing in a spatial dimension and second environment feature data affecting icing in a time dimension; performing correlation analysis on the first environment feature data and the second environment feature data to obtain a feature correlation degree between the first environment feature data and the second environment feature data; constructing, based on the first environment feature data and the second environment feature data, a three-dimensional environment model of the target power grid region, in a case where the feature correlation degree meets a feature correlation condition; and the three-dimensional environment model is used for icing prediction of the target power grid region.

[0139] In an embodiment, the computer program, when executed by the processor, also implements the following steps: obtaining initial environment data of the target power grid region under different data sources; performing data format conversion on each initial environment data respectively to obtain intermediate environment data conforming to a target data format; and performing space-time alignment on each intermediate environment data to obtain multi-source environment data after space-time alignment.

[0140] In an embodiment, the computer program, when executed by the processor, also implements the following steps: obtaining a timestamp of each intermediate environment data; determining a time alignment weight of each intermediate environment data based on a time difference between the timestamp and a preset alignment time; and performing time alignment on each intermediate environment data according to the time alignment weight to obtain the multi-source environment data.

[0141] In an embodiment, the computer program, when executed by the processor, also implements the following steps: performing spatial position alignment on each intermediate environment data to obtain each target environment data after spatial position alignment; and performing interpolation on each target environment data to obtain the multi-source environment data.

[0142] In an embodiment, the computer program, when executed by the processor, also implements the following steps: obtaining a plurality of candidate environment data of the target power grid region under different data sources; selecting, based on an icing prediction requirement of the target power grid region, target environment data matched with the icing prediction requirement from the plurality of candidate environment data; and taking the target environment data and historical icing data of the target power grid region together as initial environment data of the target power grid region.

[0143] In an embodiment, the computer program, when executed by the processor, further implements the following steps: acquiring feature constraint conditions for the first environment feature data and the second environment feature data, respectively.

[0144] In an embodiment, the computer program, when executed by the processor, further implements the following steps: in a case where the feature correlation degree meets the feature correlation condition, and the first environment feature data and the second environment feature data respectively meet the respective corresponding feature constraint conditions, constructing a three-dimensional environment model of the target power grid region based on the first environment feature data and the second environment feature data.

[0145] In an embodiment, a computer program product is provided, comprising a computer program which, when executed by a processor, implements the following steps: acquiring multi-source environment data of a target power grid region; performing feature extraction on the multi-source environment data to obtain first environment feature data affecting icing in a spatial dimension and second environment feature data affecting icing in a time dimension; performing correlation analysis on the first environment feature data and the second environment feature data to obtain a feature correlation degree between the first environment feature data and the second environment feature data; in a case where the feature correlation degree meets a feature correlation condition, constructing a three-dimensional environment model of the target power grid region based on the first environment feature data and the second environment feature data; and the three-dimensional environment model is used for icing prediction of the target power grid region.

[0146] In an embodiment, the computer program, when executed by the processor, further implements the following steps: acquiring initial environment data of the target power grid region under different data sources; performing data format conversion on each initial environment data to obtain intermediate environment data conforming to a target data format; and performing space-time alignment on each intermediate environment data to obtain multi-source environment data after space-time alignment.

[0147] In an embodiment, the computer program, when executed by the processor, further implements the following steps: acquiring a timestamp of each intermediate environment data; determining a time alignment weight of each intermediate environment data based on a time difference between the timestamp and a preset alignment time; and performing time alignment on each intermediate environment data according to the time alignment weight to obtain the multi-source environment data.

[0148] In an embodiment, the computer program, when executed by the processor, further implements the following steps: performing spatial position alignment on each intermediate environment data to obtain each target environment data after spatial position alignment; and performing interpolation on each target environment data to obtain the multi-source environment data.

[0149] In an embodiment, the computer program, when executed by the processor, further implements the following steps: obtaining a plurality of candidate environmental data of the target power grid area under different data sources; based on the icing prediction requirement of the target power grid area, screening target environmental data matched with the icing prediction requirement from the plurality of candidate environmental data; and taking the target environmental data and the historical icing data of the target power grid area as initial environmental data of the target power grid area.

[0150] In an embodiment, the computer program, when executed by the processor, further implements the following steps: obtaining feature constraint conditions for the first environmental feature data and the second environmental feature data, respectively.

[0151] In an embodiment, the computer program, when executed by the processor, further implements the following steps: in a case where the feature correlation degree meets the feature correlation condition, and the first environmental feature data and the second environmental feature data meet the respective corresponding feature constraint conditions, constructing a three-dimensional environmental model of the target power grid area based on the first environmental feature data and the second environmental feature data.

