Micro meteorological station intelligent site selection method and system based on low-altitude air route

Abstract

The invention discloses an intelligent site selection method and system for a micro meteorological station based on a low-altitude air route, and relates to the field of meteorological station site selection, and the main scheme is as follows: numbering and dividing numbering areas in which micro meteorological base stations need to be established, and obtaining the data of each numbering area in the numbering areas; preprocessing the data of each numbered area, and obtaining an evaluation index set through data analysis; integrating into a comprehensive vector in a channel stacking manner, and then analyzing the comprehensive vector to obtain the site selection probability of each numbered region; performing comprehensive analysis on the comprehensive vector, the site selection probability and the evaluation index set of each numbered region through a GBDT model to obtain a site selection score of each numbered region; selecting a site selection site of the micro weather station; the safety guarantee effect of the micro-meteorological station on the low-altitude air route is fully considered, the problem that site selection of the micro-meteorological station and the low-altitude air route requirement are disjointed in the prior art is solved, and powerful meteorological support is provided for development of low-altitude economy.

Description

Technical Field

[0001] The present invention relates to the technical field of meteorological station site selection, and specifically to an intelligent site selection method and system for a micro-meteorological station based on low-altitude air routes. Background Art

[0002] The existing site selection methods for micro-meteorological stations are usually based on experience or simple data analysis. For example, in the site selection of power micro-meteorological stations, the site selection distance is often determined by analyzing the relationship between the partial correlation coefficient of micro-meteorological data and the horizontal distance of the site.

[0003] However, with the rapid development of the low-altitude economy, the application of low-altitude air routes is becoming more and more extensive. As an important means of low-altitude meteorological monitoring, the site selection of micro-meteorological stations has a crucial impact on the safety and efficiency of low-altitude air routes. Micro-meteorological stations can monitor the meteorological conditions along low-altitude air routes in real time, such as wind speed, wind direction, visibility, etc., providing accurate meteorological information for aircraft and ensuring flight safety. However, in some complex areas, the monitoring data of micro-meteorological stations are difficult to meet the needs of refined meteorological services for low-altitude air routes. Therefore, the development of the low-altitude economy has an increasing demand for low-altitude meteorological services, and the existing micro-meteorological station site selection methods lack consideration for adapting to the development of the low-altitude economy. Summary of the Invention

[0004] Aiming at the deficiencies of the existing technology, the present invention provides an intelligent site selection method for a micro-meteorological station based on low-altitude air routes, which at least solves the problem in the existing technology that the meteorological station cannot provide sufficient support for low-altitude air routes due to the lack of reference to specific low-altitude air route data during the site selection process.

[0005] To achieve the above objectives, the present invention is realized through the following technical solutions: An intelligent site selection method for a micro-meteorological station based on low-altitude air routes, comprising: Step 1: Number and divide the numbered areas where micro-meteorological base stations need to be established, and obtain the meteorological condition data, topographic and geomorphic data, low-altitude air route data, and surrounding environment data of each numbered area within the numbered areas; Step 2: Preprocess the meteorological condition data, topographic and geomorphic data, low-altitude air route data, and surrounding environment data of each numbered area, and obtain an evaluation index set through data analysis; Step 3: Integrate the preprocessed data into a comprehensive vector by means of channel stacking, and then analyze the comprehensive vector to obtain the site selection probability of each numbered area; Step 4: Through the GBDT model, comprehensively analyze the comprehensive vector, site selection probability, and evaluation index set of each numbered area to obtain the site selection score of each numbered area; Step 5: According to the data analysis in Step 4, select the site for the micro-meteorological station from several numbered areas.

[0006] In the above - mentioned preferred solution of an intelligent siting method for micro - meteorological stations based on low - altitude air routes, the evaluation index set includes the degree of impact on low - altitude air route operation, regional terrain index, low - altitude air route attraction value, and environmental impact index; Normalize the meteorological condition data, topographic and geomorphic data, low - altitude air route data, and surrounding environment data; Analyze the meteorological condition data through a multiple linear regression model to obtain the degree of impact on low - altitude air route operation in each numbered area; Analyze the topographic and geomorphic data through a terrain index model to obtain the regional terrain index of each numbered area; Analyze the low - altitude air route data through a gravity model to obtain the low - altitude air route attraction value; Analyze the surrounding environment data through an environmental impact index model to obtain the environmental impact index.

