Urban heat island effect decision-making method and system based on satellite data and artificial intelligence

Abstract

The invention discloses an urban heat island effect decision-making method based on satellite data and artificial intelligence, and relates to the technical field of artificial intelligence, and the method comprises the steps: obtaining satellite remote sensing data and urban basic data of a target urban region, the urban basic data comprising population density distribution, historical intervention effects, infrastructure information and urban function partition data; preprocessing the satellite remote sensing data to generate a standardized surface temperature distribution image, and based on the surface temperature distribution image, identifying an urban heat island area and dividing the intensity grade of the urban heat island area; the method has the advantages that satellite remote sensing data and urban multi-dimensional basic data are fused, dynamic scoring and optimization algorithms are combined, the heat island area is accurately recognized, the optimal intervention strategy is generated, the problems that a traditional method is insufficient in data utilization and poor in decision-making scientificity are solved, and the method has the advantages of improving treatment accuracy and resource utilization efficiency.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a method and system for making decisions about urban heat island effects based on satellite data and artificial intelligence. Background Art

[0002] With the rapid advancement of global urbanization, the scale of cities continues to expand, and urban populations are becoming increasingly dense. The urban heat island effect, a typical urban environmental problem, has become increasingly prominent. This effect causes temperatures in urban areas to be significantly higher than those in surrounding rural areas. This not only increases the risk of heat disasters in cities and negatively impacts residents' living comfort, but also leads to a significant increase in urban energy consumption. For example, the frequency and duration of use of cooling equipment such as air conditioners increases, leading to energy waste and increased carbon emissions. Furthermore, high temperatures can exacerbate air pollution, posing a potential threat to residents' health, particularly their respiratory and cardiovascular systems.

[0003] Traditional interventions to combat the urban heat island effect are mostly based on local observational data or empirical judgment, lacking comprehensiveness and systematicity. These methods often struggle to accurately identify the areas of urban heat island severity and accurately predict future trends in heat island intensity, making it difficult to develop forward-looking and targeted intervention strategies.

[0004] On the other hand, cities are complex systems, with different functional zones, uneven population density, and varying infrastructure conditions across different regions. Previous urban heat island intervention measures have also had varying results. However, traditional approaches often fail to fully consider these multi-dimensional key factors when making decisions, resulting in unsatisfactory intervention results and even irrational resource allocation, with some areas receiving excessive intervention and others receiving insufficient intervention.

[0005] Therefore, a decision-making method and system for urban heat island effect based on satellite data and artificial intelligence is proposed. Summary of the Invention

[0006] In view of the above-mentioned existing technical conditions, this application is proposed. The embodiments of this application provide an urban heat island effect decision-making method based on satellite data and artificial intelligence, which can improve the accuracy of urban heat island effect management and enhance the scientific nature of intervention strategies.

[0007] According to one aspect of the present application, a method for making decisions on urban heat island effects based on satellite data and artificial intelligence is provided, including: obtaining satellite remote sensing data and urban basic data of the target urban area, the urban basic data including population density distribution, historical intervention effects, infrastructure information and urban functional zoning data; preprocessing the satellite remote sensing data to generate a standardized surface temperature distribution image; based on the surface temperature distribution image, identifying urban heat island areas and dividing their intensity levels; predicting the future intensity change trend of the heat island area through a prediction model; constructing an intervention response model to simulate the temperature impact of different intervention measures on the heat island area; constructing a dynamic scoring function to calculate the governance priority of each heat island area based on the heat island intensity level, change trend, population density distribution, infrastructure information and historical intervention effects of the heat island area; generating an optimal intervention scheduling strategy based on the simulation results of the intervention response model, the governance priority and the preset governance resource constraints.

[0008] According to another aspect of the present application, an urban heat island effect decision-making system based on satellite data and artificial intelligence is provided, including: a data acquisition module for acquiring satellite remote sensing data and urban basic data of a target urban area, the urban basic data including population density distribution, historical intervention effects, infrastructure information and urban functional zoning data; a data processing module for preprocessing the satellite remote sensing data to generate a standardized surface temperature distribution image; a heat island area classification module for identifying urban heat island areas and dividing their intensity levels based on the surface temperature distribution image; a heat island trend prediction module for predicting the future intensity change trend of the heat island area through a time series prediction model based on spatial adjacency; an intervention measure simulation module for constructing an intervention response model to simulate the temperature impact of different intervention measures on the heat island area; a governance priority ranking module for constructing a dynamic scoring function to calculate the governance priority of each heat island area based on the heat island intensity level, change trend, population density distribution, infrastructure information and historical intervention effects of the heat island area; and an intervention scheduling strategy output module for generating an optimal intervention scheduling strategy based on the simulation results of the intervention response model, the governance priority and the preset governance resource constraints.

[0009] According to another aspect of the present application, an electronic device is provided, comprising a memory and a processor, wherein the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions, which implement the steps of the above-described method when executed by the processor.

[0010] According to another aspect of the present application, a computer storage medium is provided, on which computer executable instructions are stored. When the computer executable instructions are executed by a processor, the steps of the above method are implemented.

[0011] Compared with the existing technology, the urban heat island effect decision-making method based on satellite data and artificial intelligence according to the embodiment of the present application can accurately identify heat island areas and generate optimal intervention strategies by integrating satellite remote sensing data with multi-dimensional basic urban data, combining dynamic scoring and optimization algorithms, solving the problems of insufficient data utilization and poor scientific decision-making in traditional methods, and has the advantages of improving governance accuracy and resource utilization efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] The above and other purposes, features, and advantages of the present application will become more apparent through a more detailed description of the embodiments of the present application in conjunction with the accompanying drawings. The accompanying drawings are intended to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the drawings, the same reference numerals generally represent the same components or steps.

[0013] Figure 1 The figure is a flow chart of the urban heat island effect decision-making method based on satellite data and artificial intelligence of the present invention.

[0014] Figure 2 This is a flow chart of the urban heat island effect decision-making method based on satellite data and artificial intelligence in the present invention, which identifies heat island areas and divides them into intensity levels.

