A multidimensional evaluation system for urban sustainable development based on SDGs

Abstract

The present invention relates to the technical field of prediction and evaluation systems, and specifically to a multi-dimensional evaluation system for urban sustainable development based on SDGs. A regional division module, the regional division module divides the analysis area of ​​the city according to the administrative area of ​​the current evaluated city, and also includes a carbon emission analysis module, a green analysis module, and a comprehensive evaluation module. The comprehensive analysis module of the present invention predicts the total cumulative carbon emissions of the target city through mathematical model analysis, and combines the total cumulative carbon emissions of the target city with the city's greening level to predict its total carbon absorption. Then, by combining the two, a city carbon emission index for judging the comprehensive level of urban carbon emissions can be obtained. The higher the city carbon emission index, the more greenhouse gases the city emits, and the more unfavorable it is for the sustainable development of the city, so that it is convenient for the evaluator to evaluate the governance level and measures of carbon emissions in the urban area according to the parameter.

Description

Technical Field

[0001] The present invention relates to the technical field of prediction and evaluation systems, and particularly to a multi-dimensional evaluation system for urban sustainable development based on SDGs. Background Art

[0002] SDGs, namely Sustainable Development Goals, which are called sustainable development goals in Chinese, are a series of global development goals formulated by the United Nations in 2015. SDGs are 17 global development goals officially adopted by the United Nations at the Sustainable Development Summit on September 25, 2015, aiming to comprehensively address the development issues in the three dimensions of society, economy and environment from 2015 to 2030 and shift to the path of sustainable development. SDGs include 17 specific goals, covering multiple aspects such as poverty, hunger, health, education, gender equality, water and sanitation, energy, economic growth, infrastructure, inequality, cities and human settlements, consumption and production patterns, climate change, oceans and marine resources, terrestrial ecosystems, peace and justice, and global partnerships.

[0003] As a specific goal, SDSs provides a clear direction and action guide for global sustainable development. However, since it only provides a reference basis for each country in terms of macro concepts, in the actual implementation process, when it comes to each city in each country, it is necessary to comprehensively analyze various evaluation factors collected in combination with the development level of the city. One important evaluation factor lies in the assessment of the city's response to carbon emissions. The existing evaluation systems only unilaterally statistically predict the urban carbon emissions, without considering the improvement of the urban greening level, resulting in the obtained effect evaluation value often being lower than the actual effect evaluation value of the measures, causing inaccurate results. Summary of the Invention

[0004] In view of the above-mentioned drawbacks of the prior art, the present invention provides a multi-dimensional evaluation system for urban sustainable development based on SDGs, which can effectively solve the problem that it is difficult to reasonably evaluate the sustainable development level of the urban climate action in combination with the greening level in the prior art.

[0005] To achieve the above object, the present invention is realized through the following technical solutions:

[0006] The present invention provides a multi-dimensional evaluation system for urban sustainable development based on SDGs, at least including: a regional division module, which divides the analysis area of the city according to the administrative region of the city to be evaluated currently, and also includes: a carbon emission analysis module, a green analysis module, and a comprehensive evaluation module;

[0007] The carbon emission analysis module obtains the total carbon emission value and the population statistics quantity for each year within the analysis area, divides the total carbon emission value of each year by the population statistics quantity to obtain the per capita annual carbon emission value for each year, calculates the per capita carbon emission change value for each year, performs curve fitting based on the historical total carbon emission value and the per capita carbon emission change value, and predicts the cumulative carbon emission prediction value for this year;

[0008] The green analysis module, which includes a coverage measurement unit and a coverage analysis unit. The coverage measurement unit processes the remote sensing image within the urban analysis area to obtain a rectangular unit image, calculates the green coverage rate of the city through the unit image. The coverage analysis unit then counts the tree species within the rectangular unit image through a drone, analyzes the carbon absorption capacity of the green plants within the analysis area, and calculates the carbon absorption capacity prediction value used to judge the carbon absorption capacity within the analysis area;

[0009] The comprehensive analysis module regularly obtains the carbon absorption capacity prediction value Xs and draws a line graph of the carbon absorption capacity prediction value Xs changing with time, calculates the area value of the image enclosed by the line graph of Xs changing with time and the coordinate axis within the interval from the beginning of this year to the current date, obtains the current cumulative coefficient prediction value, and subtracts the cumulative coefficient prediction value from the carbon emission cumulative prediction value to obtain the urban carbon emission index within the analysis area.

