Transmission line tower level icing thickness gridding method and system

Abstract

The invention relates to the technical field of power transmission line monitoring, and discloses a power transmission line tower-level icing thickness gridding method, which comprises the following steps: acquiring icing monitoring data and meteorological geographic information data; performing interpolation and preprocessing on the data; performing artificial intelligence model training and testing based on the interpolated and preprocessed data; using a Bayesian parameter optimization method to optimize artificial intelligence model parameters; and generating a southern region live lattice point distribution diagram based on the optimal artificial intelligence model parameters. According to the method, through combination of multi-source data and utilization of deep learning and Bayesian optimization technologies, fine grid prediction of the power transmission line tower-level icing thickness is realized, the prediction precision is improved, and an important basis is provided for power transmission line icing risk assessment and preventive maintenance.

Description

Technical Field

[0001] The invention relates to the technical field of power grids, and in particular to a gridding method based on ice thickness at tower level of power transmission lines. Background Art

[0002] Transmission line icing disaster is one of the most serious meteorological disasters in the power system, which seriously threatens the safe and stable operation of the power grid.

[0003] To reduce the losses caused by icing disasters, it is necessary to accurately predict the thickness of conductor ice, and the occurrence and development of conductor ice is closely related to the changes in meteorological elements. At present, the Southern Power Grid has carried out tower-level numerical forecasting to achieve numerical forecasting of icing phenomena at differentiated spatial scales. However, the atmosphere is a highly nonlinear system, and the icing forecast results are very sensitive to the setting of the initial icing field. The lack of an initial icing field based on observation data leads to a high underreporting rate of icing forecasts, and this error will accumulate over time, eventually leading to distortion of the forecast results. Establishing an initial icing numerical forecast field by integrating geographic information data and real-time icing observation data is an effective means to achieve accurate icing forecasts for transmission lines. However, although an icing monitoring system has been deployed in the five southern provinces, the limited monitoring data cannot reflect the distribution of icing in the entire area. At the same time, the terrain conditions in the five southern provinces are complex, and the icing growth characteristics are affected by multiple factors such as micrometeorology and microtopography. Therefore, how to make full use of icing monitoring data to optimize the initial icing forecast field is a great challenge. The establishment of the initial icing grid field capability for numerical forecasting of power grids that integrates self-built icing monitoring is a prerequisite for realizing differentiated spatial-scale numerical forecasting at the tower level. Based on the refined spatial distribution output of real-time icing, the precise spatial distribution of icing at the point-to-point tower level can be given. At the same time, the forecast error can be minimized by improving the initial icing forecast field, which can not only effectively guide the implementation of anti-icing and ice-melting work, but also optimize resource allocation and material mobilization, reduce ice-melting costs, and improve the accuracy and efficiency of ice-melting. Summary of the invention

[0004] In view of the problems existing in the prior art, the present invention is proposed.

[0005] Therefore, the problem to be solved by the present invention is how to accurately output the gridded actual ice thickness value in real time.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, an embodiment of the present invention provides a method for gridding ice thickness at the transmission line tower level, which includes obtaining ice monitoring data and meteorological and geographic information data;

[0008] Interpolate and preprocess the data;

[0009] Training and testing of AI models based on interpolated and preprocessed data;

[0010] Optimize AI model parameters using Bayesian parameter optimization methods;

[0011] Generate a real-time grid distribution map of the southern region based on the optimal artificial intelligence model parameters.

[0012] As a preferred solution of the method for gridding ice thickness at the transmission line tower level described in the present invention, the ice monitoring data obtained includes the accumulated ice value data of multiple sets of ice terminals at multiple time periods; the meteorological geographic information data includes temperature, humidity, wind speed, wind direction, micro-topography geographic information, elevation and ice area distribution map data.

[0013] As a preferred solution of the gridding method of ice thickness at the transmission line tower level of the present invention, the interpolation and preprocessing of data includes:

[0014] Perform spatial interpolation on ice monitoring data to obtain preliminary gridded ice thickness data;

[0015] Normalize meteorological and geographic information data;

[0016] The interpolated ice thickness data are matched with the normalized meteorological and geographic information data in time and space to obtain the training sample set.