[0015] Figure 3 This is a flow chart of the output optimal scheduling strategy of the urban heat island effect decision-making method based on satellite data and artificial intelligence in the present invention.

[0016] Figure 4 This is a block diagram of the urban heat island effect decision-making system based on satellite data and artificial intelligence of the present invention.

[0017] Figure 5 The present invention is a block diagram of an electronic device. DETAILED DESCRIPTION

[0018] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described herein.

[0019] Application Overview Traditional methods for addressing the urban heat island effect suffer from a single data source, resulting in inaccurate identification of heat island areas and the inability to implement multi-dimensional dynamic priority assessments. The lack of a fusion processing mechanism for satellite remote sensing data and urban infrastructure data leads to a low degree of standardization of surface temperature distribution images, resulting in discrepancies in the classification of heat island intensity levels across different spatial units. Historical intervention effects are disconnected from real-time heat island trend forecasts, and intervention response models fail to establish a dynamic association between semantic attributes and governance measures, resulting in a lack of spatiotemporal adaptability in resource scheduling strategies.

[0020] For example, when a megacity used traditional methods, thermal infrared satellite data was not corrected for radiation and atmosphere, resulting in large errors in surface temperature inversion, which prevented the accurate identification of heat island areas in some high-density residential areas. Urban functional zoning data and infrastructure information were stored as static layers, without spatial overlay analysis with real-time population density distribution. Intervention simulations considered only a single temperature drop parameter, ignoring the unique thermal environment characteristics of important transportation hubs. The governance priority calculation used a fixed weight coefficient, which made it impossible to automatically adjust the scoring function structure based on the growth rate of the heat island trend, resulting in pre-typhoon emergency resource scheduling deviating from the actual risk distribution.

[0021] If these issues are not addressed, urban heat island governance will remain in a passive response mode for a long time, with temperature regulation in key areas lagging behind the nonlinear characteristics of climate change. The synergistic deterioration of the thermal environment between established infrastructure districts and emerging commercial areas will intensify, the phase difference between peak energy consumption and urban heat island intensity fluctuations will widen, and the spatial and temporal imbalances in cooling equipment loads will continue to accumulate. The conflicting allocation of cross-sector governance resources will cause intervention measures to miss their optimal implementation window, causing heat stress indices around key public buildings to exceed safety thresholds and accelerating the degradation of urban ecosystem services.

[0022] In the face of the above problems, this application first addresses the defects of single data source and insufficient fusion processing, proposes to integrate satellite remote sensing data with multi-dimensional urban basic data, and improve the accuracy of heat island area identification by establishing a standardized processing process; to address the problem of disconnection between historical intervention and prediction, a priority evaluation mechanism based on dynamic trend prediction is designed, and resource allocation is optimized in combination with real-time population density and infrastructure distribution; to solve the lack of association between semantic attributes and governance measures, a semantic attribute mapping table is constructed to dynamically screen and adapt intervention measures; finally, by introducing a multi-constraint optimization model, the spatiotemporal adaptation of governance resources and heat island area characteristics is achieved.

[0023] Exemplary Methods The urban heat island effect decision-making method based on satellite data and artificial intelligence includes: obtaining satellite remote sensing data and urban basic data of the target urban area, the urban basic data including population density distribution, historical intervention effects, infrastructure information and urban functional zoning data; preprocessing the satellite remote sensing data to generate a standardized surface temperature distribution image; based on the surface temperature distribution image, identifying the urban heat island area and dividing its intensity level; predicting the future intensity change trend of the heat island area through a prediction model; constructing an intervention response model to simulate the temperature impact of different intervention measures on the heat island area; constructing a dynamic scoring function to calculate the governance priority of each heat island area according to the heat island intensity level, change trend, population density distribution, infrastructure information and historical intervention effects of the heat island area; generating the optimal intervention scheduling strategy based on the simulation results of the intervention response model, governance priority and preset governance resource constraints.

[0024] Among them, satellite remote sensing data refers to the surface electromagnetic wave information obtained by satellite sensors. Specifically, it can be achieved by using thermal infrared band data from satellites such as Landsat and MODIS, providing large-scale and periodically updated temperature monitoring capabilities for urban heat island effect monitoring.

[0025] Preprocessing refers to the operation of calibrating and converting the original satellite data. Specifically, radiation correction can be used to eliminate sensor errors, atmospheric correction can be used to remove aerosol interference, and geographic coordinate reprojection can be used to unify the spatial reference to generate a standardized surface temperature distribution image to eliminate data bias.

[0026] Among them, the division of heat island regional intensity levels refers to the classification of the spatial range of heat islands according to the degree of temperature anomaly. Specifically, the background temperature comparison method can be used to calculate the difference between the grid temperature and the natural surface temperature, and the threshold range can be used to divide the levels into mild, moderate and severe levels to achieve quantitative representation of the spatial distribution of heat islands.

[0027] Among them, the prediction model refers to a machine learning model based on time series data analysis. Specifically, it can use an LSTM neural network combined with a spatial adjacency weight matrix. By training the model through historical heat island intensity data and intervention records, it can predict future heat island development trends to support forward-looking decision-making.

[0028] Among them, the intervention response model refers to a calculation model that simulates the thermal effects of artificial measures. Specifically, the multivariate regression analysis method can be used to establish a quantitative relationship between the type of intervention measures, construction parameters and temperature drop, and the influence range can be calculated through the gridded heat conduction equation to provide a basis for the selection of measures.

[0029] Among them, the dynamic scoring function refers to a multi-factor weighted evaluation algorithm. Specifically, the weight coefficients of heat island intensity, population density, and infrastructure density can be determined through the hierarchical analysis method, and a linear weighted scoring system can be constructed in combination with the historical intervention effect deviation rate to achieve scientific sorting of governance priorities.

[0030] Among them, the optimal intervention scheduling strategy refers to the decision-making optimization plan under resource constraints. Specifically, an integer programming model can be used to maximize the total thermal risk reduction as the objective function, and combined with budget constraints and construction time constraints to solve the optimal combination of measures and execution sequence to improve resource utilization efficiency.

