A distribution network tower intelligent selection method, device, equipment, medium and product

Abstract

The present invention discloses a method, device, equipment, medium and product for intelligent selection of distribution network pole towers, and relates to the field of distribution network design automation technology. The method is to first obtain the grid line information and pole tower model information of each pole tower layout node in the historical distribution network design data, and then extract the multidimensional pole tower feature vector of each node and the index value on multiple pole tower selection influencing factor indicators according to the acquisition results, and determine the weight coefficient of each indicator according to the extraction results, and then apply the extraction results and the weight coefficients of each indicator to construct a machine learning model and perform model training to obtain a pole tower selection model, and finally import the index values ​​of the target node on multiple indicators into the model, and output the corresponding pole tower model, so that the designer does not need to select manually, which greatly reduces the demand for human resources and improves the selection efficiency, and avoids human errors, thereby helping designers to be more convenient and efficient in designing distribution networks.

Description

Technical Field

[0001] The present invention belongs to the technical field of distribution network design automation, and specifically relates to a distribution network pole tower intelligent selection method, device, equipment, medium and product. Background Art

[0002] With the rapid development of the power industry, the design of distribution networks has become increasingly complex and sophisticated. The traditional distribution network design method requires designers to manually select the model of the tower based on multiple factors such as the characteristics of the distribution network line, the topography and environment of the tower location, etc. This is not only time-consuming and labor-intensive, but also prone to human errors. Therefore, in order to improve the design efficiency and accuracy, there is an urgent need for an automated technology that can intelligently select distribution network towers, so that designers can be more convenient and efficient when designing distribution networks. Summary of the invention

[0003] The purpose of the present invention is to provide a distribution network tower intelligent selection method, device, computer equipment, computer readable storage medium and computer program product to solve the problems of existing distribution network design methods being time-consuming, labor-intensive and prone to human errors.

[0004] In order to achieve the above object, the present invention adopts the following technical solutions:

[0005] In a first aspect, a method for intelligent selection of distribution network towers is provided, comprising:

[0006] Obtaining grid line information and tower model information of each tower arrangement node in the historical distribution network design data;

[0007] Extracting the index values ​​of the various pole tower arrangement nodes on multiple pole tower selection influencing factor indexes according to the power grid line information of the various pole tower arrangement nodes;

[0008] Extracting a multi-dimensional tower feature vector of each tower arrangement node according to the tower model information of each tower arrangement node;

[0009] Determine the weight coefficient of each tower selection influencing factor index in the multiple tower selection influencing factor indexes according to the index value of each tower layout node on the multiple tower selection influencing factor indexes and the multi-dimensional tower feature vector of each tower layout node;

[0010] Apply the index values ​​of each tower layout node on the multiple tower selection influencing factor indicators, the tower model information of each tower layout node, and the weight coefficients of each tower selection influencing factor indicator to construct a machine learning model and perform model training to obtain a tower selection model in which the model input items are the index values ​​on the multiple tower selection influencing factor indicators and the model output items are the tower model information;

[0011] Acquire the grid line information of the target tower arrangement node in the distribution network to be designed;

[0012] Extracting the index value of the target tower arrangement node on the multiple tower selection influencing factor indexes according to the power grid line information of the target tower arrangement node;

[0013] The index values ​​of the target pole tower arrangement node on the multiple pole tower selection influencing factor indexes are imported into the pole tower selection model, and the pole tower model information of the target pole tower arrangement node is obtained by output.

[0014] Based on the above invention content, a new scheme for automatic tower selection based on historical distribution network design data and machine learning models is provided, that is, firstly, the power grid line information and tower model information of each tower layout node in the historical distribution network design data are obtained, and then the multidimensional tower feature vector of each node and the index value on multiple tower selection influencing factor indicators are extracted according to the acquisition results, and the weight coefficient of each indicator is determined according to the extraction result, and then the extraction result and the weight coefficient of each indicator are applied to construct a machine learning model and perform model training to obtain a tower selection model, and finally the index value of the target tower layout node on multiple indicators is imported into the model, and the corresponding tower model information is output, so that the manual selection of the designer is not required, which greatly reduces the human resource demand and improves the selection efficiency, and avoids human errors, thereby making it more convenient and efficient for designers to design the distribution network, and facilitating practical application and promotion.

[0015] In a possible design, the power grid line information includes the location of the pole tower, the environmental parameters of the pole tower, the voltage of the pole tower support line and / or the number of pole tower support loops, and the multiple pole tower selection influencing factor indicators include the length of the pole tower connection line segment, the angle of the pole tower connection line segment, the environmental parameters of the pole tower, the voltage of the pole tower support line and / or the number of the pole tower support loops;

[0016] And / or, the tower model information corresponds to the tower bearing capacity, tower applicable scenarios, tower materials, tower anti-corrosion methods and / or tower appearance requirements, and the multi-dimensional tower feature vector includes the characteristic value of the tower bearing capacity, the characteristic value of the tower applicable scenarios, the characteristic value of the tower material, the characteristic value of the tower anti-corrosion method and / or the characteristic value of the tower appearance requirements.

[0017] In a possible design, when the power grid line information includes the location of the pole tower and the multiple pole tower selection influencing factor indicators include the length of the pole tower connection line segment, the index value of each pole tower layout node on the multiple pole tower selection influencing factor indicators is extracted according to the power grid line information of each pole tower layout node, including:

[0018] Obtaining the location of a pole tower arrangement node and at least one adjacent pole tower arrangement node, wherein the adjacent pole tower arrangement node refers to a preceding pole tower arrangement node located before the pole tower arrangement node along the direction of the power grid line or a succeeding pole tower arrangement node located after the pole tower arrangement node along the direction of the power grid line;

[0019] For each adjacent pole tower arrangement node in the at least one adjacent pole tower arrangement node, the corresponding connecting line segment length value is calculated according to the following formula: :

[0020]

[0021] In the formula, represents the horizontal coordinate of the position of the tower at the certain tower arrangement node, represents the vertical coordinate of the location of the tower at the certain tower arrangement node, represents the elevation coordinate of the tower at the location of the tower arrangement node, represents the horizontal coordinate of the position of the tower at the adjacent tower arrangement node, represents the ordinate of the location of the tower at the adjacent tower arrangement node, Indicates the elevation coordinates of the tower at the location of the adjacent tower arrangement node;

[0022] The maximum length value among all the connection line segment length values ​​is used as the index value of the certain tower arrangement node on the tower connection line segment length;