[0007] In the above - mentioned preferred solution of an intelligent siting method for micro - meteorological stations based on low - altitude air routes, the formula for calculating the degree of impact on low - altitude air route operation is: ; Where Y is the degree of impact on low - altitude air route operation; X1, X2, X3, X4, and X5 are the wind speed value, wind direction value, temperature value, humidity value, air pressure value, and precipitation value respectively; ɑ0, ɑ1, ɑ2, ɑ3, ɑ4, and ɑ5 are the regression coefficients of the wind speed value, wind direction value, temperature value, humidity value, air pressure value, and precipitation value respectively, and ξ is the error term.

[0008] In the above - mentioned preferred solution of an intelligent siting method for micro - meteorological stations based on low - altitude air routes, the formula for calculating the regional terrain index is: ; Where TI represents the regional terrain index, H is the altitude, S represents the slope, A represents the aspect, b is the weight coefficient of H; c is the weight coefficient of S, and f is the weight coefficient of A.

[0009] In the above - mentioned preferred solution of an intelligent siting method for micro - meteorological stations based on low - altitude air routes, the formula for calculating the low - altitude air route attraction value through the gravity model is: , Where T represents the low - altitude air route attraction value; O represents the number of low - altitude air routes passing through this numbered area, Q represents the number of flight tasks passing through this numbered area, D near represents the distance between this numbered area and the nearest low - altitude air route; k is a constant, is the impedance coefficient.

[0010] In the above - mentioned preferred solution of an intelligent siting method for micro - meteorological stations based on low - altitude air routes, the formula for calculating the environmental impact index is: ; Among them, EI represents the environmental impact index, B represents the average building height, and V represents the vegetation coverage; d is the weight coefficient of B; e is the weight coefficient of V.

[0011] In the above-mentioned preferred scheme of an intelligent siting method for micrometeorological stations based on low-altitude airways, the convolutional layer of the convolutional neural network CNN is used to organize the preprocessed meteorological condition data, topographic and geomorphic data, low-altitude airway data, and surrounding environment data into one-dimensional arrays respectively, and then the one-dimensional arrays are stacked according to the channel dimension to generate a comprehensive vector.

[0012] In the above-mentioned preferred scheme of an intelligent siting method for micrometeorological stations based on low-altitude airways, the method for obtaining the siting probability of each numbered area by analyzing the comprehensive vector is as follows: The convolutional neural network CNN is used to analyze the comprehensive vector. In the convolutional neural network CNN, the output layer is designed with three neurons so that the output layer can output three probability values. The comprehensive vector is input into the fully connected layer, and the fully connected layer maps the comprehensive vector to the output layer to obtain the output score vector kz kx ; The output score vector kz is kx converted into a probability distribution through the Softmax function. The formula is: , , , Among them, is a constant; kz kx is the output score of the kx-th neuron; is the probability output by the first neuron, is the probability output by the second neuron, is the probability output by the third neuron.

[0013] In the above-mentioned preferred scheme of an intelligent siting method for micrometeorological stations based on low-altitude airways, the method for obtaining the siting score by analyzing the comprehensive vector, siting probability, and evaluation index set of each numbered area is as follows: The comprehensive vector, siting probability, and evaluation index set of each numbered area are concatenated into a vector in sequence, and then , is input into the GBDT model to obtain the siting score. The calculation formula is as follows: , Among them, DF(ST) is the siting score of this numbered area, is the scaling factor of the j-th tree model; h j (ST) is the output value of the j-th tree model.

[0014] The present invention provides an intelligent site selection method for a micro-meteorological station based on a low-altitude airway, having the following beneficial effects: (1) By preprocessing the meteorological condition data, topographic and geomorphic data, low-altitude airway data, and surrounding environment data of the numbered areas and integrating them into a comprehensive vector, the problems of scattered siting data and lack of systematicness in the prior art are solved. Traditional site selection methods often only consider a single factor or simply superimpose several factors, and cannot comprehensively and deeply analyze the comprehensive conditions of site selection. However, in this solution, various data are integrated together through channel stacking to construct a multi-dimensional comprehensive vector, making the site selection analysis more comprehensive and scientific. This not only improves the accuracy of site selection, but also better adapts to the complex low-altitude environment, fully considering the interaction of various influencing factors, and providing more powerful data support for subsequent site selection decisions.

