Urban climate risk monitoring system based on multi-source data fusion and early warning application thereof

Abstract

The present application relates to the technical field of smart city, specifically to a city climate risk monitoring system based on multi-source data fusion and early warning application thereof, comprising the following contents: data reconstruction module, precursor identification module, trend chain extraction module, grade assignment module and risk diffusion module.In the present application, the reliability of multi-source meteorological data is effectively enhanced by evaluating the credibility level of multiple data sources, the accuracy of data fusion and the reliability of spatial positioning are improved, the sensitivity of risk precursor events and the accuracy of factor propagation prediction are enhanced by deep identification of trend consistency and response delay relationship, the stability and spatial continuity of risk grade division are optimized by combining regional spatial continuity factors, the spatial impedance characteristics are included in the diffusion path prediction, the accurate deduction of the spatial diffusion path of city climate risk and the dynamic marking of the risk area are realized, and the accuracy and dynamics of city climate risk monitoring and prevention and control decision-making are enhanced.

Description

Technical Field

[0001] This invention relates to the field of smart city technology, and in particular to an urban climate risk monitoring system based on multi-source data fusion and its early warning application. Background Technology

[0002] The field of smart city technology encompasses various technologies such as the Internet of Things, cloud computing, big data, geographic information systems, remote sensing, and artificial intelligence. By sensing multiple types of urban data in real time and integrating and analyzing information from multiple sources, it enables urban operational status monitoring, accurate risk identification, and management decision support. Smart cities take urban information digitization, intelligent management decision-making, and refined urban governance as their core content. They systematically integrate sensing devices from different sources, spatial data acquisition systems, and artificial intelligence analysis platforms to comprehensively realize refined urban operation monitoring and risk management.

[0003] Among them, the urban climate risk monitoring system based on multi-source data fusion refers to the use of satellite remote sensing data, ground meteorological observation data, UAV remote sensing data, IoT sensing device data, and socio-economic statistical data to accurately identify and dynamically monitor urban rainstorms and floods, heat waves, air pollution, and various specific climate risk factors through data fusion analysis methods. The system achieves accurate identification and dynamic monitoring of urban climate risks through various methods such as satellite image interpretation, remote sensing image data processing, IoT sensor data acquisition, geographic information data spatial analysis, and machine learning model training.

[0004] In practical applications, traditional urban climate risk monitoring technologies often limit the fusion of multi-source data to simple spatial data overlay analysis. They fail to thoroughly examine the observation density and reliability of the data sources themselves, resulting in insufficient data fusion accuracy and biases in fusion results due to differences in data reliability. Urban climate risk identification often focuses on single indicators, lacking detailed analysis of the response relationships between multiple factors. This fails to accurately reveal the formation process and triggering paths of urban disasters, reducing the accuracy of disaster identification. Regional risk level assignment relies on the intensity distribution of a single factor, neglecting the spatial continuity and mutual influence between risk levels in adjacent areas. The spatial distribution of risk classification results is unreasonable. When predicting risk diffusion paths, they rarely consider spatial impedance factors such as surface features, building distribution, and topographic differences, leading to deviations in risk propagation predictions from reality. This results in inaccurate urban risk management decisions, insufficient risk warning and response capabilities, and increased urban management costs and disaster prevention difficulties. Summary of the Invention

[0005] To address the technical problems existing in the prior art, embodiments of the present invention provide an urban climate risk monitoring system based on multi-source data fusion and its early warning application. The technical solution is as follows:

[0006] On the one hand, an urban climate risk monitoring system based on multi-source data fusion is provided, which includes:

[0007] The data reconstruction module acquires multi-source meteorological monitoring data, extracts the observation density, interval value, and deviation value of multiple data sources, analyzes the reliability level of multiple data sources, and combines location information to perform data fusion on the multi-source monitoring data to generate a spatial reconstruction dataset.

[0008] The precursor identification module calls the spatial reconstruction dataset to obtain time series data of various climate factors. By analyzing the consistency characteristics of the changing trends of each data in multiple regions, it marks areas of risk concern and generates precursor disturbance trend intervals.

[0009] The trend chain extraction module calls the precursor disturbance trend interval, extracts meteorological factor monitoring sequence data of the risk concern area, analyzes the change sequence of each climate factor, identifies the response delay relationship between multiple climate factors, identifies the risk propagation trigger sequence, and generates risk trend activation path.

[0010] The risk level assignment module calls the risk trend triggering path, and calculates and adjusts the risk levels of multiple regions by analyzing the intensity of climate data changes in multiple regions and combining the spatial continuity between risk levels of adjacent regions, based on the frequency and amplitude of data changes and fluctuations, to generate spatial distribution values ​​of risk levels.

[0011] As a further aspect of the present invention, the spatial reconstruction dataset includes grid location units, fused observations, and confidence level coefficients; the precursor disturbance trend interval includes disturbance time period, disturbance factor type, and risk concern area number; the risk trend triggering path includes factor triggering sequence, response delay time difference, and trend path number; and the risk level spatial distribution value specifically includes level numerical unit, level change amplitude, and level continuity index.

[0012] As a further aspect of the present invention, the data reconstruction module includes:

[0013] The difference analysis submodule acquires multi-source meteorological monitoring data, calculates the density value of observation points from multiple data sources based on the number and spatial distribution of observation points from multiple data sources, analyzes the observation interval difference of data sources, calculates the data numerical deviation of each data source, and generates a data difference coefficient.