[0023] And / or, when the power grid line information includes the location of the pole tower and the multiple pole tower selection influencing factor indicators include the pole tower connection line segment angle, according to the power grid line information of each pole tower layout node, the index value of each pole tower layout node on the multiple pole tower selection influencing factor indicators is extracted, including:

[0024] Obtaining the location of the tower of a certain tower arrangement node, at least one adjacent tower arrangement node, and at least one next-neighboring tower arrangement node, wherein the adjacent tower arrangement node refers to a previous tower arrangement node located before the certain tower arrangement node along the direction of the power grid line or a subsequent tower arrangement node located after the certain tower arrangement node along the direction of the power grid line, and the next-neighboring tower arrangement node refers to another previous tower arrangement node located before the previous tower arrangement node along the direction of the power grid line or another subsequent tower arrangement node located before the subsequent tower arrangement node along the direction of the power grid line, and the at least one adjacent tower arrangement node corresponds one-to-one to the at least one next-neighboring tower arrangement node;

[0025] For each pair of corresponding adjacent tower arrangement nodes and next-neighboring tower arrangement nodes in the at least one adjacent tower arrangement node and the at least one next-neighboring tower arrangement node, the corresponding connecting line segment angle value is calculated according to the following formula: :

[0026]

[0027] In the formula, represents the horizontal coordinate of the position of the tower at the certain tower arrangement node, represents the vertical coordinate of the location of the tower at the certain tower arrangement node, represents the elevation coordinate of the tower at the location of the tower arrangement node, represents the horizontal coordinate of the position of the tower at the adjacent tower arrangement node, represents the ordinate of the location of the tower at the adjacent tower arrangement node, represents the elevation coordinate of the tower at the location of the adjacent tower arrangement node, represents the horizontal coordinate of the position of the tower at the node where the next adjacent tower is arranged, represents the ordinate of the location of the tower at the node where the next adjacent tower is arranged, Indicates the elevation coordinates of the tower at the location of the node of the next neighboring tower arrangement;

[0028] The maximum angle value among all the connection line segment angle values ​​is used as the index value of the certain pole tower layout node on the pole tower connection line segment angle.

[0029] In a possible design, according to the index value of each tower arrangement node on the multiple tower selection influencing factor indexes and the multi-dimensional tower feature vector of each tower arrangement node, the weight coefficient of each tower selection influencing factor index in the multiple tower selection influencing factor indexes is determined, including:

[0030] For a certain tower selection influencing factor index among the multiple tower selection influencing factor indexes, the corresponding weight coefficient is calculated according to the following formula: :

[0031]

[0032] In the formula, Represents the total number of tower layout nodes, Indicates less than or equal to A positive integer, Indicates The index value of a tower layout node on the index of the tower selection influencing factor, It represents the average value of all tower layout nodes on the influencing factor index of a tower selection. Indicates the The multi-dimensional tower feature vector of the tower arrangement nodes has , represents the total number of dimensions of the multi-dimensional tower feature vector, Indicates less than or equal to A positive integer, Indicates the The first node of the tower arrangement The eigenvalue of the tower, represents the average value of the multi-dimensional tower feature vectors of all tower arrangement nodes and has , Represents the first node of all tower layout nodes The average value of the tower characteristics.

[0033] In a possible design, when the machine learning model adopts a CART decision tree model, the index value of each tower layout node on the multiple tower selection influencing factor indicators, the tower model information of each tower layout node and the weight coefficient of each tower selection influencing factor indicator are applied to construct a machine learning model and perform model training, including:

[0034] Arrange the influencing factors of tower selection in descending order according to the weight coefficient to obtain an index sequence, and in the CART decision tree model, the index sequence is the first index sequence from the root node to the leaf node. node layers, using the first to last node in the index sequence The influencing factors of tower selection are divided into sub-nodes, among which: Indicates less than or equal to A positive integer, The total number of indicators representing the multiple influencing factors of tower selection;

[0035] The sub-node classification criterion in the CART decision tree model is: for each non-root node, the corresponding category is determined according to the pole tower models of all pole tower arrangement nodes in the corresponding sample set, wherein the non-root node refers to a node that is not the root node;

[0036] The stopping condition for sub-node division in the CART decision tree model is: for each non-root node, determine whether the frequency of occurrence of a certain tower model in the corresponding sample set exceeds the preset frequency threshold and whether the total number of samples exceeds the preset sample number threshold. If both are true, the corresponding node is taken as a leaf node, and the tower model with the highest frequency of occurrence is taken as the category of the leaf node.

[0037] In a possible design, after outputting the tower model information of the target tower arrangement node, the method further includes:

[0038] According to the pole tower model information of the target pole tower arrangement node, the distribution network pole tower at the target pole tower arrangement node is automatically drawn in the distribution network design drawing, and the pole tower related information is marked, wherein the pole tower related information includes the power grid line information and / or the pole tower model information of the target pole tower arrangement node.

[0039] In a second aspect, a distribution network tower intelligent selection device is provided, comprising a node information acquisition unit, an influence index extraction unit, a feature vector extraction unit, an index weight determination unit, a model construction training unit and a selection model application unit;

[0040] The node information acquisition unit is used to acquire the power grid line information and tower model information of each tower arrangement node in the historical distribution network design data;

[0041] The influencing index extraction unit is communicatively connected to the node information acquisition unit, and is used to extract the index value of each tower arrangement node on multiple tower selection influencing factor indicators according to the power grid line information of each tower arrangement node;

[0042] The feature vector extraction unit is communicatively connected to the node information acquisition unit, and is used to extract the multi-dimensional tower feature vector of each tower arrangement node according to the tower model information of each tower arrangement node;

[0043] The indicator weight determination unit is respectively connected to the influencing indicator extraction unit and the feature vector extraction unit for communication, and is used to determine the weight coefficient of each tower selection influencing factor indicator in the multiple tower selection influencing factor indicators according to the indicator value of each tower arrangement node on the multiple tower selection influencing factor indicators and the multi-dimensional tower feature vector of each tower arrangement node;

[0044] The model construction training unit is communicatively connected to the indicator weight determination unit, and is used to apply the indicator values ​​of each tower arrangement node on the multiple tower selection influencing factor indicators, the tower model information of each tower arrangement node, and the weight coefficients of each tower selection influencing factor indicator to construct a machine learning model and perform model training to obtain a tower selection model in which the model input items are the indicator values ​​on the multiple tower selection influencing factor indicators and the model output items are the tower model information;

[0045] The node information acquisition unit is further used to acquire the grid line information of the target tower arrangement node in the distribution network to be designed;

[0046] The influencing index extraction unit is further used to extract the index value of the target tower arrangement node on the multiple tower selection influencing factor indicators according to the power grid line information of the target tower arrangement node;

[0047] The selection model application unit is respectively communicatively connected to the influencing index extraction unit and the model construction training unit, and is used to import the index value of the target tower layout node on the multiple tower selection influencing factor indicators into the tower selection model, and output the tower model information of the target tower layout node.