[0017] As a preferred solution of the method for gridding ice thickness at the transmission line tower level of the present invention, the artificial intelligence model training and testing based on interpolated and preprocessed data includes:

[0018] A deep neural network model was constructed, with the input layer including meteorological and geographic information features and the output layer being the ice thickness;

[0019] Divide the training sample set into training set and test set;

[0020] Use the training set to train the deep neural network model;

[0021] Use the test set to evaluate the performance of the trained model.

[0022] As a preferred solution of the method for gridding ice thickness at the transmission line tower level described in the present invention, the method of optimizing the artificial intelligence model parameters using the Bayesian parameter optimization method includes:

[0023] Define the hyperparameter space of the deep neural network model, including the number of hidden layers, the number of neurons in each layer, the learning rate, etc.

[0024] Construct a Gaussian process regression model to fit the relationship between hyperparameters and model performance;

[0025] Based on the Gaussian process regression model, the Bayesian optimization algorithm is used to search for the optimal hyperparameter combination;

[0026] Retrain the deep neural network model using optimal hyperparameters.

[0027] As a preferred solution of the gridding method for ice thickness at the transmission line tower level described in the present invention, the method of generating a real-time grid distribution map of the southern region based on the optimal artificial intelligence model parameters includes:

[0028] Divide the southern region into 1km×1km grids;

[0029] Collect meteorological and geographic information data at each grid point;

[0030] The collected data is input into the optimized deep neural network model to obtain the predicted value of ice thickness at each grid point;

[0031] Based on the prediction results, a grid distribution map of ice thickness at the tower level for transmission lines in the southern region is generated.

[0032] As a preferred solution of the gridding method for ice thickness at the transmission line tower level of the present invention, the gridding distribution map of ice thickness at the transmission line tower level in the southern region includes:

[0033] Divide the southern region into grids of predetermined size; collect meteorological elements and geographic information data for each grid point; use the optimized artificial intelligence model to predict the ice thickness at each grid point; and generate a gridded distribution map of ice thickness based on the prediction results.

[0034] In a second aspect, an embodiment of the present invention provides a gridding system based on ice thickness at the transmission line tower level, which includes a data acquisition module for acquiring ice monitoring data and meteorological and geographic information data;

[0035] Data preprocessing module, used to interpolate and preprocess data;

[0036] Model training module, used for training and testing AI models based on interpolated and preprocessed data;

[0037] Parameter optimization module, used to optimize the parameters of artificial intelligence models using Bayesian parameter optimization method;

[0038] The distribution map generation module is used to generate a real-time grid distribution map of the southern region based on the optimal artificial intelligence model parameters.

[0039] In a third aspect, an embodiment of the present invention provides a computer device, including a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program instructions are executed by the processor, the steps of the method for gridding ice thickness at the transmission line tower level as described in the first aspect of the present invention are implemented.

[0040] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program instructions are executed by a processor, the steps of the method for gridding ice thickness at the transmission line tower level as described in the first aspect of the present invention are implemented.

[0041] The beneficial effects of the present invention are as follows: the present invention combines ice monitoring data and meteorological and geographic information data, and uses a deep learning model to achieve refined grid prediction of ice thickness at the tower level of power transmission lines. The Bayesian parameter optimization method is used to improve the prediction accuracy of the model. The resulting grid distribution map of ice thickness in the southern region provides an important basis for transmission line ice risk assessment and preventive maintenance, and helps to improve the safe and stable operation level of the power grid. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0043] Figure 1 A flow chart of a method for gridding ice thickness at the tower level of a transmission line provided in an embodiment of the present invention;

[0044] Figure 2 Provide a technical roadmap for the gridding method of ice thickness at the tower level for transmission lines;

[0045] Figure 3 Scatter density diagram for LightGBM and XGBoost training, validation and testing;

[0046] Figure 4 Schematic diagram of bilinear interpolation. DETAILED DESCRIPTION

[0047] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0048] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0049] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0050] Example 1

[0051] Reference Figure 1-2 , which is the first embodiment of the present invention, and provides a gridding method based on ice thickness at the transmission line tower level, comprising:

[0052] S1: Obtain ice cover monitoring data and meteorological geographic information data;

[0053] Furthermore, ice cover monitoring data include ice cover thickness and ice cover weight; meteorological and geographic information data include temperature, humidity, wind speed, altitude, and terrain undulation.