[0031] The core innovation of this application lies in establishing a decision-making framework for multi-dimensional data fusion. Through the collaborative analysis of satellite remote sensing and urban basic data, a complete technical chain including heat island identification, trend prediction, and measure simulation is constructed, and a dynamic scoring mechanism and resource optimization model are introduced to realize a closed-loop decision-making process from heat island monitoring to precise intervention.

[0032] The working process and principle of this application are to obtain satellite remote sensing data and urban basic data of the target urban area. Satellite remote sensing data contains thermal infrared band information; urban basic data includes population density distribution, historical intervention effects, infrastructure information and urban functional zoning data. The satellite remote sensing data is preprocessed, including radiation correction, atmospheric correction, geographic coordinate system unification and spatial grid division, and a standardized surface temperature distribution image is generated through a temperature inversion algorithm. Based on the surface temperature distribution image, combined with urban functional zoning data, the reference background temperature is calculated, the heat island area is identified and the intensity level is divided; a prediction model is used to predict the future intensity change trend of the heat island area; an intervention response model is constructed to simulate the temperature impact of different intervention measures on the heat island area; based on the multi-dimensional information of the heat island area, a dynamic scoring function is constructed to calculate the governance priority. Finally, based on the simulation results of the intervention response model, the governance priority and the preset governance resource constraints, the optimal intervention scheduling strategy is generated. Each step works together to achieve accurate identification, prediction and targeted intervention decisions of the urban heat island effect.

[0033] In some of the above-mentioned solutions of this application, the preprocessing process requires the generation of standardized surface temperature distribution images, but the existing methods are insufficient in data accuracy and spatial consistency, resulting in reduced reliability of subsequent heat island identification and trend prediction.

[0034] This application further proposes that the preprocessing includes the following steps: performing radiation correction and atmospheric correction on the thermal infrared band in the satellite remote sensing data to obtain surface radiation brightness data; reprojecting the radiation brightness data to a unified geographic coordinate system and dividing the urban space units according to a preset grid; based on the radiation brightness data, calculating the surface temperature value of each grid through a temperature inversion algorithm to generate a standardized surface temperature distribution image.

[0035] Among them, radiation correction and atmospheric correction are parameter calibration through ground measured data to eliminate the influence of sensor response deviation and atmospheric absorption and scattering effects.

[0036] Among them, the reprojection operation uses bilinear interpolation to unify remote sensing images of different resolutions into the UTM projection coordinate system, and the grid is divided into regular sections of 500 meters × 500 meters.

[0037] Among them, the temperature inversion algorithm adopts the split window algorithm, which uses the radiation difference between two adjacent thermal infrared bands to invert the surface temperature.

[0038] Specifically, the thermal infrared band data is first converted into radiance values ​​using a radiation calibration formula, where the calibration coefficient is derived from the calibration parameter file of the satellite payload. The atmospheric correction module calculates the atmospheric upwelling and downwelling radiation components using the MODTRAN atmospheric transfer model, removing the atmospheric influence from the apparent radiation to obtain the true surface radiation. During the reprojection process, the coordinate transformation matrix is ​​used to perform geometric correction on the original image to ensure spatial alignment of remote sensing data at different phases. The grid division uses a vector rasterization tool to generate a regular grid covering the study area, with each grid serving as an independent calculation unit. When performing temperature inversion, the surface temperature value is calculated using the split window equation using the radiance values ​​of bands 10 and 11 for Landsat8 data, where the surface emissivity is estimated using the NDVI threshold method. The resulting temperature distribution image is saved in GeoTIFF format, containing the geographic coordinates and temperature numerical attributes of each grid cell.

[0039] As a preferred embodiment, the solution of this application is specifically implemented as follows: The preprocessing process includes the following steps: First, radiometric and atmospheric corrections are performed on the thermal infrared bands in the satellite remote sensing data. Radiometric correction converts the raw digital values ​​into radiance values ​​using sensor calibration parameters. Atmospheric correction uses the MODTRAN model, incorporating atmospheric parameters such as water vapor content and aerosol optical depth, to eliminate the effects of atmospheric absorption and scattering, ultimately obtaining surface radiance data.

[0040] Secondly, the radiance data were reprojected into a unified geographic coordinate system, the UTM projection was selected as the unified coordinate system, and the urban space units were divided into 500 m × 500 m grids.

[0041] Finally, temperature inversion is performed using a single-channel algorithm based on the radiance data. Specifically, the Planck function is used to establish a model for the relationship between radiance and surface temperature, and the surface emissivity parameter is introduced for correction. The surface temperature value is calculated for each 500m x 500m grid, generating a standardized surface temperature distribution image.

[0042] Through the above technical solution, this application achieves systematic preprocessing of satellite remote sensing data. Radiometric and atmospheric corrections improve data accuracy. A unified coordinate system and gridding facilitate subsequent analysis. The temperature inversion algorithm converts radiometric brightness into intuitive temperature data. The resulting standardized surface temperature distribution image lays the foundation for subsequent heat island area identification and analysis, improving the reliability and accuracy of the overall method.

[0043] In some of the aforementioned solutions, after preprocessing to generate a surface temperature distribution image, it is necessary to identify heat island areas and classify their intensity levels. However, relying solely on temperature data can lead to biased reference background temperature selection, such as misclassifying green spaces within urban built-up areas as non-heat island areas, affecting the accuracy of heat island intensity classification.

[0044] This application further proposes to identify heat island areas and divide them into intensity levels, including: labeling the semantic attributes of each grid based on urban functional zoning data, extracting the temperature data of all grids in the non-built-up areas on the edge of the city and calculating the average temperature as the reference background temperature; calculating the difference between the temperature of each grid and the reference background temperature; identifying grids whose difference exceeds a threshold as heat island areas, and dividing them into mild, moderate or severe levels according to the difference range.

[0045] Among them, semantic attribute annotation divides the grid into types such as commercial areas, residential areas or ecological protection areas through urban functional zoning data in the geographic information system.

[0046] Among them, non-built-up areas are defined as undeveloped land or natural vegetation-covered areas outside the city, and their temperature data collection range must cover at least 15% of the total number of grids.