[0010] Furthermore, the carbon emission analysis module obtains the total carbon emission value and the population statistics quantity for each year within the analysis area, divides the total carbon emission value of each year by the population statistics quantity to obtain the per capita annual carbon emission value for each year, and calculates the per capita carbon emission change value Bh n , and the calculation formula is where C 人均 n and C 人均 n-1 respectively represent the per capita annual carbon emission value of the current year and the per capita annual carbon emission value of the previous year within the analysis area, draw a line graph of the per capita carbon emission change value within the analysis area changing with the year and perform curve fitting on the line graph. The curve fitting process is as follows:

[0011] Perform unary linear regression analysis and unary nonlinear regression analysis on each point in the line graph to obtain the unary regression fitting straight line equation and the unary regression fitting curve equation, and calculate the determination coefficient R 2 value of each fitting equation.

[0012] Furthermore, the carbon emission analysis module predicts the total carbon emission value of the analysis area for this year. The prediction process is as follows:

[0013] Set a determination threshold. When any one or more of the determination coefficients of the two fitting equations are greater than the determination threshold, the fitting equation with the largest determination coefficient is taken as the fitting curve. According to the fitting curve, calculate the predicted value Bh' of the per capita carbon emission change in the analysis area this year, and extract the per capita annual carbon emission value of the analysis area last year and the current total population Num P , and substitute them into the formula for calculation to obtain the total predicted carbon emission value this year where represents the per capita annual carbon emission value last year;

[0014] When the determination coefficients of the two fitting equations are both less than the determination threshold, plot a line graph of the total carbon emission values over time for each year and perform curve fitting. Select the fitting equation with the largest R 2 value, and compare its R 2 value with the determination threshold. When its R 2 value is greater than or equal to the determination threshold, calculate the total predicted carbon emission value this year through this fitting equation; when its R 2 value is less than the determination threshold, take the total carbon emission value last year as the total predicted carbon emission value this year

[0015] Extract the total predicted carbon emission value this year and the current month, and substitute them into the formula for calculation to obtain the current cumulative predicted carbon emission value C now , where x is the current month.

[0016] Furthermore, the steps for the coverage measurement unit to analyze and calculate the green coverage rate are as follows:

[0017] S1: Obtain the color satellite remote sensing image within the analysis area and record it as the area image. Divide the area image into multiple unit images with a length-width ratio of 4:3 or 16:9 rectangular grid lines;

[0018] S2: Perform separate image analysis on each unit image to determine the green coverage rate within each unit image: The specific process is as follows:

[0019] S3: Gray-scale the unit image to obtain a gray-scale image. Each pixel point in the gray-scale image corresponds to a gray value. The gray value of the pixel point reflects the original pixel color of the pixel point. Through gray-scale processing, the unit image can be transformed into an image with a smaller toner span and easier to analyze. There is a preset green gray interval, and the green gray interval contains the gray values of the leaves of various green plants in nature. Mark the pixel points located within the green gray interval as green points, extract all the green points in the gray-scale image, and form an image containing only green points and record it as the green distribution map. Perform S4 on the green distribution map;

[0020] S4: Divide the green distribution map into multiple 2*2 pixel squares. The pixel squares are squares composed of 4 pixel points. If the number of green points in a pixel square is less than 3, then the green points in that pixel square are removed; otherwise, they are retained.