[0014] The data source rating submodule calls the data difference coefficient, and based on the data difference coefficient, grades and evaluates the credibility of each data source, identifying the credibility level value of each data source.

[0015] The data fusion submodule calls the credibility level value of each data source, calculates the weight coefficient of multiple data sources based on the credibility level value, and combines the location information corresponding to the meteorological monitoring data to perform data fusion on the multi-source meteorological monitoring data and establish a spatial reconstruction dataset.

[0016] As a further aspect of the present invention, the specific formula for data fusion of multi-source meteorological monitoring data is as follows:

[0017]

[0018] Calculate the fused data values;

[0019] Among them, V s′ V represents the fused data value at spatial location s′. i′,s′ Let C be the original monitoring value of the i′-th data source at location s′. i′ Let D be the trust level value of the i′-th data source. i′,s′ Let C be the spatial distance from the monitoring point of the i′-th data source to the spatial location s′. j′ Let D be the trust level value of the j′-th data source. j′,s′ Let be the spatial distance from the monitoring point of the j′-th data source to the spatial location s′, n′ be the total number of data sources participating in spatial data fusion, i′ be the index number of each data source in the weighted calculation, j′ be the index number of each data source in the weighted normalization calculation, and s′ be the number of the location point to be calculated in the spatial fusion area.

[0020] As a further aspect of the present invention, the precursor identification module includes:

[0021] The trend extraction submodule calls the spatial reconstruction dataset to obtain time series data of each climate factor in multiple areas of the city and establishes a data change trend curve.

[0022] The consistency analysis submodule calls the data change trend curve, extracts the fluctuation direction and duration of each climate factor trend curve, analyzes the consistency of the fluctuation direction and duration of multiple data trend curves, and generates a trend consistency coefficient.

[0023] The regional marking submodule calls the trend consistency coefficient to mark risk-concern areas and establish precursor disturbance trend intervals based on the consistency of the changing trends of each meteorological monitoring data.

[0024] As a further aspect of the present invention, the trend chain extraction module includes:

[0025] The sequence extraction submodule calls the precursor disturbance trend interval, extracts real-time monitoring sequence values ​​of various climate factors in the region based on the coordinates of the risk concern area, including temperature, humidity, precipitation, and wind speed, and arranges the data according to the timestamp to establish factor monitoring sequence data;

[0026] The delay determination submodule calls the factor monitoring sequence data, calculates the correlation between the numerical changes of each climate factor based on the numerical changes of each data sequence, analyzes the response relationship of multiple climate factors, and obtains the response delay coefficient.

[0027] The triggering sorting submodule calls the response delay coefficient to analyze the triggering order of various climate factors in risk propagation and establish a risk trend activation path.

[0028] As a further aspect of the present invention, the level assignment module includes:

[0029] The intensity calculation submodule calls the risk trend activation path, analyzes the frequency and amplitude of changes in climate data values ​​in each region, calculates the intensity of change, and obtains data change intensity information.

[0030] The continuity analysis submodule calls the data change intensity information to detect the boundary relationship between adjacent areas. By comparing the change intensity between adjacent areas, it analyzes the spatial relationship of regional risk levels and obtains the spatial continuity coefficient.

[0031] The risk level adjustment submodule calls the spatial continuity coefficient, calculates the risk level benchmark value for multiple regions based on the change intensity value, adjusts the risk level in combination with the spatial relationship of regional risk levels, and establishes the spatial distribution value of risk level.

[0032] As a further aspect of the present invention, the system further includes:

[0033] The risk diffusion module calls the spatial distribution value of the risk level, obtains the surface material type, building density and topographic elevation difference of each region, analyzes the spatial impedance characteristics between regions, calculates the spatial diffusion direction and path of urban climate risk, marks the diffusion risk area, and generates climate risk prediction results.

[0034] The climate risk prediction results specifically refer to the diffusion direction and path, risk impact boundary, and spatial propagation intensity.

[0035] As a further aspect of the present invention, the risk diffusion module includes:

[0036] The impedance calculation submodule calls the risk level spatial distribution value to obtain the surface material type, building density, and terrain elevation difference of each area. Based on the surface material friction coefficient, building density barrier value, and terrain elevation difference value, it calculates the spatial impedance value and establishes the spatial impedance coefficient.

[0037] The specific formula for calculating the spatial impedance value is as follows:

[0038]

[0039] Calculate the overall spatial impedance value;

[0040] Among them, Z t′ The spatial composite impedance value of region t′, f t′ B represents the average coefficient of friction of the surface material in region t′. t′ H represents the building density barrier value within region t′. max,t′ H represents the highest topographic elevation of region t′. min,t′ H represents the lowest topographic elevation of region t′. avg W represents the average elevation difference across all regions. f The weight representing the coefficient of friction of surface materials, W b W represents the weight of the building density barrier value. h The weights represent the differences in terrain elevation, and t′ is the index of the region.

[0041] The directional analysis submodule calls the spatial impedance coefficient, analyzes the spatial diffusion direction of climate risk based on the obstruction strength of diffusion risk propagation between multiple regions, identifies diffusion paths, and generates risk diffusion prediction results.