[0015] (2) By introducing the GBDT model to comprehensively analyze the comprehensive vector, site selection probability, and evaluation index set of each numbered area, the problems of inaccurate site selection evaluation and insufficient reliability in the prior art are solved. Traditional site selection evaluation methods may be based on simple rules or empirical formulas, and it is difficult to handle complex non-linear relationships and large amounts of data analysis. The GBDT model, as a powerful machine learning algorithm, can automatically learn the complex patterns and rules in the data, and comprehensively and deeply mine and analyze various indicators of site selection. The site selection score obtained through the GBDT model can more accurately reflect the actual site selection value of each numbered area, effectively avoiding the interference of human factors and subjective judgments, and improving the scientificity and reliability of site selection decisions.

[0016] (3) According to the score-based site selection calculation results, the site selection location of the micro-meteorological station is selected from several numbered areas, solving the problems of lack of quantitative basis for site selection decisions and difficulty in objective comparison in the prior art. When traditional methods select the site selection location, they often rely on simple index sorting based on experience, lacking a systematic and quantitative evaluation system. However, in this solution, by constructing an evaluation index set and calculating the site selection score, a clear quantitative standard is provided for site selection decisions. This makes the comparison between different numbered areas more intuitive and objective, and can quickly and accurately screen out the most suitable site selection location, improving the efficiency and quality of the site selection work, ensuring that the site selection of the micro-meteorological station can meet the meteorological monitoring needs of the low-altitude airway to the greatest extent, and guaranteeing the safety and efficiency of low-altitude flight.

[0017] (3) The entire intelligent siting method of micro-meteorological stations based on low-altitude airways fully considers the safety guarantee role of micro-meteorological stations for low-altitude airways, and solves the problem of the disconnection between the siting of micro-meteorological stations and the needs of low-altitude airways in the prior art. Traditional siting methods often do not closely combine with the actual operation of low-altitude airways, resulting in the monitoring data of micro-meteorological stations being unable to effectively serve low-altitude flights. However, in this solution, by using low-altitude airway data as one of the important siting bases, it ensures that the siting of micro-meteorological stations can accurately serve the meteorological guarantee needs of low-altitude airways, improving the safety and operation efficiency of low-altitude airways. At the same time, this intelligent siting method can also be dynamically adjusted and optimized according to the changes of low-altitude airways and the operation of micro-meteorological stations, with strong adaptability and foresight, providing strong meteorological support for the development of the low-altitude economy. Brief Description of the Drawings

[0018] Figure 1 It is a schematic diagram of the steps of an intelligent siting method of a micro-meteorological station based on low-altitude airways according to the present invention. Detailed Embodiments

[0019] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0020] Embodiment 1:

[0021] Please refer to Figure 1 , the present invention provides an intelligent siting method of a micro-meteorological station based on low-altitude airways, including: Step 1: Number and divide the numbered areas where micro-meteorological base stations need to be established, and obtain the meteorological condition data, topographic and geomorphic data, low-altitude airway data, and surrounding environment data of each numbered area within the numbered areas.

[0022] Specifically, the area where the micro-meteorological base station needs to be established is divided into square or rectangular grid small areas at intervals of longitude and latitude. For example, a small area can be divided every 0.01 degrees in the longitude direction and every 0.01 degrees in the latitude direction, so that each area point presents as a regular grid on the map, facilitating geographical positioning and data management. The number of each small area can correspond to the matching longitude and latitude ranges, such as area s1 corresponding to "N39.99 - 40.00, E116.30 - 116.31".

[0023] It should be noted that meteorological condition data can be obtained from the national meteorological department, other meteorological agencies, or meteorological satellite data and meteorological radars, and can at least include wind speed, temperature, humidity, air pressure, precipitation, etc.; topographic and geomorphic data can be obtained through a geographic information system database or field measurement, and can at least include altitude, slope, aspect, etc.; low-altitude airway data can be obtained through the aviation management department and aviation data service providers through application and cooperation, and can at least include the number of airways, traffic flow, etc. By importing the location data of each numbered area and low-altitude airway data into GIS software and using the spatial analysis function of GIS, the estimated distance between each meteorological station to be set and the low-altitude airway can be calculated, which is used as the distance between the meteorological station and the nearest low-altitude airway; the surrounding environment data can be obtained through the urban planning department or field investigation, and can at least include building height, vegetation coverage, water area, etc.