[0054] It should be noted that the ice monitoring data is obtained through ice monitoring equipment installed on key towers, the meteorological data comes from meteorological station observations and meteorological numerical forecast models, and the geographic information data comes from the digital elevation model (DEM).

[0055] S2: interpolate and preprocess the data;

[0056] Furthermore, the data is interpolated and preprocessed, including:

[0057] Perform spatial interpolation on ice monitoring data to obtain preliminary gridded ice thickness data;

[0058] Normalize meteorological and geographic information data;

[0059] The interpolated ice thickness data are matched with the normalized meteorological and geographic information data in time and space to obtain the training sample set.

[0060] It should be noted that the spatial interpolation uses the Kriging interpolation method, which can take into account spatial autocorrelation and improve interpolation accuracy. The normalization process uses the minimum-maximum normalization method to unify the features of different dimensions into the [0, 1] interval to facilitate model training.

[0061] S3: Training and testing of AI models based on interpolated and preprocessed data;

[0062] Furthermore, AI model training and testing based on interpolated and preprocessed data include:

[0063] A deep neural network model was constructed, with the input layer including meteorological and geographic information features and the output layer being the ice thickness;

[0064] Divide the training sample set into training set and test set;

[0065] Use the training set to train the deep neural network model;

[0066] Use the test set to evaluate the performance of the trained model.

[0067] It should be noted that the deep neural network model uses a multi-layer perceptron (MLP) structure, which can effectively capture the nonlinear relationship between features. The model training uses the back propagation algorithm, and the loss function selects the mean square error (MSE).

[0068] S4: Optimize AI model parameters using Bayesian parameter optimization method;

[0069] Furthermore, the Bayesian parameter optimization method is used to optimize the parameters of the artificial intelligence model, including:

[0070] Define the hyperparameter space of the deep neural network model, including the number of hidden layers, the number of neurons in each layer, the learning rate, etc.

[0071] Construct a Gaussian process regression model to fit the relationship between hyperparameters and model performance;

[0072] Based on the Gaussian process regression model, the Bayesian optimization algorithm is used to search for the optimal hyperparameter combination;

[0073] Retrain the deep neural network model using optimal hyperparameters.

[0074] It should be noted that Bayesian optimization can find the optimal hyperparameter combination more efficiently than grid search and random search. The Gaussian process regression model can quantify hyperparameter uncertainty and help strike a balance between exploration and utilization.

[0075] S5: Generate a real-time grid distribution map of the southern region based on the optimal AI model parameters.

[0076] Furthermore, based on the optimal AI model parameters, a real-time grid distribution map of the southern region is generated, including:

[0077] Divide the southern region into 1km×1km grids;

[0078] Collect meteorological and geographic information data at each grid point;

[0079] The collected data is input into the optimized deep neural network model to obtain the predicted value of ice thickness at each grid point;

[0080] Based on the prediction results, a grid distribution map of ice thickness at the tower level for transmission lines in the southern region is generated.

[0081] It should be noted that the 1km×1km grid resolution can better reflect the spatial distribution characteristics of ice thickness, and it also matches the resolution of commonly used meteorological numerical forecast models. The generated distribution map can use different colors to represent different ice thickness levels, and intuitively display the distribution of ice risk.

[0082] The above is a schematic scheme of a method for gridding ice thickness at the transmission line tower level in this embodiment. It should be noted that the technical scheme of the system based on gridding ice thickness at the transmission line tower level and the technical scheme based on the above-mentioned method for gridding ice thickness at the transmission line tower level belong to the same concept. The details of the technical scheme of the system based on gridding ice thickness at the transmission line tower level in this embodiment that are not described in detail can all be referred to the description of the technical scheme based on the above-mentioned method for gridding ice thickness at the transmission line tower level.

[0083] In this embodiment, the gridding system based on the ice thickness of the transmission line tower level includes:

[0084] A data acquisition module is used to acquire ice monitoring data and meteorological and geographic information data;

[0085] Data preprocessing module, used to interpolate and preprocess data;

[0086] Model training module, used for training and testing AI models based on interpolated and preprocessed data;

[0087] Parameter optimization module, used to optimize the parameters of artificial intelligence models using Bayesian parameter optimization method;

[0088] The distribution map generation module is used to generate a real-time grid distribution map of the southern region based on the optimal artificial intelligence model parameters.