[0047] Among them, the difference calculation adopts the absolute temperature difference method, and the threshold is set at 2.5℃ as the baseline for heat island judgment; in the level classification, 2.5-4.0℃ is mild, 4.0-6.0℃ is moderate, and more than 6.0℃ is severe.

[0048] Specifically, supported by functional zoning data, we first screen out non-built-up area grids on the edge of the city, such as farmland, forest or water areas, to exclude the impact of densely built-up areas on the background temperature. When calculating the average temperature of such areas, a weighted average algorithm is used, with the area of ​​a single grid as the weight coefficient. The temperature difference value of each urban grid objectively reflects the degree of thermal anomaly by comparing it with the background temperature. When the temperature difference of a grid reaches 2.5°C, it is determined to have entered the heat island area. The levels are further divided according to the amplitude of the temperature difference. For example, if the temperature difference of a commercial area grid is 5.2°C, it is classified as a moderate heat island area. This division method combines the functional attributes of urban space to eliminate the interference of different surface cover types on the temperature benchmark, ensuring that the intensity level reflects the actual thermal environment pressure.

[0049] As a preferred embodiment, the solution of this application is specifically implemented as follows: Based on the semantic attributes of each grid cell labeled with urban functional zoning data, temperature data for all grid cells in non-built-up areas on the urban fringe were extracted, and the average temperature was calculated as the reference background temperature. Specifically, non-built-up areas such as farmland, woodland, and water within 5 kilometers of the urban fringe were divided into 500-meter by 500-meter grid cells. The temperature data for these grid cells were extracted, and their arithmetic mean was calculated as the reference background temperature.

[0050] Next, calculate the difference between the temperature of each grid cell and the reference background temperature. For example, for each 500m x 500m grid cell within the urban built-up area, subtract the reference background temperature obtained in the previous step from its temperature value to obtain the temperature difference.

[0051] Therefore, grids with temperature differences exceeding the threshold are identified as heat island areas and classified as mild, moderate, or severe based on the range of the difference. Specifically, 2.5°C, 4°C, and 6°C can be set as the thresholds for classification. When the temperature difference is between 2.5°C and 4°C, the grid is identified as a mild heat island area; when the temperature difference is between 4°C and 6°C, the grid is identified as a moderate heat island area; and when the temperature difference exceeds 6°C, the grid is identified as a severe heat island area.

[0052] Through the above technical solution, this application can accurately identify urban heat island areas and objectively classify their intensity levels. By using non-built-up areas on the urban fringe as a reference background, the impact of uneven temperature distribution within the city on heat island identification is avoided. By setting multi-level thresholds, a refined classification of heat island intensity is achieved, providing more accurate basic data for subsequent intervention decisions. In addition, the grid-based analysis method improves the spatial accuracy of heat island identification and intensity classification, which is conducive to the formulation of more targeted heat island control measures.

[0053] In some of the above-mentioned solutions of this application, although the intensity level of the current heat island area can be identified and the grid can be divided, it is difficult to accurately predict the dynamic change trend of the future intensity of the heat island area by relying solely on the current temperature difference data, resulting in a lack of foresight in intervention decisions and the possibility of a mismatch between resource allocation and future heat island risks.

[0054] This application further proposes that the prediction model is a time series neural network model based on spatial adjacency. The input of the prediction model includes historical intervention effects, heat island intensity change data of the current grid and adjacent grids, and the output is the prediction result of the heat island intensity level of the current grid in the future time window.

[0055] Among them, the spatial adjacency relationship captures the spatial diffusion law of the heat island effect by introducing the heat island intensity change data of adjacent grids.

[0056] Among them, the temporal neural network model adopts a long short-term memory network structure to perform time series modeling on historical intervention effects and heat island intensity change data, and learn the evolution pattern of heat island intensity over time.

[0057] Among them, the input historical intervention effect data includes the type of past intervention measures, implementation time and corresponding temperature drop records, which are used to quantify the inhibitory effect of intervention measures on heat island intensity.

[0058] The output results represent the heat island intensity in the future time window in the form of discrete levels, which is convenient for direct connection with subsequent priority calculations.

[0059] As a preferred embodiment, the solution of this application is specifically implemented as follows: The prediction model is a time-series neural network model based on spatial adjacency. The inputs include historical intervention effects and heat island intensity changes for the current grid and adjacent grids. The output is a prediction of the heat island intensity level for the current grid in the future time window.

[0060] Specifically, the prediction model uses a long-short-term memory network structure, with the input layer containing multiple feature channels. The historical intervention effect channel records the types of intervention measures implemented in a grid over the past six months and their effectiveness scores. The current grid heat island intensity change channel records the monthly heat island intensity level sequence for the grid over the past 12 months. The adjacent grid heat island intensity change channel records the monthly heat island intensity level sequence for the eight grids spatially adjacent to the current grid over the past 12 months.

[0061] The LSTM network consists of 64 hidden units and uses a three-layer stacked structure to extract temporal features. The network's output layer uses a softmax activation function to predict the probability distribution of the heat island intensity level for the current grid cell over the next three months. Training uses a cross-entropy loss function, and the Adam optimizer is used for parameter updates, with a learning rate set to 0.001. To improve the model's generalization ability, the training dataset is constructed from historical data using a sliding window method, and dropout technology is applied to prevent overfitting.

[0062] After the model training is completed, the historical data of the target grid and its adjacent grids can be input to predict the changing trend of the heat island intensity level of the grid in the next three months, providing a decision-making basis for the subsequent formulation of intervention strategies.

[0063] Through the above technical solution, this application achieves accurate prediction of the urban heat island effect. The prediction model makes full use of spatial adjacency and historical intervention effect information to improve the accuracy of the prediction. The structural design based on the time series neural network enables the model to effectively capture the long-term and short-term changes in the intensity of the heat island. The output probability distribution form provides decision makers with richer trend information. The time span of the prediction results covers the short-term and medium-term, meeting the decision-making needs of different time scales. This forward-looking prediction provides a scientific basis for the formulation of targeted intervention measures, which helps to improve the efficiency and effectiveness of urban heat island effect governance.

