Urban heat island structural characteristics analysis system and method based on three-dimensional simulation

Abstract

The present invention discloses a system and method for analyzing the structural characteristics of an urban heat island based on three-dimensional simulation, which belongs to the technical field of three-dimensional modeling. The method comprises selecting multiple locations as collection points; recording temperature information and additional information of the collection points at multiple historical time points, wherein the additional information comprises meteorological data at the historical time points, shadow coverage time, surface coverage data within a first distance range, and urban building data; updating the three-dimensional city map in combination with the urban renewal data to obtain a future city map; establishing a prediction model; obtaining additional information of a target point in the future city map at a predicted time point, calculating the predicted temperature of the target point based on the additional information and the prediction model, and obtaining a city temperature distribution map; defining a target point whose predicted temperature is greater than a critical threshold as a high-risk point, and defining an area within a second distance range of the high-risk point as a high-risk area. The present invention can obtain the distribution of the urban heat island effect in the future, thereby providing help and reference for urban planning.

Description

Technical Field

[0001] The present invention belongs to the technical field of three-dimensional modeling, and in particular relates to a method and system for analyzing the structural characteristics of an urban heat island based on three-dimensional simulation. Background Art

[0002] The urban heat island effect refers to the phenomenon that the temperature in urban areas is higher than that in surrounding suburbs due to human activities or dense urban buildings. The heat island effect will have a significant impact on urban climate, residents' health, energy consumption and environmental quality. Therefore, studying the distribution and evolution of the urban heat island effect is of great significance to promoting the sustainable development of cities.

[0003] The following methods are proposed in the prior art to analyze the urban heat island effect. For example, the Chinese patent document with the publication number CN105868393A discloses a method for making and displaying a digital map of a micro-heat island in the city center. The method is based on GIS and a city coordinate system, and the urban thermal environment data and the urban spatial morphology data and image information are superimposed to form a digital map that directly reflects the characteristics of the urban thermal environment, and the rapid acquisition and access of massive data are achieved by combining with a computer database, and the temperature change of the urban micro-heat island space is directly reflected. For another example, the Chinese patent document with the publication number CN116502458A discloses a method for simulating an urban heat island based on a surface building index. The method obtains urban building data and vegetation data, and constructs an urban simulation model and a vegetation simulation model according to the building index and the vegetation index respectively, and then obtains the spatial distribution of vegetation and buildings according to the map software and the model, and constructs a panoramic model of the city, so that the urban landscape can be accurately simulated and restored, and finally the corresponding heat release is obtained according to the building and vegetation types to simulate the urban heat island.

[0004] The above are all based on satellite remote sensing images or collected point cloud images to model and analyze the existing urban heat island structure. In practical applications, it is also necessary to obtain the future urban heat island structure based on the existing urban characteristics and future urban planning. The above existing technologies do not eliminate methods for solving this problem. Summary of the invention

[0005] To solve the above problems, the present invention provides a system and method for analyzing the structural characteristics of urban heat islands based on three-dimensional simulation to solve the problems in the prior art.

[0006] In order to achieve the above-mentioned purpose of the invention, the present invention proposes a method for analyzing the structural characteristics of urban heat islands based on three-dimensional simulation, comprising:

[0007] Obtaining a three-dimensional city map of a target city, and selecting a plurality of locations as collection points according to the building distribution of the three-dimensional city map;

[0008] Recording temperature information and additional information of the collection point at multiple historical time points, the additional information including meteorological data at the historical time point, shadow coverage time before the historical time point, surface coverage data and urban building data within a first distance range of the collection point;

[0009] Determine a predicted time point, obtain city renewal data at the predicted time point, and update the three-dimensional city map in combination with the city renewal data to obtain a future city map;

[0010] analyzing the temperature information and the additional information to establish a prediction model;

[0011] Arranging a plurality of target points in the future city map, acquiring the additional information of the target points at the predicted time points, calculating the predicted temperatures of the target points based on the additional information and the prediction model, and importing the predicted temperatures into the future city map to obtain a city temperature distribution map;

[0012] A critical threshold is set, and the target point whose predicted temperature is greater than the critical threshold is defined as a high-risk point, and the area within the second distance range of the high-risk point is defined as a high-risk area.