[0042] The path marking submodule calls the risk diffusion prediction results, marks the diffusion risk areas by predicting the spatial diffusion trajectory of urban climate risk, and establishes climate risk prediction results.

[0043] On the other hand, an urban climate risk monitoring and early warning application based on multi-source data fusion is provided, which is used to carry out an urban climate risk monitoring system based on multi-source data fusion.

[0044] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following:

[0045] By conducting credibility level assessments on multiple data sources, the reliability of multi-source meteorological data is effectively enhanced, improving the accuracy of data fusion and spatial positioning reliability. By utilizing in-depth identification of trend consistency and response delay relationships, the sensitivity of risk precursor events and the accuracy of factor propagation prediction are enhanced. Combined with regional spatial continuity factors, the stability and spatial continuity of risk level classification are optimized. By incorporating spatial impedance characteristics into diffusion path prediction, accurate extrapolation of urban climate risk spatial diffusion paths and dynamic marking of risk areas are achieved, thereby enhancing the precision and dynamism of urban climate risk monitoring and prevention and control decisions. Attached Figure Description

[0046] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0047] Figure 1 This is a system flowchart of the present invention;

[0048] Figure 2 This is a schematic diagram of the system framework of the present invention. Detailed Implementation

[0049] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0050] In embodiments of the present invention, words such as "exemplarily," "for example," etc., are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" in the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the word "exemplary" is intended to present the concept in a concrete manner. Furthermore, in embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one.

[0051] In the embodiments of this invention, the terms "image" and "picture" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, their intended meanings are consistent. Similarly, the terms "of," "corresponding (relevant)," and "corresponding" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, their intended meanings are consistent.

[0052] In this embodiment of the invention, sometimes a subscript such as W1 may be written in a non-subscript form such as W1. When the difference is not emphasized, the meaning they express is the same.

[0053] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0054] This invention provides an urban climate risk monitoring system based on multi-source data fusion. Please refer to [link / reference]. Figures 1 to 2 This invention provides a technical solution: an urban climate risk monitoring system based on multi-source data fusion, comprising:

[0055] The data reconstruction module acquires multi-source meteorological monitoring data, extracts the observation density, interval value, and deviation value of multiple data sources, analyzes the reliability level of multiple data sources, and combines location information to perform data fusion on the multi-source monitoring data to generate a spatial reconstruction dataset.

[0056] The precursor identification module calls the spatial reconstruction dataset to obtain time series data of various climate factors. By analyzing the consistency characteristics of the changing trends of each data in multiple regions, it marks areas of risk concern and generates precursor disturbance trend intervals.

[0057] The trend chain extraction module calls the precursor disturbance trend interval, extracts meteorological factor monitoring sequence data of the risk concern area, analyzes the change sequence of each climate factor, identifies the response delay relationship between multiple climate factors, identifies the risk propagation trigger sequence, and generates risk trend activation path.

[0058] The risk level assignment module calls the risk trend triggering path, and based on the frequency and amplitude of data changes, it analyzes the intensity of climate data changes in multiple regions, combines the spatial continuity between risk levels in adjacent regions, calculates and adjusts the risk levels of multiple regions, and generates spatial distribution values ​​of risk levels.

[0059] The risk diffusion module calls the spatial distribution value of the risk level, obtains the surface material type, building density and topographic elevation difference of each region, analyzes the spatial impedance characteristics between regions, calculates the spatial diffusion direction and path of urban climate risk, marks the diffusion risk areas, and generates climate risk prediction results.

[0060] The spatial reconstruction dataset includes grid location units, fused observations, and confidence level coefficients. The precursor disturbance trend interval includes disturbance time period, disturbance factor type, and risk concern area number. The risk trend triggering path includes factor triggering sequence, response delay time difference, and trend path number. The spatial distribution value of the risk level specifically includes level numerical unit, level change amplitude, and level continuity index. The climate risk prediction result specifically refers to diffusion direction path, risk impact boundary, and spatial propagation intensity.

[0061] The data reconstruction module includes:

[0062] The difference analysis submodule acquires multi-source meteorological monitoring data, calculates the density value of observation points from multiple data sources based on the number and spatial distribution of observation points from multiple data sources, analyzes the observation interval difference of data sources, calculates the data numerical deviation of each data source, and generates a data difference coefficient.

[0063] The difference analysis submodule acquires multi-source meteorological monitoring data, obtaining the number and spatial location of observation points from each data source within the observation area. The observation area is divided into 100 equally sized grid cells, each with an area of ​​1 square kilometer. The number of observation points from each data source within each cell is extracted, and the observation density value is calculated. Normalization is then performed based on the maximum and minimum observation point densities. The observation time interval values ​​for each data source are collected (e.g., 10 minutes for data source A, 15 minutes for data source B, and 20 minutes for data source C). The absolute difference between each pair of data sources is calculated to obtain the time interval difference, which is then normalized as a time consistency index. Meteorological observation values ​​from different data sources at the same location and time are extracted (e.g., A = 25.0℃, B = 26.0℃, C = 24.5℃). The average value is calculated first, and then the deviation of each observation value is calculated to obtain the data consistency index. Using these three indices as input, weighting coefficients are set as follows: observation density weight 0.4, observation time interval weight 0.3, and numerical deviation weight 0.3. These are then weighted and synthesized to form the data difference coefficient, calculated using the following formula:

[0064] D s =0.4·M s +0.3·T s +0.3·P s ;

[0065] Among them, D s M is the data difference coefficient of data source s. s T is the observation density index normalized from data source s. s P is the observation interval index after normalization of data source s. s This is the numerical deviation index after normalization of the data source s.

