A method, device, terminal equipment and computer-readable storage medium for evaluating wind resistance of distribution network tower structures

Abstract

The present invention discloses a method, apparatus, terminal device, and computer-readable storage medium for evaluating the wind resistance of distribution network tower structures. The method comprises: obtaining meteorological and topographical data of the distribution network tower to be evaluated; the meteorological data includes temperature, humidity, wind speed, and rainfall intensity; the topographical data includes land type, slope, windward or leeward slope, and the angle between the prevailing wind direction and the line direction in winter; inputting the meteorological and topographical data into a neural network model, so that the model generates soil shear strength based on the meteorological and topographical data; determining the ultimate overturning moment and the corresponding critical wind speed based on the soil shear strength; and determining the wind resistance of the distribution network tower to be evaluated based on the critical wind speed. By implementing the present invention, the impact of rainfall on soil can be taken into account, improving the accuracy and reliability of the evaluation.

Description

Technical Field

[0001] The present invention relates to the field of power engineering performance evaluation, and in particular to a method, device, terminal equipment and computer-readable storage medium for evaluating the wind resistance of a distribution network tower structure. Background Art

[0002] Distribution network towers are a crucial component of transmission lines, supporting them and bearing the weight of cables. In severe weather conditions, wind is a major external load. Inadequate wind resistance can lead to tower collapse, line breakage, or cable damage, resulting in power outages and serious impacts on users and grid operations. Therefore, evaluating and improving the wind resistance of distribution network towers is crucial to ensuring power supply reliability.

[0003] At present, there are the following methods for evaluating the wind resistance of distribution network tower structures: (1) using a probabilistic framework to calculate the vulnerability of distribution towers under wind loads, obtain the probability of different damage levels of the tower structure, and evaluate the wind resistance of the tower structure; (2) using the JC method to calculate the component reliability index of the tower structure under different wind load design conditions, and use the reliability of the structure under different wind loads to evaluate the wind resistance of the tower structure; (3) evaluating the wind vibration response of the tower structure based on the collapse ratio and wind vibration coefficient of the risk point, thereby quantifying the wind resistance of the tower structure. These methods are based on the structural characteristics of the distribution network tower structure itself and the influence of wind loads to evaluate its wind resistance, while ignoring the influence of other external environmental factors. However, soil characteristics and conditions are crucial to the stability and bearing capacity of the tower foundation, which in turn affects the wind resistance of the entire tower structure. Rainfall will deteriorate the soil at the distribution tower foundation and reduce its strength. The existing evaluation methods ignore the effects of rainwater infiltration and soil erosion. Therefore, it is urgent to further improve the existing evaluation methods and consider the impact of rainfall on soil. Summary of the Invention

[0004] The embodiments of the present invention provide a method, apparatus, terminal device, and computer-readable storage medium for evaluating the wind resistance of a distribution network tower structure, which can take into account the impact of rainfall on soil and improve the accuracy and reliability of the evaluation.

[0005] An embodiment of the present invention provides a method for evaluating the wind resistance of a distribution network tower structure, comprising:

[0006] Obtain meteorological and topographic data for the distribution network towers to be assessed; meteorological data includes temperature, humidity, wind speed, and rainfall intensity; topographic data includes land type, slope, windward or leeward slope, and the angle between the prevailing wind direction and the line direction in winter;

[0007] inputting meteorological data and topographic data into the neural network model so that the neural network model generates soil shear strength according to the meteorological data and topographic data;

[0008] Determine the ultimate overturning moment and the corresponding critical wind speed based on the soil shear strength;

[0009] According to the critical wind speed, the wind resistance capacity of the distribution network tower to be evaluated is determined.

[0010] Furthermore, before inputting the meteorological data and terrain data into the neural network model, the following steps are also included:

[0011] The meteorological data and topographic data are standardized and missing values ​​are supplemented to obtain preprocessed meteorological data and topographic data.

[0012] Furthermore, the meteorological data and terrain data are standardized and missing values ​​are supplemented to obtain preprocessed meteorological data and terrain data, including:

[0013] Z-Score standardization was used to transform meteorological and topographic data into a normal distribution with a mean of 0 and a standard deviation of 1;

[0014] Linear interpolation is used to supplement the missing values ​​of the normal distribution to obtain the preprocessed meteorological and topographic data.