[0048] In a third aspect, the present invention provides a computer device comprising a memory, a processor and a transceiver which are communicatively connected in sequence, wherein the memory is used to store a computer program, the transceiver is used to send and receive messages, and the processor is used to read the computer program to execute the intelligent selection method for distribution network towers as described in the first aspect or any possible design of the first aspect.

[0049] In a fourth aspect, the present invention provides a computer-readable storage medium having instructions stored thereon, which, when executed on a computer, executes the method for intelligent selection of distribution network towers as described in the first aspect or any possible design of the first aspect.

[0050] In a fifth aspect, the present invention provides a computer program product, including a computer program or instructions, which, when executed by a computer, implements the method for intelligent selection of distribution network towers as described in the first aspect or any possible design of the first aspect.

[0051] Beneficial effects of the above scheme:

[0052] (1) The present invention creatively provides a new scheme for automatic pole tower selection based on historical distribution network design data and machine learning models, that is, first obtaining the grid line information and pole tower model information of each pole tower layout node in the historical distribution network design data, then extracting the multi-dimensional pole tower feature vector of each node and the index value of multiple pole tower selection influencing factor indicators according to the acquisition results, and determining the weight coefficient of each indicator according to the extraction results, and then applying the extraction results and the weight coefficient of each indicator to construct a machine learning model and perform model training to obtain a pole tower selection model, and finally importing the index values ​​of the target pole tower layout node on multiple indicators into the model, and outputting the corresponding pole tower model information, so that there is no need for designers to manually select, which greatly reduces the demand for human resources and improves the selection efficiency, and avoids human errors, thereby making it more convenient and efficient for designers to design distribution networks, and facilitating practical application and promotion;

[0053] (2) This scheme is highly efficient, that is, it can complete the complex tower selection work in a short time, greatly improving the design efficiency;

[0054] (3) This scheme is accurate, that is, by comprehensively considering multiple parameters, it ensures that the selected tower model best meets actual needs and reduces the possibility of human error;

[0055] (4) This solution has the feature of automation, that is, node line information and parameters are automatically obtained through the GIS engine, which reduces the tedious steps of manual input and realizes fully automated operation;

[0056] (5) This solution is user-friendly, that is, through a simple and intuitive interface, users can easily make fine-tuning and provide feedback, which improves the user experience;

[0057] (6) This solution has self-learning capabilities, that is, it can continuously optimize the intelligent selection model based on user feedback and gradually improve the accuracy and reliability of selection. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. 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 paying creative work.

[0059] Figure 1 A schematic diagram of a flow chart of a method for intelligent selection of distribution network towers provided in an embodiment of the present invention.

[0060] Figure 2 An exemplary diagram of the positional relationship between a certain pole tower arrangement node and an adjacent pole tower arrangement node and a next adjacent pole tower arrangement node provided in an embodiment of the present invention.

[0061] Figure 3 A schematic diagram of the structure of a distribution network tower intelligent selection device provided in an embodiment of the present invention.

[0062] Figure 4 A schematic diagram of the structure of a computer device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0063] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the present invention will be briefly introduced below in combination with the drawings and the description of the embodiments or the prior art. Obviously, the following description of the structures of the drawings is only some embodiments of the present invention. For ordinary technicians in this field, other embodiments can be obtained based on these embodiments without creative work. It should be noted that the description of these embodiments is used to help understand the present invention, but does not constitute a limitation of the present invention.

[0064] It should be understood that although the terms first and second, etc. may be used herein to describe various objects, these objects should not be limited by these terms. These terms are only used to distinguish one object from another object. For example, a first object can be referred to as a second object, and similarly, a second object can be referred to as a first object without departing from the scope of the exemplary embodiments of the present invention.

[0065] It should be understood that the term "and / or" that may appear in this document is merely a description of the association relationship between associated objects, indicating that there may be three relationships. For example, A and / or B can indicate three situations: A exists alone, B exists alone, or A and B exist at the same time. For another example, A, B and / or C can indicate the existence of any one of A, B and C or any combination of them. The term " / and" that may appear in this document describes another type of association object relationship, indicating that there may be two relationships. For example, A / and B can indicate two situations: A exists alone or A and B exist at the same time. In addition, the character " / " that may appear in this document generally indicates that the previous and next associated objects are in an "or" relationship.

[0066] Example

[0067] like Figure 1 As shown, the distribution network pole tower intelligent selection method provided in the first aspect of this embodiment can be, but is not limited to, executed by a computer device with certain computing resources, such as a server, a personal computer (PC, which refers to a multi-purpose computer with a size, price and performance suitable for personal use; desktops, laptops, small laptops, tablets and ultrabooks are all personal computers), smart phones, personal digital assistants (PDA) or wearable devices. Figure 1 As shown, the distribution network tower intelligent selection method may include but is not limited to the following steps S1 to S8.

[0068] S1. Obtaining the grid line information and tower model information of each tower arrangement node in the historical distribution network design data.

