A tower installation risk assessment method and system based on communication network

By building a risk factor network and prediction model and using IoT devices to conduct real-time risk assessment, the problems of low efficiency and poor accuracy of risk assessment in tower installation are solved, and the safety and reliability of tower installation are improved.

CN118627914BActive Publication Date: 2024-12-03BEIJING HUAXUN COMM TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411095993.5
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-12
Publication Date
2024-12-03
Estimated Expiration
2044-08-12

AI Technical Summary

Technical Problem

In the prior art, tower installation risk assessment is inefficient and poorly accurate, and cannot effectively respond to the network dependence, electromagnetic compatibility and network security challenges of communication networks.

Method used

The original data of the tower installation site is collected through IoT devices, a risk factor network is built, degree distribution, clustering coefficient and shortest path length are calculated, key nodes are determined, risk prediction models are built based on these structural characteristics, and real-time risk assessment is carried out.

Benefits of technology

It realizes a comprehensive, accurate and real-time risk assessment of tower installation, improves the accuracy and robustness of risk prediction, helps to discover and prevent risks in advance, and enhances the safety and reliability of tower installation.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118627914B_ABST
    Figure CN118627914B_ABST
Patent Text Reader

Abstract

The present application provides a tower installation risk assessment method and system based on a communication network. Among them, the original data is collected; multiple potential risk factors affecting the tower installation are extracted from the original data and used as nodes of the risk factor network; the risk factor network is constructed; the degree distribution, clustering coefficient and shortest path length of the risk factor network are calculated to obtain the structural characteristics of the risk factor network; according to the structural characteristics of the risk factor network, the key nodes of the risk factor network are determined, and the key nodes are used as target risk factors; based on the target risk factors, a risk prediction model is constructed; real-time data is collected through Internet of Things devices and input into the risk prediction model, so as to perform risk assessment based on real-time data through the risk prediction model, and output the risk assessment results. The technical solution provided by the present application can improve the efficiency and accuracy of tower installation risk assessment.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The embodiments of the present application relate to the field of data processing technology, and in particular to a tower installation risk assessment method and system based on a communication network. Background Art

[0002] In the context of the current rapid development of communication technology, especially the widespread application of communication networks, towers, as a key component of supporting infrastructure, face unprecedented challenges and risks in their installation and operation. Traditional tower installation risk assessment methods mainly rely on engineers' experience and preliminary design data. This assessment method has obvious shortcomings when facing the new needs of the times. Specifically, tower installation based on communication networks has the following major background defects:

[0003] First, the network dependence has increased significantly. With its ultra-high speed and ultra-low latency, the communication network requires that the tower not only be able to carry more antenna units in terms of structural design, but also have highly integrated functions to support large-scale network equipment deployment. This high degree of dependence on the network makes the tower system extremely complex. Any small fault point may cause a chain reaction, seriously affecting the stability and service quality of the entire network.

[0004] Secondly, electromagnetic compatibility has become a new challenge. The network uses high-frequency signal transmission, and its electromagnetic radiation characteristics have changed significantly compared with traditional communication frequency bands. This may not only interfere with surrounding electronic equipment, such as civil aviation navigation systems, satellite communication systems, etc., but may also be reversely interfered by these systems, affecting the quality and coverage of the signal. If the electromagnetic compatibility problem is not handled properly, it will seriously restrict the actual application effect of the network and user experience.

[0005] Finally, cybersecurity threats are becoming increasingly prominent. The openness and extensive connectivity of the network have greatly improved the efficiency of information transmission, but they have also provided more entry points for hacker attacks. Once a malicious network attack occurs, not only will user data be at risk of being leaked, but more seriously, the tower and its associated communication equipment may be remotely controlled, causing physical damage or service interruption.

[0006] In summary, the risk assessment of tower installation based on communication networks must go beyond the traditional assessment framework and introduce advanced monitoring methods, data analysis technologies and intelligent early warning systems to cope with emerging challenges such as network dependence, electromagnetic compatibility and network security, and ensure the safe and stable operation of infrastructure. Summary of the invention

[0007] The embodiments of the present application provide a tower installation risk assessment method and system based on a communication network, so as to solve the problems of poor efficiency and low accuracy of tower installation risk assessment in the prior art.

[0008] In a first aspect, an embodiment of the present application provides a tower installation risk assessment method based on a communication network, comprising:

[0009] Collect raw data from tower installation sites through IoT devices;

[0010] Extracting multiple potential risk factors affecting the installation of the iron tower from the original data, and using the multiple potential risk factors as nodes of a risk factor network;

[0011] Taking the factor relationships among the multiple potential risk factors as edges of the risk factor network to construct a risk factor network;

[0012] By calculating the degree distribution, clustering coefficient and shortest path length of the risk factor network, the structural characteristics of the risk factor network are obtained;

[0013] Determining key nodes of the risk factor network according to structural characteristics of the risk factor network, and using the key nodes as target risk factors;

[0014] Based on the target risk factors, construct a risk prediction model;

[0015] Real-time data is collected through the Internet of Things device and input into the risk prediction model, so that risk assessment is performed based on the real-time data through the risk prediction model, and the risk assessment result is output.

[0016] Optionally, before constructing the risk factor network by calculating the degree distribution, clustering coefficient and shortest path length of the risk factor network, the method further includes:

[0017] Extracting node features from the risk factor network, and training a degree distribution prediction model based on the node features, so as to predict the degree of each node in the risk factor network through the trained degree distribution prediction model to obtain a degree distribution;

[0018] The degree refers to the number of edges that directly connect the node to other nodes, and the degree distribution refers to the distribution of the degrees of all nodes in the risk factor network.

[0019] Optionally, before constructing the risk factor network by calculating the degree distribution, clustering coefficient and shortest path length of the risk factor network, the method further includes:

[0020] Determine a heuristic function, wherein the heuristic function is used to calculate an expected cost from a current node to a target node;

[0021] Setting basic parameters of the ACO algorithm, wherein the basic parameters include at least: the number of ants, the evaporation rate of pheromones, and the importance weight of pheromones;

[0022] In each iteration cycle, according to the pheromone concentration of the current node and the heuristic function, the ant selects the next node to construct a path, wherein the pheromone concentration is determined according to the evaporation rate of the pheromone and the importance weight of the pheromone;

[0023] At the end of each iteration, the pheromone concentration of each node is updated according to the path quality of the path until the maximum number of iterations is reached or the shortest path is found;

[0024] The shortest path length is extracted from the shortest path.

[0025] Optionally, before constructing the risk factor network by calculating the degree distribution, clustering coefficient and shortest path length of the risk factor network, the method further includes:

[0026] By formula group:

[0027] , calculating the clustering coefficient of each of the nodes in the risk factor network;

[0028] in, Represented as the dynamic weight between node i and node j at time point t; Represented as the state vector of node i at time point t, Represented as the state vector of node j at time point t; It is represented as the state change vector of node i, the difference between the state at time point t and the previous time point; It is represented as the state change vector of node j, the difference between the state at time point t and the previous time point; It is represented as the distance between node i and node j; function 𝑓 is represented as a nonlinear function, which is used to calculate the dynamic weight;

[0029] in, It is expressed as the dynamic spatiotemporal weighted clustering coefficient of node i at time t; It is represented as the weighted degree of node i at time point t; It is represented as the dynamic weight between node j and node l at time point t, It is represented as the dynamic weight between node l and node i at time point t; σ is represented as the scale parameter that controls the influence of distance on weight;

[0030] in, It is expressed as the final dynamic spatiotemporal weighted clustering coefficient of node i at time t; It is expressed as a weight factor, which determines the relative importance of the current clustering coefficient and the predicted clustering coefficient; Represented as node i at time point The predicted dynamic spatiotemporal weighted clustering coefficient; Expressed as the difference between the predicted future time point and the current time point.