[0015] Furthermore, the neural network model is obtained in the following way:

[0016] Obtain historical meteorological data and historical terrain data of distribution network towers at several historical moments and the soil shear strength at the corresponding historical moments;

[0017] For the historical meteorological data and historical topographic data at each historical moment, standardization processing and missing value supplementation are performed on the historical meteorological data and historical topographic data to obtain pre-processed historical meteorological data and historical topographic data;

[0018] The preprocessed historical meteorological data and historical terrain data at several historical moments and their corresponding soil shear strength at historical moments are respectively input into the initial physical information neural network model for iterative training until the loss function converges, thereby obtaining a trained neural network model.

[0019] Furthermore, the loss function includes: data error term and physical information error term;

[0020] Data error terms, specifically: mean square error;

[0021] The physical information error term is determined as follows:

[0022] Based on the pre-processed historical meteorological data and historical topographic data at the current historical moment input into the physical information neural network model, the mathematical shear strength of the soil is calculated using the Moore-Coulomb criterion;

[0023] Calculate the residual of soil strength based on the predicted shear strength generated during the current iterative training of the physical information neural network model and the corresponding mathematical shear strength;

[0024] The residual of soil strength is used as the physical information error term in the current iteration process of the physical information neural network model.

[0025] Furthermore, the ultimate overturning moment and the corresponding critical wind speed are determined based on the soil shear strength, including:

[0026] Calculate the soil erosion thickness based on the universal soil loss equation, where the soil erodibility factor is determined based on the soil shear strength.

[0027] According to the structure of distribution network tower, soil shear strength and soil scour thickness, a numerical model of distribution network tower is constructed based on artificial boundary method.

[0028] The numerical model of the distribution network tower is solved by the equivalent strain wedge method, and the ultimate overturning moment of the distribution network tower and the corresponding critical wind speed are obtained.

[0029] Based on the above method embodiment, the present invention provides a corresponding device embodiment, including: an environmental data acquisition module, a soil shear strength determination module, a critical wind speed determination module, and a wind resistance assessment module;

[0030] The environmental data acquisition module is used to obtain meteorological and topographic data of the distribution network towers to be evaluated. Meteorological data includes temperature, humidity, wind speed, and rainfall intensity. Topographic data includes land type, slope, windward or leeward slope, and the angle between the prevailing wind direction and the line direction in winter.

[0031] a soil shear strength determination module, for inputting meteorological data and topographic data into a neural network model so that the neural network model generates soil shear strength based on the meteorological data and topographic data;

[0032] Critical wind speed determination module, used to determine the ultimate overturning moment and the corresponding critical wind speed based on the soil shear strength;

[0033] The wind resistance evaluation module is used to determine the wind resistance of the distribution network tower to be evaluated based on the critical wind speed.

[0034] Furthermore, the device for evaluating the wind resistance of a distribution network tower structure further includes: a pre-processing module;

[0035] The preprocessing module includes: performing standardization processing on meteorological data and terrain data and supplementing missing values ​​to obtain preprocessed meteorological data and terrain data.

[0036] Based on the above-mentioned method embodiment, the present invention provides a corresponding terminal device embodiment, including: a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, the steps of the distribution network tower structure wind resistance evaluation method as described in the present invention are implemented.

[0037] Based on the above-mentioned method embodiment, the present invention provides a corresponding computer-readable storage medium embodiment, including: a stored computer program, which, when the computer program is running, controls the device where the computer-readable storage medium is located to execute the steps of the distribution network tower structure wind resistance assessment method as described in the present invention.

[0038] Compared with the prior art, the beneficial effects of the embodiment of this solution are:

[0039] The present invention obtains meteorological data and topographic data of a distribution network tower to be evaluated, wherein the meteorological data includes temperature, humidity, wind speed and rainfall intensity, and the topographic data includes land type, slope, windward slope or leeward slope, and the angle between the main wind direction and the line direction in winter; then, the meteorological data and topographic data are standardized and missing values ​​are supplemented to obtain pre-processed meteorological data and topographic data; the pre-processed meteorological data and topographic data are input into a neural network model so that the neural network generates soil moisture content and shear strength based on the pre-processed meteorological data and topographic data, and the soil strength after rainfall erosion and infiltration is predicted by the neural network model, thereby taking into account the influence of rainwater infiltrating the soil and soil being eroded by rainwater; then, the ultimate overturning moment and the corresponding critical wind speed are determined based on the soil moisture content and shear strength; and the wind resistance of the distribution network tower to be evaluated is determined based on the critical wind speed, thereby achieving the evaluation of the wind resistance of the distribution network tower based on the soil conditions.