[0013] Further, selecting the collection point comprises the following steps:

[0014] Acquire the building type of each urban building in the three-dimensional urban map, where the building types include industrial type, medical type and commercial type, and set a corresponding judgment threshold for each type of urban building;

[0015] The three-dimensional city map is divided into a plurality of first grid areas, and the first grid areas satisfying at least one preset condition are screened and defined as first areas, wherein the preset condition includes that the number of urban buildings of the same type in the first grid area exceeds the corresponding judgment threshold, or the average height of the urban buildings is greater than the first threshold, or the number of urban buildings is greater than the second threshold;

[0016] The first area is divided again into multiple second grid areas, and the second grid areas that meet at least one of the preset conditions are screened and defined as second areas. This step is repeated until an Nth area is obtained by division, and in the Nth area, the number of the Nth areas corresponding to each of the preset conditions is greater than the third threshold and less than the fourth threshold, and the center point of the Nth area is used as the collection point.

[0017] Furthermore, establishing the prediction model comprises the following steps:

[0018] The meteorological data includes meteorological temperature, wind force level and wind direction; the urban building data includes the number of urban buildings of each building type within the first distance range and the average height of all urban buildings; the surface coverage data includes the green area and water coverage area within the first distance range; a BP neural network model is established, the additional information of each collection point is used as the input feature of the neural network, the corresponding temperature information is used as the output feature of the neural network model, and the trained neural network is used as the prediction model.

[0019] Furthermore, obtaining the building type of each urban building in the three-dimensional city map includes the following steps:

[0020] Obtain an urban land use planning map, which includes the purpose of each plot of land, including industrial land, medical land and commercial land, and match the urban land use planning map and the three-dimensional city map based on longitude and latitude to determine the corresponding building type according to the purpose of the plot of land where the urban building is located.

[0021] Furthermore, obtaining the three-dimensional city map includes the following steps:

[0022] Acquire an aerial image of the target city, extract the plane coordinates, height value, and brightness value of each pixel point in the aerial image, and sequentially connect the pixel points whose positions are adjacent and whose height value difference is within a fifth threshold value based on the plane coordinates, so as to draw a plurality of first contours in the aerial image;

[0023] Calculating an average value of the height values ​​of each first contour and each pixel point therein, setting the first contour whose average value is greater than a sixth threshold as a second contour, the second contour being the contour of the top surface of the urban building, and generating a three-dimensional model of the urban building based on the height values ​​of each pixel point on the edge of the second contour;

[0024] Acquire the brightness value of the second contour and each pixel point therein, cluster the brightness values, and divide the pixel points into multiple categories according to the clustering result;

[0025] The area covered by the pixel points of the same category is defined as a sub-area, the average value of the height value in each sub-area is calculated, and the top surface model of the urban building is generated based on the average value and the position of each sub-area.

[0026] Furthermore, if there is a sub-region where the average value of the brightness value is lower than a seventh threshold, the top surface model is constructed according to the average value of the height values ​​of the pixel points in the corresponding sub-region.

[0027] Furthermore, obtaining the shadow coverage time of the acquisition point includes the following steps:

[0028] The sunrise time of the day is obtained, and the time period between the sunrise time and the current time point is divided into multiple sub-time periods, and the shadow coverage of each urban building within the first distance range of the collection point during the sub-time period is simulated. If it is determined that the collection point is covered in the sub-time period according to the shadow coverage situation, the corresponding sub-time period is defined as a covered time period, and all the covered time periods between the sunrise time and the current time point are accumulated as the shadow coverage time of the collection point.

[0029] Further, obtaining the shadow coverage condition includes the following steps:

[0030] A vertical section facing the sun is determined in an urban building, a section shape of the top surface model in the vertical section is obtained and defined as a cross-sectional shape, a shadow length and a projection direction of each urban building in the sub-time period are calculated, and the cross-sectional shape is projected onto the ground based on the shadow length and the projection direction to form the shadow coverage of the urban building.

[0031] The present invention also provides a system for analyzing the structural characteristics of urban heat islands based on three-dimensional simulation, which is used to implement the above-mentioned method for analyzing the structural characteristics of urban heat islands based on three-dimensional simulation, and the system comprises:

[0032] A first analysis module is used to obtain a three-dimensional city map of a target city and select multiple locations as collection points according to the building distribution of the three-dimensional city map;

[0033] An information collection module, used to record the temperature information and additional information of the collection point at multiple historical time points, wherein the additional information includes meteorological data at the historical time point, shadow coverage time before the historical time point, surface coverage data within a first distance range of the collection point, and urban building data;

[0034] A map updating module, which determines a predicted time point, obtains city renewal data at the predicted time point, and updates the three-dimensional city map in combination with the city renewal data to obtain a future city map;

[0035] A prediction module, configured to analyze the temperature information and the additional information to establish a prediction model, arrange multiple target points in the future city map, obtain the additional information of the target points at the prediction time point, calculate the predicted temperature of the target points based on the additional information and the prediction model, import the predicted temperature into the future city map, and obtain a city temperature distribution map;

[0036] The second analysis module is provided with a critical threshold value. The second analysis module defines the target point whose predicted temperature is greater than the critical threshold value as a high-risk point, and defines an area within a second distance range of the high-risk point as a high-risk area.