[0031] Optionally, determining the key nodes of the risk factor network according to the structural characteristics of the risk factor network includes:

[0032] Simulating the diffusion degree of risk factors according to the structural characteristics of the risk factor network;

[0033] Calculate the benefit of each node in the risk factor network by presetting the benefit matrix;

[0034] Calculate the comprehensive influence of each node according to the diffusion degree of the risk factor and the income of each node;

[0035] According to the ranking of the comprehensive influence of each node, the top K% nodes are identified as key nodes;

[0036] The step of simulating the diffusion of risk factors according to the structural characteristics of the risk factor network comprises:

[0037] By formula: , simulating the diffusion degree of risk factors; among them, It is expressed as the rate of change of risk exposure of node i over time; It is expressed as the exposure degree of node i to the risk factor; It is represented as the rate at which node i naturally reduces its risk exposure; Expressed as the risk transmission rate; It is represented as the weighted sum of the impact of node i’s neighbor node j on its risk exposure, where: Represented as the set of neighbor nodes of node i; Represented as the dynamic weight between node i and node j at time point t; It is represented as the risk exposure degree of neighbor node j at time point t;

[0038] The evaluating the benefit of each node in the risk factor network includes:

[0039] Through the profit matrix:

[0040] , calculate the benefit of each node in the risk factor network; where, Expressed as the revenue Ri of node i; It represents the benefit obtained by node i when node i chooses to cooperate and all its neighboring nodes also choose to cooperate; b represents the benefit obtained by node i when node i chooses to cooperate and at least one of its neighboring nodes chooses to betray, and b benefit is lower than a benefit; c represents the benefit obtained by node i when node i chooses to betray and all its neighboring nodes choose to cooperate, and c benefit is higher than b benefit and lower than a benefit; d represents the benefit obtained by node i when node i chooses to betray and at least one of its neighboring nodes chooses to betray, and d benefit is the lowest;

[0041] The comprehensive influence of each node is calculated according to the diffusion degree of the risk factor and the income of each node, including:

[0042] By formula: , calculate the comprehensive influence of each node; where, It is expressed as the comprehensive influence of node i, It is expressed as a weight factor between 0 and 1, which is used to balance the rate of change of the risk exposure of node i over time and the influence of the income Ri of each node on the comprehensive influence; It is expressed as the absolute value integral of the state change rate of node i in the time interval [0, T], It is expressed as the benefit of node i in the cooperative game, which is determined according to the behavior of node i and its neighboring nodes;

[0043] The step of identifying the top K% of nodes as key nodes according to the ranking of the comprehensive influence of each node includes:

[0044] By formula: , identify the first K% of nodes as key nodes, where Represented as a set of key nodes, Represented as node i, is represented as the set of all nodes in the risk factor network; It is represented as the comprehensive influence of node i; Represented as a preset threshold, used to filter key nodes that meet the conditions.

[0045] Optionally, determining the key nodes of the risk factor network according to the structural characteristics of the risk factor network includes:

[0046] According to the structural characteristics of the risk factor network, the degree centrality, betweenness centrality, closeness centrality and eigenvector centrality of each of the nodes are calculated;

[0047] Calculate the comprehensive centrality score of each node based on the degree centrality, betweenness centrality, closeness centrality and eigenvector centrality of each node;

[0048] If the comprehensive centrality score of the node meets the preset threshold, the node is determined to be a key node;

[0049] Among them, the degree centrality refers to the number of edges directly connected to the node; the betweenness centrality refers to the frequency of the node appearing in all shortest paths in the risk factor network; the closeness centrality refers to the inverse of the average distance of the node to all other nodes in the risk factor network; and the eigenvector centrality refers to the importance of the nodes adjacent to the node.

[0050] Optionally, the calculating of the degree centrality, betweenness centrality, closeness centrality and eigenvector centrality of each of the nodes includes:

[0051] By formula: , calculate the degree centrality of each of the nodes, where, Represented as the node The degree of It is expressed as the total number of nodes in the risk factor network;

[0052] By formula: , calculate the betweenness centrality of each of the nodes, where, Represented as a node To Node The number of shortest paths, Represented as nodes in all shortest paths of the risk factor network The frequency of occurrence, is represented as the set of all nodes in the risk factor network;

[0053] By formula: , calculate the closeness centrality of each of the nodes, where, is represented by the total number of nodes in the risk factor network, Represented as a node To Node The shortest path length;

[0054] By formula: , calculate the eigenvector centrality of each of the nodes, where, Represented as a node The set of all adjacent nodes of is the main eigenvalue, Represented as a node The eigenvector centrality of the node Belongs to the node The adjacent nodes of

[0055] The step of calculating the comprehensive centrality score of each node based on the degree centrality, betweenness centrality, closeness centrality and eigenvector centrality of each node comprises:

[0056] By formula: , calculate the comprehensive centrality score of each of the nodes;

[0057] in, Represented as a node The comprehensive centrality score of Respectively expressed as The weight of Respectively expressed as The average value of Respectively expressed as The standard deviation of .

[0058] In a second aspect, an embodiment of the present application provides a tower installation risk assessment system based on a communication network, comprising:

[0059] A collection module is used to collect raw data of the tower installation site through IoT devices;

[0060] An identification module is used to extract multiple potential risk factors that affect the installation of the iron tower from the original data, and use the multiple potential risk factors as nodes of the risk factor network; use the factor relationships between the multiple potential risk factors as edges of the risk factor network to construct the risk factor network; calculate the degree distribution, clustering coefficient and shortest path length of the risk factor network to obtain the structural characteristics of the risk factor network; determine the key nodes of the risk factor network according to the structural characteristics of the risk factor network, and use the key nodes as target risk factors;

[0061] A construction module, used to construct a risk prediction model based on the target risk factors;

[0062] An assessment module is used to collect real-time data through the Internet of Things device and input it into the risk prediction model, so as to perform risk assessment based on the real-time data through the risk prediction model and output the risk assessment result.

[0063] In a third aspect, an embodiment of the present application provides a computing device, including a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement the tower installation risk assessment method based on the communication network as described in the first aspect above.

[0064] In a fourth aspect, an embodiment of the present application provides a computer storage medium storing a computer program, which, when executed by a computer, implements the tower installation risk assessment method based on a communication network as described in the first aspect above.

[0065] In an embodiment of the present application, raw data of a tower installation site is collected through an Internet of Things device; multiple potential risk factors affecting the tower installation are extracted from the raw data, and the multiple potential risk factors are used as nodes of a risk factor network; the factor relationships between the multiple potential risk factors are used as edges of the risk factor network to construct a risk factor network; the degree distribution, clustering coefficient and shortest path length of the risk factor network are calculated to obtain the structural characteristics of the risk factor network; based on the structural characteristics of the risk factor network, key nodes of the risk factor network are determined, and the key nodes are used as target risk factors; based on the target risk factors, a risk prediction model is constructed; real-time data is collected through the Internet of Things device and input into the risk prediction model, so as to perform risk assessment based on the real-time data through the risk prediction model, and output the risk assessment results.

[0066] This application proposes a tower installation risk assessment method based on a communication network, which has the following significant beneficial effects compared to traditional risk assessment methods:

[0067] The original data of the tower installation site is collected through IoT devices to ensure the real-time and accuracy of the data, providing a solid foundation for subsequent analysis. Multiple potential risk factors affecting tower installation are used as nodes of the network, and the relationships between these risk factors are used as edges to construct a sophisticated risk factor network. By calculating the structural characteristics of the risk factor network, such as degree distribution, clustering coefficient, and shortest path length, we can gain an in-depth understanding of the topological structure of the network and the interaction between risk factors. Based on the identified target risk factors, a deep learning model is constructed for risk prediction, which improves the accuracy and robustness of the prediction. Real-time data collected by IoT devices is combined with risk prediction models to conduct real-time risk assessment and timely identify potential risks.

[0068] In summary, the present application scheme can provide a comprehensive, accurate and real-time risk assessment system, which not only helps to discover and prevent risks in advance, but also provides scientific decision-making support for decision makers, thereby improving the safety and reliability of tower installation.