[0040] In summary, the present invention considers the impact of rainfall on the soil environment, utilizes meteorological and topographic data of distribution network towers, calculates the moisture content and shear strength of the soil, and determines the wind resistance of the distribution network towers accordingly, thereby improving the accuracy and reliability of the assessment. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 1 is a flow chart of a method for evaluating the wind resistance of a distribution network tower structure provided by one embodiment of the present invention;

[0042] Figure 2 1 is a schematic diagram of linear interpolation calculation provided by an embodiment of the present invention;

[0043] Figure 3 1 is a flow chart of a neural network model training process provided by one embodiment of the present invention;

[0044] Figure 4 is a schematic structural diagram of a strain wedge provided by an embodiment of the present invention;

[0045] Figure 5 This is a cross-sectional diagram of a distribution network tower flushing pit provided by an embodiment of the present invention.

[0046] Figure 6 It is a structural schematic diagram of a device for evaluating the wind resistance of a distribution network tower structure provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0047] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0048] like Figure 1 As shown, an embodiment of the present invention provides a method for evaluating the wind resistance of a distribution network tower structure, the method comprising at least the following steps:

[0049] Step S1: Acquire meteorological and topographic data of the distribution network tower to be evaluated; meteorological data includes temperature, humidity, wind speed, and rainfall intensity; topographic data includes land type, slope, windward or leeward slope, and the angle between the prevailing wind direction and the line direction in winter;

[0050] For step S1, meteorological data and topographic data of the distribution network tower to be evaluated are collected from the meteorological station, wherein the meteorological data include but are not limited to temperature, humidity, wind speed and rainfall intensity, and the topographic data include but are not limited to land type, slope, windward slope or leeward slope, and the angle between the main wind direction in winter and the line direction.

[0051] Step S2: inputting meteorological data and topographic data into the neural network model, so that the neural network model generates soil shear strength according to the meteorological data and topographic data;

[0052] In a preferred embodiment, before inputting the meteorological data and the terrain data into the neural network model, the method further comprises:

[0053] The meteorological data and topographic data are standardized and missing values ​​are supplemented to obtain preprocessed meteorological data and topographic data.

[0054] In a preferred embodiment, the meteorological data and the topographic data are standardized and missing values ​​are supplemented to obtain pre-processed meteorological data and topographic data, including:

[0055] Z-Score standardization was used to transform meteorological and topographic data into a normal distribution with a mean of 0 and a standard deviation of 1;

[0056] Linear interpolation is used to supplement the missing values ​​of the normal distribution to obtain the preprocessed meteorological and topographic data.

[0057] In step S2, to improve data quality and prediction accuracy, the collected meteorological and topographic data are preprocessed before being input into the neural network model. Data inconsistency and scale differences are addressed through methods such as standardization and missing value processing. Specifically, the Z-Score standardization method is first used to standardize the meteorological and topographic data, converting the data into a normal distribution with a mean of 0 and a standard deviation of 1. The mathematical representation is as follows:

[0058]

[0059] Where z represents the mean, x represents the value of the data, μ represents the expected value, σ represents the standard deviation, and N represents the number of data in each group. Data standardization can eliminate the scale differences between different data and make them comparable.

[0060] Secondly, for the treatment of missing values, such as Figure 2 As shown, linear interpolation is used for completion. The principle of linear interpolation is based on the linear relationship between known data points. Assuming that the point to be completed is x, and the two known data points are (x0, y0) and (x1, y1), according to the linear interpolation formula:

[0061]

[0062] You can get:

[0063]

[0064] Filling missing values ​​with linear interpolation can preserve the integrity and accuracy of the data to the greatest extent. Through the preprocessing steps, preprocessed meteorological and terrain data are obtained to ensure data consistency and availability.

[0065] Next, the preprocessed meteorological and topographic data are fed into the neural network model. These data serve as input features to describe the soil's environmental conditions and topographic characteristics. Meteorological data, such as temperature, humidity, wind speed, and rainfall intensity, reflect the soil's climatic environment and influence its shear strength. Topographic data, such as land type, slope, windward or leeward slope, and the angle between the prevailing winter wind direction and the line's orientation, describe the soil's physical characteristics and stress response. This data is fed into the neural network's input layer, and after calculation and learning in the hidden layer, an output layer is generated, representing the predicted soil shear strength.

[0066] The following is a detailed description of the training process of the neural network model:

[0067] like Figure 3 As shown in Figure 2, the training process of the neural network model includes the following steps:

[0068] Step S201: Obtain historical meteorological data and historical terrain data of distribution network towers at several historical moments and the soil shear strength at the corresponding historical moments;

[0069] In step S201, historical meteorological and topographical data for distribution network towers at various historical times are obtained from a meteorological database. The historical meteorological data includes temperature, humidity, wind speed, and rainfall intensity. The historical topographical data includes land type, slope, windward or leeward slope, and the angle between the prevailing winter wind direction and the line's orientation. Next, soil shear strength data for the historical times is obtained through field observations, soil sampling and testing, or historical records.