[0069] In the step S1, the historical distribution network design data refers to all the distribution network design drawings that have passed the verification, which specifically include a large number of pole tower layout nodes and the pole tower design content at each pole tower layout node: the power grid line information and the pole tower model information, etc., wherein the power grid line information is used as the constraint condition content of the pole tower selection, specifically including but not limited to the location of the pole tower (that is, it can be a three-dimensional coordinate including elevation coordinates and horizontal and vertical coordinates represented by longitude and latitude), the environmental parameters of the pole tower (such as parameters such as ambient temperature and ambient humidity), the voltage of the pole tower supporting line (such as 500kV, 330kV, 220kV, 110kV, 66kV, 35kV, 10kV and 380V, etc.) and / or the number of circuits supported by the pole tower (such as double circuits, three circuits and four circuits, etc.); the pole tower model information is used as the specific result of the pole tower selection, which can be but not limited to the corresponding pole tower bearing capacity (that is, the tower can The maximum load that the pole can withstand. In the design of the distribution network, it is necessary to select a suitable pole tower according to the electrical parameters and load requirements of the line to ensure that the pole tower can safely bear the load of the line, so it can be represented by the specific value of the maximum load), applicable scenarios of the pole tower (that is, since different types of pole towers are suitable for different scenarios, such as mountainous areas, plains or cities, etc., the applicable scenario information can help designers select a suitable pole tower type according to actual conditions to meet the layout requirements of the line in different terrains and environments), the material used for the pole tower (that is, since the material of the pole tower directly affects its strength, durability and service life, the material information can help designers select a pole tower that meets the design requirements to ensure the safe and reliable operation of the line), the anti-corrosion method of the pole tower (which is of great significance for designers to select suitable pole towers and design distribution networks) and / or the appearance requirements of the pole tower (which is also of great significance for designers to select suitable pole towers and design distribution networks), etc. In addition, in order to automatically obtain the power grid line information and tower model information of each tower layout node, so as to reduce the tedious steps of manual input and achieve the purpose of fully automatic operation, preferably, the power grid line information and tower model information of each tower layout node in the historical distribution network design data can be automatically obtained through a GIS (Geographic Information System or Geo-Information system) engine; the aforementioned GIS engine is a software or technical platform for processing, analyzing and visualizing geographic spatial data, and can have multiple functions, including data processing, visualization, spatial analysis and application development, etc. Therefore, based on the existing GIS engine, conventional modifications can be made to achieve the purpose of automatically obtaining the aforementioned information.

[0070] S2. According to the power grid line information of each tower arrangement node, extract the index value of each tower arrangement node on multiple tower selection influencing factor indicators.

[0071] In step S2, the multiple indices of factors affecting tower selection include but are not limited to tower connection line segment length, tower connection line segment angle, environmental parameters of the tower, tower support line voltage and / or the number of tower support loops, etc., wherein the tower environment parameters, tower support line voltage and / or the number of tower support loops, etc. can be directly extracted from the power grid line information, while the tower connection line segment length and / or the tower connection line segment angle need to be converted based on the tower location in the power grid line information. Considering that there is a height difference between the nodes of the two towers, that is, the elevation information of the two nodes is inconsistent, the elevation information needs to be taken into account when calculating the length of the connecting line segment between the two nodes (that is, the distance between the two nodes) and the angle of the connecting line segment, and it cannot be simply discussed in plane coordinates (this is contrary to the actual construction environment on site); at the same time, considering that a tower (that is, a node) has two adjacent nodes and connecting line segments in front and behind, except for the head pole and the tail pole, so it is necessary to calculate the length and angle of the two connecting line segments in front and behind respectively, and then compare them, and take the largest length and angle.

[0072] In the step S2, specifically, when the power grid line information includes the location of the pole tower and the multiple pole tower selection influencing factor indicators include the length of the pole tower connection line segment, the index value of each pole tower layout node on the multiple pole tower selection influencing factor indicators is extracted according to the power grid line information of each pole tower layout node, including but not limited to the following steps S211 to S213.

[0073] S211. Obtain the location of a certain tower layout node and at least one adjacent tower layout node, wherein the adjacent tower layout node refers to a preceding tower layout node that is located before the certain tower layout node along the direction of the power grid line or a succeeding tower layout node that is located after the certain tower layout node along the direction of the power grid line.

[0074] In step S211, Figure 2 As shown, Indicates a certain pole tower arrangement node, there are two adjacent pole tower arrangement nodes, respectively and In addition, if the pole designed at the pole tower layout node is the head pole, then there will be only one adjacent pole tower layout node: the rear pole tower layout node; and if the pole designed at the pole tower layout node is the tail pole, then there will be only one adjacent pole tower layout node: the front pole tower layout node.

[0075] S212. For each adjacent pole tower arrangement node in the at least one adjacent pole tower arrangement node, the corresponding connecting line segment length value is calculated according to the following formula: :

[0076]

[0077] In the formula, represents the horizontal coordinate of the position of the tower at the certain tower arrangement node, represents the vertical coordinate of the location of the tower at the certain tower arrangement node, represents the elevation coordinate of the tower at the location of the tower arrangement node, represents the horizontal coordinate of the position of the tower at the adjacent tower arrangement node, represents the ordinate of the location of the tower at the adjacent tower arrangement node, Indicates the elevation coordinates of the tower at the location of the adjacent tower arrangement node.

[0078] S213. Using the maximum length value among all the connection line segment length values ​​as the index value of the certain tower layout node on the tower connection line segment length.

[0079] In the step S2, specifically, when the power grid line information includes the location of the pole tower and the multiple pole tower selection influencing factor indicators include the pole tower connection line segment angle, the index value of each pole tower layout node on the multiple pole tower selection influencing factor indicators is extracted according to the power grid line information of each pole tower layout node, including but not limited to the following steps S221 to S223.

[0080] S221. Obtain the tower locations of a certain tower layout node, at least one adjacent tower layout node and at least one next-neighboring tower layout node, wherein the adjacent tower layout node refers to a previous tower layout node located before the certain tower layout node along the direction of the power grid line or a subsequent tower layout node located after the certain tower layout node along the direction of the power grid line, and the next-neighboring tower layout node refers to another previous tower layout node located before the previous tower layout node along the direction of the power grid line or another subsequent tower layout node located before the subsequent tower layout node along the direction of the power grid line, and the at least one adjacent tower layout node corresponds one-to-one to the at least one next-neighboring tower layout node.

[0081] In step S221, Figure 2 As shown, Indicates a certain pole tower arrangement node, there are two adjacent pole tower arrangement nodes, respectively and Indicated; there are also two nodes for the arrangement of the next adjacent tower, respectively and Indicates; among them, and One-to-one correspondence and a pair of corresponding adjacent tower layout nodes and next adjacent tower layout nodes, and One-to-one correspondence and another pair of corresponding adjacent pole tower arrangement nodes and next adjacent pole tower arrangement nodes. In addition, if the pole designed at the pole tower arrangement node is the first pole or the second first pole, then the adjacent pole tower arrangement node and the next adjacent pole tower arrangement node will only have one: the rear pole tower arrangement node and the other rear pole tower arrangement node; and if the pole designed at the pole tower arrangement node is the tail pole or the second tail pole, then the adjacent pole tower arrangement node and the next adjacent pole tower arrangement node will also only have one: the front pole tower arrangement node and the other front pole tower arrangement node.