[0069] These and other aspects of the present application will become more clearly understood in the description of the following embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0071] Figure 1 A flow chart of a tower installation risk assessment method based on a communication network provided in an embodiment of the present application;

[0072] Figure 2 A schematic diagram of the structure of a tower installation risk assessment system based on a communication network provided in an embodiment of the present application;

[0073] Figure 3 A schematic diagram of the structure of a computing device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0074] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.

[0075] In some of the processes described in the specification and claims of this application and the above-mentioned figures, multiple operations that appear in a specific order are included, but it should be clearly understood that these operations may not be executed in the order in which they appear in this article or executed in parallel. The serial numbers of the operations, such as 101, 102, etc., are only used to distinguish between different operations, and the serial numbers themselves do not represent any execution order. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions of "first", "second", etc. in this article are used to distinguish different messages, devices, modules, etc., do not represent the order of precedence, and do not limit the "first" and "second" to be different types.

[0076] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of this application.

[0077] Figure 1 A flow chart of a tower installation risk assessment method based on a communication network provided in an embodiment of the present application, such as Figure 1 As shown, the method includes:

[0078] 101. Collect raw data of tower installation sites through IoT devices;

[0079] In this step, the raw data includes at least: geographic information, climate conditions, soil type, earthquake frequency, tower design parameters, material properties, and construction plan. Specifically, IoT devices are deployed at the tower installation site and surrounding areas to collect various environmental and engineering data in real time. These devices may include sensors, cameras, weather stations, geological detectors, etc., which can monitor and record geographic coordinates, temperature and humidity, wind speed and direction, soil composition, seismic activity, tower design parameters (such as height, load capacity), material properties (such as strength, corrosion resistance), and construction progress.

[0080] In the embodiment of the present application, it is assumed that a tower installation project is located in an earthquake-prone area, and the IoT devices are deployed within a radius of 10 kilometers, including:

[0081] Geographic Information System (GIS) sensors that record GPS coordinates and terrain features every 10 minutes.

[0082] The weather station uploads temperature, humidity, wind speed and direction data every 5 minutes.

[0083] Soil tester, records soil pH, salinity and moisture content every day.

[0084] Earthquake monitors monitor seismic activity in real time and send alerts immediately once an earthquake is detected.

[0085] Construction site surveillance cameras, which continuously record video, are used to monitor construction quality and safety measures.

[0086] 102. Extracting multiple potential risk factors affecting the installation of the iron tower from the original data, and using the multiple potential risk factors as nodes of a risk factor network;

[0087] In this step, we first analyze the raw data to identify all potential risk factors related to tower installation. These factors can be geological conditions, climate variables, design specifications, material properties, construction techniques, etc. Each identified risk factor will become a node in the risk factor network.

[0088] In the present embodiment, it is assumed that the original data includes soil sample analysis reports, annual meteorological data, tower design drawings, building material specifications, and construction schedules. From these data, we identify the following potential risk factors as nodes:

[0089] Soil stability (node ​​A);

[0090] Annual average wind speed (node ​​B);

[0091] Tower height (node ​​C);

[0092] Material tensile strength (node ​​D);

[0093] Construction delay rate (node ​​E);

[0094] 103. Using the factor relationships between the multiple potential risk factors as edges of the risk factor network to construct a risk factor network;

[0095] In this step, we analyze the interrelationships between these risk factors, and if there is a causal relationship or correlation between two risk factors, we add an edge to the risk factor network to connect these two nodes. This helps to understand the interactions between risk factors.

[0096] In the embodiment of the present application, in the risk factor network, we find that there is a relationship between node B (annual average wind speed) and node C (tower height), because a taller tower may be affected by stronger winds, so an edge is added between node B and node C. Similarly, we find that there may be an indirect relationship between node D (material tensile strength) and node E (construction delay rate), because material problems may cause construction delays, so an edge is also added between node D and node E.

[0097] 104. Calculating the degree distribution, clustering coefficient, and shortest path length of the risk factor network to obtain structural characteristics of the risk factor network;

[0098] In this step, we calculate the degree distribution, clustering coefficient, and shortest path length of the risk factor network, and these structural features help identify which nodes or risk factors in the network are the most critical.

[0099] In this embodiment of the present application, in our risk factor network, we calculated:

[0100] Node A (soil stability) has a degree of 2, which means it has a direct relationship with nodes B (annual average wind speed) and C (tower height).

[0101] The clustering coefficient was 0.5, indicating that there was a certain degree of clustering in the network and risk factors tended to be associated with each other in small groups.

[0102] The average shortest path length is 2, which means that from any risk factor in the network to another risk factor, it only takes an average of 2 intermediate nodes to reach the destination.

[0103] 105. Determine key nodes of the risk factor network according to structural characteristics of the risk factor network, and use the key nodes as target risk factors.

[0104] In the last step, we determine which risk factors play key roles in the risk factor network based on the calculated structural features. These key nodes may have high degree, high betweenness centrality or be in an important position in the network. We identify these nodes as target risk factors for priority attention and management.

[0105] In the present embodiment, in our network, Node A (soil stability) and Node B (annual average wind speed) show high betweenness centrality because they are located at the intersection of multiple paths in the network, which indicates that these two risk factors have a wide range of impacts on the stability of the entire system. Therefore, we identify Node A and Node B as target risk factors that require special attention and monitoring.

[0106] Through the above steps, we not only constructed a risk factor network, but also identified key risk factors, providing a key direction for subsequent tower installation risk assessment.

[0107] 106. Constructing a risk prediction model based on the target risk factors;

[0108] In this step, based on the identified risk factors, suitable predictive models are developed or selected to assess their potential impact on the tower installation process and long-term stability. The models may be based on statistical, physical principles or machine learning algorithms.

[0109] In the embodiment of the present application, an integrated model based on machine learning is constructed to predict the impact of earthquakes on the stability of iron towers. The model uses historical earthquake data, soil type and iron tower design parameters as input features, and predicts the response of the iron tower under different earthquake intensities through training. In a specific embodiment, the model selects the random forest algorithm, and the training set contains earthquake data and iron tower response records from the past 10 years, and the features include earthquake magnitude, depth, distance, soil type, iron tower height and load capacity.

[0110] 107. Collect real-time data through the Internet of Things device and input it into the risk prediction model, so as to perform risk assessment based on the real-time data through the risk prediction model and output the risk assessment result.

[0111] In this step, the data collected in real time is input into the risk prediction model to dynamically assess the risk level of the tower under current conditions. The risk assessment results output by the model can be used for real-time early warning, decision support and emergency response.

[0112] In the embodiment of the present application, the real-time data stream is preprocessed and input into the earthquake impact prediction model. For example, when an earthquake with a magnitude of 5.5 is detected, the model predicts the maximum displacement, stress distribution and potential damage of the tower. Assuming that the model predicts that the maximum displacement of the tower under the current conditions is 0.2 meters, the stress is concentrated at the bottom connection, and the probability of damage is 5%, this information is immediately sent to the project manager so that appropriate measures can be taken, such as suspending construction, checking structural integrity or initiating emergency plans.

[0113] Through the above steps, the embodiment of the present application provides a systematic and intelligent tower installation risk assessment method, which can effectively integrate and analyze multi-source data and provide a scientific basis for engineering safety.

[0114] Optionally, before constructing the risk factor network by calculating the degree distribution, clustering coefficient and shortest path length of the risk factor network, it is also necessary to calculate the degree distribution of the risk factor network. Specifically, the following process may be included:

[0115] Extracting node features from the risk factor network, and training a degree distribution prediction model based on the node features, so as to predict the degree of each node in the risk factor network through the trained degree distribution prediction model to obtain a degree distribution;

[0116] The degree refers to the number of edges that directly connect the node to other nodes, and the degree distribution refers to the distribution of the degrees of all nodes in the risk factor network.

[0117] In this step, node characteristics: In the risk factor network, each node has its own characteristics, which may include the type of risk factor represented by the node, attributes, historical data, interaction intensity with other risk factors, etc. For example, a node representing soil stability may have characteristics such as soil type, bearing capacity, and groundwater level.