[0066] Setting M A =0.6, T A =0.2, P A =0.1, then:

[0067] D A =0.4·0.6+0.3·0.2+0.3·0.1=0.24+0.06+0.03=0.33;

[0068] Generate the data difference coefficient.

[0069] The data source rating submodule calls the data difference coefficient, and based on the data difference coefficient, grades and evaluates the credibility of each data source, identifying the credibility level value of each data source.

[0070] The data source rating submodule calls the data difference coefficient, sets the confidence level range to four levels, and defines the ranges as follows: [0, 0.25) for level 1, [0.25, 0.5) for level 2, [0.5, 0.75) for level 3, and [0.75, 1.0] for level 4. It reads the difference coefficient of each data source and determines the range it falls into, assigning a corresponding confidence level value. The calculation formula is as follows:

[0071]

[0072] Among them, L s D represents the trust level value of data source s. s The coefficient of variation for data source s.

[0073] Setting D B =0.72, then 0.5≤D B <0.75, corresponding to a trust level of:

[0074] L B =3;

[0075] Identify the trust level value for each data source.

[0076] The data fusion submodule calls the credibility level value of each data source, calculates the weight coefficient of multiple data sources based on the credibility level value, and combines the location information corresponding to the meteorological monitoring data to perform data fusion on the multi-source meteorological monitoring data and establish a spatial reconstruction dataset.

[0077] The specific formula for data fusion of multi-source meteorological monitoring data is as follows:

[0078]

[0079] Calculate the fused data values;

[0080] Among them, V s′ V represents the fused data value at spatial location s′. i′,s′ Let C be the original monitoring value of the i′-th data source at location s′. i′ Let D be the trust level value of the i′-th data source. i′,s′ Let C be the spatial distance from the monitoring point of the i′-th data source to the spatial location s′. j′ Let D be the trust level value of the j′-th data source. j′,s′Let be the spatial distance from the monitoring point of the j′-th data source to the spatial location s′, n′ be the total number of data sources participating in spatial data fusion, i′ be the index number of each data source in the weighted calculation, j′ be the index number of each data source in the weighted normalization calculation, and s′ be the number of the location point to be calculated in the spatial fusion area.

[0081] formula:

[0082]

[0083] Detailed explanation of the formula and its calculation derivation:

[0084] The formula is used to calculate the fusion value of multi-source meteorological monitoring data at spatial location point s′, and the result is used to establish the temperature fusion output value of the corresponding location in the spatial reconstruction dataset;

[0085] Parameter meanings and settings:

[0086] V s′ The fused meteorological value for location point s′;

[0087] C i′ The reliability level values ​​for the i′-th data source are set as C1 = 0.9, C2 = 0.6, and C3 = 0.3, reflecting the sampling stability and temporal integrity of the data source.

[0088] D i′,s′ Let D be the distance from the sensor of the i′-th data source to the location point s′, in meters. 1,s′ =120, D 2,s′ =100, D 3,s′ =80, reflecting the suppressive effect of spatial distribution on data weights;

[0089] V i′,s′ Let V be the raw temperature value collected by the i′-th data source near location s′. 1,s′ =32.0℃, V 2,s′ =30.5℃, V 3,s′ =31.0℃;

[0090] n′ represents the number of data sources participating in the fusion, which is set to 3;

[0091] i′ represents the weight index and normalized index number involved in the calculation;

[0092] s′ represents the current location point number to be merged;

[0093] Substitute the parameters into the formula to calculate:

[0094]

[0095] 0.0075 + 0.0060 + 0.00375 = 0.01725;

[0096]

[0097] V s′ ≈0.4348×32.0+0.3478×30.5+0.2174×31.0=13.91+10.61+

[0098] 6.74 = 31.26℃;

[0099] The result of 31.26 indicates that the representative temperature value of the current spatial location point after fusing multi-source meteorological monitoring data is 31.26℃, which is the fused output value of this point in the spatial reconstruction dataset.

[0100] The precursor identification module includes:

[0101] The trend extraction submodule calls the spatial reconstruction dataset to obtain time series data of each climate factor in multiple areas of the city and establishes a data change trend curve.

[0102] The trend extraction submodule calls the spatial reconstruction dataset to obtain time series data of each climate factor in multiple areas of the city. It calls the meteorological factor observations (temperature, wind speed, precipitation, surface heat flux, etc.) for a specific urban area from the spatial reconstruction dataset, extracting the hourly observations of each factor in chronological order to form multiple factor time series. The extraction process is based on each meteorological factor, with a sampling time span of 72 hours, resulting in continuous observations at 72 time nodes. Combining the time series of each factor, the first-order difference value is calculated at each time node to obtain the variable's change amplitude. The positive or negative change of the difference value is used as a fluctuation direction indicator. Combined with the maximum duration of the same fluctuation direction in the continuous time series, a trend segment is constructed. If a meteorological factor shows a continuous upward trend for 12 hours, the trend segment direction is upward, and the duration is 12 hours. Trend segments of the same meteorological factor in all areas are merged to obtain the trend curve for each climate factor in each area. The trend curve is stored by area number, with time on the x-axis and the numerical change amplitude on the y-axis, forming a continuous change trajectory. Let the value of each node in the trend curve be the meteorological factor value Y of a certain area at time t. r,t The formula for calculating the difference between two adjacent times is:

[0103] ΔY r,t =Y r,t -Y r,t-1 ;

[0104] Where, ΔY r,t Y represents the magnitude of change of region r over time t. r,tY is the observed value at the current time. r,t-1 This is the value observed at the previous moment.