[0082] S222. For each pair of corresponding adjacent tower arrangement nodes and next-neighboring tower arrangement nodes in the at least one adjacent tower arrangement node and the at least one next-neighboring tower arrangement node, the corresponding connecting line segment angle value is calculated according to the following formula :

[0083]

[0084] In the formula, represents the horizontal coordinate of the position of the tower at the certain tower arrangement node, represents the vertical coordinate of the location of the tower at the certain tower arrangement node, represents the elevation coordinate of the tower at the location of the tower arrangement node, represents the horizontal coordinate of the position of the tower at the adjacent tower arrangement node, represents the ordinate of the location of the tower at the adjacent tower arrangement node, represents the elevation coordinate of the tower at the location of the adjacent tower arrangement node, represents the horizontal coordinate of the position of the tower at the node where the next adjacent tower is arranged, represents the ordinate of the location of the tower at the node where the next adjacent tower is arranged, Indicates the elevation coordinate of the tower where the node of the next neighboring tower is located.

[0085] S223. Using the maximum angle value among all the connection line segment angle values ​​as the index value of the certain tower layout node on the tower connection line segment angle.

[0086] S3. Extracting a multi-dimensional tower feature vector of each tower arrangement node according to the tower model information of each tower arrangement node.

[0087] In the step S3, since the pole tower model information corresponds to but is not limited to the pole tower bearing capacity, pole tower applicable scenarios, pole tower materials, pole tower anti-corrosion methods and / or pole tower appearance requirements, the multi-dimensional pole tower feature vector specifically includes but is not limited to the characteristic value of the pole tower bearing capacity (which may be specifically the specific value of the maximum load), the characteristic value of the pole tower applicable scenario (which may be specifically the scene characteristic value obtained based on the existing one-hot encoding method), the characteristic value of the material used for the pole tower (which may be specifically the scene characteristic value obtained based on the existing label encoding method), the characteristic value of the pole tower anti-corrosion method (which may be specifically the method characteristic value obtained by encoding using the existing glove word embedding encoding technology) and / or the characteristic value of the pole tower appearance requirement (which may be specifically the requirement characteristic value obtained by encoding using the existing glove word embedding encoding technology), etc.

[0088] S4. Determine the weight coefficient of each tower selection influencing factor index in the multiple tower selection influencing factor indexes according to the index value of each tower layout node on the multiple tower selection influencing factor indexes and the multi-dimensional tower feature vector of each tower layout node.

[0089] In step S4, the weight coefficients of the various tower selection influencing factors are used to prepare for the subsequent construction of a machine learning model. Considering that the Pearson correlation coefficient can measure the correlation between two variables X and Y (the specific value is between -1 and 1, where 1 represents a complete positive correlation, 0 represents no linear correlation, and -1 represents a complete negative correlation), the calculation formula of the Pearson correlation coefficient can be used as a reference. For a certain tower selection influencing factor indicator among the multiple tower selection influencing factors, the corresponding weight coefficient can be calculated according to the following formula: :

[0090]

[0091] In the formula, Represents the total number of tower layout nodes, Indicates less than or equal to A positive integer, Indicates The index value of a tower layout node on the index of the tower selection influencing factor, It represents the average value of all tower layout nodes on the influencing factor index of a tower selection. Indicates the The multi-dimensional tower feature vector of the tower arrangement nodes has , represents the total number of dimensions of the multi-dimensional tower feature vector, Indicates less than or equal to A positive integer, Indicates the The first node of the tower arrangement The characteristic value of the tower, represents the average value of the multi-dimensional tower feature vectors of all tower arrangement nodes and has , Indicates the first node of all tower layout nodes Based on the above calculation formula, if the weight coefficient If it is close to 1, it means that the certain tower selection influencing factor index has a strong correlation with the tower selection. Therefore, the greater the weight coefficient of the certain tower selection influencing factor index is, the greater the weight coefficient is. Close to 0, it indicates that the certain pole tower selection influencing factor index has a weak correlation with the pole tower selection, so the weight coefficient of the certain pole tower selection influencing factor index is smaller (Note: the sum of the weight coefficients of the various pole tower selection influencing factor indexes in this embodiment does not necessarily need to be equal to 1). In addition, the weight coefficients of the various pole tower selection influencing factor indexes can also be determined by using the entropy weight method or other methods according to the index values ​​of the various pole tower layout nodes on the multiple pole tower selection influencing factor indexes and the multi-dimensional pole tower feature vectors of the various pole tower layout nodes.

[0092] S5. Apply the index values ​​of each tower layout node on the multiple tower selection influencing factor indicators, the tower model information of each tower layout node and the weight coefficient of each tower selection influencing factor indicator to construct a machine learning model and perform model training to obtain a tower selection model in which the model input items are the index values ​​on the multiple tower selection influencing factor indicators and the model output items are the tower model information.

[0093] In step S5, the machine learning model refers to a mathematical model based on a machine learning algorithm (for example, a machine learning algorithm based on a support vector machine, K nearest neighbor method, stochastic gradient descent method, multilayer perceptron, decision tree, back propagation neural network or radial basis function network, etc.) for processing data and making predictions; in the field of machine learning, there are many different types of models to choose from, and each model has its applicable scenarios and advantages. In order to make full use of the weight coefficients of the indicators of the factors affecting the selection of the tower, so as to improve the accuracy of the prediction results, the machine learning model preferably adopts a CART (Classification and Regression Tree) decision tree model for predicting the tower model, wherein the specific formula of the CART decision tree model can be expressed as follows:

[0094]

[0095] In the formula, Represents the input sample The predicted category, express The leaf node where represents the indicator function, Indicates The true category of samples, Indicates the index of the category.

[0096] In the step S5, it is considered that in the traditional decision tree algorithm CART, the division of child nodes is usually based on features and thresholds, that is, the data set is divided according to whether a certain value of the feature in the data set is greater than or less than a certain threshold. However, in the scenario of intelligent selection of pole towers, the selection goal is to predict the pole tower model of new data based on known pole tower models, rather than dividing the data set based on features and thresholds. Therefore, in order to meet the needs of intelligent selection of pole towers, this embodiment modifies the CART algorithm, that is, preferably, the index values ​​of each pole tower layout node on the multiple pole tower selection influencing factor indicators, the pole tower model information of each pole tower layout node, and the weight coefficients of each pole tower selection influencing factor indicators are applied to construct a machine learning model and perform model training, including but not limited to the following points (A) to (C).