[0118] Degree distribution prediction model: This is a machine learning model that predicts the degree of each node in the risk factor network. The model makes predictions by learning the relationship between node features and node degrees. Commonly used models include decision trees, random forests, support vector machines, or deep learning models.

[0119] Degree: refers to the number of edges directly connected to a node, that is, the connectivity of the node. In the risk factor network, the degree reflects the "social activity" of the node. The higher the degree, the greater the influence of the node in the network.

[0120] Degree distribution: describes the degree distribution of all nodes in the network, usually expressed as a histogram or probability distribution function. Degree distribution can reveal the topological structure of the network, such as whether the network exhibits scale-free characteristics (a few nodes have a large number of edges, and most nodes have only a small number of edges).

[0121] In the embodiment of this application, it is assumed that we are building a tower installation risk factor network, which contains 20 risk factor nodes. In order to predict the degree of each node, we first extract features from each node, such as the geological conditions, climate factors, design parameters, etc. represented by the node. These features may include:

[0122] Characteristics of geological condition nodes: soil type, soil density, groundwater level.

[0123] Characteristics of climate factor nodes: average wind speed, precipitation, temperature change.

[0124] Characteristics of design parameter nodes: tower height, material strength, and seismic resistance level.

[0125] Training degree distribution prediction model

[0126] Data preparation: Collect historical data of each node, including node characteristics and actual degree.

[0127] For example, for the geological condition node, we may have the following data as shown in Table 1:

[0128] Table 1

[0129] Node ID Soil Type Soil density Groundwater level Actual degree A Clay 1.8g / cm³ 2m 4 B sand 1.6g / cm³ 1m 3

[0130] Model training: Choose an appropriate machine learning model, such as random forest, and train it using the above data. The model will learn how to predict the degree based on the node features.

[0131] Suppose we use the random forest model to train and obtain a trained degree distribution prediction model.

[0132] Degree prediction: Input the features of all nodes in the network into the trained degree distribution prediction model to predict the degree of each node.

[0133] Degree distribution: Summarize the predicted degrees of all nodes, plot the degree distribution histogram, or calculate the degree distribution function to understand the topological structure characteristics of the network.

[0134] For example, the predicted degree distribution might be as shown in Table 2 below:

[0135] Table 2

[0136] degree Number of nodes 1 3 2 5 3 6 4 3 5 2 6 1

[0137] Through the above steps, we are able to predict the degree of each node in the risk factor network and construct the degree distribution, which helps to deeply understand the structural characteristics of the network and provide a basis for subsequent risk assessment and management strategies.

[0138] Optionally, before constructing the risk factor network by calculating the degree distribution, clustering coefficient and shortest path length of the risk factor network, it is also necessary to calculate the shortest path length of the risk factor network. Specifically, the following process may be included:

[0139] Determine a heuristic function, which is used to calculate the expected cost from the current node to the target node; set basic parameters of the ACO algorithm, which at least include: the number of ants, the evaporation rate of pheromones, and the importance weight of pheromones; in each iteration cycle, according to the pheromone concentration of the current node and the heuristic function, make the ants select the next node to construct a path, and the pheromone concentration is determined according to the evaporation rate of the pheromone and the importance weight of the pheromone; at the end of each iteration, update the pheromone concentration of each node according to the path quality of the path until the maximum number of iterations is reached or the shortest path is found; extract the shortest path length from the shortest path.

[0140] In this step, the heuristic function: In the Ant Colony Optimization (ACO) algorithm, the heuristic function is a strategy used to guide ants to choose the next node. It usually estimates the expected cost of reaching the target node from the current node based on some distance or cost metric. The design of the heuristic function should reflect the characteristics of the problem so that ants can explore the optimal solution more efficiently.

[0141] The basic parameters of the ACO algorithm are: the number of ants determines the degree of parallelism of the search; the evaporation rate of pheromones simulates the phenomenon that pheromones gradually disappear over time in the real world, which helps to prevent the algorithm from falling into local optimality; the importance weight of pheromones affects the degree of ant's reliance on historical information when choosing a path.

[0142] Pheromone concentration: In the ACO algorithm, pheromones are left on each path. A high pheromone concentration means that this path is frequently chosen by ants, thus guiding subsequent ants to be more likely to choose this path.

[0143] Path quality: This is a measure of the quality of a path and is usually inversely proportional to the cost or length of the path. In the problem of finding the shortest path, path quality is often represented by the inverse of the path length.

[0144] Maximum number of iterations: One of the termination conditions of the ACO algorithm. When the preset maximum number of iterations is reached, the algorithm stops running even if the optimal solution is not found.

[0145] In the embodiment of the present application, a risk factor network is considered, which includes multiple nodes, each node represents a potential risk factor, and the edge represents the correlation between different risk factors. Our goal is to find the shortest path between any two nodes to evaluate the efficiency of risk propagation.

[0146] Determine the heuristic function: Assume that the heuristic function is , which represents the expected cost from node i to node j, can be simply defined as the inverse of the Euclidean distance between two nodes, that is:

[0147]

[0148] in, It is expressed as the straight-line distance from node i to node j.

[0149] Set the basic parameters of the ACO algorithm: set the number of ants to 100, the evaporation rate of pheromone to 0.5 (meaning that after each iteration, the pheromone concentration will be reduced by half), and the importance weight of pheromone to 1.2.

[0150] Path construction: In each iteration cycle, each ant starts from the starting point and moves to the next node according to the pheromone concentration 𝜏 of the current node. 𝑖𝑗 and the heuristic function , use the probability formula to select the next node j, the formula is:

[0151]

[0152] in, Expressed as the importance weight of pheromone, Expressed as the importance weight of the heuristic function, Represented as the neighbor set of node i.

[0153] Update pheromone concentration: At the end of each iteration, the pheromone concentration on each edge is updated according to the length of the path found by the ants. The new pheromone concentration The calculation formula is:

[0154]

[0155] in, Expressed as the evaporation rate of the pheromone, It is expressed as the total amount of pheromone left by the ant on the edge 𝑖𝑗 in this iteration, and its calculation formula is: , where 𝑚 is the number of ants passing through edge 𝑖𝑗, 𝑄 is a constant, is the path length of ant k.

[0156] Termination condition: The algorithm runs until a maximum number of iterations is reached (e.g., 500) or a shortest path is found that satisfies a certain condition.

[0157] Extract the shortest path length: When the algorithm terminates, select the shortest path from all recorded paths as the shortest path and read its length.

[0158] Through the above embodiments, we can effectively use the ACO algorithm to calculate the length of the shortest path between any two points in the risk factor network, which is of great significance for understanding the propagation path and speed of risks.

[0159] Optionally, before constructing the risk factor network by calculating the degree distribution, clustering coefficient and shortest path length of the risk factor network, it is also necessary to calculate the clustering coefficient of the risk factor network. Specifically, the following process may be included:

[0160] By formula group:

[0161] , calculating the clustering coefficient of each of the nodes in the risk factor network;

[0162] in, Represented as the dynamic weight between node i and node j at time point t; Represented as the state vector of node i at time point t, Represented as the state vector of node j at time point t; It is represented as the state change vector of node i, the difference between the state at time point t and the previous time point; It is represented as the state change vector of node j, the difference between the state at time point t and the previous time point; It is represented as the distance between node i and node j; function 𝑓 is represented as a nonlinear function, which is used to calculate the dynamic weight;

[0163] in, It is expressed as the dynamic spatiotemporal weighted clustering coefficient of node i at time t; It is represented as the weighted degree of node i at time point t; It is represented as the dynamic weight between node j and node l at time point t, It is represented as the dynamic weight between node l and node i at time point t; σ is represented as the scale parameter that controls the influence of distance on weight;

[0164] in, It is expressed as the final dynamic spatiotemporal weighted clustering coefficient of node i at time t; It is expressed as a weight factor, which determines the relative importance of the current clustering coefficient and the predicted clustering coefficient; Represented as node i at time point The predicted dynamic spatiotemporal weighted clustering coefficient; Expressed as the difference between the predicted future time point and the current time point.