[0021] S5: Calculate the number of green points in the green distribution map after denoising divided by the total number of pixel points in the unit image to obtain the green coverage value in the unit image. Calculate the sum of the green coverage values of all unit images in the regional image to obtain the green coverage rate in the analysis area.

[0022] Furthermore, the coverage measurement unit analyzes the accuracy of the green coverage rate. The specific analysis process is as follows:

[0023] Statistically analyze the floor area and height of high-rise buildings with a building height greater than or equal to 20M in the analysis area. Calculate the product of the floor area and height of each high-rise building as its volume reference value. Calculate the average value of all high-rise building volume reference values and denote it as the volume average value. Obtain the acquisition time of the regional image. Obtain the solar altitude angle at that time in the analysis area according to the acquisition time. Substitute the volume average value V and the solar altitude angle H′ into the formula for calculation to obtain the deviation judgment value Dt of each unit image, where ρ 1 and ρ 2 are both preset weight factors, and judge the accuracy of the green coverage rate through the deviation judgment value.

[0024] Furthermore, the coverage analysis unit obtains all the green coverage rates and their deviation judgment values W i collected in the most recent month, obtains the acquisition time of each green coverage rate, calculates the time difference between the acquisition time and the current time in days and denotes it as T i , where i represents the serial number of the green coverage rate. Substitute the deviation judgment value W i and the time difference T i into the formula for calculation to obtain the selection reference value, where η 1 and η 2 are both preset weight factors, and select the green coverage rate with the largest selection reference value as the current coverage rate FG.

[0025] Furthermore, the coverage analysis unit presets a minimum coverage threshold, removes the unit images with green coverage values lower than the minimum coverage threshold among all unit images, denotes the remaining unit images as analysis images, conducts aerial photography on all analysis images through a drone, obtains the high-definition images corresponding to each analysis image in the corresponding area and denotes them as regional high-definition images, and adjusts the unit length in the rectangular grid lines and the drone shooting conditions so that the shapes of the green plant leaves and branches in the captured regional high-definition images can be clearly displayed;

[0026] Intelligently identify the tree leaves and branches in the green-covered area of the high-definition regional map through an image recognition algorithm to determine their species, mark different species of trees in the same high-definition regional map, and count the number of each type of tree. There is a preset set of tree species absorption capacities, which includes multiple tree species and each tree species corresponds to a carbon absorption capacity value. According to the formula Calculate the green absorption value GL corresponding to each high-definition regional map, where p and q respectively represent the tree species label and the total number of tree species in the region, and Zl p 、U p respectively represent the number of tree species with label p and its corresponding carbon absorption capacity value;

[0027] Calculate and analyze the sum of the green absorption values of all high-definition regional maps in the region, denoted as GL all and substitute it into the formula Xs = λ 1 FG + λ 2 GL all to calculate the current predicted carbon absorption capacity value Xs in the analysis region, where λ 1 、λ 2 are both preset weight factors.

[0028] The technical solution provided by the present invention has the following beneficial effects compared with the known prior art:

[0029] 1. The comprehensive analysis module of the present invention analyzes and predicts the total cumulative carbon emissions of the target city through a mathematical model, and uses the predicted value as the analysis value for predicting the total carbon emissions of the city and subsequent analysis. Compared with collecting actual carbon emission data, it does not require the collection and calculation of a huge amount of data, is more convenient and fast, and is easier to implement. Improving the measurement efficiency by sacrificing a certain degree of accuracy is more suitable for real-time calculation and time series analysis. In addition, by combining the total cumulative carbon emissions of the target city with the urban greening level to predict its total carbon absorption, and then by combining the two, a city carbon emission index can be obtained for judging the comprehensive carbon emission level of the city. The higher the city carbon emission index, the more greenhouse gases the city emits, and the more unfavorable it is to the sustainable development of the city. Thus, it is convenient for evaluators to evaluate the governance level and measures of carbon emissions in the city area according to this parameter.