[0105] Set Y r,10 =25.6, Y r,9 =24.9, then:

[0106] ΔY r,10 =25.6-24.9=0.7;

[0107] This indicates that the factor is trending upward at that point in time. This operation is performed on each factor, each region, and each point in time to establish a data change trend curve.

[0108] The consistency analysis submodule calls the data change trend curve, extracts the fluctuation direction and duration of each climate factor trend curve, analyzes the consistency of the fluctuation direction and duration of multiple data trend curves, and generates a trend consistency coefficient.

[0109] The consistency analysis submodule calls the data change trend curves to extract the fluctuation direction and duration of each climate factor's trend curve. It then extracts the trend segments of each climate factor by region in a structured manner, defining the trend segment direction as continuous increase or decrease, and the duration as the length of the trend segment. The extraction method involves recording the start time, end time, and direction type of continuous positive or negative segments from the trend curve. After obtaining the trend segment combinations, it calculates the trend consistency coefficient based on the consistency of trend direction and duration overlap of different climate factors within the same time period. The trend consistency coefficient assumes consistent direction, with the proportion of duration overlap used as the weighting factor. The formula is as follows:

[0110]

[0111] Among them, C r Let δ be the trend consistency coefficient for region r, N be the total number of climate factors, and δ be the trend consistency coefficient for region r. f This indicates whether factor f is aligned with the direction of the principal factor during this time period (values ​​are 0 or 1). This represents the duration of overlap between the trend segment of factor f and the trend segment of the main factor. This represents the total duration of the trend segment for this factor.

[0112] Given region r, among the three factors, wind speed and temperature have the same direction and overlap for 6 hours, the overall trend period is 8 hours, and precipitation has a different direction. Then:

[0113]

[0114] Obtain the trend consistency coefficient.

[0115] The regional marking submodule calls the trend consistency coefficient to mark risk-concern areas and establish precursor disturbance trend intervals based on the consistency of the changing trends of each meteorological monitoring data.

[0116] The regional labeling submodule calls the trend consistency coefficient to label risk areas based on the trend consistency of climate factors in each region. The consistency coefficient threshold range is set as follows: below 0.4 indicates no risk area, 0.4 to 0.7 indicates a slightly risky area, and above 0.7 indicates a strongly risky area. The trend consistency coefficient of each region is extracted and compared with the above range for judgment. Combined with spatial location, the labeling results are mapped to a map coordinate system to generate precursor disturbance trend ranges, forming regional spatial distribution data. The judgment criterion uses the following formula:

[0117]

[0118] Among them, R r C is the attention level label for region r. r Let r be the trend consistency coefficient for region r.

[0119] If the trend consistency coefficient for region A is set to 0.73, then:

[0120] R A =2;

[0121] The area was identified as a region of high concern, and a trend range for precursory disturbances was established.

[0122] The trend chain extraction module includes:

[0123] The sequence extraction submodule calls the precursor disturbance trend interval, extracts real-time monitoring sequence values ​​of various climate factors in the region based on the coordinates of the risk concern area, including temperature, humidity, precipitation, and wind speed, and arranges the data according to the timestamp to establish factor monitoring sequence data;

[0124] The sequence extraction submodule calls the aforementioned precursor disturbance trend interval and extracts real-time monitoring values ​​of temperature, humidity, precipitation, and wind speed within the risk-concerned area based on the coordinates of the area. It obtains temperature, humidity, precipitation, and wind speed data for area A from the risk-concern start time 00:00 to 24:00 the following day. The data is synchronously sampled at 1-hour intervals to form continuous observation data. Assuming that the temperature in area A at 00:00 is 24.5℃, humidity is 80%, precipitation is 2mm / h, and wind speed is 1.5m / s, the data is then extracted hourly up to 24 hours. The data is stored separately, and the timestamps corresponding to each type of meteorological data are retrieved. All factor data are arranged by time index. For example, in region A, the temperature data for the 5th hour is 25.7℃, humidity is 78%, precipitation is 1.5mm / h, and wind speed is 1.2m / s. The data record for the 5th hour is indexed and aligned with the data for the preceding and following time periods. This process is repeated until all 24-hour data is arranged. During this process, the extracted temperature, humidity, precipitation, and wind speed data are used to form a set of sequences at continuous time nodes. The data at time t in each meteorological factor sequence is denoted as X. i,t Where i represents a specific climate factor and t represents a time index, the difference between two adjacent points can be determined using the following formula:

[0125] ΔX i,t =X i,t -X i,t-1 ;

[0126] Where, ΔX i,t Let X be the change of the i-th meteorological factor between time t and time t-1. i,t Let X be the observation data at time t. i,t-1 Given the observation data at time t-1, and setting the temperature in region A to 25.7℃ in the 5th hour and 25.2℃ in the 4th hour, the temperature change is as follows:

[0127] ΔX 温度,5 =25.7-25.2=0.5℃;

[0128] Establish factor monitoring sequence data.