[0037] Compared with the prior art, the beneficial effects of the present invention are at least as follows:

[0038] The present invention first establishes a three-dimensional city map that highly restores the target city, with the purpose of obtaining the specific distribution of urban buildings and greening information, and then determining representative collection points. Then, temperature sensors are arranged at the collection points to obtain a large amount of accurate temperature information, and the two-dimensional city map and the three-dimensional city map are combined to obtain additional information of the collection points. Finally, this information is input into the neural network model for training to obtain a prediction model. After obtaining the additional information of the target point to be analyzed later, the prediction model combines this additional information to predict the corresponding temperature situation. That is, the correlation between various urban distribution characteristics and the urban heat island effect is implicit in the prediction model.

[0039] Before predicting the future urban heat island distribution, the present invention first updates the current three-dimensional city map according to the urban renewal data at the time point to be predicted to obtain the future city map, then selects multiple target points in the future city map, obtains additional information of the target points, and combines the prediction model to obtain the temperature data of the target points. After obtaining the temperature data, the distribution of the urban heat island effect can be established, thereby helping relevant personnel to evaluate the urban planning scheme. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 It is a flowchart of the steps of the method for analyzing the structural characteristics of urban heat islands based on three-dimensional simulation of the present invention;

[0041] Figure 2 A schematic diagram showing the principle of dividing the grid area according to the present invention;

[0042] Figure 3 A schematic diagram of the shadow coverage of urban buildings is generated for the present invention;

[0043] Figure 4 It is a structural schematic diagram of the urban heat island structural characteristics analysis system based on three-dimensional simulation of the present invention. DETAILED DESCRIPTION

[0044] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0045] It is understood that the terms "first", "second", etc. used in this application may be used herein to describe various elements, but unless otherwise specified, these elements are not limited by these terms. These terms are only used to distinguish a first element from another element. For example, without departing from the scope of this application, a first xx script may be referred to as a second xx script, and similarly, a second xx script may be referred to as a first xx script.

[0046] like Figure 1 As shown, a method for analyzing the structural characteristics of urban heat islands based on three-dimensional simulation includes:

[0047] S1: Obtain a three-dimensional city map of a target city, and select multiple locations as collection points according to the building distribution of the three-dimensional city map.

[0048] S2: Recording temperature information and additional information of the collection point at multiple historical time points, wherein the additional information includes meteorological data at the historical time point, shadow coverage time before the historical time point, surface coverage data within a first distance range of the collection point, and urban building data.

[0049] The present invention establishes a three-dimensional city map by aerial photography, and the specific establishment method is described later. The three-dimensional city map includes information such as the building type and building height of each building in the city. Then, multiple locations are selected as collection points according to the distribution of the buildings. Temperature sensors are set at the collection points, and the actual temperature of the collection points is obtained by the temperature sensors. It is remotely sent to the mobile terminal through the wireless network, so that the temperature information of multiple subsequent time points can be obtained by deploying the sensors once. In addition, since data is collected by deploying temperature sensors, a large amount of data for training the prediction model can be quickly obtained, so that the prediction accuracy of the subsequent prediction model is guaranteed. Compared with the existing method of obtaining temperature by remote sensing, the temperature obtained by the sensor is more accurate. Compared with the method of handheld thermometer, this method eliminates the need for frequent on-site temperature measurement in the future. Both of these traditional acquisition methods cannot obtain a large amount of data for training the prediction model in a short time.

[0050] In addition, the meteorological data in the additional information of the collection point can be obtained through the meteorological forecast station, and the shadow coverage time is obtained through simulation and calculation. The specific acquisition method is described later. The first distance range is set to within 3 km of the collection point in this embodiment. The surface coverage data includes vegetation coverage and river distribution, which are obtained through the existing two-dimensional city map. The urban building data includes building height and three-dimensional shape of the building, which are obtained through the three-dimensional city map.

[0051] S3: Determine a predicted time point, obtain city renewal data at the predicted time point, and update the three-dimensional city map in combination with the city renewal data to obtain a future city map.

[0052] S4: Analyze the temperature information and the additional information to establish a prediction model.