[0070] After collecting meteorological data, topographic data, and soil shear strength data, a dataset is created. Each data sample contains meteorological data, topographic data, and the corresponding soil shear strength at a historical moment. This dataset can be used to train and test prediction models.

[0071] Step S202: For each historical moment in time, perform standardization processing on the historical meteorological data and historical terrain data and supplement missing values ​​to obtain pre-processed historical meteorological data and historical terrain data;

[0072] In step S202, the historical meteorological and topographic data at each historical moment are normalized using the Z-Score method. By calculating the difference between each data point and the mean of the feature it belongs to and dividing it by the standard deviation of the feature, the data is transformed into a normal distribution with a mean of 0 and a standard deviation of 1. Linear interpolation is performed for missing values ​​in the historical meteorological and topographic data to make the dataset more complete and improve the accuracy of subsequent modeling and prediction, ultimately resulting in a preprocessed dataset.

[0073] Step S203: The pre-processed historical meteorological data and historical terrain data at several historical moments and the soil shear strength at the corresponding historical moments are respectively input into the initial physical information neural network model for iterative training until the loss function converges to obtain a trained neural network model.

[0074] In a preferred embodiment, the loss function includes: a data error term and a physical information error term;

[0075] Data error terms, specifically: mean square error;

[0076] The physical information error term is determined as follows:

[0077] Based on the pre-processed historical meteorological data and historical topographic data at the current historical moment input into the physical information neural network model, the mathematical shear strength of the soil is calculated using the Moore-Coulomb criterion;

[0078] Calculate the residual of soil strength based on the predicted shear strength generated during the current iterative training of the physical information neural network model and the corresponding mathematical shear strength;

[0079] The residual of soil strength is used as the physical information error term in the current iteration process of the physical information neural network model.

[0080] For step S203, first, a physical information neural network (Physics-Informed Neural Networks, PINN) model is constructed, and then the preprocessed data set obtained in step S202 is input into the initial physical information neural network model for iterative training until the loss function converges to obtain a trained neural network model.

[0081] The PINN architecture consists of an input layer, a hidden layer, and an output layer. The input layer receives the preprocessed dataset, the output layer represents the soil shear strength, and the hidden layer is the core of the neural network. Each neuron processes the input signal through a series of linear layers with a standard tanh activation function. Through these hidden layers, the model can learn the complex nonlinear relationship between the input data and the soil shear strength.

[0082] It should be noted that PINN is a novel network structure that combines traditional mathematical and physical models with modern deep learning techniques. It is similar to traditional fully connected neural network structures, but differs from them in the implementation of the model's loss function. The PINN model embeds physical information in the definition of the loss function, which consists of two main terms: data-fidelity loss and physics-informed loss.

[0083] The data error term measures the difference between the network's predicted output and the actual observed data, aiming to ensure the network fits the data as closely as possible. In this example, the mean squared error (MSE) is used as the data error term to optimize the model by minimizing the gap between the network's predicted values ​​and the true observed values. The physical information error term evaluates whether the predicted results satisfy the laws of physics. This part of the loss function is formed by substituting the network's predicted physical quantities into the corresponding physical laws, resulting in a residual error. This ensures that the predicted results adhere to the laws of physics. PINN incorporates physical knowledge as a priori, enabling it to generalize well to new problems outside the training data distribution and maintain reasonable predictions of physical phenomena even in data-scarce environments. In this example, the mathematical shear strength of the soil is calculated using the Moore-Coulomb criterion. The predicted shear strength is then compared with the mathematical shear strength to calculate the residual of the soil strength, which is used as the physical information error term.

[0084] After obtaining the trained neural network model, the test set can be used to cross-validate the predicted data and the corresponding experimental data. Specifically, the data in the test set is input into the neural network model for prediction to obtain the model prediction results. Then, these prediction results are compared with the corresponding experimental data. By calculating the error between them, the accuracy of the model can be evaluated. If the error between the two is within 10%, it can be considered that the model's prediction results have good credibility.

[0085] Step S3: Determine the ultimate overturning moment and the corresponding critical wind speed based on the soil shear strength;

[0086] In a preferred embodiment, determining the ultimate overturning moment and the corresponding critical wind speed based on the soil shear strength includes:

[0087] Calculate the soil erosion thickness based on the universal soil loss equation, where the soil erodibility factor is determined based on the soil shear strength.

[0088] According to the structure of distribution network tower, soil shear strength and soil scour thickness, a numerical model of distribution network tower is constructed based on artificial boundary method.