[0089] This embodiment further provides a computing device, which is applicable to the case of gridding ice thickness at the tower level of a transmission line, and includes:

[0090] A memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute computer executable instructions to implement the gridding method of ice thickness at the transmission line tower level as proposed in the above embodiment.

[0091] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the method for gridding ice thickness at the tower level of a transmission line is implemented as proposed in the above embodiment.

[0092] The storage medium proposed in this embodiment and the method for implementing ice thickness gridding based on the transmission line tower level proposed in the above embodiment belong to the same inventive concept. The technical details not fully described in this embodiment can be found in the above embodiment, and this embodiment has the same beneficial effects as the above embodiment.

[0093] Through the above description of the implementation methods, the technicians in the relevant field can clearly understand that the present invention can be implemented by means of software and necessary general hardware, and of course it can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory (FLASH), hard disk or optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform the methods of various embodiments of the present invention.

[0094] Furthermore, the present embodiment also provides a gridding method based on ice thickness at the transmission line tower level, including a data acquisition module for acquiring ice monitoring data and meteorological and geographic information data;

[0095] Data preprocessing module, used to interpolate and preprocess data;

[0096] Model training module, used for training and testing AI models based on interpolated and preprocessed data;

[0097] Parameter optimization module, used to optimize the parameters of artificial intelligence models using Bayesian parameter optimization method;

[0098] The distribution map generation module is used to generate a real-time grid distribution map of the southern region based on the optimal artificial intelligence model parameters.

[0099] This embodiment also provides a computer device, which is suitable for the case of a gridding method for ice thickness at the transmission line tower level, including a memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute the computer executable instructions to implement the gridding method for ice thickness at the transmission line tower level proposed in the above embodiment.

[0100] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used 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 and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0101] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the method for gridding ice thickness at the tower level of a transmission line is implemented as proposed in the above embodiment.

[0102] In summary, the present invention combines ice monitoring data and meteorological and geographic information data, and uses a deep learning model to achieve refined grid prediction of ice thickness at the transmission line tower level. The Bayesian parameter optimization method is used to improve the prediction accuracy of the model. The resulting grid distribution map of ice thickness in the southern region provides an important basis for transmission line ice risk assessment and preventive maintenance, and helps to improve the safe and stable operation level of the power grid.

[0103] Example 2

[0104] Reference Figure 1-Figure 4 , which is the second embodiment of the present invention, and this embodiment provides a gridding method based on ice thickness at the transmission line tower level. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0105] S1: Obtain the cumulative ice value data of 2581 ice terminals from 04:00 on January 24, 2024 to 15:00 on February 26, 2024 (a total of 104 hours), and use Kriging interpolation to a resolution of 90m;

[0106] S2: Obtain the 1km resolution temperature, humidity, wind speed, wind direction meteorological elements and 90m micro-topography geographic information elements, elevation, and ice area distribution map data for the southern region provided by the China Public Service Center, and interpolate all data to 90m resolution;

[0107] S3: interpolate the above data to 2581 ice-covered terminals using bilinear interpolation method;

[0108] S4: Match the interpolation data and the actual ice-covered tower monitoring data in time as the training and test data sets of the artificial intelligence model. Divide them in chronological order. The first 74 time points are used for training and verification. Among these 74 time points, 1,806 terminals are selected for training and 775 terminals are used for verification; all terminals in the last 30 time points are used for testing. The verification and test results of different artificial intelligence models are shown in the following table (the model prediction results are not less than 0);

[0109] S5: Use the Bayesian parameter optimization method to find the best parameters for artificial intelligence model training and predict the ice thickness of the tower on the test set;

[0110] S6: Extend the optimal AI model to the 90m grid to form a real-time grid distribution map of the southern region.

[0111] It can be seen that by obtaining the cumulative ice value data of 2581 sets of ice terminals and the 1km resolution temperature, humidity, wind speed, wind direction meteorological elements and 90-meter micro-topography geographic information elements, elevation, and ice area distribution map data of the southern region provided by the China Public Service Center, and through a series of processing, it is possible to more accurately output the actual grid distribution map of the southern region, provide scientific support for the forecast of icing on transmission lines, and have important scientific significance and application value.