[0064] In some of the above-mentioned schemes of this application, when constructing the intervention response model, if the intervention measures are selected only based on the basic properties of the heat island area, it is difficult to accurately reflect the actual thermal response effects of different measures, and there is a lack of quantitative analysis of historical intervention data, resulting in a deviation between the simulation results and the actual temperature drop.

[0065] This application further proposes to screen all suitable intervention measures from a preset semantic attribute-intervention measure mapping table based on the semantic attributes of the heat island area grid; based on the historical intervention effects, the thermal response parameters of each intervention measure are determined through regression analysis, and the thermal response parameters include temperature drop and impact range.

[0066] Among them, the semantic attribute-intervention measure mapping table is pre-established through urban functional zoning data and engineering experience, such as mapping roof greening in commercial areas, mapping high-reflective road pavement in transportation hubs, and mapping ventilation corridor planning in industrial areas.

[0067] Among them, the regression analysis adopts a multivariate linear model, taking the type, implementation intensity, and meteorological conditions of historical intervention measures as independent variables, and temperature change data as dependent variables, and fitting to obtain the temperature drop coefficient and spatial impact radius parameters.

[0068] The impact range is calculated using the kernel density estimation algorithm, expanding outward from the intervention point, and the kernel function bandwidth is set to 200 meters.

[0069] As a preferred embodiment, the solution of this application is specifically implemented as follows: Constructing an intervention response model involves the following steps: First, based on the semantic attributes of the heat island grid, appropriate intervention measures are selected from a pre-set semantic attribute-intervention mapping table. For example, for commercial area grids, options include increasing green space and installing reflective coatings; for industrial area grids, options include optimizing industrial layout and adding ventilation corridors.

[0070] Next, based on historical intervention results, regression analysis was used to determine the thermal response parameters for each intervention measure. These parameters included the magnitude of the temperature drop and the range of impact. Specifically, data on past interventions and their effectiveness were collected, and a regression model was constructed to analyze the relationship between these interventions and temperature changes. For example, for measures that increased green space, a thermal response parameter was calculated: every 100 square meters of green space added resulted in a 0.5°C temperature drop within a 50-meter radius.

[0071] Through the above technical solution, this application can select appropriate intervention measures based on the characteristics of different regions and quantify the effects of each measure based on historical data. This can more accurately predict the impact of intervention measures, provide a basis for developing targeted heat island control strategies, and improve the precision and effectiveness of urban heat island effect management.

[0072] In some of the above-mentioned solutions in this application, although the temperature impact of different measures is simulated through intervention response models, there is a lack of a multi-dimensional comprehensive evaluation mechanism, which makes it difficult to quantify the urgency of governance in different heat island areas, resulting in difficulty in accurately matching resource allocation with actual regional needs.

[0073] This application further proposes steps for calculating governance priorities, including calculating the current heat island intensity level, heat island trend growth rate, population density level, important infrastructure distribution density and historical response effect score for each heat island area, where the historical response effect score is the deviation rate between the actual temperature drop and the simulated temperature drop; constructing a scoring function to calculate the comprehensive priority value, the scoring function is:

[0074] in, is the heat island intensity level, is the trend growth rate, is the population density level, is the infrastructure density, Score historical responses, 、 、 、 and is the weight coefficient determined by the hierarchical analysis method; the governance priority of each heat island area is obtained according to the priority value.

[0075] The current heat island intensity level is determined by the difference between the grid temperature and the background temperature, with larger differences indicating higher levels. The heat island trend growth rate is calculated using the future intensity change data output by the forecast model, with higher growth rates corresponding to higher values. The population density level is divided according to the number of people per unit area within the grid, with higher densities indicating higher levels. The infrastructure distribution density is calculated based on the number and coverage of key medical and transportation facilities, with higher densities indicating higher values. The historical response effectiveness score is calculated by comparing the deviation rate between the actual temperature drop from historical interventions and the simulated drop rate, with lower deviation rates indicating higher scores. The weight coefficients are determined using the Analytic Hierarchy Process (AHP), which constructs a judgment matrix based on expert judgments on the importance of the indicators. The weights of each indicator are then calculated after a consistency test.

[0076] Specifically, when constructing a judgment matrix using the Analytic Hierarchy Process (AHP), experts compared the importance of each of the five indicators based on urban management objectives. After generating the matrix, eigenvectors were calculated to determine initial weights, which were then refined after consistency testing to obtain final weight coefficients. After standardization, the indicator data for each heat island region was substituted into a scoring function to calculate a comprehensive priority value. Higher priority values ​​indicate a higher heat risk and greater urgency for governance. For example, if a region has a heat island intensity level of 3, a trend growth rate of 2, a population density of 4, an infrastructure density of 3, and a historical response score of 5, the weight coefficients are 0.3, 0.2, 0.25, 0.15, and 0.1, respectively, resulting in a comprehensive priority value of 3 × 0.3 + 2 × 0.2 + 4 × 0.25 + 3 × 0.15 + 5 × 0.1 = 3.25. All regions are ranked according to this calculation, and resources are allocated preferentially to the highest-ranked areas, thereby maximizing heat risk reduction benefits within limited resources.

[0077] As a preferred embodiment, the solution of this application is specifically implemented as follows: The following indicators are calculated for each heat island area: current heat island intensity level, heat island trend growth rate, population density level, important infrastructure distribution density and historical response effect score. The historical response effect score is the deviation rate between the actual temperature drop and the simulated temperature drop.

[0078] Construct a scoring function to calculate the comprehensive priority value. The scoring function is:

[0079] in, is the heat island intensity level, is the trend growth rate, is the population density level, is the infrastructure density, Score historical responses, 、 、 、 and is the weight coefficient determined by the hierarchical analysis method.

[0080] The governance priority of each heat island area is obtained by sorting according to the priority value.

[0081] Specifically, each heat island area is first evaluated. The heat island intensity level can be divided into mild, moderate and severe, and assigned values ​​of 1, 2, and 3 respectively. The heat island trend growth rate is calculated through a forecast model and expressed as the annual temperature increase. The population density level is divided into low, medium and high levels based on the population density distribution data, and assigned values ​​of 1, 2, and 3 respectively. The distribution density of important infrastructure is calculated as the number of important infrastructure per unit area. The historical response effect score is expressed in percentage, and the calculation formula is: (actual temperature drop - simulated temperature drop) / simulated temperature drop × 100%.