[0089] The numerical model of the distribution network tower is solved by the equivalent strain wedge method, and the ultimate overturning moment of the distribution network tower and the corresponding critical wind speed are obtained.

[0090] For step S3, the universal soil loss equation is a mathematical model based on the physical laws of soil scour and the principles of hydrodynamics. In this equation, the soil erodibility factor is considered one of the key factors determining soil scour thickness. A larger value for the soil erodibility factor indicates a lower shear strength of the soil, making scour more likely to occur. The following is the mathematical expression of the universal soil loss equation:

[0091] A=R×K×L×S×C×P

[0092] Where A represents the average annual soil loss per unit area (t / (hm 2 ·a)), R represents the rainfall erosivity factor (MJ·mm / (hm 2 ·h)), K represents the soil erodibility factor, L and S represent the slope length factor and slope factor, LS is also called the topographic factor, C represents the vegetation cover and management factor, and P represents the soil and water conservation measures factor. When no soil and water conservation measures are taken, P = 1.

[0093] The rainfall erosivity factor R can be calculated based on the rainfall data obtained in step S1 using the following formula:

[0094] R r =α(P r ) β

[0095] α=21.586β -7.182

[0096]

[0097] Among them, R r represents the rainfall erosivity factor on day r (MJ·mm / (hm 2 h)), P r represents the daily rainfall on day r (mm), when P r When <12mm, P r The value is 0, P d12 Indicates daily rainfall (mm) when the daily rainfall is ≥12 mm, P y12 represents the annual rainfall (mm) with daily rainfall ≥ 12 mm, α and β represent the statistical coefficients for calculating rainfall erosion factors, and are dimensionless.

[0098] The soil erodibility factor K can be determined based on the soil shear strength obtained in step S2. Soil shear strength refers to the maximum resistance that soil can withstand when subjected to shear. It is one of the important indicators of soil mechanical properties and reflects the soil's stability and anti-erosion ability. Specifically, the soil shear strength can be combined with other parameters (such as rainfall intensity, slope, soil type, etc.) to calculate the value of the soil erodibility factor K using an empirical formula or statistical model.

[0099] The terrain factor LS can be calculated based on the terrain data obtained in step S1 using the following formula:

[0100]

[0101] Where λ represents the projected slope length, τ represents the ratio of rill erosion to surface erosion, and θ represents the slope gradient.

[0102] The vegetation cover factor C can be calculated based on the vegetation coverage using the following formula:

[0103]

[0104] Among them, FVC stands for Fractional Vegetation Cover, NDVI stands for Normalized Difference Vegetation Index, max and NDVI min Indicates the maximum and minimum values ​​of NDVI during vegetation growth. For distribution network towers, NDVI can be used. max =1, NDVI min = 0. NDVI values ​​can be directly obtained from domestic data platforms such as the Geospatial Data Cloud.

[0105] After calculating the soil loss amount using the universal soil loss equation, the soil erosion thickness S under rainfall erosion can be calculated using the following formula: d :

[0106]

[0107] Among them, S d It represents the thickness of soil washed away by rainfall, and A represents the average annual soil loss per unit area (t / (hm 2 a)), ρ represents the density of the soil, which is generally taken as 1.35 kg / m 3 , or use local measured data, T represents the duration of rainfall.

[0108] Based on the structure of the distribution tower, soil shear strength, and soil scour thickness, a numerical model of the distribution tower was constructed using the artificial boundary method in ANSYS software. Specifically, a certain soil range was considered, and artificial bottom and side boundaries were introduced to divide the soil-structure model into a finite domain and an infinite domain. For non-rigid foundations, a cohesive boundary was selected as the artificial bottom boundary. This boundary simulates the interaction between the soil and the structure, accounting for soil stress transfer and deformation. For rigid foundations, a fixed boundary was selected as the artificial bottom boundary. This fixed boundary simulates the constraints imposed by the rigid foundation on the structure, accounting for soil stress transfer and deformation. Artificial side boundaries can be selected as either a bounding boundary or a vertical displacement constraint boundary. Both boundary conditions accurately simulate the free-field response of the soil. The artificial boundary method demonstrates excellent performance in simulating soil-structure interaction in dynamic response analysis of soil-structure interaction.