[0030] 2. When analyzing the collected regional images, the coverage measurement unit of the present invention can analyze and calculate a deviation judgment value for judging the accuracy of the green coverage rate according to the current lighting conditions and the high-rise building data in the region. Through this value, the accuracy of the green coverage rate can be judged. Compared with directly collecting and using the green coverage rate in the prior art, the coverage analysis unit can screen out a more accurate green coverage rate for subsequent analysis by combining the deviation judgment value and the timeliness of the image. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the accompanying drawings required for the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can be obtained based on these drawings.

[0032] Figure 1 It is the overall module block diagram of the present invention. Specific embodiments

[0033] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts fall within the scope of protection of the present invention.

[0034] The following further describes the present invention with reference to embodiments.

[0035] Refer to Figure 1 , a multi-dimensional assessment system for urban sustainable development based on SDGs, at least including: a regional division module, which divides the analysis area of the city according to the administrative region of the city to be evaluated currently, and also includes a carbon emission analysis module, a green analysis module, and a comprehensive assessment module.

[0036] The carbon emission analysis module obtains the total carbon emission value and the population statistics quantity for each year in the analysis area, divides the total carbon emission value of each year by the population statistics quantity to obtain the per capita annual carbon emission value for each year, and calculates the per capita carbon emission change value Bh n The calculation formula is where C 人均 n and C 人均 n-1 respectively represent the per capita annual carbon emission value in the current year and the per capita annual carbon emission value in the previous year in the analysis area, draw a line graph of the per capita carbon emission change value in the analysis area changing with the year and perform curve fitting on the line graph. The curve fitting process is as follows:

[0037] Perform unary linear regression analysis and unary non-linear regression analysis on each point in the line graph to obtain the unary regression fitting straight line equation and the unary regression fitting curve equation, and calculate the determination coefficient of each fitting equation, that is, the R 2 value;

[0038] Further, predict the total carbon emission value of the analysis area this year. The prediction process is as follows:

[0039] Set a determination threshold. When any one or more of the determination coefficients of the two fitting equations are greater than the determination threshold (indicating the existence of a reliable prediction equation), take the fitting equation with the largest determination coefficient as the fitting curve of the line graph. Calculate the predicted value Bh' of the per capita carbon emission change in the analysis area this year according to the fitting curve of the line graph. Extract the per capita annual carbon emission value of the analysis area last year and the current total population Num P , and substitute it into the formula for calculation to obtain the total predicted carbon emission value this year where represents the per capita annual carbon emission value last year;

[0040] When the determination coefficients of the two fitting equations are both less than the determination threshold, plot the line graph of the total carbon emission value changing with time for each year and perform curve fitting. Select the fitting equation with the largest R 2 value, and compare its R 2 value with the determination threshold. When its R 2 value is greater than or equal to the determination threshold (indicating that this fitting equation is a reliable prediction equation), calculate the total predicted carbon emission value this year through this fitting equation; when its R 2 value is less than the determination threshold, take the total carbon emission value last year as the total predicted carbon emission value this year (In the case where prediction cannot be made through curve fitting, select the total carbon emission value of the most recent year as the prediction value this year, which can minimize the prediction error);

[0041] Furthermore, extract the total predicted carbon emission value this year and the current month, and substitute them into the formula for calculation to obtain the current cumulative predicted carbon emission value C now , where x is the current month, used to predict the total carbon emission value within the analysis area so far this year.

[0042] It should be noted that the total carbon emission value is obtained by the statistical agency through calculating the carbon emissions within three scopes to comprehensively obtain the total carbon emission value within the region. Scope 1 within the three scopes refers to all direct emissions within the urban jurisdiction area, mainly including carbon emissions generated by transportation, construction, industrial production processes, agriculture, forestry and land use change, and waste treatment activities; Scope 2 refers to the indirect emissions related to energy occurring outside the urban jurisdiction area, mainly including carbon emissions generated by the purchased electricity, heating, and / or refrigeration, etc. to meet urban consumption; Scope 3 refers to other indirect emissions caused by urban internal activities, generated outside the jurisdiction area but not included in Scope 2, including carbon emissions in the production, transportation, use, and waste treatment links of all items purchased by the town from outside the jurisdiction area. The carbon emission values within these scopes can all be calculated according to the urban data released by the National Bureau of Statistics based on existing technologies.