[0053] The prediction model is constructed by a neural network algorithm. Its function is to input predetermined features, such as meteorological data and surface cover data, into the prediction model. The prediction model analyzes these features to predict the temperature information of the collection point. After the prediction model is established, when the future urban heat island effect needs to be analyzed, the time point to be predicted is first determined, such as 15:00 on June 1, 2030, and then the corresponding urban renewal data is obtained through the database. The urban renewal data is provided by relevant departments or obtained from urban planning maps or outlines published on the Internet, which include which areas will build or renovate urban buildings, and which areas will demolish existing urban buildings and replace them with functional buildings, such as railway stations or parks. The existing three-dimensional city map is updated based on these data. Specifically, if a new building is planned in a certain area, the building is added to the three-dimensional city map. The new buildings in the future city map can be replaced by a unified preset model, and the height generated is determined according to the parameters in the planning scheme.

[0054] S5: Arrange multiple target points in the future city map, obtain the additional information of the target points at the predicted time points, calculate the predicted temperature of the target points based on the additional information and the prediction model, import the predicted temperature into the future city map, and obtain a city temperature distribution map.

[0055] S6: Setting a critical threshold, defining the target point whose predicted temperature is greater than the critical threshold as a high-risk point, and defining an area within a second distance range of the high-risk point as a high-risk area.

[0056] The target points can be automatically and evenly spaced in the future city map, or they can be manually laid out, and then additional information of the target is obtained. The method for obtaining additional information of the target point is the same as that of the collection point. The additional information of the target point is then input into the prediction model to obtain the predicted temperature of the target point. Finally, the predicted temperature obtained is marked in the future city map to obtain the city temperature distribution map. Here, the critical threshold is set to 40°C. When the predicted temperature of the target point is higher than 40°C, the target point is set as a high point, and the area within the second distance range around it is set as a high-risk area. The second distance range is, for example, 1KM. The temperature change of the target point in the model can be obtained by multiple prediction time point models. If a large number of high-risk areas appear in the city and the number of occurrences is frequent, the urban planning situation should be reviewed to avoid serious problems with the urban heat island effect.

[0057] The present invention first establishes a three-dimensional city map that highly restores the target city, with the purpose of obtaining the specific distribution of urban buildings and greening information, and then determining representative collection points. Then, temperature sensors are arranged at the collection points to obtain a large amount of accurate temperature information, and the two-dimensional city map and the three-dimensional city map are combined to obtain additional information of the collection points. Finally, this information is input into the neural network model for training to obtain a prediction model. After obtaining the additional information of the target point to be analyzed later, the prediction model combines this additional information to predict the corresponding temperature situation. That is, the correlation between various urban distribution characteristics and the urban heat island effect is implicit in the prediction model.

[0058] Before predicting the future distribution of urban heat islands, the present invention first updates the current three-dimensional city map according to the urban renewal data at the time point to be predicted to obtain the future city map, and then selects multiple target points in the future city map, obtains additional information of the target points, and obtains the temperature data of the target points in combination with the prediction model. After obtaining the temperature data, the distribution of the urban heat island effect can be established, thereby helping relevant personnel to evaluate urban planning programs.

[0059] It is particularly noteworthy that the present invention can be used to obtain the distribution of urban heat island effects in the future, thereby providing help and reference for urban planning.

[0060] In this embodiment, selecting the collection point includes the following steps:

[0061] Acquire the building type of each urban building in the three-dimensional urban map, where the building types include industrial type, medical type and commercial type, and set a corresponding judgment threshold for each type of urban building;

[0062] The three-dimensional city map is equally divided into a plurality of first grid areas, and the first grid areas satisfying at least one preset condition are screened and defined as first areas, wherein the preset condition includes that the number of urban buildings of the same type in the first grid area exceeds the corresponding judgment threshold, or the average height of the urban buildings is greater than the first threshold, or the number of urban buildings is greater than the second threshold.

[0063] When generating a 3D city map, the building type of each city building has been marked in advance, and then different judgment thresholds are set for each building type, such as the judgment threshold of 6 for industrial type, 2 for medical type, and 1 for commercial type. Figure 2As shown, it is assumed that the three-dimensional city map D is evenly divided into 9 first grid areas, and then each first grid area is detected to determine whether the first grid area meets at least one preset condition. The meaning of the preset condition here is that if the number of urban buildings of the same type in the first grid area exceeds the corresponding judgment threshold, the area is considered to have a gathering area of ​​buildings of the corresponding type, if the average height of the urban buildings is greater than the first threshold, it is considered to be a gathering area of ​​high-rise buildings, and if the number of urban buildings exceeds the second threshold, the first threshold is 50, for example, it is considered to be a densely built area.