[0109] Then, the numerical model of the distribution network tower is solved by the equivalent strain wedge method. The strain wedge method is an extension of the one-dimensional elastic foundation reaction force method (Beam on elastic foundation method, BEF) in three-dimensional space. Figure 4 As shown in Figure 2, in the strain wedge method, the soil in front of the pile is assumed to form a three-dimensional wedge in the compressed area. As the horizontal force on the structure increases, the size of the wedge and the stress-strain curve of the soil will change. When the pile top is subjected to a horizontal load, the soil around the pile will resist the pile body. The action of the soil around the pile is equivalent to a nonlinear spring on the foundation beam. The governing equation of the pile-soil system is:

[0110]

[0111] Where EI represents the bending stiffness of the pile, y represents the horizontal displacement of the pile, z represents the depth below the ground surface, k represents the foundation reaction modulus, k = p / y, p represents the soil resistance provided by the unit pile length (N / m). Under the action of rainfall erosion, the foundation soil of the distribution network tower will form scour pits, such as Figure 5 As shown in the figure, S tw =S w +S d / tanθ.

[0112] It should be noted that the strain wedge method takes into account the continuity and constitutive model of the soil, and can also consider pile characteristics such as pile top constraints, pile body material, cross-sectional shape and flexural stiffness. This method does not rely on empirical parameters based on field tests and can provide more accurate soil-structure interaction analysis results.

[0113] The pile-soil system governing equation is used to solve the stress-strain curve of the excavated body, and the overturning moment of the distribution network tower is solved using the following formula:

[0114]

[0115] Among them, M u represents the ultimate overturning moment of the distribution network tower, h represents the foundation depth of the pole, b0 represents the calculated width of the pole, t represents the height of the pole rotation point from the ground, p z represents the earth pressure at depth z.

[0116] A dynamic analysis of the finite element model at different wind speeds was performed to obtain the bending moment at the base of the pole, thereby establishing a corresponding relationship between the critical wind speed and the ultimate overturning bending moment. Specifically, using finite element software, the pole and the surrounding soil layer were established as a finite element model. The model should include the geometry of the pole, material properties, and the properties of the soil layer. For different wind speeds, corresponding wind loads were applied to the pole, and a dynamic analysis was performed. The stress-strain curve of the soil body was solved using the pile-soil system control equation, and the overturning moment of the distribution network tower was solved. By gradually increasing the wind speed, the ultimate overturning bending moment at different wind speeds was obtained. Then, based on the obtained ultimate overturning bending moment and the corresponding wind speed, a corresponding relationship between the critical wind speed and the ultimate overturning bending moment was established.

[0117] To further explain, the wind load applied in the dynamic analysis is considered to be the superposition of average wind and fluctuating wind. Fluctuating wind is simulated using the Davenport wind speed spectrum. Average wind is the average value of wind direction and wind speed over a certain period of time and can be calculated using the following formula:

[0118]

[0119] Among them, v represents the average wind speed, v 10 It represents the 10-minute average wind speed at a height of 10 m, z represents the height, and γ represents the power exponent related to the ground roughness.

[0120] Finally, according to the corresponding relationship between critical wind speed and ultimate overturning bending moment, the critical wind speed of distribution network tower is obtained.

[0121] Step S4: Determine the wind resistance of the distribution network tower to be evaluated based on the critical wind speed.

[0122] For step S4, based on the critical wind speed obtained in step S3, the corresponding wind resistance is determined according to a preset range. In this embodiment, when the critical wind speed does not exceed the first preset wind speed threshold, the corresponding wind resistance is determined to be level three wind resistance. When the critical wind speed exceeds the first preset wind speed threshold and the critical wind speed does not exceed the second preset wind speed threshold, the corresponding wind resistance is determined to be level two wind resistance. When the critical wind speed exceeds the second preset wind speed threshold, the corresponding wind resistance is determined to be level one wind resistance. The distribution network tower corresponding to the level one wind resistance can withstand stronger wind forces and provide higher structural stability and reliability.

[0123] This paper uses a PINN neural network model to predict soil strength after rainfall erosion and infiltration, taking into account the effects of rainwater infiltration and soil erosion. Unlike traditional neural network models, the PINN model embeds physical information in the loss function to ensure that the predicted data meets physical laws. This approach greatly reduces errors and significantly improves prediction accuracy. In addition, the present invention also uses the equivalent strain wedge method to solve the structure's ultimate overturning moment and wind speed tolerance. Unlike traditional methods, this method does not rely on empirical parameters from field experiments and has better applicability. By using the equivalent strain wedge method, a structure's wind resistance can be more accurately assessed, providing a more reliable basis for engineering design and decision-making.