[0082] Furthermore, the weight coefficients of each indicator were determined through the hierarchical analysis method. First, a hierarchical structure model was established, with the target layer defined as "heat island control priority" and the criterion layer including the above five indicators. Secondly, a judgment matrix was constructed, and the importance of each indicator was compared pairwise using the expert scoring method. Then, the eigenvectors were calculated and consistency tested to obtain the weight coefficients of each indicator. 、 、 、 and .

[0083] Therefore, the index value of each heat island area is substituted into the scoring function to calculate the comprehensive priority value Finally, press The values ​​are sorted from large to small to obtain the governance priority ranking results of each heat island area.

[0084] Through the above technical solutions, this application has achieved a scientific assessment and ranking of the priorities for urban heat island regional governance. This method comprehensively considers multiple key factors such as heat island intensity, changing trends, population density, infrastructure distribution and historical intervention effects, avoiding the limitations of single indicator evaluation. By constructing a dynamic scoring function, the weights of each factor can be flexibly adjusted to adapt to the actual conditions of different cities. At the same time, the introduction of the hierarchical analysis method to determine the weight coefficient improves the objectivity and scientific nature of the evaluation process. This method helps city managers to more accurately identify heat island areas that are in urgent need of governance, optimize resource allocation, and improve the pertinence and effectiveness of heat island effect governance.

[0085] In some of the above-mentioned solutions in this application, although the governance priority of each heat island area is calculated through a dynamic scoring function, the problem of resource allocation conflict still exists in the actual scheduling process. For example, if resources are allocated only in sequence according to priority, it may not be possible to coordinate the synergistic effects of multi-region governance globally, resulting in inefficient resource utilization and failure to maximize the overall heat risk reduction effect. In addition, the implementation costs and expected effects of different intervention measures vary, and there are constraints on construction resources and time windows between regions. A simple sequential allocation strategy is difficult to meet actual governance needs.

[0086] This application further proposes constructing an optimization model with the goal of maximizing the overall heat risk reduction value of all heat island areas, where the heat risk reduction value of each area is jointly determined by its governance priority value and the expected temperature drop of the corresponding intervention measures; setting constraints for the optimization model, including total governance budget constraints, timing constraints of the intervention time windows of each area, and construction resource allocation constraints; solving the optimization model, and outputting the optimal intervention sequence and adaptation measure combination scheme for each heat island area as the optimal scheduling strategy.

[0087] The optimization model's objective function is defined as the sum of all regional heat risk reduction values. This value is weighted by multiplying the priority value by the temperature drop, ensuring that combinations of high-priority areas and highly effective interventions receive higher weights. The total budget constraint within the constraints is controlled by summing the costs of the selected interventions for each region. Timing constraints avoid resource conflicts by setting implementation windows for interventions in different regions. Construction resource allocation constraints limit the number of tasks that can be executed in parallel based on the number of available construction teams or equipment.

[0088] Specifically, the optimization model uses a mixed integer linear programming method to solve the problem, converting the regional governance sequence, measure selection and resource allocation into decision variables. For example, for a certain heat island area, if the "roof greening" measure is selected, its cost, expected temperature drop and required construction period will be used as input parameters to participate in the objective function calculation, and will be limited by the current available budget and the number of construction resources. During the solution process, the model iteratively adjusts the combination and time schedule of intervention measures in different areas to find the maximum value of the objective function that meets all constraints. The optimal scheduling strategy thus output not only includes the specific intervention measures for each area, but also clearly stipulates the time sequence and resource allocation plan for the implementation of the measures to ensure that the benefits of heat island effect governance are maximized under limited resource conditions.

[0089] As a preferred embodiment, the solution of this application is specifically implemented as follows: Generating the optimal scheduling strategy includes the following steps: An optimization model was constructed to maximize the overall heat risk reduction across all heat island regions. The heat risk reduction for each region was determined by the priority of the intervention and the expected temperature drop from the corresponding intervention.

[0090] Constraints are set for the optimization model, including total governance budget constraints, timing constraints for intervention time windows in each region, and construction resource allocation constraints.

[0091] Solve the optimization model and output the optimal intervention sequence and combination of adaptation measures for each heat island area as the optimal scheduling strategy.

[0092] Specifically, the objective function of the optimization model can be expressed as:

[0093] Among them, Pi is the governance priority value of the i-th heat island area, and ΔTi is the expected temperature drop of the corresponding intervention measures.

[0094] Constraints include: ∑Ci≤B, where Ci is the intervention cost of the i-th region and B is the total budget constraint.

[0095] Ti≤Tj, if the intervention in region i must be completed before that in region j.

[0096] ∑Ri≤R, where Ri is the amount of construction resources required for the i-th area and R is the total available resources.

[0097] Furthermore, a genetic algorithm was used to solve the optimization model. The population size was set to 100, the number of iterations was 1000, the crossover probability was 0.8, and the mutation probability was 0.1. Through continuous iterative optimization, the optimal intervention sequence and combination of adaptation measures for each heat island region were ultimately determined.

[0098] Therefore, the optimization model based on multi-dimensional constraints can generate more reasonable and efficient intervention scheduling strategies.

[0099] Through the above technical solution, this application can generate an optimal heat island intervention scheduling strategy under limited resource constraints. This strategy fully considers multiple factors such as the governance priority of each heat island area, the effectiveness of intervention measures, budget constraints, timing constraints, and resource allocation, and can maximize the reduction of heat island risks. At the same time, the optimization model's solution process is highly automated, which can quickly generate scheduling plans and improve decision-making efficiency. In addition, this method has strong flexibility and adaptability, and can adjust constraints and optimization goals according to actual conditions to meet the specific needs of different cities.