[0097] (A) Arrange the influencing factors of tower selection in descending order according to the weight coefficient to obtain an index sequence, and in the CART decision tree model, the index sequence is arranged along the direction from the root node to the leaf node. node layers, using the first to last node in the index sequence The influencing factors of tower selection are divided into sub-nodes, among which: Indicates less than or equal to A positive integer, The total number of indices representing the multiple influencing factors of tower selection. Specifically, at the root node layer (i.e., the first node layer along the direction from the root node to the leaf node in the CART decision tree model), the influencing factor with the largest weight is selected for the initial sub-node division; at the intermediate node layer (i.e., ), under each branch, further sub-node division is performed layer by layer using secondary influencing factors until each leaf node corresponds to a specific tower model, forming a decision path (i.e., factors with large influencing factors are divided into large classification decision paths, and factors with small influencing factors are refined layer by layer to ultimately determine the tower model of each leaf node).

[0098] (B) The sub-node classification criterion in the CART decision tree model is: for each non-root node, the corresponding category is determined according to the tower model of all tower layout nodes in the corresponding sample set, wherein the non-root node refers to a node that is not the root node. In other words, each leaf node represents a specific tower model, rather than dividing the data set according to features and thresholds. Specifically, during the construction of the decision tree, the frequency of occurrence of different tower models in the sample set of non-root nodes is counted, and then the tower model with the highest frequency is selected as the category of the node. The advantage of this modification is that the decision tree can be more directly matched with the tower model, so that the decision tree can more accurately predict the tower model of new data without the need for complex division and calculation of features and thresholds, thereby improving the efficiency and accuracy of the algorithm, and also making the algorithm easier to understand and explain.

[0099] (C) The sub-node division stopping condition in the CART decision tree model is: for each non-root node, determine whether the frequency of occurrence of a certain tower model in the corresponding sample set exceeds the preset frequency threshold and whether the total number of samples exceeds the preset sample number threshold. If both are true, the corresponding node is used as a leaf node, and the tower model with the highest frequency of occurrence is used as the category of the leaf node. The modification of the above-mentioned stopping condition is to ensure that the algorithm can determine the final category according to the tower model in the sample set when reaching the leaf node, so as to adapt to the needs of intelligent selection of towers, that is, considering that the stopping condition of the traditional decision tree algorithm CART is usually determined according to the purity of the node (that is, whether the category of the samples in the node is consistent) or whether the number of samples in the node is less than a certain threshold, however, in the scenario of intelligent selection of towers, the selection goal is to predict the tower model of new data based on the known tower model, so it is necessary to modify the stopping condition so that the algorithm can determine the final category according to the tower model in the sample set when reaching the leaf node. In addition, the preset frequency threshold can be specifically exemplified as 70%, and the preset sample number threshold can be specifically exemplified as 50.

[0100] In step S5, based on the above points (A) to (C), the modification advantage is that the tower model of the new data can be predicted more accurately, because the algorithm can directly determine the final category according to the tower model in the sample set at the leaf node, without the need for complex calculation and division according to features and thresholds. In addition, other construction and training points of the CART decision tree model are existing technical means and will not be repeated here.

[0101] S6. Obtain the grid line information of the target tower arrangement node in the distribution network to be designed.

[0102] In step S6, in order to reduce the tedious steps of manual input, the power grid line information of the target tower arrangement node can also be automatically obtained through the GIS engine.

[0103] S7. According to the power grid line information of the target pole tower arrangement node, extract the index value of the target pole tower arrangement node on the multiple pole tower selection influencing factor indexes.

[0104] In step S7, the specific extraction method can refer to the aforementioned step S2, which will not be repeated here.

[0105] S8. Importing the index values ​​of the target pole tower arrangement node on the multiple pole tower selection influencing factor indexes into the pole tower selection model, and outputting the pole tower model information of the target pole tower arrangement node.

[0106] After step S8, in order to achieve the purpose of further automatic design based on the intelligent selection result of the pole tower, preferably, after the pole tower model information of the target pole tower arrangement node is output, the method also includes but is not limited to: according to the pole tower model information of the target pole tower arrangement node, the distribution network pole tower at the target pole tower arrangement node is automatically drawn in the distribution network design diagram, and the tower related information is marked, wherein the tower related information includes but is not limited to the power grid line information and / or the pole tower model information of the target pole tower arrangement node. In addition, if the user fine-tunes or modifies the distribution network pole tower at the target pole tower arrangement node, the fine-tuning or modification result can also be used as a new historical distribution network design data, and then the pole tower selection model is optimized and trained based on the aforementioned steps S1 to S6, so as to further improve the accuracy of the intelligent selection result of the model and the user's satisfaction.

[0107] Therefore, based on the intelligent selection method for distribution network towers described in the aforementioned steps S1 to S8, a new solution for automatic tower selection based on historical distribution network design data and machine learning models is provided, that is, firstly, the grid line information and tower model information of each tower layout node in the historical distribution network design data are obtained, and then the multidimensional tower feature vector of each node and the index value on multiple tower selection influencing factor indicators are extracted according to the acquisition results, and the weight coefficient of each indicator is determined according to the extraction result, and then the extraction result and the weight coefficient of each indicator are applied to construct a machine learning model and perform model training to obtain a tower selection model, and finally the index value of the target tower layout node on multiple indicators is imported into the model, and the corresponding tower model information is output, so that the manual selection of the designer is not required, which greatly reduces the human resource demand and improves the selection efficiency, and avoids human errors, thereby making it more convenient and efficient for designers to design the distribution network, and facilitating practical application and promotion.