[0043] The green analysis module, which includes a coverage measurement unit and a coverage analysis unit. The coverage measurement unit processes the remote sensing image within the urban analysis area to obtain a rectangular unit image, calculates the green coverage rate of the city through the unit image. The coverage analysis unit then counts the tree species within the rectangular unit image through a drone, analyzes the carbon absorption capacity of the green plants within the analysis area, and calculates the current carbon absorption capacity prediction value within the analysis area to judge the carbon absorption capacity within the analysis area.

[0044] The coverage measurement unit obtains the color satellite remote sensing image within the analysis area and records it as the area image. The area image is divided into multiple unit images with rectangular grid lines having an aspect ratio of 4:3 or 16:9. Each rectangle within the rectangular grid lines is denoted as a unit rectangle. Each unit image at the center position of the analysis area is rectangular and has the same shape and size, while the area of each unit image at the edge position of the analysis area is less than or equal to the area of the unit rectangle. Furthermore, the divided unit images can all be completely displayed by a standard picture frame with a display ratio of 4:3 or 16:9, so that each unit image can be analyzed separately. Compared with the analysis of the complete area image, dividing it into unit images and then analyzing can ensure that the analysis results are more accurate and reliable and contribute to accurate positioning.

[0045] Furthermore, perform separate image analysis on each unit image to determine the green coverage rate within each unit image: The specific process is as follows:

[0046] The unit image is grayscale processed to obtain a grayscale image. Each pixel point in the grayscale image corresponds to a grayscale value, and the grayscale value of the pixel point reflects the original pixel color of the pixel point. Through grayscale processing, the unit image can be converted into an image with a smaller toner span and easier to analyze. There is a preset green grayscale interval, which contains the grayscale values of the leaf colors of various green plants in nature. Therefore, when green plants are photographed by a remote sensing satellite and are within the grayscale image, the grayscale values of the pixel points corresponding to their leaves are within the green grayscale interval. By screening the pixel points whose grayscale values are within the green grayscale interval, the pixel points corresponding to the leaf colors can be screened out. The pixel points located within the green grayscale interval are denoted as green points. All the green points in the grayscale image are extracted, and the image composed only of green points is denoted as the green distribution map. The green distribution map is denoised to remove the discrete green points in the green distribution map. The denoising process is as follows:

[0047] The green distribution map is divided into multiple 2*2 pixel squares. The pixel square is a square composed of 4 pixel points. If the number of green points in the pixel square is less than 3, the green points in the pixel square are removed, otherwise they are retained. Through denoising, it can not only reduce the influence of the green noise generated in the image acquisition process on the calculation of the green coverage rate in the unit image, but also reduce the influence of the green proportion of non-green plants on the calculation of the green coverage rate;

[0048] Calculate the number of green points in the denoised green distribution map divided by the total number of pixel points in the unit image to obtain the green coverage value in the unit image. Calculate the sum of the green coverage values of all unit images in the regional image to obtain the green coverage rate in the analysis area.

[0049] Furthermore, the accuracy of the green coverage rate is analyzed. The analysis process is as follows:

[0050] Statistically analyze the floor area and height of high-rise buildings with a building height greater than or equal to 20M in the analysis area. Calculate the product of the floor area and height of each high-rise building as its volume reference value. Calculate the average value of all high-rise building volume reference values and denote it as the volume average value,

[0051] It should be noted that the volume reference value can reflect the volume size of each high-rise building, and the volume average value is used to reflect the average volume size of high-rise buildings in the analysis area. The larger the volume average value, the larger the average volume of high-rise buildings in the analysis area, and the larger the shadow area generated when irradiated by sunlight. Generally speaking, the direct factor affecting the shadow area generated by sunlight irradiation is the area of the building body facing the sun. Based on the principle that buildings are built facing the sun during construction, the building volume can also indirectly reflect the size of the shadow area formed under sunlight irradiation.