[0129] The delay determination submodule calls the factor monitoring sequence data, calculates the correlation between the numerical changes of each climate factor based on the numerical changes of each data sequence, analyzes the response relationship of multiple climate factors, and obtains the response delay coefficient.

[0130] The delay determination submodule calls the factor monitoring sequence data, performs numerical difference calculation on each climate factor sequence according to the changes in the values ​​of each climate factor in the sequence, extracts the data difference between adjacent time points of each sequence, calls the difference data of temperature, humidity, precipitation, and wind speed within the same time range, and calculates the correlation coefficient of any two climate factor difference values ​​using the Pearson correlation coefficient calculation formula:

[0131]

[0132] Among them, R xy The correlation coefficient is the difference between meteorological factors x and y. x,t and X y,t Let x and y be the differences between factors x and y at time t, respectively. and To calculate the mean of the corresponding difference series, where T is the total time period, the temperature and wind speed difference series in region A are defined as observation difference data for 5 consecutive hours. The temperature difference series are 0.4℃, 0.3℃, 0.5℃, 0.2℃, and 0.1℃, and the wind speed difference series are 0.25m / s, 0.2m / s, 0.3m / s, 0.15m / s, and 0.1m / s. Calculate the mean of the temperature difference series.

[0133] Calculate the mean of the wind speed difference series

[0134]

[0135] Calculate the correlation coefficient R between temperature and wind speed. xy :

[0136]

[0137] (0.1)(0.05)+(0)(0)+(0.2)(0.1)+(-0.1)(-0.05)+(-0.2)(-0.1)=0.005+0+0.02+0.005+0.02=0.05;

[0138]

[0139] Therefore, the correlation coefficient between temperature and wind speed in region A is calculated to be 1, indicating a strong correlation. After the correlation calculation is completed, for the two climate factor sequences that have reached a strong correlation, the numerical difference data of the two climate factors are called and compared hourly. Starting from the initial moment, the time point when the first climate factor shows a significant change (the numerical difference is greater than the set change threshold, such as the temperature difference exceeding 0.3℃) is used as the starting reference point. The numerical change of the other climate factor difference value is continuously checked hourly, and the time point when the factor shows a significant change (the wind speed difference exceeds the set change threshold, such as exceeding 0.2m / s) is recorded. Then, the two time points are called and the numerical values ​​are compared. The time of the factor that shows a significant change later is subtracted from the time of the factor that shows a significant change earlier to obtain the time difference between the two factors, which is the response delay coefficient. For example, if the threshold temperature for significant change is set to 0.3℃ and the wind speed to 0.2m / s, in region A, the temperature difference value in the 3rd hour is 0.5℃, which exceeds the temperature threshold of 0.3℃, and the temperature factor is determined to be the first to show a significant change. In the 4th hour, the wind speed difference value first exceeds the threshold to 0.25m / s. At this time, the significant change in wind speed is 1 hour later than the significant change in temperature. Therefore, the response delay coefficient between temperature and wind speed is determined to be 1 hour.

[0140] The triggering sorting submodule calls the response delay coefficient to analyze the triggering order of various climate factors in risk propagation and establish a risk trend activation path.

[0141] The triggering and sorting submodule calls the response delay coefficients, specifically the response delay coefficients between temperature, humidity, precipitation, and wind speed. It compares the magnitudes of these response delay coefficients among the climate factors, determining that the smaller value indicates an earlier response and the larger value indicates a relative lag. This clarifies the triggering order of the factors. For example, assuming the temperature-wind speed delay coefficient in region A is 0.5 hours, the precipitation-wind speed delay coefficient is 2 hours, and the temperature-precipitation delay coefficient is 1 hour, the triggering order is arranged from shortest to longest delay time. The shortest delay time is determined to be temperature-wind speed (0.5 hours), followed by temperature-precipitation (1 hour delay) and finally precipitation-wind speed (2 hours delay). The triggering order of each climate factor is then called, and the propagation sequence of climate risk within the region is structured according to the triggering order, forming a continuous triggering link between risk factors. The evolution path of risk factors within the region is recorded based on the order of the triggering links, stored as a data chain of trigger sequence, thus establishing a risk trend activation path.

[0142] The level assignment module includes:

[0143] The intensity calculation submodule calls the risk trend activation path, analyzes the frequency and amplitude of changes in climate data values ​​in each region, calculates the intensity of change, and obtains data change intensity information.

[0144] The intensity calculation submodule calls the risk trend activation path. Based on the temperature, humidity, precipitation, and wind speed monitoring data within region A, it retrieves hourly monitoring values ​​over a 24-hour period. It calculates the difference between each data point within each hour, accumulates the absolute values ​​of these differences, and divides them by the number of data points within the time period to obtain the fluctuation amplitude. For example, if the temperature data for region A is 24.5℃ in the first hour, 25.0℃ in the second hour, and 25.7℃ in the third hour, with differences of 0.5℃ and 0.7℃ respectively, then the temperature fluctuation amplitude is (0.5 + 0.7) / 2 = 0.6℃. It then retrieves the number of times the data change within the same time period, using the number of times the numerical difference exceeds the corresponding threshold per unit time as the change frequency. Assuming the temperature change threshold is 0.3℃, then both differences in region A exceed the threshold, with a change frequency of 2 times. Multiplying the fluctuation amplitude by the change frequency yields the change intensity value. The calculation formula is as follows:

[0145] I = F × A;

[0146] Where I is the intensity of change, F is the frequency of change, and A is the amplitude of fluctuation. Taking the temperature data above as an example, the frequency of change F = 2, and the amplitude of fluctuation A = 0.6℃. Substituting these values ​​into the formula:

[0147] I = 2 × 0.6 = 1.2;

[0148] The same calculation process was performed on temperature, humidity, precipitation, and wind speed respectively, and the data were then summarized to obtain information on the intensity of data changes.