[0165] In this step, dynamic weight calculation refers to calculating the dynamic weight between node i and node j through the nonlinear function 𝑓 This weight takes into account the node status, status change and distance (which can be understood as geographical distance), reflecting the strength and change of the connection between nodes.

[0166] Weighted degree calculation refers to calculating the weighted degree of node i at time point t , which is the sum of the dynamic weights of the edges directly connected to node i.

[0167] Dynamic spatiotemporal weighted clustering coefficient calculation refers to calculating the dynamic spatiotemporal weighted clustering coefficient of node i at time point t. , which reflects the weighted density of closed loops (triangular structures) between node i and its neighbor nodes.

[0168] Predicted clustering coefficient calculation: Combine the current clustering coefficient and the predicted clustering coefficient to calculate the final dynamic spatiotemporal weighted clustering coefficient of node i at time point t .

[0169] In this embodiment of the application, assume that we are evaluating a risk factor network for tower installation in a communication network. The network contains 5 nodes, representing different risk factors, such as weather conditions, geological stability, construction quality, etc. We will analyze the data collected by the IoT devices:

[0170] Dynamic weight calculation:

[0171] State vector: ,in, represents temperature, Represents humidity, Represents wind speed.

[0172] State change vector: .

[0173] Geographic distance: Assume that the geographical distance between node 1 and node 2 is 10km.

[0174] Non-linear functions:

[0175] ,in, is the squared Euclidean distance between state vectors, and are scale parameters represented as state vector and geographic distance, respectively;

[0176] Assumptions: =10, =100, then the dynamic weight between node 1 and node 2 is:

[0177] ;

[0178] Weighted degree calculation: Assuming that at time point t, node 1 is connected to node 2 and node 3, and the dynamic weights between them are 0.8 and 0.6 respectively, then the weighted degree of node 1 is: =0.8+0.6=1.4;

[0179] Calculation of dynamic spatiotemporal weighted clustering coefficient: Assuming that at time point t, node 1 is connected to node 2 and node 3, node 2 is connected to node 3, and the dynamic weights between them are 0.8, 0.6, and 0.7, respectively. The geographical distances between node 1 and node 2 and node 3 are 10 km and 15 km, respectively. The dynamic spatiotemporal weighted clustering coefficient of node 1 is: .

[0180] Calculation of predicted clustering coefficient: Assuming 𝛼=0.7, and the predicted dynamic spatiotemporal weighted clustering coefficient of node 1 at time point t+1 is 0.3, then the final dynamic spatiotemporal weighted clustering coefficient of node 1 at time point t is:

[0181] ;

[0182] Through the above examples, we can see how the dynamic spatiotemporal weighted clustering coefficient is calculated, and how to combine the current clustering coefficient and the predicted clustering coefficient to obtain the final dynamic spatiotemporal weighted clustering coefficient. These calculation results help to more accurately evaluate the dynamic characteristics of nodes in the network and provide a basis for risk assessment and network optimization.

[0183] Optionally, in the embodiment of the present application, in step 105, "determining key nodes of the risk factor network according to structural characteristics of the risk factor network" may include the following process:

[0184] As a possible solution, the diffusion degree of risk factors can be simulated according to the structural characteristics of the risk factor network; the benefit of each node in the risk factor network can be calculated by presetting the benefit matrix; the comprehensive influence of each node can be calculated according to the diffusion degree of the risk factors and the benefit of each node; according to the ranking of the comprehensive influence of each node, the top K% nodes can be identified as key nodes;

[0185] Specifically, simulating the diffusion of risk factors according to the structural characteristics of the risk factor network includes:

[0186] By formula: , simulating the diffusion degree of risk factors; among them, It is expressed as the rate of change of risk exposure of node i over time; It is expressed as the exposure degree of node i to the risk factor; It is represented as the rate at which node i naturally reduces its risk exposure; Expressed as the risk transmission rate; It is represented as the weighted sum of the impact of node i’s neighbor node j on its risk exposure, where: Represented as the set of neighbor nodes of node i; Represented as the dynamic weight between node i and node j at time point t; It is represented as the risk exposure degree of neighbor node j at time point t;

[0187] The evaluating the benefit of each node in the risk factor network includes:

[0188] Through the profit matrix:

[0189] , calculate the benefit of each node in the risk factor network; where, Expressed as the revenue Ri of node i; It represents the benefit obtained by node i when node i chooses to cooperate and all its neighboring nodes also choose to cooperate; b represents the benefit obtained by node i when node i chooses to cooperate and at least one of its neighboring nodes chooses to betray, and b benefit is lower than a benefit; c represents the benefit obtained by node i when node i chooses to betray and all its neighboring nodes choose to cooperate, and c benefit is higher than b benefit and lower than a benefit; d represents the benefit obtained by node i when node i chooses to betray and at least one of its neighboring nodes chooses to betray, and d benefit is the lowest;

[0190] The comprehensive influence of each node is calculated according to the diffusion degree of the risk factor and the income of each node, including:

[0191] By formula: , calculate the comprehensive influence of each node; where, It is expressed as the comprehensive influence of node i, It is expressed as a weight factor between 0 and 1, which is used to balance the rate of change of the risk exposure of node i over time and the influence of the income Ri of each node on the comprehensive influence; It is expressed as the absolute value integral of the state change rate of node i in the time interval [0, T], It is expressed as the benefit of node i in the cooperative game, which is determined according to the behavior of node i and its neighboring nodes;

[0192] The step of identifying the top K% of nodes as key nodes according to the ranking of the comprehensive influence of each node includes:

[0193] By formula: , identify the first K% of nodes as key nodes, where Represented as a set of key nodes, Represented as node i, is represented as the set of all nodes in the risk factor network; It is represented as the comprehensive influence of node i; Represented as a preset threshold, used to filter key nodes that meet the conditions.

[0194] In this step, the risk factor diffusion is simulated: through the differential equation Simulate the diffusion process of risk factors in the network. This takes into account the state changes of nodes, natural decay rates, and influences from neighbors.

[0195] Payoff matrix calculation: The payoff matrix of the cooperative game is used to calculate the payoff of the node, which reflects the payoff of the node and its neighbors under cooperation and defection behaviors.

[0196] Comprehensive influence calculation: Combine the degree of spread of risk factors and node benefits to calculate the comprehensive influence of each node. This takes into account the drastic degree of node status change and the benefits in social interaction.

[0197] Key node identification: Based on the ranking of the nodes’ comprehensive influence, the top K% of nodes are identified as key nodes. This helps identify the nodes in the network that have the greatest impact on risk propagation.

[0198] In this embodiment of the application, assume that we are evaluating a risk factor network for tower installation in a communication network. The network contains 5 nodes, representing different risk factors, such as weather conditions, geological stability, construction quality, etc. We will analyze the data collected by the IoT devices:

[0199] Risk factor diffusion simulation:

[0200] State change rate: It represents the rate of change of the risk exposure of node i over time.

[0201] Risk exposure: Represents the risk exposure degree of node i at time t.

[0202] Natural decay rate: 𝛽 represents the rate at which node i naturally reduces its risk exposure.

[0203] Risk transmission rate: 𝛾 represents the risk transmission rate.

[0204] Weighted sum of neighbor influence: represents the weighted sum of the impact of node i’s neighbor node j on its risk exposure, where is the set of neighbor nodes of node i.

[0205] Assume that 𝛽=0.1, 𝛾=0.2, the risk exposure of node 1 at time point t is 0.5, the risk exposure of its neighbor nodes 2 and 3 are 0.3 and 0.4 respectively, and the dynamic weights between them are 0.8 and 0.6 respectively. Then the rate of change of the risk exposure of node 1 over time is:

[0206] ;

[0207] Profit matrix calculation:

[0208] Profit Matrix: The profit matrix Ri is determined based on the behavior of the node and its neighbors.

[0209] Assume that if node i chooses to cooperate and all its neighbors also choose to cooperate, the payoff a = 10. If node i chooses to cooperate but at least one neighbor chooses to defect, the payoff b = 5. If node i chooses to defect and all its neighbors choose to cooperate, the payoff c = 7. If node i chooses to defect and at least one neighbor also chooses to defect, the payoff d = 2.