[0052] Obtain the acquisition time of the regional image, obtain the solar altitude angle at that time in the analysis area according to the acquisition time, and use the volume average value and the solar altitude angle H′ and substitute them into the formula for calculation to obtain the deviation judgment value Dt of each unit image, where ρ 1 and ρ 2 are both preset weight factors. Judge the accuracy of the green coverage rate through the deviation judgment value. As the solar altitude angle decreases and the volume average value increases, the shadow area generated by the sun shining on the building will increase, so that part of the green area will be covered in the regional image, resulting in a decrease in the accuracy of the green coverage rate calculation. On the contrary, the decrease in the solar altitude angle and the decrease in the volume average value will improve the accuracy of the green coverage rate calculation. Therefore, the deviation judgment value can help judge the accuracy of the green coverage rate.

[0053] The coverage analysis unit obtains all the green coverage rates and their deviation judgment values W collected within the last month i , obtains the acquisition time of each green coverage rate, and calculates the time difference between the acquisition time and the current time in days and records it as T i , where i represents the serial number of the green coverage rate. Substitute the deviation judgment value W i and the time difference T i into the formula for calculation to obtain the selection reference value, where η 1 and η 2 are both preset weight factors. Select the green coverage rate with the largest selection reference value as the current coverage rate FG. By combining time to select a suitable green coverage rate as the current coverage rate, the influence of the timeliness on the analysis result of the image acquisition data can be effectively reduced;

[0054] A minimum coverage threshold is preset. Exclude the unit images with green coverage values lower than the minimum coverage threshold among all unit images, and record the remaining unit images as analysis images. Conduct aerial photography on all analysis images by a drone to obtain high-definition images corresponding to each analysis image area and record them as regional high-definition images. By excluding small-scale green coverage, the number of pictures that the drone needs to take can be reduced to improve the image acquisition efficiency. Adjust the unit length in the rectangular grid line and the drone shooting conditions so that the shapes of the green plant leaves and branches in the captured regional high-definition images can be clearly displayed. The specific operation is to reduce the ratio of the unit length in the rectangular grid line to the actual length or increase the altitude and photo resolution of the drone aerial photography. Through the above operations, the captured pictures can display the image features of the same area with more pixel points, so that the objects in the pictures are clearer and easier to identify and recognize;

[0055] The intelligent recognition is carried out on the tree leaves and branches in the green coverage area of the regional high-definition map through an image recognition algorithm to determine their types. Different types of trees in the same regional high-definition map are marked and the quantity of each type of tree is counted. There is a preset tree species absorption capacity set, which contains multiple tree species and each tree species corresponds to a carbon absorption capacity value. According to the formula Calculate the green absorption value GL corresponding to each regional high-definition map, where p and q respectively represent the tree species label number and the total number of tree species in the region, and Zl p , U p respectively represent the quantity of the tree species with label p and its corresponding carbon absorption capacity value (the amount of carbon dioxide that can be absorbed by each tree species per day under the current regional environment). Calculate and analyze the sum of the green absorption values of all regional high-definition maps in the region, denoted as GL all , substitute it into the formula Xs = λ 1 FG + λ 2 GL all to calculate the current carbon absorption capacity prediction value Xs in the analysis region, where λ 1 , λ 2 are both preset weight factors.

[0056] It should be noted that identifying tree species based on leaves and branches through image recognition technology is an existing technology, and will not be elaborated here too much.