[0149] The continuity analysis submodule calls the data change intensity information to detect the boundary relationship between adjacent areas. By comparing the change intensity between adjacent areas, it analyzes the spatial relationship of regional risk levels and obtains the spatial continuity coefficient.

[0150] The continuity analysis submodule retrieves data change intensity information. Based on the boundary coordinates of region A, it obtains the boundary coordinates of its neighboring region B, performs position matching on the boundary coordinates to confirm the boundary relationship between regions, and retrieves the change intensity values ​​corresponding to regions A and B. Assuming the change intensity of region A is 1.2 and that of region B is 0.8, it compares the magnitudes of the two change intensity values, calculates the absolute value of the difference between the two change intensity values, and divides it by the mean of the two change intensity values ​​to obtain the spatial continuity coefficient. The specific formula is as follows:

[0151]

[0152] Where C is the spatial continuity coefficient, I A Let I be the intensity of change in region A.B To represent the intensity of change in region B, substitute the above hypothetical data:

[0153]

[0154] The spatial continuity coefficient value is between 0 and 1. We set 0 to 0.3 as strong spatial continuity, 0.3 to 0.7 as medium continuity, and above 0.7 as weak continuity. Therefore, the spatial continuity between region A and region B is medium, and we obtain the spatial continuity coefficient.

[0155] The risk level adjustment submodule calls the spatial continuity coefficient, calculates the risk level benchmark value for multiple regions based on the change intensity value, adjusts the risk level in combination with the spatial relationship of regional risk levels, and establishes the spatial distribution value of risk level.

[0156] The risk level adjustment submodule calls the spatial continuity coefficient to set the baseline value for the regional risk level, which is calculated based on the intensity of change. First, it calls the intensity of change values ​​for three factors: temperature, precipitation, and wind speed. Then, it weights and sums these three intensity values ​​according to preset weighting coefficients. The weighting coefficients are set to 0.4 for temperature, 0.3 for precipitation, and 0.3 for wind speed. For example, in region A, the intensity of temperature change is 0.5℃, the intensity of precipitation change is 1.2 mm / h, and the intensity of wind speed change is 0.8 m / s. The formula for calculating the baseline risk level is:

[0157] L = 0.4 × T + 0.3 × R + 0.3 × W;

[0158] Where L is the baseline value for the risk level, and T, R, and W are the intensity values ​​of changes in temperature, precipitation, and wind speed, respectively. Substituting the data, we get:

[0159] L=0.4×0.5+0.3×1.2+0.3×0.8=0.2+0.36+0.24=0.8;

[0160] Next, the spatial relationship of risk levels in adjacent areas is invoked. By comparing the spatial continuity coefficient of an area with the baseline value of its risk level, for pairs of areas with a continuity coefficient greater than or equal to 2.5, if the difference between the baseline values ​​of the risk levels in the areas exceeds 0.3, the baseline value of the risk level in the area with the lower baseline value is increased by 0.1 until the difference is less than or equal to 0.3. For example, if the spatial continuity coefficient of area A and area B is 2.7, the baseline value of risk level A is 0.8, and the baseline value of risk level B is 1.2, then the baseline value of risk level in area A is adjusted to 0.9. After the baseline values ​​of risk levels in all adjacent areas are adjusted, the spatial distribution value of risk levels is formed.

[0161] The risk diffusion module includes:

[0162] The impedance calculation submodule calls the risk level spatial distribution value to obtain the surface material type, building density, and terrain elevation difference of each area. Based on the surface material friction coefficient, building density barrier value, and terrain elevation difference value, it calculates the spatial impedance value and establishes the spatial impedance coefficient.

[0163] The specific formula for calculating the spatial impedance value is as follows:

[0164]

[0165] Calculate the overall spatial impedance value;

[0166] Among them, Z t′ The spatial composite impedance value of region t′, f t′ B represents the average coefficient of friction of the surface material in region t′. t′ H represents the building density barrier value within region t′. max,t′ H represents the highest topographic elevation of region t′. min,t′ H represents the lowest topographic elevation of region t′. avg W represents the average elevation difference across all regions. f The weight representing the coefficient of friction of surface materials, W b W represents the weight of the building density barrier value. h The weight represents the difference in terrain elevation, and t′ is the index of the region.

[0167] formula:

[0168]

[0169] Detailed explanation of the formula and its calculation derivation:

[0170] The formula is used to calculate the spatial composite impedance value of each area of ​​the city, and the result is used to measure the degree to which climate risks are hindered in the process of propagation in the area.