[0112] Specifically, the above step S3 includes:

[0113] Mathematically, bilinear interpolation represents the linear interpolation extension of an interpolation function containing two variables. The core idea of ​​this method is to perform linear interpolation in two directions respectively.

[0114] like Figure 4 As shown, in order to find the value of the unknown function f at point P = (x, y), it is assumed that the known function f at Q 11 =(x1,y1),Q 12 =(x1,y2),Q 21 =(x2,y1) and Q 22 =The values ​​at the four points (x2, y2).

[0115] First, linear interpolation is performed in the x direction to obtain equations (1) and (2), and then linear interpolation is performed in the y direction to obtain equation (3). In this way, the result of the function f(x, y) is obtained, as shown in equation (4).

[0116]

[0117]

[0118]

[0119]

[0120] In this way, the above formula can be used to interpolate the ice thickness, ice area distribution map, elevation geographic information and meteorological element data after Kriging interpolation with a resolution of 90m to ice detection towers with different longitudes and latitudes.

[0121] The following is a detailed description of this method with an application scenario example:

[0122] The cumulative ice value data of 2581 ice-covered terminals from 04:00 on January 24, 2024 to 15:00 on February 26, 2024 (a total of 104 time periods) were obtained, and the resolution was interpolated to 90m using Kriging; the 1km resolution temperature, humidity, wind speed, wind direction meteorological elements and 90-meter micro-topography geographic information elements, elevation, and ice area distribution map data for the southern region provided by the China Public Service Center were obtained, and all interpolated to 90m resolution; the above data were clipped using areas where the temperature grid contour line was less than 0; the above data were interpolated to 2581 ice-covered terminals using the bilinear interpolation method; the interpolated data and the actual ice-covered tower monitoring data were matched in time as the training and test data sets of the artificial intelligence model, and divided into chronological order, the first 74 time points were used for training and verification, of which 1806 terminals were selected for training and 775 terminals were used for verification; all terminals in the last 30 time points were used for testing. The verification and test results of different artificial intelligence models are shown in the following table (the model prediction results are not less than 0); the Bayesian parameter optimization method is used to find the best parameters for artificial intelligence model training, and the ice thickness of the tower is predicted on the test set. The optimal artificial intelligence model is extended to the 90m grid to form a real-time grid distribution map of the southern region.

[0123] The specific steps are as follows:

[0124] Obtain the cumulative ice value data of 2581 ice terminals from 04:00 on January 24, 2024 to 15:00 on February 26, 2024 (a total of 104 hours), and use Kriging interpolation to 90m resolution;

[0125] Obtain the 1km resolution temperature, humidity, wind speed, wind direction meteorological elements and 90m micro-topography geographic information elements, elevation, and ice area distribution map data for the southern region provided by the China Public Service Center, and interpolate all of them to 90m resolution;

[0126] The above data are clipped using the area where the temperature grid contour line is less than 0; the above data are interpolated to 2581 ice-covered terminals using the bilinear interpolation method;

[0127] The interpolation data and the actual ice-covered tower monitoring data are matched in time as the training and test data sets of the artificial intelligence model. They are divided in chronological order. The first 74 time points are used for training and verification. Among these 74 time points, 1,806 terminals are selected for training and 775 terminals are used for verification; all terminals are used for testing at the last 30 time points. The verification and test results of different artificial intelligence models are shown in the following table (the model prediction results are not less than 0);

[0128] The Bayesian parameter optimization method is used to find the optimal parameters for training the artificial intelligence model and predict the ice thickness of the tower on the test set.

[0129] The optimal artificial intelligence model is extended to the 90m grid to form a real-time grid distribution map of the southern region.

[0130] Based on Example 1, this example further compares and analyzes the performance of different artificial intelligence models. The model performance evaluation indicators include root mean square error (RMSE), mean absolute error (MAE) and determination coefficient (R2). The results are shown in the following table

[0131]

[0132]

[0133] From the results, we can see that the LightGBM model performs best in this task, with the lowest RMSE and MAE, and the highest R 2 value;

[0134] This example further verifies the effectiveness of Bayesian parameter optimization. Taking the LightGBM model as an example, the model performance before and after using Bayesian optimization is compared:

[0135]

[0136] The results show that the prediction accuracy of the model has been significantly improved through Bayesian parameter optimization, with RMSE and MAE reduced by 10.13% and 10.29% respectively. At the same time, the training time of the optimized model has been reduced by 20.83%, improving the computational efficiency.