[0024] In this step, numbering and partitioning the numbered areas where micro-meteorological base stations need to be established can solve the problems of unclear and unsystematic partitioning of the siting areas in traditional siting methods. Traditional methods often rely on manual experience or simple geographical partitioning, lacking scientific area partitioning standards and methods. Through numbering and partitioning, the boundaries of each area can be made clear and the scope can be defined, facilitating subsequent refined data collection and analysis. Obtaining meteorological condition data, topographic and geomorphic data, low-altitude airway data, and surrounding environment data for each numbered area within the numbered area can solve the problems of incomplete and inaccurate data acquisition in the existing technology. Traditional siting methods usually only consider a single factor or simply superimpose several factors, and cannot comprehensively and deeply analyze the comprehensive conditions of siting. However, this solution can comprehensively and deeply analyze the siting conditions of each numbered area by obtaining multi-dimensional data, improving the scientificity and accuracy of siting.

[0025] Step 2: Preprocess the meteorological condition data, topographic and geomorphic data, low-altitude airway data, and surrounding environment data for each numbered area, and obtain an evaluation index set through data analysis.

[0026] Specifically, the preprocessing includes at least data normalization and alignment processing. By normalizing the meteorological condition data, topographic and geomorphic data, low-altitude airway data, and surrounding environment data, the values of each data type can be normalized to the range of 0 to 1. For example, after normalizing the wind speed, temperature, humidity, air pressure, precipitation, etc. in the meteorological condition data, several indicators such as wind speed (unit: m / s), temperature (unit: °C), humidity (unit: %), air pressure (unit: hPa), and precipitation (unit: mm) are selected. Suppose in a certain numbered area, the values of these indicators after normalization processing (such as mapping the data to the [0,1] interval) are: wind speed 0.6, temperature 0.4, humidity 0.5, air pressure 0.7, precipitation 0.2, etc. The meteorological condition data, topographic and geomorphic data, low-altitude airway data, and surrounding environment data of each numbered area are preprocessed, including data normalization and alignment processing. Through the normalization processing, the values of different data types are mapped to the same interval range, making the data comparable and processable, improving the compatibility and processability of the data, providing a basis for subsequent comprehensive analysis, and avoiding the problems in traditional site selection methods where the data types are diverse and the dimensions are different, resulting in difficult data processing and inability to directly conduct comprehensive analysis.

[0027] Furthermore, the evaluation index set includes: the degree of influence on low-altitude airway operation, regional terrain index, low-altitude airway attraction value, and environmental impact index.

[0028] Specifically, the degree of influence on low-altitude airway operation can be obtained by using a multiple linear regression model to analyze the relationship between meteorological elements and low-altitude airway operation. The calculation method adopted can be: ; where Y si is the degree of influence on low-altitude airway operation; X1, X2, X3, X4, and X5 are the wind speed value, wind direction value, temperature value, humidity value, air pressure value, and precipitation value respectively; ɑ0, ɑ1, ɑ2, ɑ3, ɑ4, and ɑ5 are the regression coefficients of the wind speed value, wind direction value, temperature value, humidity value, air pressure value, and precipitation value respectively, and ξ is the error term. By collecting historical meteorological data and low-altitude airway operation data, and using methods such as the least squares method to estimate the regression coefficients, the quantitative relationship between meteorological elements and low-altitude airway operation can be obtained. Here, ɑ0 can take a value of 0.5, ɑ1 can take a value of 0.3, ɑ2 can take a value of 0.2, ɑ3 can take a value of 0.1, ɑ4 can take a value of 0.15, and ɑ5 can take a value of 0.25.

[0029] Through the multiple linear regression model, the impact of meteorological elements on the operation of low-altitude air routes can be quantified, providing a scientific evaluation basis, improving the scientificity and accuracy of site selection, and avoiding the problem that the traditional site selection method cannot quantify the impact of meteorological elements on the operation of low-altitude air routes, resulting in the lack of scientific basis for site selection decisions.

[0030] Specifically, the regional terrain index can be used to quantify the impact of topography and geomorphology by using the terrain index model. The formula can be: ; Among them, TI represents the regional terrain index, H is the altitude, S represents the slope, A represents the aspect, b is the weight coefficient of H, which can take a value of 0.5; c is the weight coefficient of S, which can take a value of 0.3, and f is the weight coefficient of A, which can take a value of 0.2. Through the calculation and analysis of the terrain index, the impact of topography and geomorphology of different sites on the siting of micro-meteorological stations can be evaluated, providing a basis for site selection in terms of topography and geomorphology.