[0064] The first area is divided again into multiple second grid areas, and the second grid areas that meet at least one of the preset conditions are screened and defined as second areas. This step is repeated until an Nth area is obtained by division, and in the Nth area, the number of the Nth areas corresponding to each of the preset conditions is greater than the third threshold and less than the fourth threshold, and the center point of the Nth area is used as the collection point.

[0065] Usually the first grid area is divided into larger areas, so most of the first grid areas will meet at least one preset condition, such as Figure 2 As shown, except for the first grid areas A1 and A2, the remaining first grid areas all meet at least one preset condition, so they are further divided into multiple second grid areas, for example, the first grid area is further divided into second grid areas B1-B4, and then each second grid area is analyzed again to determine whether it meets at least one preset condition. The scope of the corresponding area is narrowed by continuous subdivision, and finally an area that only meets one or two preset conditions is obtained. This implementation also adds a restriction condition to the division process. The specific restriction condition is that after each N-1 area is divided to obtain the Nth grid area, the number of the Nth area that meets each condition is obtained. Here, the third threshold is set to 10 and the fourth threshold is set to 30. For example, in the Nth grid area, there are 12 grid areas that meet the number of industrial-type urban buildings exceeding 5, which is between the third threshold and the fourth threshold. Similarly, at this time, the number of other grid areas that meet the preset conditions is between the third threshold and the fourth threshold, so the grid area will no longer be subdivided. This condition can limit the number of generated collection points.

[0066] This method only selects the building type and quantity in the grid area. The specific positioning process is time-consuming by continuously dividing the grid area. Compared with directly analyzing the entire city 3D map, this method allows each grid area to be analyzed in parallel, which speeds up the overall analysis speed. In the analysis process, the grid areas that do not meet the preset conditions are continuously discarded, thereby further speeding up the analysis speed.

[0067] Establishing the prediction model comprises the following steps:

[0068] The meteorological data includes meteorological temperature, wind force level and wind direction; the urban building data includes the number of urban buildings of each building type within the first distance range and the average height of all urban buildings; the surface coverage data includes the green area and water coverage area within the first distance range; a BP neural network model is established, the additional information of each collection point is used as the input feature of the neural network, the corresponding temperature information is used as the output feature of the neural network model, and the trained neural network is used as the prediction model.

[0069] The meteorological data at the current time is obtained from the meteorological forecast department, and the meteorological data at the future predicted time point is determined according to the meteorological data of the same period in previous years. Based on the above description, there are 7 features that need to be input into the BP neural network, and the BP neural network outputs temperature information. The neural network of this embodiment only has one hidden layer, the number of input layer neurons is set to 7, the number of output layer neurons is set to 1, the training function uses traingdm, the activation function uses sigmoid, the learning rate is set to 0.001, and when the mean square error MSE is set to 0.1, in other embodiments, building materials and ground materials can also be input into the neural network as additional information to further improve the accuracy of the prediction.

[0070] In this embodiment, obtaining the building type of each urban building in the three-dimensional city map includes the following steps:

[0071] Obtain an urban land use planning map, which includes the purpose of each plot of land, including industrial land, medical land and commercial land, and match the urban land use planning map and the three-dimensional city map based on longitude and latitude to determine the corresponding building type according to the purpose of the plot of land where the urban building is located.

[0072] Specifically, first obtain the outline shape of each urban building in the three-dimensional city map, where the outline shape is the shape from a bird's-eye view, obtain the longitude and latitude of the centroid of each outline shape, obtain the location with the same longitude and latitude in the urban land use planning map, obtain the purpose of the plot where the location is located, and the building type can be determined based on the purpose. For example, for a plot used for industrial purposes, the corresponding building type is industrial.

[0073] In this embodiment, obtaining the three-dimensional city map includes the following steps:

[0074] Acquire an aerial image of the target city, extract the plane coordinates, height value and brightness value of each pixel point in the aerial image, and connect the pixel points with adjacent positions and whose height value difference is within the fifth threshold range in sequence based on the plane coordinates to draw multiple first contours in the aerial image.

[0075] Calculate the average value of the height values ​​of each first contour and each pixel point inside it, set the first contour whose average value is greater than a sixth threshold as the second contour, the second contour is the contour of the top surface of the urban building, and generate a three-dimensional model of the urban building based on the height values ​​of each pixel point on the edge of the second contour.