[0100] Exemplary Systems Figure 4The diagram shows an urban heat island effect decision-making system based on satellite data and artificial intelligence according to an embodiment of the present application, including: a data acquisition module for acquiring satellite remote sensing data and urban basic data of the target urban area, the urban basic data including population density distribution, historical intervention effects, infrastructure information and urban functional zoning data; a data processing module for preprocessing the satellite remote sensing data to generate a standardized surface temperature distribution image; a heat island area classification module for identifying urban heat island areas and dividing their intensity levels based on the surface temperature distribution image; a heat island trend prediction module for predicting the future intensity change trend of the heat island area through a time series prediction model based on spatial adjacency; an intervention measure simulation module for constructing an intervention response model to simulate the temperature impact of different intervention measures on the heat island area; a governance priority sorting module for constructing a dynamic scoring function to calculate the governance priority of each heat island area based on the heat island intensity level, change trend, population density distribution, infrastructure information and historical intervention effects of the heat island area; an intervention scheduling strategy output module for generating an optimal intervention scheduling strategy based on the simulation results of the intervention response model, the governance priority and the preset governance resource constraints.

[0101] In one example, the data processing module preprocesses data including: performing radiation correction and atmospheric correction on the thermal infrared band in the satellite remote sensing data to obtain surface radiation brightness data; reprojecting the radiation brightness data to a unified geographic coordinate system and dividing the urban space units according to a preset grid; based on the radiation brightness data, calculating the surface temperature value of each grid through a temperature inversion algorithm to generate a standardized surface temperature distribution image.

[0102] In one example, the heat island area division module identifies heat island areas and divides them into intensity levels, including: labeling the semantic attributes of each grid based on urban functional zoning data, extracting the temperature data of all grids in the non-built-up areas on the edge of the city and calculating the average temperature as the reference background temperature; calculating the difference between the temperature of each grid and the reference background temperature; identifying grids whose difference exceeds a threshold as heat island areas, and dividing them into mild, moderate or severe levels according to the difference range.

[0103] In one example, the prediction model of the heat island trend prediction module is a time series neural network model based on spatial adjacency. The input of the prediction model includes historical intervention effects, heat island intensity change data of the current grid and adjacent grids, and the output is the prediction result of the heat island intensity level of the current grid in the future time window.

[0104] In one example, the intervention measure simulation module constructs an intervention response model, including: screening all suitable intervention measures from a preset semantic attribute-intervention measure mapping table based on the semantic attributes of the heat island area grid; and determining the thermal response parameters of each intervention measure through regression analysis based on historical intervention effects. The thermal response parameters include temperature drop and impact range.

[0105] In one example, the governance priority ranking module calculates governance priorities by: calculating the following indicators for each heat island area: current heat island intensity level, heat island trend growth rate, population density level, important infrastructure distribution density and historical response effect score, where the historical response effect score is the deviation rate between the actual temperature drop and the simulated temperature drop; constructing a scoring function to calculate the comprehensive priority value, the scoring function is:

[0106] in, is the heat island intensity level, is the trend growth rate, is the population density level, is the infrastructure density, Score historical responses, 、 、 、 and is the weight coefficient determined by the hierarchical analysis method; the governance priority of each heat island area is obtained according to the priority value.

[0107] In one example, the intervention scheduling strategy output module generates the optimal scheduling strategy and includes: constructing an optimization model with the goal of maximizing the overall heat risk reduction value of all heat island areas, where the heat risk reduction value of each area is determined by its governance priority value and the expected temperature drop of the corresponding intervention measures; setting constraints for the optimization model, including the total governance budget limit, the timing constraints of the intervention time window of each area, and the construction resource allocation limit; solving the optimization model and outputting the optimal intervention sequence and adaptation measure combination scheme for each heat island area as the optimal scheduling strategy.

[0108] Exemplary electronic devices Figure 5 The figure shows an electronic device according to an embodiment of the present application. The electronic device can be the mobile device itself, or a stand-alone device independent of the mobile device, which can communicate with the mobile device to receive collected input signals from the mobile device and send the selected target driving behavior to the mobile device.

[0109] Figure 5 The figure shows a block diagram of an electronic device according to an embodiment of the present application.

[0110] like Figure 5 As shown, the electronic device includes one or more processors and memory.

[0111] The processor may be a central processing unit (CPU) or other forms of processing units having data processing capabilities and / or instruction execution capabilities, and may control other components in the electronic device to perform desired functions.

[0112] The memory may include one or more computer program products, which may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and / or cache memory. Non-volatile memory may include, for example, read-only memory (ROM), a hard disk, flash memory, etc. One or more computer program instructions may be stored on the computer-readable storage medium, and the processor may execute the program instructions to implement the driving behavior decision-making method of each embodiment of the present application described above and / or other desired functions.

[0113] In one example, the electronic device may further include an input device and an output device, and these components are interconnected via a bus system and / or other forms of connection mechanisms (not shown).

[0114] Of course, to simplify, Figure 5 Only some of the components in the electronic device related to the present application are shown, and components such as a bus, an input / output interface, etc. are omitted. In addition, the electronic device may further include any other appropriate components according to specific application scenarios.

[0115] Exemplary computer storage media The embodiment of the present application may also be a computer-readable storage medium having stored thereon a computer program indicating The computer program instructions, when executed by a processor, cause the processor to execute the steps of the driving behavior decision-making method according to various embodiments of the present application described in the above “Exemplary Method” section of this specification.

[0116] Computer readable storage media can adopt any combination of one or more readable media. The readable medium can be a readable signal medium or a readable storage medium. The readable storage medium can, for example, include but is not limited to a system, device or component of electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination thereof. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof.

[0117] The basic principles of the present application have been described above in conjunction with specific embodiments. However, it should be noted that the advantages, strengths, and effects mentioned in this application are merely illustrative and not restrictive, and it should not be assumed that these advantages, strengths, and effects are required of each embodiment of this application. In addition, the specific details disclosed above are merely illustrative and facilitating understanding, and are not restrictive. The above details do not limit this application to necessarily being implemented using the above specific details.