[0031] Through the terrain index model, the impact of topography and geomorphology on the siting of micro-meteorological stations can be comprehensively quantified, providing a scientific basis in terms of topography and geomorphology, improving the adaptability and reliability of site selection, and avoiding the situation that the traditional site selection method does not consider the impact of topography and geomorphology comprehensively enough, resulting in the site selection may not be able to make full use of the terrain advantages or avoid the terrain disadvantages.

[0032] By using the gravity model to analyze the low-altitude air route data, the attraction values of low-altitude air routes to different regions can be obtained. The formula can be: , Among them, T represents the low-altitude air route attraction value; O represents the number of low-altitude air routes passing through this numbered area, Q represents the number of flight tasks passing through this numbered area, and D near represents the distance between this numbered area and the nearest low-altitude air route; k is a constant, is the impedance coefficient; the constant k is used to adjust the overall scale of the flow and can be selected as a constant, such as 0.1. The impedance coefficient represents the impact degree of distance on the attraction value, which can be determined and adjusted through fitting analysis of historical data. For example, if is selected, it means a greater influence.

[0033] Through the gravity model, the attraction values of low-altitude air routes to different regions can be quantified, providing a scientific basis for the demand of low-altitude air routes, improving the pertinence and effectiveness of site selection, and solving the limitation problem that the traditional site selection method insufficiently considers the attraction of low-altitude air routes, resulting in the site selection may not be able to effectively serve the actual needs of low-altitude air routes.

[0034] Specifically, the environmental impact index can be used to quantify the impact of the surrounding environment by using the environmental impact index model, and the formula can be: ; Among them, EI represents the environmental impact index, B represents the average building height, V represents the vegetation coverage; d is the weight coefficient of B, which can take a value of 0.6; e is the weight coefficient of V, which can take a value of 0.4.

[0035] Through the environmental impact index model, the impact of the surrounding environment on the siting of the micrometeorological station can be comprehensively quantified, providing a scientific basis in terms of the environment, improving the adaptability and sustainability of the siting, and solving the problem that the traditional siting method does not consider the impact of the surrounding environment comprehensively enough, resulting in the siting may not be fully integrated into the surrounding environment or avoid environmental interference.

[0036] Step 3: Integrate the preprocessed data into a comprehensive vector through channel stacking, and then analyze the comprehensive vector to obtain the siting probability of each numbered area.

[0037] Specifically, the normalized and aligned meteorological condition data, topographic and geomorphic data, low-altitude airway data, and surrounding environment data are respectively organized into one-dimensional arrays through the convolutional layer. For example, each index of the meteorological condition data organized into a one-dimensional array can be [0.6, 0.4, 0.5, 0.7, 0.2]. All data groups are converted into four one-dimensional arrays in the same principle; then the four one-dimensional arrays are stacked according to the channel dimension, which can be achieved by using the numpy library in Python to form a comprehensive vector.

[0038] In the convolutional neural network CNN of this solution, the output layer needs to be designed with three neurons so that the output layer can output three probability values. The first neuron corresponds to low suitability, the second neuron corresponds to medium suitability, and the third neuron corresponds to high suitability, which is achieved by using the Softmax function; after obtaining the comprehensive vector, use the pooling layer to reduce the dimension of the comprehensive vector, input the comprehensive vector into the fully connected layer, and the fully connected layer maps the comprehensive vector to the score of the output layer. Then the output score vector is: , where kz is the output score vector, which contains three elements , and ; W is the weight matrix of the fully connected layer, and its dimension is the number of neurons × the sum of the lengths of each array in the one-dimensional array. For example, the number of neurons is 3, and the sum of the lengths of each array in the one-dimensional array is 20, then the dimension of W is 3×20. is the bias term, whose dimension is the same as the number of neurons. The dimension and values of the weight matrix determine how the input vector is mapped to the score vector of the output layer, thus affecting the calculation of the final site selection probability.

[0039] The output score vector kz is converted kx into a probability distribution through the Softmax function. The formula can be: , , , where, is a constant; it can take the value of 3.14; kz kx is the output score of the kx-th neuron; is the probability output by the first neuron, is the probability output by the second neuron, is the probability output by the third neuron.