[0057] The comprehensive analysis module regularly obtains the carbon absorption capacity prediction value Xs and draws a line graph of the change of the carbon absorption capacity prediction value Xs over time (the unit of the time axis is days). Calculate the area value of the image enclosed by the line graph of the change of Xs over time from the beginning of this year to the current date interval to obtain the current cumulative coefficient prediction value. Subtract the cumulative coefficient prediction value from the carbon emission cumulative prediction value to obtain the urban carbon emission index. The higher the urban carbon emission index, the more greenhouse gases are emitted by the city, and the more unfavorable it is to the sustainable development of the city. Thus, it is convenient for evaluators to evaluate the governance level and measures of carbon emissions in the urban area according to this parameter.

[0058] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements will not make the essence of the corresponding technical solutions deviate from the protection scope of the technical solutions of the embodiments of the present invention.

Claims

1. A multi-dimensional evaluation system for urban sustainable development based on SDGs, comprising a regional division module, which divides the analysis area of ​​the city according to the administrative area of ​​the current evaluated city, characterized in that: Also includes: Carbon emission analysis module, green analysis module, and comprehensive assessment module; The carbon emission analysis module obtains the total carbon emission value and population statistics of each year in the analysis area, divides the total carbon emission value of each year by the population statistics to obtain the annual per capita carbon emission value, calculates the annual per capita carbon emission change value, and performs curve fitting based on the historical total carbon emission value and per capita carbon emission change value to predict the cumulative carbon emission forecast value for this year; Green analysis module: The green analysis module includes a coverage measurement unit and a coverage analysis unit. The coverage measurement unit processes the remote sensing image in the urban analysis area to obtain a rectangular unit image, and calculates the green coverage rate of the city through the unit image. The coverage analysis unit counts the tree species in the rectangular unit image through a drone, analyzes the carbon absorption capacity of green plants in the area, and calculates the carbon absorption capacity prediction value used to judge the carbon absorption capacity in the analysis area; The steps of analyzing and calculating the green coverage rate by the coverage measurement unit are as follows: S1: Obtain a color satellite remote sensing image within the analysis area as a regional image, and divide the regional image into a plurality of unit images using rectangular grid lines with an aspect ratio of 4:3 or 16:9; S2: Perform a separate image analysis on each unit image to determine the green coverage within each unit image. The specific process is as follows: S3: grayscale the unit image to obtain a grayscale image. Each pixel in the grayscale image corresponds to a grayscale value. The grayscale value of the pixel reflects the original pixel color of the pixel. The unit image can be converted into an image with a small color span and easy to analyze through grayscale processing. A green grayscale interval is preset. The green grayscale interval contains the grayscale values ​​of the leaves of various green plants in nature. The pixels in the green grayscale interval are recorded as green points. All green points in the grayscale image are extracted to form an image containing only green points, which is recorded as a green distribution map. The green distribution map is subjected to S4; S4: Divide the green distribution map into multiple 2*2 pixel blocks, where the pixel blocks are squares consisting of 4 pixels. If the number of green points in a pixel block is less than 3, the green points in the pixel block are removed, otherwise they are retained. S5: Calculate the number of green points in the green distribution map after noise removal and divide it by the total number of pixels in the unit image to obtain the green coverage value in the unit image, calculate the sum of the green coverage values ​​of all unit images in the regional image, and obtain the green coverage rate in the analysis area; The coverage measurement unit analyzes the accuracy of the green coverage rate, and the analysis process is as follows: The floor area and height of the high-rise buildings with a height of 20 meters or more in the statistical analysis area are calculated, and the product of the floor area and height of each high-rise building is calculated as its volume reference value. The average value of the volume reference values ​​of all high-rise buildings is calculated and recorded as the volume average value. The acquisition time of the regional image is obtained, and the solar altitude angle of the analysis area at that time is obtained according to the acquisition time. The deviation judgment value of each unit image is calculated according to the volume average value and the solar altitude angle, and the accuracy of the green coverage rate is judged by the deviation judgment value; The coverage analysis unit obtains all green coverage rates and their deviation judgment values ​​collected in the last month, obtains the collection time of each green coverage rate, calculates the time difference between the collection time and the current time in days, calculates the selection reference value according to the deviation judgment value and the time difference, and selects the green coverage rate with the largest selection reference value as the current coverage rate FG; The coverage analysis unit is preset with a minimum coverage threshold, and unit images with green coverage values ​​lower than the minimum coverage threshold are eliminated from all unit images, and the retained unit images are recorded as analysis images. All analysis images are photographed by drones, and high-definition images of the area corresponding to each analysis image are obtained and recorded as regional high-definition images. The unit lengths in the rectangular grid lines and the drone shooting conditions are adjusted so that the shapes of leaves and branches of green plants in the captured regional high-definition images can be clearly displayed; The image recognition algorithm is used to intelligently identify the leaves and branches of trees in the green coverage area in the regional high-definition image to determine their types. Different types of trees in the same regional high-definition image are marked and the number of each type of tree is counted. A tree species absorption capacity set is preset. The tree species absorption capacity set contains multiple tree species and each tree species corresponds to a carbon absorption capacity value. According to the formula The green absorption value GL corresponding to the high-definition image of each area is calculated, where p and q represent the tree species number and the total number of tree species in the area, respectively. They represent the number of tree species labeled p and their corresponding carbon absorption capacity values; The sum of the green absorption values ​​of all high-definition images in the analysis area is calculated and recorded as , substitute into the formula The current carbon absorption capacity prediction value Xs in the analysis area is calculated, where All are preset weight factors; The comprehensive analysis module regularly obtains the carbon absorption capacity prediction value Xs and draws a line graph of the carbon absorption capacity prediction value Xs changing with time, calculates the image area value enclosed by the line of Xs changing with time and the coordinate axis from the beginning of this year to the current date, obtains the current cumulative coefficient prediction value, and subtracts the cumulative coefficient prediction value from the carbon emission cumulative prediction value to obtain the urban carbon emission index in the analysis area.