[0108] like Figure 3 As shown, the second aspect of this embodiment provides a virtual device for implementing the intelligent selection method for distribution network towers described in the first aspect, including a node information acquisition unit, an influence index extraction unit, a feature vector extraction unit, an index weight determination unit, a model construction training unit and a selection model application unit;

[0109] The node information acquisition unit is used to acquire the power grid line information and tower model information of each tower arrangement node in the historical distribution network design data;

[0110] The influencing index extraction unit is communicatively connected to the node information acquisition unit, and is used to extract the index value of each tower arrangement node on multiple tower selection influencing factor indicators according to the power grid line information of each tower arrangement node;

[0111] The feature vector extraction unit is communicatively connected to the node information acquisition unit, and is used to extract the multi-dimensional tower feature vector of each tower arrangement node according to the tower model information of each tower arrangement node;

[0112] The indicator weight determination unit is respectively connected to the influencing indicator extraction unit and the feature vector extraction unit for communication, and is used to determine the weight coefficient of each tower selection influencing factor indicator in the multiple tower selection influencing factor indicators according to the indicator value of each tower arrangement node on the multiple tower selection influencing factor indicators and the multi-dimensional tower feature vector of each tower arrangement node;

[0113] The model construction training unit is communicatively connected to the indicator weight determination unit, and is used to apply the indicator values ​​of each tower arrangement node on the multiple tower selection influencing factor indicators, the tower model information of each tower arrangement node, and the weight coefficients of each tower selection influencing factor indicator to construct a machine learning model and perform model training to obtain a tower selection model in which the model input items are the indicator values ​​on the multiple tower selection influencing factor indicators and the model output items are the tower model information;

[0114] The node information acquisition unit is further used to acquire the grid line information of the target tower arrangement node in the distribution network to be designed;

[0115] The influencing index extraction unit is further used to extract the index value of the target tower arrangement node on the multiple tower selection influencing factor indicators according to the power grid line information of the target tower arrangement node;

[0116] The selection model application unit is respectively communicatively connected to the influencing index extraction unit and the model construction training unit, and is used to import the index value of the target tower layout node on the multiple tower selection influencing factor indicators into the tower selection model, and output the tower model information of the target tower layout node.

[0117] The working process, working details and technical effects of the aforementioned device provided in the second aspect of this embodiment can be referred to the intelligent selection method for distribution network towers described in the first aspect, and will not be repeated here.

[0118] like Figure 4 As shown, the third aspect of this embodiment provides a computer device for executing the intelligent selection method for distribution network pole towers as described in the first aspect, including a memory, a processor and a transceiver that are sequentially connected in communication, wherein the memory is used to store a computer program, the transceiver is used to send and receive messages, and the processor is used to read the computer program to execute the intelligent selection method for distribution network pole towers as described in the first aspect. For example, the memory may include, but is not limited to, a random access memory (RAM), a read-only memory (ROM), a flash memory, a first-in-first-out memory (FIFO) and / or a first-in-last-out memory (FILO), etc.; the processor may include, but is not limited to, a microprocessor of the STM32F105 series. In addition, the computer device may also include, but is not limited to, a power module, a display screen and other necessary components.

[0119] The working process, working details and technical effects of the aforementioned computer equipment provided in the third aspect of this embodiment can be referred to the intelligent selection method for distribution network towers described in the first aspect, and will not be repeated here.

[0120] In a fourth aspect of this embodiment, there is provided a computer-readable storage medium storing instructions including the method for intelligent selection of distribution network pole towers as described in the first aspect, that is, the computer-readable storage medium stores instructions, and when the instructions are executed on a computer, the method for intelligent selection of distribution network pole towers as described in the first aspect is executed. The computer-readable storage medium refers to a carrier for storing data, which may include but is not limited to computer-readable storage media such as floppy disks, optical disks, hard disks, flash memories, USB flash drives, and / or memory sticks, and the computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices.

[0121] The working process, working details and technical effects of the aforementioned computer-readable storage medium provided in the fourth aspect of this embodiment can be referred to the intelligent selection method for distribution network towers as described in the first aspect, and will not be repeated here.

[0122] In a fifth aspect, the present embodiment provides a computer program product, including a computer program or an instruction, wherein when the computer program or the instruction is executed by a computer, the method for intelligent selection of distribution network towers as described in the first aspect is implemented. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device.

[0123] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the protection scope of the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for intelligent selection of distribution network towers. It is characterized in that include: Obtaining grid line information and tower model information of each tower arrangement node in the historical distribution network design data; Extracting the index values ​​of the various pole tower arrangement nodes on multiple pole tower selection influencing factor indexes according to the power grid line information of the various pole tower arrangement nodes; Extracting a multi-dimensional tower feature vector of each tower arrangement node according to the tower model information of each tower arrangement node; Determine the weight coefficient of each tower selection influencing factor index in the multiple tower selection influencing factor indexes according to the index value of each tower layout node on the multiple tower selection influencing factor indexes and the multi-dimensional tower feature vector of each tower layout node; Apply the index values ​​of each tower layout node on the multiple tower selection influencing factor indicators, the tower model information of each tower layout node, and the weight coefficients of each tower selection influencing factor indicator to construct a machine learning model and perform model training to obtain a tower selection model in which the model input items are the index values ​​on the multiple tower selection influencing factor indicators and the model output items are the tower model information; Acquire the grid line information of the target tower arrangement node in the distribution network to be designed; Extracting the index value of the target pole tower arrangement node on the multiple pole tower selection influencing factor indexes according to the power grid line information of the target pole tower arrangement node; Importing the index value of the target pole tower arrangement node on the multiple pole tower selection influencing factor indexes into the pole tower selection model, and outputting the pole tower model information of the target pole tower arrangement node; According to the pole tower model information of the target pole tower arrangement node, the distribution network pole tower at the target pole tower arrangement node is automatically drawn in the distribution network design diagram, and the pole tower related information is marked, wherein the pole tower related information includes the power grid line information and / or the pole tower model information of the target pole tower arrangement node; If the user fine-tunes or modifies the distribution network tower where the node is arranged on the target tower, the fine-tuning or modification result will be used as a new historical distribution network design data and used to optimize the tower selection model.

2. The intelligent selection method for distribution network towers according to claim 1, It is characterized in that The power grid line information includes the location of the pole tower, the environmental parameters of the pole tower, the voltage of the pole tower support line and / or the number of pole tower support loops, and the multiple pole tower selection influencing factor indicators include the length of the pole tower connection line segment, the angle of the pole tower connection line segment, the environmental parameters of the pole tower, the voltage of the pole tower support line and / or the number of the pole tower support loops; And / or, the tower model information corresponds to the tower bearing capacity, tower applicable scenarios, tower materials, tower anti-corrosion methods and / or tower appearance requirements, and the multi-dimensional tower feature vector includes the characteristic value of the tower bearing capacity, the characteristic value of the tower applicable scenarios, the characteristic value of the tower material, the characteristic value of the tower anti-corrosion method and / or the characteristic value of the tower appearance requirements.