[0076] Specifically, after taking aerial images of the target city from two viewpoints, the three-dimensional coordinates of each point in the aerial image are determined using a stereo matching algorithm. The three-dimensional coordinates include plane coordinates and height coordinates. The value of the height coordinate is the value of the distance between the pixel and the ground. The height value of each pixel is determined based on the height coordinate. After the aerial image is processed into a grayscale image, the grayscale value of the pixel is used as the brightness value. The higher the grayscale value, the higher the corresponding brightness value. Multiple first contours can be obtained by sequentially connecting pixel points with adjacent plane coordinates and height value differences within the fifth threshold range. Pixel points that meet this condition are generally the roof contours of aerial images and the terrain contours on the ground.

[0077] Then, in order to distinguish the two, the average value of the height value of each pixel point in each contour is calculated, and the first contour with an average value greater than the sixth threshold is selected as the second contour representing the top surface shape of the urban building. The sixth threshold is, for example, 3 meters. Finally, the second contour is placed in the space for constructing the three-dimensional map of the city according to the average height of each pixel point of the second contour, and then the second contour is stretched downward to construct a three-dimensional model of the urban building.

[0078] The brightness value of each pixel point in the second contour and inside the second contour is obtained, the brightness values ​​are clustered, and the pixels are divided into multiple categories according to the clustering results, and the differences in the brightness values ​​of the pixels in the same category are all within a preset value range.

[0079] The area covered by the pixel points of the same category is defined as a sub-area, the average value of the height value in each sub-area is calculated, and the top surface model of the urban building is generated based on the average value and the position of each sub-area.

[0080] Since the top surfaces of urban buildings are diverse, considering that the stereo matching algorithm may have errors in calculating the height of each pixel, in order to further improve the fineness of the building model and the accuracy of the subsequent calculation of the shadow coverage time, the present invention also clusters the brightness values ​​of each pixel point inside each second contour. Generally, the farther the position of a certain roof is from the ground, the greater the brightness value of its pixel point, so the brightness values ​​of the pixels in the same plane should be closer; this embodiment uses the nearest neighbor propagation algorithm for clustering. For a second contour and the pixels inside it, if all the pixels are clustered into one type after clustering, it indicates that the roof is a plane. When generating the building model, the roof is directly processed as a plane, which can avoid the appearance of a wavy roof.

[0081] If there are multiple types, it means that the structure of the roof is more three-dimensional and there are multiple planes with different heights. In this case, the planes formed by pixels of the same type are counted as sub-regions. The placement of the sub-region in space is determined according to the average value of the position and height value of each sub-region, and then the sub-region is stretched downward based on the contour shape to construct a three-dimensional top surface model of the roof.

[0082] In this embodiment, if the average value of the brightness value of the sub-region is lower than the seventh threshold, the top surface model is constructed according to the average value of the height values ​​of the pixel points in the corresponding sub-region.

[0083] If the average brightness value of a sub-area is too low, it is considered that it may be blocked by the shadows of surrounding buildings, resulting in a low brightness value. In this case, the sub-area is treated as a plane and its height value is determined according to the average height value of each pixel inside.

[0084] Obtaining the shadow coverage time of the acquisition point comprises the following steps:

[0085] The sunrise time of the day is obtained, and the time period between the sunrise time and the current time point is divided into multiple sub-time periods, and the shadow coverage of each urban building within the first distance range of the collection point during the sub-time period is simulated. If it is determined that the collection point is covered in the sub-time period according to the shadow coverage situation, the corresponding sub-time period is defined as a covered time period, and all the covered time periods between the sunrise time and the current time point are accumulated as the shadow coverage time of the collection point.

[0086] The time covered by the building shadow will also affect the temperature collected by the temperature sensor. The present invention adopts a simulation method to obtain the time when the temperature sensor is covered by the building shadow. Specifically, assuming that the shadow coverage time before 15:00 on the same day is to be obtained, first obtain the sunrise time of the day, for example, 7 o'clock, and divide 7 o'clock to 15:00 into 8 sub-time periods, each sub-time period is 1 hour long, and then determine whether the location of the collection point in each sub-time period is covered by the building shadow. If covered, the corresponding sub-time period is used as the covered time period. Finally, count how many covered segments there are in the 8 sub-time periods. If there are 5 covered time periods, the corresponding shadow coverage time is 5 hours.

[0087] In this embodiment, obtaining the shadow coverage situation includes the following steps:

[0088] A vertical section facing the sun is determined in an urban building, a section shape of the top surface model in the vertical section is obtained and defined as a cross-sectional shape, a shadow length and a projection direction of each urban building in the sub-time period are calculated, and the cross-sectional shape is projected onto the ground based on the shadow length and the projection direction to form the shadow coverage of the urban building.