[0210] Based on the above assumptions, when node 1 chooses to cooperate, node 2 chooses to cooperate, and node 3 chooses to betray, the benefit of node 1 is b=5.

[0211] Comprehensive influence calculation:

[0212] Comprehensive influence: through formula , calculate the comprehensive influence of each node.

[0213] Weight factor: 𝛼=0.6, used to balance the impact of state change and benefits.

[0214] State change points: Represents the absolute value integral of the state change rate of node i in the time interval [0,T].

[0215] Node payoff: Ri is determined based on the behavior of the node and its neighbors.

[0216] Assuming that in the time interval [0,10], the absolute value integral of the state change rate of node 1 is 2.5, then the comprehensive influence of node 1 is: ;

[0217] Key node identification:

[0218] Key node set: through the formula Identify the top K% nodes as key nodes.

[0219] The preset threshold value: θ=3 is used to filter out key nodes that meet the conditions.

[0220] Assuming we wish to identify the top 20% of nodes as key nodes, and the combined influence of all nodes in the network is 3.5, 2.0, 4.0, 1.5, and 3.0 respectively, then the top 20% of nodes will be nodes with a combined influence greater than or equal to 3.0, i.e., nodes 1 and 3.

[0221] Through the above examples, we can see how to identify key nodes by simulating the degree of diffusion of risk factors, calculating the benefits of nodes, and calculating the comprehensive influence of nodes. These calculation results help to more accurately assess which nodes in the network have the greatest impact on risk propagation, providing a basis for risk assessment and network optimization.

[0222] As another possible implementation scheme, the degree centrality, betweenness centrality, closeness centrality and eigenvector centrality of each node may be calculated according to the structural characteristics of the risk factor network; based on the degree centrality, betweenness centrality, closeness centrality and eigenvector centrality of each node, the comprehensive centrality score of each node may be calculated; if the comprehensive centrality score of the node meets a preset threshold, the node is determined to be a key node;

[0223] Among them, the degree centrality refers to the number of edges directly connected to the node; the betweenness centrality refers to the frequency of the node appearing in all shortest paths in the risk factor network; the closeness centrality refers to the inverse of the average distance of the node to all other nodes in the risk factor network; and the eigenvector centrality refers to the importance of the nodes adjacent to the node.

[0224] Specifically, the degree centrality, betweenness centrality, closeness centrality and eigenvector centrality of each of the nodes are calculated, including:

[0225] By formula: , calculate the degree centrality of each of the nodes, where, Represented as the node The degree of It is expressed as the total number of nodes in the risk factor network;

[0226] By formula: , calculate the betweenness centrality of each of the nodes, where, Represented as a node To Node The number of shortest paths, Represented as nodes in all shortest paths of the risk factor network The frequency of occurrence, is represented as the set of all nodes in the risk factor network;

[0227] By formula: , calculate the closeness centrality of each of the nodes, where, is represented by the total number of nodes in the risk factor network, Represented as a node To Node The shortest path length;

[0228] By formula: , calculate the eigenvector centrality of each of the nodes, where, Represented as a node The set of all adjacent nodes of is the main eigenvalue, Represented as a node The eigenvector centrality of the node Belongs to the node The adjacent nodes.

[0229] Specifically, based on the degree centrality, betweenness centrality, closeness centrality, and eigenvector centrality of each node, a comprehensive centrality score of each node is calculated, including:

[0230] By formula: , calculate the comprehensive centrality score of each of the nodes;

[0231] in, Represented as a node The comprehensive centrality score of Respectively expressed as The weight of Respectively expressed as The average value of Respectively expressed as The standard deviation of .

[0232] In this step, degree centrality measures the number of edges directly connected to node v in its network, expressed as the degree of node v divided by the total number of nodes in the network minus 1. Degree centrality reflects the connectivity of the node. The larger the degree, the more important the role of the node in information dissemination.

[0233] Betweenness centrality: measures the frequency of node v appearing in all shortest paths, expressed as the number of shortest paths passing through node v divided by the total number of shortest paths. Nodes with high betweenness centrality play a bridging role in information or resource transfer.

[0234] Closeness centrality: measures the inverse of the average distance from node v to all other nodes in the network. Nodes with high closeness centrality mean that they can interact or exchange information with other nodes more quickly.

[0235] Eigenvector centrality: measures the importance of node v to its neighboring nodes, usually calculated based on the principal eigenvector of the adjacency matrix of the network. Eigenvector centrality reflects the status and influence of a node in a complex network.

[0236] Comprehensive centrality score: Combine degree centrality, betweenness centrality, closeness centrality, and eigenvector centrality by weighted average to get the comprehensive centrality score of node v. The higher the score, the more critical the node.

[0237] In the embodiment of the present application, it is assumed that we have the following risk factor network, which includes 5 nodes, representing soil stability, wind speed, rainfall, tower design, and construction quality, respectively. The network structure is as follows:

[0238] Node A (soil stability) connects B (wind speed) and C (rainfall);

[0239] Node B (wind speed) connects A (soil stability), C (rainfall), and D (tower design);

[0240] Node C (rainfall) connects A (soil stability), B (wind speed), and E (construction quality);

[0241] Node D (tower design) connects B (wind speed) and E (construction quality);

[0242] Node E (construction quality) connects C (rainfall) and D (tower design);

[0243] We calculate the centrality index for each node:

[0244] Calculate degree centrality: Node A has a degree of 2, so ; Similarly, the degree centrality of other nodes can be calculated;

[0245] Calculate betweenness centrality: Assume that there are two shortest paths from node A to node E, one of which passes through node B and the other does not pass through node B, then the betweenness centrality of node B is ; Similarly, the betweenness centrality of other nodes can be calculated;

[0246] Calculate closeness centrality: Assume that the average shortest path length from node A to all other nodes in the network is 2, then ; Similarly, the proximity centrality of other nodes can be calculated;

[0247] Calculate the eigenvector centrality: Solve the main eigenvector based on the adjacency matrix. This application assumes that the eigenvector centrality of node A is 0.8; the eigenvector centrality of other nodes can be calculated similarly;

[0248] Calculate the comprehensive centrality score: Assume that all the centrals are =0.2, =0.3, , =0.2, and the mean and standard deviation of each centrality are , , =2, , , , , .

[0249] Then the comprehensive centrality score of node A is: ; The comprehensive centrality scores of other nodes can be calculated.

[0250] Through the above steps, we can identify the key nodes in the risk factor network. For example, if the preset threshold is 0.5, the nodes with a comprehensive centrality score greater than or equal to 0.5 will be identified as key nodes. This helps us focus on the factors that play a core role in risk communication and management, and provides a basis for formulating effective risk control strategies.

[0251] It should be noted that the process of solving the main eigenvector based on the adjacency matrix to obtain the eigenvector centrality involves linear algebra and eigenvalue decomposition. The eigenvector centrality is calculated by finding the eigenvector corresponding to the maximum eigenvalue of the network adjacency matrix, and the elements of the eigenvector correspond to the centrality scores of each node in the network. Since this part of the content belongs to the existing technology, this application will not be expanded in detail.

[0252] Figure 2 A schematic diagram of the structure of a tower installation risk assessment system based on a communication network provided in an embodiment of the present application, such as Figure 2 As shown, the system includes:

[0253] The collection module 21 is used to collect the original data of the tower installation site through the Internet of Things device;

[0254] The identification module 22 is used to extract multiple potential risk factors that affect the installation of the iron tower from the original data, and use the multiple potential risk factors as nodes of the risk factor network; use the factor relationships between the multiple potential risk factors as edges of the risk factor network to construct the risk factor network; calculate the degree distribution, clustering coefficient and shortest path length of the risk factor network to obtain the structural characteristics of the risk factor network; determine the key nodes of the risk factor network according to the structural characteristics of the risk factor network, and use the key nodes as target risk factors;

[0255] A construction module 23 is used to construct a risk prediction model based on the target risk factors;

[0256] The evaluation module 24 is used to collect real-time data through the Internet of Things device and input it into the risk prediction model, so as to perform risk evaluation based on the real-time data through the risk prediction model and output the risk evaluation result.