[0137] Through the above embodiments, the gridding method for ice thickness at the transmission line tower level provided by the present invention can effectively integrate multi-source data and use advanced machine learning algorithms to achieve high-precision ice thickness prediction. This method not only improves the prediction accuracy, but also generates a high-resolution grid distribution map of ice thickness, providing important decision support for power grid operation and disaster prevention and mitigation.

[0138] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A gridding method based on ice thickness at the transmission line tower level, characterized by: include, Obtain ice cover monitoring data and meteorological and geographic information data; Interpolate and preprocess the data; Training and testing of AI models based on interpolated and preprocessed data; Optimize AI model parameters using Bayesian parameter optimization methods; Generate a real-time grid distribution map of the southern region based on the optimal artificial intelligence model parameters.

2. The method for gridding ice thickness at the transmission line tower level according to claim 1, characterized in that: The ice cover monitoring data obtained includes the accumulated ice cover value data of multiple sets of ice cover terminals at multiple time periods; the meteorological geographic information data includes temperature, humidity, wind speed, wind direction, micro-topography geographic information, elevation and ice area distribution map data.

3. The method for gridding ice thickness at the transmission line tower level according to claim 2, characterized in that: The interpolation and preprocessing of data includes: Perform spatial interpolation on ice monitoring data to obtain preliminary gridded ice thickness data; Normalize the meteorological and geographic information data; The interpolated ice thickness data is matched with the normalized meteorological and geographic information data in time and space to obtain the training sample set.

4. The method for gridding ice thickness at the transmission line tower level according to claim 3 is characterized in that: The artificial intelligence model training and testing based on interpolated and preprocessed data includes: A deep neural network model is constructed, the input layer includes meteorological and geographical information features, and the output layer is ice thickness; Divide the training sample set into training set and test set; Use the training set to train the deep neural network model; Use the test set to evaluate the performance of the trained model.

5. The method for gridding ice thickness at the transmission line tower level according to claim 4, characterized in that: The method of optimizing artificial intelligence model parameters by using the Bayesian parameter optimization method includes: Define the hyperparameter space of the deep neural network model, including the number of hidden layers, the number of neurons in each layer, the learning rate, etc.; Construct a Gaussian process regression model to fit the relationship between hyperparameters and model performance; Based on the Gaussian process regression model, the Bayesian optimization algorithm is used to search for the optimal hyperparameter combination; Retrain the deep neural network model using optimal hyperparameters.

6. The method for gridding ice thickness at the transmission line tower level according to claim 5, characterized in that: The method of generating a real-time grid distribution map of the southern region based on the optimal artificial intelligence model parameters includes: Divide the southern region into 1km×1km grids; Collect meteorological and geographic information data at each grid point; The collected data is input into the optimized deep neural network model to obtain the predicted value of ice thickness at each grid point; Based on the prediction results, a grid distribution map of ice thickness at the tower level for transmission lines in the southern region is generated.

7. The method for gridding ice thickness at the transmission line tower level according to claim 6, characterized in that: The grid distribution map of ice thickness at the tower level of power transmission lines in the southern region includes: Divide the southern region into grids of predetermined size; Collect meteorological elements and geographic information data at each grid point; The optimized artificial intelligence model is used to predict the ice thickness at each grid point; a grid distribution map of ice thickness is generated based on the prediction results.

8. A gridding system for ice thickness at the tower level of a power transmission line, based on the gridding method for ice thickness at the tower level of a power transmission line according to any one of claims 1 to 7, characterized in that: Also includes, A data acquisition module is used to acquire ice monitoring data and meteorological and geographic information data; Data preprocessing module, used to interpolate and preprocess data; Model training module, used for training and testing AI models based on interpolated and preprocessed data; Parameter optimization module, used to optimize the parameters of artificial intelligence models using Bayesian parameter optimization method; The distribution map generation module is used to generate a real-time grid distribution map of the southern region based on the optimal artificial intelligence model parameters.

9. 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 for gridding ice thickness at the transmission line tower level according to any one of claims 1 to 7 are implemented.

10. 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 for gridding ice thickness at the transmission line tower level according to any one of claims 1 to 7 are implemented.

Images