[0152] It should be noted that the user information (including but not limited to user equipment information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the present application are all information and data authorized by the user or authorized by all parties, and the collection, use and processing of the relevant data need to comply with relevant regulations.

[0153] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer readable storage medium, and when executed, can include the processes of the above-mentioned embodiment methods. Any reference to memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. The non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical storage, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. The volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration but not limitation, the RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The database involved in the embodiments provided in the present application can include at least one of a relational database and a non-relational database. The non-relational database can include a distributed database based on a block chain, etc., without being limited thereto. The processor involved in the embodiments provided in the present application can be a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., without being limited thereto.

[0154] The technical features of the above embodiments can be combined arbitrarily. In order to make the description simple, all possible combinations of the technical features in the above embodiments are not described, but as long as the combinations of the technical features do not exist, they should be considered as the scope of the present application.

[0155] The above-described embodiments are merely illustrative of several embodiments of the present application, and the description is relatively specific and detailed, but should not be understood as a limitation on the scope of the patent. It should be noted that for those skilled in the art, without departing from the concept of the present application, a number of modifications and improvements can be made, which are all within the scope of the present application. Therefore, the scope of protection of the present application should be subject to the appended claims.

Claims

1. A method for initial field modeling of power grid icing prediction based on multi-source data, characterized in that: The method comprises: Acquire multi-source environmental data of the target power grid area; Performing feature extraction on the multi-source environmental data to obtain first environmental feature data affecting icing in a spatial dimension and second environmental feature data affecting icing in a temporal dimension; Performing a correlation analysis on the first environmental feature data and the second environmental feature data to obtain a feature correlation degree between the first environmental feature data and the second environmental feature data; When the feature correlation degree meets the feature correlation condition, a three-dimensional environmental model of the target power grid area is constructed based on the first environmental feature data and the second environmental feature data; the three-dimensional environmental model is used to predict icing of the target power grid area.

2. The method according to claim 1, characterized in that The acquiring of multi-source environmental data of the target power grid area includes: Acquiring initial environmental data of the target power grid area from different data sources; Performing data format conversion on each of the initial environment data to obtain each intermediate environment data that conforms to the target data format; The intermediate environmental data are temporally and spatially aligned to obtain the temporally and spatially aligned multi-source environmental data.

3. The method according to claim 2, characterized in that The performing spatiotemporal alignment on each of the intermediate environmental data to obtain the spatiotemporally aligned multi-source environmental data includes: Obtaining the timestamp of each intermediate environment data; Determining a time alignment weight for each of the intermediate environment data based on a time difference between the timestamp and a preset alignment time; According to the time alignment weight, time alignment is performed on each of the intermediate environment data to obtain multi-source environment data.

4. The method according to claim 2, characterized in that The performing spatiotemporal alignment on each of the intermediate environmental data to obtain the spatiotemporally aligned multi-source environmental data further includes: Performing spatial position alignment on each of the intermediate environment data to obtain each target environment data after spatial position alignment; Interpolate each of the target environment data to obtain multi-source environment data.

5. The method according to claim 2, characterized in that The obtaining of initial environmental data of the target power grid area from different data sources includes: Acquire multiple candidate environmental data of the target power grid area from different data sources; Based on the predicted icing demand of the target power grid area, screening target environmental data that matches the predicted icing demand from the multiple candidate environmental data; The target environmental data and the historical ice coverage data of the target power grid area are used together as the initial environmental data of the target power grid area.

6. The method according to claim 1, characterized in that The method further comprises: respectively obtaining characteristic constraint conditions for the first environmental characteristic data and the second environmental characteristic data; When the feature correlation degree satisfies a feature correlation condition, constructing a three-dimensional environment model of the target power grid area based on the first environment feature data and the second environment feature data includes: When the feature correlation degree satisfies the feature correlation condition, and the first environmental feature data and the second environmental feature data respectively satisfy their corresponding feature constraint conditions, a three-dimensional environmental model of the target power grid area is constructed based on the first environmental feature data and the second environmental feature data.

7. A power grid icing prediction initial field modeling device based on multi-source data, characterized in that: The device comprises: A multi-source data acquisition module is used to obtain multi-source environmental data of the target power grid area; a feature data extraction module, configured to extract features from the multi-source environmental data to obtain first environmental feature data affecting icing in a spatial dimension and second environmental feature data affecting icing in a temporal dimension; a feature data association analysis module, configured to perform association analysis on the first environmental feature data and the second environmental feature data to obtain a feature association degree between the first environmental feature data and the second environmental feature data; A three-dimensional environment modeling module is used to construct a three-dimensional environment model of the target power grid area based on the first environment feature data and the second environment feature data when the feature correlation degree meets the feature correlation condition; the three-dimensional environment model is used to predict icing of the target power grid area.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images