[0257] Figure 2 The tower installation risk assessment system based on communication network can be executed Figure 1 The implementation principle and technical effect of the tower installation risk assessment method based on a communication network described in the embodiment shown are not repeated here. The specific way in which each module and unit performs operations in the tower installation risk assessment system based on a communication network in the above embodiment has been described in detail in the embodiment of the method, and will not be elaborated here.

[0258] In one possible design, Figure 2 The tower installation risk assessment system based on a communication network of the embodiment shown can be implemented as a computing device, such as Figure 3 As shown, the computing device may include a storage component 31 and a processing component 32;

[0259] The storage component 31 stores one or more computer instructions, wherein the one or more computer instructions are called and executed by the processing component 32 .

[0260] The processing component is used to: collect raw data of the tower installation site through the Internet of Things device; extract multiple potential risk factors affecting the tower installation from the raw data, and use the multiple potential risk factors as nodes of the risk factor network; use the factor relationships between the multiple potential risk factors as the edges of the risk factor network to construct a risk factor network; calculate the degree distribution, clustering coefficient and shortest path length of the risk factor network to obtain the structural characteristics of the risk factor network; determine the key nodes of the risk factor network according to the structural characteristics of the risk factor network, and use the key nodes as target risk factors; construct a risk prediction model based on the target risk factors; collect real-time data through the Internet of Things device and input it into the risk prediction model, so as to perform risk assessment based on the real-time data through the risk prediction model, and output the risk assessment result.

[0261] The processing component 32 may include one or more processors to execute computer instructions to complete all or part of the steps in the above method. Of course, the processing component may also be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors or other electronic components to perform the above method.

[0262] The storage component 31 is configured to store various types of data to support operations at the terminal. The storage component can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0263] Of course, a computing device may also include other components, such as input / output interfaces, communication components, etc.

[0264] The input / output interface provides an interface between the processing component and the peripheral interface module, which may be an output device, an input device, etc.

[0265] The communication component is configured to facilitate, among other things, wired or wireless communications between the computing device and other devices.

[0266] Among them, the computing device can be a physical device or an elastic computing host provided by a cloud computing platform, etc. In this case, the computing device can refer to a cloud server, and the above-mentioned processing components, storage components, etc. can be basic server resources rented or purchased from the cloud computing platform.

[0267] The present application also provides a computer storage medium storing a computer program, wherein the computer program can achieve the above-mentioned Figure 1 The illustrated embodiment is a tower installation risk assessment method based on a communication network.

[0268] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0269] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0270] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution 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 ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0271] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A tower installation risk assessment method based on a communication network, characterized in that: include: Collect raw data from tower installation sites through IoT devices; Extracting multiple potential risk factors affecting the installation of the iron tower from the original data, and using the multiple potential risk factors as nodes of a risk factor network; Taking the factor relationships among the multiple potential risk factors as edges of the risk factor network to construct a risk factor network; By calculating the degree distribution, clustering coefficient and shortest path length of the risk factor network, the structural characteristics of the risk factor network are obtained; Determine key nodes of the risk factor network according to structural characteristics of the risk factor network, and use the key nodes as target risk factors; Based on the target risk factors, construct a risk prediction model; Collecting real-time data through the IoT device and inputting it into the risk prediction model, so as to perform risk assessment based on the real-time data through the risk prediction model, and outputting the risk assessment result; Before obtaining the structural characteristics of the risk factor network by calculating the degree distribution, clustering coefficient and shortest path length of the risk factor network, the method further includes: By formula group: ; ; ; , calculating the clustering coefficient of each of the nodes in the risk factor network; in, Represented as the dynamic weight between node i and node j at time point t; Represented as the state vector of node i at time point t, Represented as the state vector of node j at time point t; It is represented as the state change vector of node i, the difference between the state at time point t and the previous time point; It is represented as the state change vector of node j, the difference between the state at time point t and the previous time point; It is represented as the distance between node i and node j; function 𝑓 is represented as a nonlinear function, which is used to calculate the dynamic weight; in, It is expressed as the dynamic spatiotemporal weighted clustering coefficient of node i at time t; It is represented as the weighted degree of node i at time point t; It is represented as the dynamic weight between node j and node l at time point t, It is represented as the dynamic weight between node l and node i at time point t; σ is represented as the scale parameter that controls the influence of distance on weight; in, It is expressed as the final dynamic spatiotemporal weighted clustering coefficient of node i at time t; It is expressed as a weight factor, which determines the relative importance of the current clustering coefficient and the predicted clustering coefficient; Represented as node i at time point The predicted dynamic spatiotemporal weighted clustering coefficient; Expressed as the gap between the predicted future time point and the current time point; Determining the key nodes of the risk factor network according to the structural characteristics of the risk factor network includes: Simulating the diffusion degree of risk factors according to the structural characteristics of the risk factor network; Calculate the benefit of each node in the risk factor network by presetting the benefit matrix; Calculate the comprehensive influence of each node according to the diffusion degree of the risk factor and the income of each node; According to the ranking of the comprehensive influence of each node, the top K% nodes are identified as key nodes; The step of simulating the diffusion degree of risk factors according to the structural characteristics of the risk factor network includes: By formula: , simulating the diffusion degree of risk factors; among them, It is expressed as the rate of change of risk exposure of node i over time; It is expressed as the exposure degree of node i to the risk factor; It is represented as the rate at which node i naturally reduces its risk exposure; Expressed as the risk transmission rate; It is represented as the weighted sum of the impact of node i’s neighbor node j on its risk exposure, where: Represented as the set of neighbor nodes of node i; Represented as the dynamic weight between node i and node j at time point t; It is represented as the risk exposure degree of neighbor node j at time point t; The evaluating the benefit of each node in the risk factor network includes: Through the profit matrix: , calculate the benefit of each node in the risk factor network; where, Expressed as the revenue Ri of node i; It represents the benefit obtained by node i when node i chooses to cooperate and all its neighboring nodes also choose to cooperate; b represents the benefit obtained by node i when node i chooses to cooperate and at least one of its neighboring nodes chooses to betray, and b benefit is lower than a benefit; c represents the benefit obtained by node i when node i chooses to betray and all its neighboring nodes choose to cooperate, and c benefit is higher than b benefit and lower than a benefit; d represents the benefit obtained by node i when node i chooses to betray and at least one of its neighboring nodes chooses to betray, and d benefit is the lowest; The comprehensive influence of each node is calculated according to the diffusion degree of the risk factor and the income of each node, including: By formula: , calculate the comprehensive influence of each node; where, It is expressed as the comprehensive influence of node i, It is expressed as a weight factor between 0 and 1, which is used to balance the rate of change of the risk exposure of node i over time and the influence of the income Ri of each node on the comprehensive influence; It is expressed as the absolute value integral of the state change rate of node i in the time interval [0, T], It is expressed as the benefit of node i in the cooperative game, which is determined according to the behavior of node i and its neighboring nodes; The step of identifying the top K% of nodes as key nodes according to the ranking of the comprehensive influence of each node includes: By formula: , identify the first K% of nodes as key nodes, where Represented as a set of key nodes, Represented as node i, is represented as the set of all nodes in the risk factor network; It is represented as the comprehensive influence of node i; Represented as a preset threshold, used to filter key nodes that meet the conditions.

2. The method according to claim 1, characterized in that: Before obtaining the structural characteristics of the risk factor network by calculating the degree distribution, clustering coefficient and shortest path length of the risk factor network, the method further includes: Extracting node features from the risk factor network, and training a degree distribution prediction model based on the node features, so as to predict the degree of each node in the risk factor network through the trained degree distribution prediction model to obtain a degree distribution; The degree refers to the number of edges that directly connect the node to other nodes, and the degree distribution refers to the distribution of the degrees of all nodes in the risk factor network.