[0089] Obtain the vertical section according to the direction of sunlight, obtain the shape of the top surface model in the vertical section, and project it on the ground to obtain the shadow area. Figure 3 As shown, F is the direction of the sun. The vertical section G is determined according to the direction, and the cross-sectional shape of the urban building D in the vertical section G is obtained. Then, the cross-sectional shape is projected onto the ground according to the projection direction to obtain the shadow area E. Then, the length of the shadow area E is adjusted proportionally according to the shadow length.

[0090] When calculating the shadow length, first obtain the longitude and latitude of the second contour centroid and the solar declination of the corresponding sub-time period. Based on the solar declination and the longitude and latitude of each urban building, calculate the solar altitude angle and solar azimuth of the corresponding urban building. The solar altitude angle is the angle between the sun's rays and the ground, and the azimuth is the angle between the sun's rays and the horizon in the north direction. The direction of the sun's radiation is determined based on the solar altitude angle and the solar azimuth. After determining the altitude angle, the shadow length can be determined. The shadow length is calculated, for example, using the following formula: L=h / tanθ, where L is the shadow length, θ is the altitude angle of the sun, and h is the height of the urban building. After determining the direction of the sun's radiation, the projection direction can be determined. After obtaining the above data, the shadows of urban buildings can be roughly simulated, and the corresponding shadow coverage can be obtained.

[0091] like Figure 4As shown, the present invention also provides a system for analyzing the structural characteristics of urban heat islands based on three-dimensional simulation, which is used to implement the above-mentioned method for analyzing the structural characteristics of urban heat islands based on three-dimensional simulation, and the system includes:

[0092] A first analysis module is used to obtain a three-dimensional city map of a target city and select multiple locations as collection points according to the building distribution of the three-dimensional city map;

[0093] An information collection module, used to record the temperature information and additional information of the collection point at multiple historical time points, wherein the additional information includes meteorological data at the historical time point, shadow coverage time before the historical time point, surface coverage data within a first distance range of the collection point, and urban building data;

[0094] A map updating module, which determines a predicted time point, obtains city renewal data at the predicted time point, and updates the three-dimensional city map in combination with the city renewal data to obtain a future city map;

[0095] A prediction module, configured to analyze the temperature information and the additional information to establish a prediction model, arrange multiple target points in the future city map, obtain the additional information of the target points at the prediction time point, calculate the predicted temperature of the target points based on the additional information and the prediction model, import the predicted temperature into the future city map, and obtain a city temperature distribution map;

[0096] The second analysis module is provided with a critical threshold value. The second analysis module defines the target point whose predicted temperature is greater than the critical threshold value as a high-risk point, and defines an area within a second distance range of the high-risk point as a high-risk area.

[0097] It should be understood that the various technical features of the above-mentioned embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the various technical features in the above-mentioned embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0098] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for analyzing the structural characteristics of urban heat islands based on three-dimensional simulation, characterized in that: include: Obtaining a three-dimensional city map of a target city, and selecting a plurality of locations as collection points according to the building distribution of the three-dimensional city map; Recording temperature information and additional information of the collection point at multiple historical time points, the additional information including meteorological data at the historical time point, shadow coverage time before the historical time point, surface coverage data and urban building data within a first distance range of the collection point; Determine a predicted time point, obtain city renewal data at the predicted time point, and update the three-dimensional city map in combination with the city renewal data to obtain a future city map; analyzing the temperature information and the additional information to establish a prediction model; Arranging a plurality of target points in the future city map, acquiring the additional information of the target points at the predicted time points, calculating the predicted temperatures of the target points based on the additional information and the prediction model, and importing the predicted temperatures into the future city map to obtain a city temperature distribution map; A critical threshold is set, and the target point whose predicted temperature is greater than the critical threshold is defined as a high-risk point, and the area within the second distance range of the high-risk point is defined as a high-risk area.

2. The method according to claim 1, characterized in that Selecting the collection point includes the following steps: Acquire the building type of each urban building in the three-dimensional urban map, where the building types include industrial type, medical type and commercial type, and set a corresponding judgment threshold for each type of urban building; The three-dimensional city map is divided into a plurality of first grid areas, and the first grid areas satisfying at least one preset condition are screened and defined as first areas, wherein the preset condition includes that the number of urban buildings of the same type in the first grid area exceeds the corresponding judgment threshold, or the average height of the urban buildings is greater than the first threshold, or the number of urban buildings is greater than the second threshold; The first area is divided again into multiple second grid areas, and the second grid areas that meet at least one of the preset conditions are screened and defined as second areas. This step is repeated until an Nth area is obtained by division, and in the Nth area, the number of the Nth areas corresponding to each of the preset conditions is greater than the third threshold and less than the fourth threshold, and the center point of the Nth area is used as the collection point.