[0040] In this step, the input data is subjected to feature extraction through the convolutional layer to capture important patterns and features in the data. The dimensionality of the comprehensive vector is reduced through the pooling layer to reduce the computational complexity while retaining important information, solving the problems that traditional methods are difficult to handle high-dimensional data, resulting in high computational complexity, low efficiency, difficult to automatically extract important features in the data, and requiring manual intervention. Through the convolutional layer and the pooling layer, high-dimensional data can be efficiently processed, the computational complexity can be reduced, important features in the data can be automatically extracted, manual intervention can be reduced, and the generalization ability of the model can be improved. The comprehensive vector output by the pooling layer is mapped to the score vector of the output layer, and a linear transformation is performed through the weight matrix W and the bias term . The output layer is designed to have three neurons, corresponding to the probability values of low suitability, medium suitability, and high suitability respectively, solving the problems that traditional site selection methods are difficult to handle multi-classification problems, unable to accurately evaluate the suitability of different site selection schemes, and difficult to map the comprehensive vector to probability values, resulting in a lack of scientific basis for site selection decisions. Through the fully connected layer and the Softmax function, the comprehensive vector can be mapped to three probability values, corresponding to low suitability, medium suitability, and high suitability respectively, providing a scientific basis for site selection evaluation. The Softmax function normalizes the output score vector into probability values, ensuring the rationality and interpretability of the probability values.

[0041] Step four: Through the GBDT model, a comprehensive analysis is performed on the comprehensive vector, site selection probability, and evaluation index set of each numbered area, and the site selection score of each numbered area can be obtained.

[0042] Specifically, the comprehensive vector, site selection probabilities of different categories, and the evaluation index set of each numbered area are concatenated into a vector in sequence, such as , and then , input it into the GBDT model to obtain the site selection score DF(ST), and the calculation formula is as follows: , where The scaling factor of the jth tree model, which is used to control the contribution of each model tree; h j (ST) is the output of the jth tree model. The settings of the tree model can be controlled by some parameters, such as the depth of the tree, the number of leaf nodes, the learning rate, etc. These parameters can be automatically adjusted through methods such as cross-validation to ensure the performance and generalization ability of the model; the site selection score is obtained by constructing multiple tree models h j (ST) through multiple rounds of iteration, and the outputs of these tree models are weighted and summed. In this way, the GBDT model can comprehensively consider various features, gradually optimize the site selection score, and help the intelligent agent better evaluate and select the optimal site selection strategy.

[0043] In this step, by splicing different data groups into a vector to form a unified input vector, the processability and analyzability of the data are improved, solving the problems in traditional site selection methods where different types of data are scattered, lacking a unified representation form, the dimensions of different data groups may be different, lacking a unified scoring standard, resulting in difficult comparison of evaluation results; through the comprehensive analysis of the GBDT model, the site selection suitability of each numbered area can be accurately evaluated, providing a scientific basis for site selection scoring. By comprehensively analyzing the input vector through the GBDT model, the site selection score DF(ST) is obtained. Through multiple rounds of iteration, multiple tree models hj(ST) are constructed, and the outputs of these tree models are weighted and summed to obtain the final site selection score. The GBDT model can comprehensively consider various features, gradually optimize the site selection score, improve the scientificity and accuracy of site selection, integrate different evaluation indicators into a unified scoring model, ensure the comparability and scientificity of the scoring results, and through multiple rounds of iteration and the construction of tree models, the GBDT model can handle complex data relationships, improve the generalization ability of the model, and solve the problem that traditional methods are difficult to handle complex data relationships, resulting in poor generalization ability of the model.

[0044] Step Five: According to the calculation results of Step Four, select the site for the micrometeorological station from several numbered areas.

[0045] Specifically, traverse and screen the calculation results of Step Four, and select the numbered area corresponding to the maximum value as the site for the micrometeorological station.

[0046] Example 2:

[0047] The content of Step 4 also includes: defining the comprehensive vector of each numbered area as the state of this area; comprehensively calculating the site selection probability and the site selection score, and taking the calculation result as the reward value to analyze the site selection value of each area through the Q-learning algorithm model; Specifically, the calculation formula for comprehensively calculating the site selection probability and the site selection score can be to normalize the site selection probability and the site selection score, and calculate the comprehensive analysis value by means of weighting. The calculation formula can be: ; where JL is the comprehensive analysis value and serves as the reward value of the Q-learning algorithm, is the weight coefficient of is the weight coefficient of is the weight coefficient of is the weight coefficient of DF(ST) and can take a value of 0.3.