3. The method according to claim 1, characterized in that Before obtaining the structural characteristics of the risk factor network by calculating the degree distribution, clustering coefficient and shortest path length of the risk factor network, the method further includes: Determine a heuristic function, wherein the heuristic function is used to calculate the expected cost from the current node to the target node; Setting basic parameters of the ACO algorithm, wherein the basic parameters include at least: the number of ants, the evaporation rate of pheromones, and the importance weight of pheromones; In each iteration cycle, according to the pheromone concentration of the current node and the heuristic function, the ant selects the next node to construct a path, wherein the pheromone concentration is determined according to the evaporation rate of the pheromone and the importance weight of the pheromone; At the end of each iteration, the pheromone concentration of each node is updated according to the path quality of the path until the maximum number of iterations is reached or the shortest path is found; The shortest path length is extracted from the shortest path.

4. The method according to claim 1, characterized in that: Determining the key nodes of the risk factor network according to the structural characteristics of the risk factor network includes: According to the structural characteristics of the risk factor network, calculating the degree centrality, betweenness centrality, closeness centrality and eigenvector centrality of each of the nodes; Calculate the comprehensive centrality score of each node based on the degree centrality, betweenness centrality, closeness centrality and eigenvector centrality of each node; If the comprehensive centrality score of the node meets the preset threshold, the node is determined to be a key node; Among them, the degree centrality refers to the number of edges directly connected to the node; the betweenness centrality refers to the frequency of the node appearing in all shortest paths in the risk factor network; the closeness centrality refers to the inverse of the average distance of the node to all other nodes in the risk factor network; and the eigenvector centrality refers to the importance of the nodes adjacent to the node.

5. The method according to claim 4, characterized in that The calculating of the degree centrality, betweenness centrality, closeness centrality and eigenvector centrality of each node includes: By formula: , calculate the degree centrality of each of the nodes, where, Represented as the node The degree of It is expressed as the total number of nodes in the risk factor network; By formula: , calculate the betweenness centrality of each of the nodes, where, Represented as a node To Node The number of shortest paths, Represented as nodes in all shortest paths of the risk factor network The frequency of occurrence, is represented as the set of all nodes in the risk factor network; By formula: , calculate the closeness centrality of each of the nodes, where, is represented by the total number of nodes in the risk factor network, Represented as a node To Node The shortest path length; By formula: , calculate the eigenvector centrality of each of the nodes, where, Represented as a node The set of all adjacent nodes of is the main eigenvalue, Represented as a node The eigenvector centrality of the node Belongs to the node The adjacent nodes of The step of calculating the comprehensive centrality score of each node based on the degree centrality, betweenness centrality, closeness centrality and eigenvector centrality of each node comprises: By formula: , calculate the comprehensive centrality score of each of the nodes; in, Represented as a node The comprehensive centrality score of Respectively expressed as The weight of Respectively expressed as The average value of Respectively expressed as The standard deviation of .

6. A tower installation risk assessment system based on a communication network, characterized in that: include: A collection module is used to collect raw data of the tower installation site through IoT devices; An identification module is used to extract multiple potential risk factors affecting the installation of the iron tower from the original data, and use the multiple potential risk factors as nodes of a risk factor network; use the factor relationships between the multiple potential risk factors as edges of the risk factor network to construct a risk factor network; and calculate the degree distribution, clustering coefficient and shortest path length of the risk factor network to obtain the structural characteristics of the risk factor network; Determine key nodes of the risk factor network according to structural characteristics of the risk factor network, and use the key nodes as target risk factors; A construction module, used to construct a risk prediction model based on the target risk factors; An assessment module, for collecting real-time data through the IoT device and inputting it into the risk prediction model, so as to perform risk assessment based on the real-time data through the risk prediction model, and output a risk assessment result; Before obtaining the structural characteristics of the risk factor network by calculating the degree distribution, clustering coefficient and shortest path length of the risk factor network, the method further includes: By formula group: ; ; ; , calculating the clustering coefficient of each of the nodes in the risk factor network; in, Represented as the dynamic weight between node i and node j at time point t; Represented as the state vector of node i at time point t, Represented as the state vector of node j at time point t; It is represented as the state change vector of node i, the difference between the state at time point t and the previous time point; It is represented as the state change vector of node j, the difference between the state at time point t and the previous time point; It is represented as the distance between node i and node j; function 𝑓 is represented as a nonlinear function, which is used to calculate the dynamic weight; in, It is expressed as the dynamic spatiotemporal weighted clustering coefficient of node i at time t; It is represented as the weighted degree of node i at time point t; It is represented as the dynamic weight between node j and node l at time point t, It is represented as the dynamic weight between node l and node i at time point t; σ is represented as the scale parameter that controls the influence of distance on weight; in, It is expressed as the final dynamic spatiotemporal weighted clustering coefficient of node i at time t; It is expressed as a weight factor, which determines the relative importance of the current clustering coefficient and the predicted clustering coefficient; Represented as node i at time point The predicted dynamic spatiotemporal weighted clustering coefficient; Expressed as the gap between the predicted future time point and the current time point; Determining the key nodes of the risk factor network according to the structural characteristics of the risk factor network includes: Simulating the diffusion degree of risk factors according to the structural characteristics of the risk factor network; Calculate the benefit of each node in the risk factor network by presetting the benefit matrix; Calculate the comprehensive influence of each node according to the diffusion degree of the risk factor and the income of each node; According to the ranking of the comprehensive influence of each node, the top K% nodes are identified as key nodes; The step of simulating the diffusion degree of risk factors according to the structural characteristics of the risk factor network includes: By formula: , simulating the diffusion degree of risk factors; among them, It is expressed as the rate of change of risk exposure of node i over time; It is expressed as the exposure degree of node i to the risk factor; It is represented as the rate at which node i naturally reduces its risk exposure; Expressed as the risk transmission rate; It is represented as the weighted sum of the impact of node i’s neighbor node j on its risk exposure, where: Represented as the set of neighbor nodes of node i; Represented as the dynamic weight between node i and node j at time point t; It is represented as the risk exposure degree of neighbor node j at time point t; The evaluating the benefit of each node in the risk factor network includes: Through the profit matrix: , calculate the benefit of each node in the risk factor network; where, Expressed as the revenue Ri of node i; It represents the benefit obtained by node i when node i chooses to cooperate and all its neighboring nodes also choose to cooperate; b represents the benefit obtained by node i when node i chooses to cooperate and at least one of its neighboring nodes chooses to betray, and b benefit is lower than a benefit; c represents the benefit obtained by node i when node i chooses to betray and all its neighboring nodes choose to cooperate, and c benefit is higher than b benefit and lower than a benefit; d represents the benefit obtained by node i when node i chooses to betray and at least one of its neighboring nodes chooses to betray, and d benefit is the lowest; The comprehensive influence of each node is calculated according to the diffusion degree of the risk factor and the income of each node, including: By formula: , calculate the comprehensive influence of each node; where, It is expressed as the comprehensive influence of node i, It is expressed as a weight factor between 0 and 1, which is used to balance the rate of change of the risk exposure of node i over time and the influence of the income Ri of each node on the comprehensive influence; It is expressed as the absolute value integral of the state change rate of node i in the time interval [0, T], It is expressed as the benefit of node i in the cooperative game, which is determined according to the behavior of node i and its neighboring nodes; The step of identifying the top K% of nodes as key nodes according to the ranking of the comprehensive influence of each node includes: By formula: , identify the first K% of nodes as key nodes, where Represented as a set of key nodes, Represented as node i, is represented as the set of all nodes in the risk factor network; It is represented as the comprehensive influence of node i; Represented as a preset threshold, used to filter key nodes that meet the conditions.

7. A computing device, characterized in that It comprises a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement the tower installation risk assessment method based on the communication network as described in any one of claims 1 to 5.

8. A computer storage medium, characterized in that: A computer program is stored, and when the computer executes the computer program, the tower installation risk assessment method based on the communication network as described in any one of claims 1 to 5 is implemented.