3. The intelligent selection method for distribution network towers according to claim 1, It is characterized in that When the power grid line information includes the location of the pole tower and the multiple pole tower selection influencing factor indicators include the length of the pole tower connection line segment, the index value of each pole tower layout node on the multiple pole tower selection influencing factor indicators is extracted according to the power grid line information of each pole tower layout node, including: Obtaining the location of a pole tower arrangement node and at least one adjacent pole tower arrangement node, wherein the adjacent pole tower arrangement node refers to a preceding pole tower arrangement node located before the pole tower arrangement node along the direction of the power grid line or a succeeding pole tower arrangement node located after the pole tower arrangement node along the direction of the power grid line; For each adjacent pole tower arrangement node in the at least one adjacent pole tower arrangement node, the corresponding connecting line segment length value is calculated according to the following formula: : In the formula, represents the horizontal coordinate of the position of the tower at the certain tower arrangement node, represents the vertical coordinate of the location of the tower at the certain tower arrangement node, represents the elevation coordinate of the tower at the location of the tower arrangement node, represents the horizontal coordinate of the position of the tower at the adjacent tower arrangement node, represents the ordinate of the location of the tower at the adjacent tower arrangement node, Indicates the elevation coordinates of the tower at the location of the adjacent tower arrangement node; The maximum length value among all the connection line segment length values ​​is used as the index value of the certain tower arrangement node on the tower connection line segment length; And / or, when the power grid line information includes the location of the pole tower and the multiple pole tower selection influencing factor indicators include the pole tower connection line segment angle, according to the power grid line information of each pole tower layout node, the index value of each pole tower layout node on the multiple pole tower selection influencing factor indicators is extracted, including: Obtaining the location of the tower of a certain tower arrangement node, at least one adjacent tower arrangement node, and at least one next-neighboring tower arrangement node, wherein the adjacent tower arrangement node refers to a previous tower arrangement node located before the certain tower arrangement node along the direction of the power grid line or a subsequent tower arrangement node located after the certain tower arrangement node along the direction of the power grid line, and the next-neighboring tower arrangement node refers to another previous tower arrangement node located before the previous tower arrangement node along the direction of the power grid line or another subsequent tower arrangement node located before the subsequent tower arrangement node along the direction of the power grid line, and the at least one adjacent tower arrangement node corresponds one-to-one to the at least one next-neighboring tower arrangement node; For each pair of corresponding adjacent tower arrangement nodes and next-neighboring tower arrangement nodes in the at least one adjacent tower arrangement node and the at least one next-neighboring tower arrangement node, the corresponding connecting line segment angle value is calculated according to the following formula: : In the formula, represents the horizontal coordinate of the position of the tower at the certain tower arrangement node, represents the vertical coordinate of the location of the tower at the certain tower arrangement node, represents the elevation coordinate of the tower at the location of the tower arrangement node, represents the horizontal coordinate of the position of the tower at the adjacent tower arrangement node, represents the ordinate of the location of the tower at the adjacent tower arrangement node, represents the elevation coordinate of the tower at the location of the adjacent tower arrangement node, represents the horizontal coordinate of the position of the tower at the node where the next adjacent tower is arranged, represents the ordinate of the location of the tower at the node where the next adjacent tower is arranged, Indicates the elevation coordinates of the tower at the location of the node of the next neighboring tower arrangement; The maximum angle value among all the connection line segment angle values ​​is used as the index value of the certain pole tower layout node on the pole tower connection line segment angle.

4. The intelligent selection method for distribution network towers according to claim 1, It is characterized in that Determining the weight coefficient of each tower selection influencing factor index in the multiple tower selection influencing factor indexes according to the index value of each tower layout node on the multiple tower selection influencing factor indexes and the multi-dimensional tower feature vector of each tower layout node, including: For a certain tower selection influencing factor index among the multiple tower selection influencing factor indexes, the corresponding weight coefficient is calculated according to the following formula: : In the formula, Represents the total number of tower layout nodes, Indicates less than or equal to A positive integer, Indicates The index value of a tower layout node on the index of the influencing factor of the tower selection, It represents the average value of all tower layout nodes on the influencing factor index of a tower selection. Indicates the The multi-dimensional tower feature vector of the tower arrangement nodes has , represents the total number of dimensions of the multi-dimensional tower feature vector, Indicates less than or equal to A positive integer, Indicates the The first node of the tower arrangement The characteristic value of the tower, represents the average value of the multi-dimensional tower feature vectors of all tower arrangement nodes and has , Indicates the first node of all tower layout nodes The average value of the tower characteristics.

5. The intelligent selection method for distribution network towers according to claim 1, It is characterized in that When the machine learning model adopts a CART decision tree model, the index value of each tower layout node on the multiple tower selection influencing factor indicators, the tower model information of each tower layout node and the weight coefficient of each tower selection influencing factor indicator are applied to construct a machine learning model and perform model training, including: Arrange the influencing factors of tower selection in descending order according to the weight coefficient to obtain an index sequence, and in the CART decision tree model, the index sequence is the first index sequence from the root node to the leaf node. node layers, using the first to last node in the index sequence The influencing factors of tower selection are divided into sub-nodes, among which: Indicates less than or equal to A positive integer, The total number of indicators representing the multiple influencing factors of tower selection; The sub-node classification criterion in the CART decision tree model is: for each non-root node, the corresponding category is determined according to the pole tower models of all pole tower arrangement nodes in the corresponding sample set, wherein the non-root node refers to a node that is not the root node; The stopping condition for sub-node division in the CART decision tree model is: for each non-root node, determine whether the frequency of occurrence of a certain tower model in the corresponding sample set exceeds the preset frequency threshold and whether the total number of samples exceeds the preset sample number threshold. If both are true, the corresponding node is taken as a leaf node, and the tower model with the highest frequency of occurrence is taken as the category of the leaf node.

6. A computer device, It is characterized in that It includes a memory, a processor and a transceiver which are communicatively connected in sequence, wherein the memory is used to store a computer program, the transceiver is used to send and receive messages, and the processor is used to read the computer program to execute the distribution network pole tower intelligent selection method as described in any one of claims 1 to 5.

7. A computer-readable storage medium, Features The computer-readable storage medium stores instructions, and when the instructions are executed on the computer, the distribution network pole tower intelligent selection method as described in any one of claims 1 to 5 is executed.

8. A computer program product comprising a computer program or instructions, It is characterized in that When the computer program or the instruction is executed by a computer, the method for intelligent selection of distribution network towers as claimed in any one of claims 1 to 5 is implemented.

Images