[0048] It should be noted that the iterative update formula for the site selection value of the Q-learning algorithm is: ; where represents the site selection value of the current numbered area. Initially, the site selection value can be randomly initialized. As the Q-learning algorithm runs, the site selection value is gradually updated through the site selection value update formula. Specifically, after each action is taken, according to the reward value and the expectation of the next site selection value, the current site selection value is updated and optimized through the site selection value update formula, reflecting the long-term value of choosing the current action in the current state; represents the maximum site selection value of the next numbered area, indicating that in the next state under, all possible actions are traversed, the site selection value of each action is calculated, and the maximum value among them is taken. This value represents the site selection value of the optimal action in the next state; is the current state feature of the current numbered area; represents the selected location of the micrometeorological station in the current state; represents the reward value obtained after taking the current action. Here =JL; represents the current learning step size, which controls the degree of fusion of new information and is usually a value between 0 and 1. Here, it can be selected as 0.6; It represents the discount factor, which is the weight of future rewards and controls the degree of emphasis on future rewards in current learning. It is usually a value between 0 and 1 and can be set to 0.5; si represents the serial number of the numbered area.

[0049] In the Q - learning algorithm model, at the beginning, a site selection value table is established. The site selection value table is initialized to zero or random values. The site selection value table records the site selection values of each state - action pair. The model selects an action in the current state and executes the action , which will give an immediate reward value and the next state . Then, update the site selection value of the current state - action pair . Then continue to interact with the environment and repeat the above steps to continuously update the site selection value table. After multiple iterations, when the site selection values in the site selection value table no longer change significantly after multiple iterations, such as the change value of the same numbered area is less than the set change threshold, it can be considered that the site selection value has converged and the algorithm has obtained the final result, that is, each numbered area has a stable site selection value as the site selection value of each numbered area.

[0050] In this solution, the comprehensive vector of each numbered area is defined as the state of the area. The site selection probability and site selection score are comprehensively calculated, and the site selection value of each area is analyzed through the Q - learning algorithm model. This method solves problems in traditional site selection methods such as scattered data, difficult calculation of reward values, inaccurate evaluation of site selection values, slow convergence speed of the model, difficult initialization of site selection values, unscientific update of site selection values, difficult comprehensive evaluation, and unreasonable weight setting. Through the application of the Q - learning algorithm, the site selection value table can be gradually optimized, the site selection value of each area can be accurately evaluated, and the scientificity and accuracy of site selection can be improved.

[0051] Comprehensively calculate the site selection score and site selection value to obtain a comprehensive evaluation value; Specifically, comprehensively calculate the site selection score and site selection value of each numbered area. The calculation formula can be: ; where represents the comprehensive evaluation value of the numbered area si; represents the site selection score of the numbered area si; represents 's weight coefficient, which can take a value of 0.4, represents 's weight coefficient, which can take a value of 0.6.

[0052] ​By comprehensively calculating the site selection score and site selection value, a scientific comprehensive evaluation value calculation method is provided, ensuring the rationality and interpretability of the comprehensive evaluation value.

[0053] Embodiment 3:

[0054] The present invention also discloses an intelligent site selection system for a micro-meteorological station based on a low-altitude airway, which is used to implement the above-mentioned intelligent site selection method for a micro-meteorological station based on a low-altitude airway.

[0055] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution.

[0056] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0057] The above is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present application can easily think of changes or substitutions, which should all be covered within the protection scope of the present application.

Claims

1. An intelligent site selection method for micro-meteorological stations based on low-altitude air routes, characterized in that, Including: Step 1: Number and divide the numbered areas where micro-meteorological base stations need to be established, and obtain the meteorological condition data, topographic and geomorphic data, low-altitude airway data, and surrounding environment data for each numbered area within the numbered areas; Step 2: Preprocess the meteorological condition data, topographic and geomorphic data, low-altitude airway data, and surrounding environment data for each numbered area, and obtain an evaluation index set through data analysis; Step 3: Integrate the preprocessed data into a comprehensive vector by means of channel stacking, and then analyze the comprehensive vector to obtain the site selection probability for each numbered area; Step 4: Through the GBDT model, comprehensively analyze the comprehensive vector, site selection probability, and evaluation index set for each numbered area to obtain the site selection score for each numbered area; Step 5: According to the data analysis in Step 4, select the site for the micro-meteorological station from several numbered areas.