[0124] like Figure 6 As shown, based on the above method embodiment, a corresponding device embodiment is provided;

[0125] An embodiment of the present invention provides a device for evaluating the wind resistance of a distribution network tower structure, comprising: an environmental data acquisition module, a soil shear strength determination module, a critical wind speed determination module, and a wind resistance evaluation module;

[0126] The environmental data acquisition module is used to obtain meteorological and topographic data of the distribution network towers to be evaluated. Meteorological data includes temperature, humidity, wind speed, and rainfall intensity. Topographic data includes land type, slope, windward or leeward slope, and the angle between the prevailing wind direction and the line direction in winter.

[0127] a soil shear strength determination module, for inputting meteorological data and topographic data into a neural network model so that the neural network model generates soil shear strength based on the meteorological data and topographic data;

[0128] Critical wind speed determination module, used to determine the ultimate overturning moment and the corresponding critical wind speed based on the soil shear strength;

[0129] The wind resistance evaluation module is used to determine the wind resistance of the distribution network tower to be evaluated based on the critical wind speed.

[0130] In a preferred embodiment, the device for evaluating the wind resistance of a distribution network tower structure further includes: a pre-processing module;

[0131] The preprocessing module includes: performing standardization processing on meteorological data and terrain data and supplementing missing values ​​to obtain preprocessed meteorological data and terrain data.

[0132] It can be understood that the above-mentioned device embodiment corresponds to the method embodiment of the present invention, which can implement the distribution network tower structure wind resistance assessment method provided by any of the above-mentioned method embodiments of the present invention.

[0133] It should be noted that the device embodiments described above are merely illustrative, and some or all of the modules may be selected according to actual needs to achieve the purpose of the present embodiment. Furthermore, in the drawings of the device embodiments provided by the present invention, the connection relationship between modules indicates that they have a communication connection, which may be implemented as one or more communication buses or signal lines. Those skilled in the art can understand and implement the present invention without inventive effort.

[0134] Based on the above-mentioned embodiment of the distribution network pole tower structure wind resistance capacity assessment method, another embodiment of the present invention provides a terminal device, which includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, the distribution network pole tower structure wind resistance capacity assessment method of any embodiment of the present invention is implemented.

[0135] For example, in this embodiment, the computer program may be divided into one or more modules, which are stored in the memory and executed by the processor to implement the present invention. The one or more module elements may be a series of computer program instruction segments capable of performing specific functions, and the instruction segments are used to describe the execution process of the computer program in the terminal device.

[0136] The terminal device may be a computing device such as a desktop computer, a notebook computer, a PDA, a cloud server, etc. The terminal device may include, but is not limited to, a processor and a memory.

[0137] The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the terminal device, connecting various parts of the entire terminal device using various interfaces and lines.

[0138] Based on the above method embodiment, another embodiment is provided: another embodiment of the present invention provides a computer-readable storage medium, including a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute the distribution network tower structure wind resistance assessment method described in any one of the above method embodiments of the present invention.

[0139] In particular, if the module / unit integrated into the distribution network tower structure wind resistance assessment device / terminal equipment is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention can implement all or part of the processes in the above-mentioned embodiment method by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, it can implement the steps of each of the above-mentioned method embodiments. In particular, the computer program includes computer program code, which can be in source code form, object code form, executable file, or some intermediate form. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal, and software distribution medium.

[0140] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A method for evaluating the wind resistance of distribution network tower structures, characterized in that: include: Obtain meteorological and topographic data of the distribution network towers to be evaluated; The meteorological data includes: temperature, humidity, wind speed and rainfall intensity; the terrain data includes: land type, slope, windward slope or leeward slope, and the angle between the main wind direction in winter and the route direction; Inputting meteorological data and topographic data into a neural network model so that the neural network model generates soil shear strength based on the meteorological data and topographic data; Determining the ultimate overturning moment and the corresponding critical wind speed based on the soil shear strength; determining the ultimate overturning moment and the corresponding critical wind speed based on the soil shear strength includes: calculating the soil scour thickness based on a universal soil loss equation; wherein the soil erodibility factor in the universal soil loss equation is determined based on the soil shear strength; constructing a numerical model of the distribution network tower based on the artificial boundary method according to the structure of the distribution network tower, the soil shear strength, and the soil scour thickness; and solving the numerical model of the distribution network tower using the equivalent strain wedge method to obtain the ultimate overturning moment and the corresponding critical wind speed of the distribution network tower; Determining the wind resistance of the distribution network tower to be evaluated based on the critical wind speed; The neural network model is obtained in the following way: Obtain historical meteorological data and historical topographic data of distribution network towers at several historical moments and the soil shear strength at the corresponding historical moments; for the historical meteorological data and historical topographic data at each historical moment, perform standardization processing on the historical meteorological data and historical topographic data and supplement missing values ​​to obtain pre-processed historical meteorological data and historical topographic data; input the pre-processed historical meteorological data and historical topographic data at several historical moments and the soil shear strength at the corresponding historical moments into an initial physical information neural network model for iterative training until the loss function converges to obtain a trained neural network model; the loss function includes: a data error term and a physical information error term; the data error term is specifically: a mean square error; the physical information error term is determined in the following manner: according to the pre-processed historical meteorological data and historical topographic data at the historical moment currently input to the physical information neural network model, the mathematical shear strength of the soil is calculated by the Moore-Coulomb criterion; according to the predicted shear strength generated during the current iterative training process of the physical information neural network model and the corresponding mathematical shear strength, the residual of the soil strength is calculated; the residual of the soil strength is used as the physical information error term in the current iterative process of the physical information neural network model.