3. The method according to claim 2, characterized in that Establishing the prediction model comprises the following steps: The meteorological data includes meteorological temperature, wind force level and wind direction; the urban building data includes the number of urban buildings of each building type within the first distance range and the average height of all urban buildings; the surface coverage data includes the green area and water coverage area within the first distance range; a BP neural network model is established, the additional information of each collection point is used as the input feature of the neural network, the corresponding temperature information is used as the output feature of the neural network model, and the trained neural network is used as the prediction model.

4. The method according to claim 2, characterized in that: Obtaining the building type of each city building in the three-dimensional city map comprises the following steps: Obtain an urban land use planning map, which includes the purpose of each plot of land, including industrial land, medical land and commercial land, and match the urban land use planning map and the three-dimensional city map based on longitude and latitude to determine the corresponding building type according to the purpose of the plot of land where the urban building is located.

5. The method according to claim 1, characterized in that Acquiring the three-dimensional city map comprises the following steps: Acquire an aerial image of the target city, extract the plane coordinates, height value, and brightness value of each pixel point in the aerial image, and sequentially connect the pixel points whose positions are adjacent and whose height value difference is within a fifth threshold value based on the plane coordinates, so as to draw a plurality of first contours in the aerial image; Calculating an average value of the height values ​​of each first contour and each pixel point therein, setting the first contour whose average value is greater than a sixth threshold as a second contour, the second contour being the contour of the top surface of the urban building, and generating a three-dimensional model of the urban building based on the height values ​​of each pixel point on the edge of the second contour; Acquire the brightness value of the second contour and each pixel point therein, cluster the brightness values, and divide the pixel points into multiple categories according to the clustering result; The area covered by the pixel points of the same category is defined as a sub-area, the average value of the height value in each sub-area is calculated, and the top surface model of the urban building is generated based on the average value and the position of each sub-area.

6. The method according to claim 5, characterized in that If the average value of the brightness value of any sub-region is lower than the seventh threshold, the top surface model is constructed according to the average value of the height values ​​of the pixel points in the corresponding sub-region.

7. The method according to claim 5, characterized in that Obtaining the shadow coverage time of the acquisition point comprises the following steps: The sunrise time of the day is obtained, and the time period between the sunrise time and the current time point is divided into multiple sub-time periods, and the shadow coverage of each urban building within the first distance range of the collection point during the sub-time period is simulated. If it is determined that the collection point is covered in the sub-time period according to the shadow coverage situation, the corresponding sub-time period is defined as a covered time period, and all the covered time periods between the sunrise time and the current time point are accumulated as the shadow coverage time of the collection point.

8. The method according to claim 7, characterized in that Obtaining the shadow coverage includes the following steps: A vertical section facing the sun is determined in an urban building, a section shape of the top surface model in the vertical section is obtained and defined as a cross-sectional shape, a shadow length and a projection direction of each urban building in the sub-time period are calculated, and the cross-sectional shape is projected onto the ground based on the shadow length and the projection direction to form the shadow coverage of the urban building.

9. A system for analyzing the structural characteristics of urban heat islands based on three-dimensional simulation, used to implement the method according to any one of claims 1 to 8, characterized in that: include: A first analysis module is used to obtain a three-dimensional city map of a target city and select multiple locations as collection points according to the building distribution of the three-dimensional city map; An information collection module, used to record the temperature information and additional information of the collection point at multiple historical time points, wherein the additional information includes meteorological data at the historical time point, shadow coverage time before the historical time point, surface coverage data within a first distance range of the collection point, and urban building data; A map updating module, which determines a predicted time point, obtains city renewal data at the predicted time point, and updates the three-dimensional city map in combination with the city renewal data to obtain a future city map; A prediction module, configured to analyze the temperature information and the additional information to establish a prediction model, arrange a plurality of target points in the future city map, obtain the additional information of the target points at the prediction time point, calculate the predicted temperature of the target points based on the additional information and the prediction model, import the predicted temperature into the future city map, and obtain a city temperature distribution map; The second analysis module is provided with a critical threshold value. The second analysis module defines the target point whose predicted temperature is greater than the critical threshold value as a high-risk point, and defines an area within a second distance range of the high-risk point as a high-risk area.

Images