2. The intelligent siting method of a micrometeorological station based on a low-altitude airway according to claim 1, wherein The evaluation index set includes the degree of influence on low-altitude airway operation, regional terrain index, low-altitude airway attraction value, and environmental impact index; Normalize the meteorological condition data, topographic and geomorphic data, low-altitude airway data, and surrounding environment data; analyze the meteorological condition data through a multiple linear regression model to obtain the degree of influence on low-altitude airway operation for each numbered area; Analyze the topographic and geomorphic data through a terrain index model to obtain the regional terrain index for each numbered area; Analyze the low-altitude airway data through a gravity model to obtain the low-altitude airway attraction value; Analyze the surrounding environment data through an environmental impact index model to obtain the environmental impact index.

3. The intelligent site selection method of a micro-meteorological station based on a low-altitude airway according to claim 2, wherein The formula for calculating the degree of influence on low-altitude airway operation is: ； Where Y is the degree of influence on low-altitude airway operation; X1, X2, X3, X4, and X5 are the wind speed value, wind direction value, temperature value, humidity value, air pressure value, and precipitation value respectively; ɑ0, ɑ1, ɑ2, ɑ3, ɑ4, and ɑ5 are the regression coefficients of the wind speed value, wind direction value, temperature value, humidity value, air pressure value, and precipitation value respectively, and ξ is the error term.

4. The intelligent siting method of a micro-meteorological station based on a low-altitude airway according to claim 3, characterized in that, The calculation formula for the regional terrain index is: ； Where TI represents the regional terrain index, H is the altitude, S represents the slope, A represents the aspect, b is the weight coefficient of H; c is the weight coefficient of S, and f is the weight coefficient of A.

5. The intelligent site selection method for a micro-meteorological station based on a low-altitude airway according to claim 4, characterized in that, The formula based on which the low-altitude airway attraction value is calculated through the gravity model is: ， Among them, T represents the attraction value of the low-altitude airway; O represents the number of low-altitude airways passing through this numbered area, Q represents the number of flight missions passing through this numbered area, and D near represents the distance between this numbered area and the nearest low-altitude airway; k is a constant, which is the impedance coefficient.

6. The intelligent site selection method for a micro-meteorological station based on a low-altitude airway according to claim 5, characterized in that, The formula for calculating the environmental impact index is: ； Where EI represents the environmental impact index, B represents the average building height, V represents the vegetation coverage; d is the weight coefficient of B; e is the weight coefficient of V.

7. The intelligent siting method of a micro-meteorological station based on a low-altitude airway according to claim 6, characterized in that, Through the convolutional layer of the convolutional neural network CNN, organize the preprocessed meteorological condition data, topographic and geomorphic data, low-altitude airway data, and surrounding environment data into one-dimensional arrays respectively, and then stack the one-dimensional arrays according to the channel dimension to generate a comprehensive vector.

8. The intelligent site selection method for a micro-meteorological station based on a low-altitude airway according to claim 7, wherein, The method for obtaining the site selection probability for each numbered area by analyzing the comprehensive vector is: Analyze the comprehensive vector through a convolutional neural network (CNN). In the CNN, the output layer is designed to have three neurons so that the output layer can output three probability values. Input the comprehensive vector into the fully connected layer, and the fully connected layer maps the comprehensive vector to the output layer to obtain the output score vector kz kx ; Convert the output score vector kz through the Softmax function kx into a probability distribution, with the formula: ， ， ， Among them, is a constant; kz kx is the output score of the kx-th neuron; is the probability of the output of the first neuron, is the probability of the output of the second neuron, is the probability of the output of the third neuron.

9. The intelligent siting method of a micro-meteorological station based on a low-altitude airway according to claim 8, wherein, The method for obtaining the site selection score by analyzing the comprehensive vector, site selection probability, and evaluation index set for each numbered area is: Concatenate the comprehensive vector, site selection probability, and evaluation index set of each numbered area in sequence into a vector , and then , input it into the GBDT model to obtain the site selection score, and the calculation formula is as follows: ， Among them, DF(ST) is the site selection score of this numbered area, is the scaling factor of the j-th tree model; h j (ST) is the output value of the j-th tree model.

10. An intelligent siting system for micro-meteorological stations based on low-altitude air routes, characterized in that: An intelligent site selection method for a micro-meteorological station based on a low-altitude airway for implementing any one of the above claims 1-9.

Images