2. The multi-dimensional evaluation system for urban sustainable development based on SDGs according to claim 1 is characterized in that: The carbon emission analysis module obtains the total carbon emission value and population statistics of each year in the analysis area, divides the total carbon emission value of each year by the population statistics to obtain the annual per capita carbon emission value of each year, and calculates the annual per capita carbon emission change value , the calculation formula is ,in They represent the annual per capita carbon emission value of the current year and the annual per capita carbon emission value of the previous year in the analysis area respectively. A line graph of the change of per capita carbon emission value in the analysis area with the year is drawn and a curve fitting is performed on the line graph. The curve fitting process is as follows: Perform univariate linear regression analysis and univariate nonlinear regression analysis on each point in the line graph to obtain the univariate regression fitting straight line equation and the univariate regression fitting curve equation, and calculate the determination coefficient of each fitting equation respectively value.

3. The multi-dimensional evaluation system for urban sustainable development based on SDGs according to claim 2 is characterized in that: The carbon emission analysis module predicts the total carbon emission value of the analysis area this year. The prediction process is as follows: A determination threshold is set. When any one or more of the determination coefficients of the two fitting equations is greater than the determination threshold, the fitting equation with the largest determination coefficient is used as the fitting curve. The predicted per capita carbon emission change value of the analysis area this year is calculated based on the fitting curve. , extract the annual per capita carbon emissions of the analysis area last year and the current total population , substitute into the formula Calculate the total carbon emissions for this year ,in It represents the annual per capita carbon emission value last year; When the coefficients of determination of the two fitting equations are both less than the determination threshold, a line graph of the total carbon emissions of each year over time is drawn and a curve fitting is performed. The fitting equation with the largest value is The value is compared with the determined threshold. When the value is greater than or equal to the determined threshold, the estimated total carbon emissions for this year are calculated through the fitting equation; When the value is less than the determined threshold, the total carbon emission value of last year is used as the estimated total carbon emission value of this year. ; Extract this year's estimated total carbon emissions and the current month, substitute into the formula Calculate the current carbon emission cumulative forecast value , where x is the current month.

Images