[0118] The block diagrams of the devices, devices, equipment, and systems involved in this application are merely illustrative examples and are not intended to require or imply that they must be connected, arranged, or configured in the manner shown in the block diagrams. As will be appreciated by those skilled in the art, these devices, devices, equipment, and systems can be connected, arranged, or configured in any manner. Words such as "include," "comprise," "have," and the like are open-ended words, meaning "including but not limited to," and can be used interchangeably therewith. The words "or" and "and" used herein refer to the words "and / or" and can be used interchangeably therewith, unless the context clearly indicates otherwise. The word "such as" used herein refers to the phrase "such as but not limited to," and can be used interchangeably therewith.

[0119] It should also be noted that in the apparatus, device, and method of the present application, each component or each step can be decomposed and / or recombined, and such decomposition and / or recombination should be regarded as equivalent solutions of the present application.

[0120] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other aspects without departing from the scope of the present application. Therefore, the present application is not intended to be limited to the aspects shown herein, but rather to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0121] The above description has been provided for the purpose of illustration and description. Furthermore, this description is not intended to limit the embodiments of the present application to the forms disclosed herein. Although a number of example aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, alterations, additions, and sub-combinations thereof.

[0122] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. Urban heat island effect decision-making method based on satellite data and artificial intelligence, characterized by: include: Obtain satellite remote sensing data and basic urban data for the target urban area, including population density distribution, historical intervention effects, infrastructure information, and urban functional zoning data; Preprocessing the satellite remote sensing data to generate a standardized surface temperature distribution image; Based on the surface temperature distribution image, identifying urban heat island areas and classifying their intensity levels; Predicting the future intensity change trend of the heat island area through a prediction model; Construct an intervention response model to simulate the effects of different intervention measures on the temperature of the heat island area; Based on the heat island intensity level, change trend, population density distribution, infrastructure information and historical intervention effects of the heat island area, a dynamic scoring function is constructed to calculate the governance priority of each heat island area; Based on the simulation results of the intervention response model, governance priorities and preset governance resource constraints, an optimal intervention scheduling strategy is generated.

2. The urban heat island effect decision-making method based on satellite data and artificial intelligence according to claim 1 is characterized in that: The pretreatment includes: Performing radiation correction and atmospheric correction on the thermal infrared band in the satellite remote sensing data to obtain surface radiation brightness data; Reproject the radiance data to a unified geographic coordinate system and divide the urban space units into preset grids; Based on the radiation brightness data, the surface temperature value of each grid is calculated using the temperature inversion algorithm to generate a standardized surface temperature distribution image.

3. The urban heat island effect decision-making method based on satellite data and artificial intelligence according to claim 2 is characterized in that: The identification of heat island areas and classification of intensity levels include: Annotate the semantic attributes of each grid based on the urban functional zoning data, extract the temperature data of all grids in the non-built-up area on the edge of the city, and calculate the average temperature as the reference background temperature; Calculating the difference between the temperature of each grid and the reference background temperature; Grids with difference values ​​exceeding the threshold are identified as heat island areas and classified as mild, moderate, or severe based on the difference value range.

4. The urban heat island effect decision-making method based on satellite data and artificial intelligence according to claim 3 is characterized by: The prediction model is a temporal neural network model based on spatial adjacency. The input of the prediction model includes historical intervention effects and heat island intensity change data of the current grid and adjacent grids. The output is the prediction result of the heat island intensity level of the current grid in the future time window.

5. The urban heat island effect decision-making method based on satellite data and artificial intelligence according to claim 4 is characterized in that: The construction of the intervention response model includes: According to the semantic attributes of the heat island area grid, all suitable intervention measures are screened from a preset semantic attribute-intervention measure mapping table; Based on historical intervention effects, the thermal response parameters of each intervention measure were determined through regression analysis. The thermal response parameters included the temperature drop and the impact range.

6. The urban heat island effect decision-making method based on satellite data and artificial intelligence according to claim 5 is characterized in that: The computing governance priorities include: The following indicators are calculated for each heat island area: current heat island intensity level, heat island trend growth rate, population density level, density of important infrastructure distribution, and historical response effect score, which is the deviation rate between the actual temperature drop and the simulated temperature drop; Construct a scoring function to calculate the comprehensive priority value, the scoring function is: Among them, the is the heat island intensity level, is the trend growth rate, is the population density level, is the infrastructure density, Score historical responses, 、 、 、 and is the weight coefficient determined by the analytic hierarchy process; The governance priority of each heat island area is obtained by sorting according to the priority value.

7. The urban heat island effect decision-making method based on satellite data and artificial intelligence according to claim 6 is characterized in that: Generating the optimal scheduling strategy includes: An optimization model was constructed to maximize the overall heat risk reduction across all heat island regions, where the heat risk reduction for each region was determined by its priority and the expected temperature reduction from the corresponding intervention measures. Setting constraints for the optimization model, including total governance budget constraints, timing constraints for intervention time windows in each region, and construction resource allocation constraints; Solve the optimization model and output the optimal intervention sequence and adaptation measure combination scheme for each heat island area as the optimal scheduling strategy.

8. An urban heat island effect decision-making system based on satellite data and artificial intelligence, applying the method according to any one of claims 1 to 7, characterized in that: include: A data acquisition module is used to obtain satellite remote sensing data and basic urban data of the target urban area, including population density distribution, historical intervention effects, infrastructure information, and urban functional zoning data; A data processing module is used to pre-process the satellite remote sensing data to generate a standardized surface temperature distribution image; a heat island area classification module, configured to identify urban heat island areas and classify their intensity levels based on the surface temperature distribution image; A heat island trend prediction module is used to predict the future intensity change trend of the heat island area through a time series prediction model based on spatial adjacency; An intervention measure simulation module is used to construct an intervention response model to simulate the effects of different intervention measures on the temperature of the heat island area; A governance priority ranking module is used to construct a dynamic scoring function to calculate the governance priority of each heat island area based on the heat island intensity level, change trend, population density distribution, infrastructure information and historical intervention effects of the heat island area; The intervention scheduling strategy output module is used to generate the optimal intervention scheduling strategy based on the simulation results of the intervention response model, the governance priority and the preset governance resource constraints.

9. An electronic device comprising a memory and a processor, characterized in that: The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer storage medium having computer-executable instructions stored thereon, characterized in that: When the computer-executable instructions are executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images