2. The method for evaluating the wind resistance of distribution network tower structures according to claim 1, characterized in that: Before feeding the meteorological and topographic data into the neural network model, it also includes: The meteorological data and terrain data are standardized and missing values ​​are supplemented to obtain preprocessed meteorological data and terrain data.

3. The method for evaluating the wind resistance of distribution network tower structures according to claim 2, characterized in that: The meteorological data and terrain data are standardized and missing values ​​are supplemented to obtain preprocessed meteorological data and terrain data, including: Z-Score standardization was used to convert the meteorological data and topographic data into a normal distribution with a mean of 0 and a standard deviation of 1; The linear interpolation method is used to supplement the missing values ​​of the normal distribution to obtain preprocessed meteorological data and terrain data.

4. A device for evaluating the wind resistance of a distribution network tower structure, characterized in that: include: Environmental data acquisition module, soil shear strength determination module, critical wind speed determination module, and wind resistance assessment module; The environmental data acquisition module is used to obtain meteorological data and terrain data of the distribution network tower to be evaluated; The meteorological data includes: temperature, humidity, wind speed and rainfall intensity; the terrain data includes: land type, slope, windward slope or leeward slope, and the angle between the main wind direction in winter and the route direction; The soil shear strength determination module is used to input meteorological data and terrain data into a neural network model so that the neural network model generates soil shear strength based on the meteorological data and terrain data; the neural network model is obtained by: obtaining historical meteorological data and terrain data at several historical moments of the distribution network tower and the soil shear strength at the corresponding historical moments; for the historical meteorological data and terrain data at each historical moment, performing standardization processing and missing value supplementation on the historical meteorological data and terrain data to obtain pre-processed historical meteorological data and terrain data; inputting the pre-processed historical meteorological data and terrain data at several historical moments and the soil shear strength at the corresponding historical moments into the initial physical model, respectively. Iterative training is performed in the physical information neural network model until the loss function converges to obtain a trained neural network model; the loss function includes: a data error term and a physical information error term; the data error term is specifically: a mean square error; the physical information error term is determined by: calculating the mathematical shear strength of the soil by the Moore-Coulomb criterion based on the preprocessed historical meteorological data and historical terrain data at the historical moment currently input to the physical information neural network model; calculating the residual of the soil strength based on the predicted shear strength generated during the current iterative training of the physical information neural network model and the corresponding mathematical shear strength; and using the residual of the soil strength as the physical information error term in the current iterative process of the physical information neural network model; The critical wind speed determination module is configured to determine the ultimate overturning moment and the corresponding critical wind speed based on the soil shear strength; determining the ultimate overturning moment and the corresponding critical wind speed based on the soil shear strength includes: calculating the soil scour thickness based on a universal soil loss equation; wherein the soil erodibility factor in the universal soil loss equation is determined based on the soil shear strength; constructing a numerical model of the distribution network tower based on the artificial boundary method according to the structure of the distribution network tower, the soil shear strength, and the soil scour thickness; and solving the numerical model of the distribution network tower using the equivalent strain wedge method to obtain the ultimate overturning moment of the distribution network tower and the corresponding critical wind speed; The wind resistance evaluation module is used to determine the wind resistance of the distribution network tower to be evaluated based on the critical wind speed.

5. The device for evaluating the wind resistance of a distribution network tower structure according to claim 4, characterized in that: Also includes: Preprocessing module; The preprocessing module includes: performing standardization processing and missing value supplementation on the meteorological data and terrain data to obtain preprocessed meteorological data and terrain data.

6. A terminal device, characterized in that: include: A processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, the method for evaluating the wind resistance of a distribution network tower structure according to any one of claims 1 to 3 is implemented.

7. A computer-readable storage medium, characterized in that include: A stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute the method for evaluating the wind resistance of the distribution network tower structure according to any one of claims 1 to 3.

Images