[0171] Parameter meanings and settings:

[0172] f t′ The coefficient of friction of the regional surface material is set to 0.65;

[0173] W f The weight of the friction coefficient of the surface material is set to 0.3;

[0174] B t′ The building density barrier value for the area is set to 0.42;

[0175] W b The weight for building density barrier value is set to 0.4;

[0176] H max,t′The highest elevation in this area is set at 82 meters.

[0177] H min,t′ The lowest elevation for this area is set at 56 meters.

[0178] H avg The average elevation difference within the city's overall area is set at 24 meters;

[0179] W h The weight for the topographic elevation difference is set to 0.3;

[0180] Substitute the parameters into the formula to calculate:

[0181]

[0182] The result of 0.688 indicates that the spatial impedance value of this area is in the medium to high range. The spatial impedance value can be used for sorting or classification marking, and can serve as a boundary judgment standard in the process of identifying risk propagation trend paths. In subsequent directional analysis, it can be compared with the impedance value of the surrounding area to determine the preferred propagation path.

[0183] The directional analysis submodule calls the spatial impedance coefficient, analyzes the spatial diffusion direction of climate risk based on the obstruction strength of diffusion risk propagation between multiple regions, identifies diffusion paths, and generates risk diffusion prediction results.

[0184] The directional analysis submodule calls the spatial impedance coefficient to analyze the resistance strength to the propagation of diffusion risk between multiple regions, and gradually determines the diffusion path:

[0185] Given the spatial impedance coefficients of any two adjacent regions, calculate the absolute difference between the impedance coefficients of the two regions:

[0186] R = |Z1 - Z2|;

[0187] Where R is the impedance difference between regions, and Z1 and Z2 are the spatial impedance coefficients of two adjacent regions, respectively.

[0188] Taking regions C and D as examples, assuming the impedance coefficient of region C is 21.115 and that of region D is 18.275, the calculation is as follows:

[0189] R = |21.115 - 18.275| = 2.84;

[0190] The diffusion barrier strength is determined based on the impedance difference. An impedance difference in the range of 0-1 is defined as low barrier, 1-3 as medium barrier, and greater than 3 as high barrier. In the example, the impedance difference between region C and region D is 2.84, which belongs to the medium barrier strength.

[0191] Further, the spatial distribution value of risk level is invoked. Assuming that the risk level of region C is 4 and the risk level of region D is 2, when the difference between the high and low risk levels exceeds 1 and the barrier strength is at a low or medium level, it is determined that the risk can spread from the high-level region to the low-level region. Therefore, it is determined that the risk spreads from region C to region D, the diffusion path is identified, and the risk diffusion prediction result is generated.

[0192] The path marking submodule calls the risk diffusion prediction results, marks the diffusion risk areas by predicting the spatial diffusion trajectory of urban climate risk, and establishes climate risk prediction results.

[0193] The path marking submodule calls the risk diffusion prediction results to gradually determine the diffusion trajectory of the risk area:

[0194] Call the coordinates of the starting point and center of the diffusion direction region, assuming the center coordinates of the starting region are (x...). s ,y s ) = (30, 40), the center coordinates of the direction region are (x d ,y d ) = (42, 55), calculate the length of the diffusion trajectory:

[0195]

[0196] Where L is the length of the diffusion trajectory, (x s ,y s (x) represents the coordinates of the center point of the starting region. d ,y d () represents the coordinates of the center point of the diffusion direction region.

[0197] Substitute the above data:

[0198]

[0199] The diffusion trajectory line is invoked and overlaid with urban spatial grids. Based on the grids traversed by the diffusion trajectory line, the corresponding urban grid numbers are determined. These grid numbers are then used to mark the regions, and the areas covered by the trajectory are marked as diffusion risk areas. Finally, climate risk prediction results are established.

[0200] The application is a city climate risk monitoring and early warning system based on multi-source data fusion.

[0201] The above embodiments can be implemented, in whole or in part, by software, hardware (such as circuits), firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of the present invention are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. A semiconductor medium can be a solid-state drive.

[0202] It should be understood that the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. A and B can be singular or plural. Additionally, the character " / " in this article generally indicates an "or" relationship between the preceding and following related objects, but it can also represent an "and / or" relationship. Please refer to the context for a more accurate understanding.

[0203] In this invention, "at least one" means one or more, and "more than one" means two or more. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of a single item or a plurality of items. For example, at least one of a, b, or c can represent: a, b, c, ab, ac, bc, or abc, where a, b, and c can be a single item or multiple items.

[0204] It should be understood that, in various embodiments of the present invention, the order of the above-mentioned process numbers does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0205] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0206] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the devices, apparatuses, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0207] In the several embodiments provided by this invention, it should be understood that the disclosed devices, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0208] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0209] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0210] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0211] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A city climate risk monitoring system based on multi-source data fusion, characterized in that, The system includes: The data reconstruction module acquires multi-source meteorological monitoring data, extracts the observation density, interval value, and deviation value of multiple data sources, analyzes the reliability level of multiple data sources, and combines location information to perform data fusion on the multi-source monitoring data to generate a spatial reconstruction dataset. The precursor identification module calls the spatial reconstruction dataset to obtain time series data of various climate factors. By analyzing the consistency characteristics of the changing trends of each data in multiple regions, it marks areas of risk concern and generates precursor disturbance trend intervals. The trend chain extraction module calls the precursor disturbance trend interval, extracts meteorological factor monitoring sequence data of the risk concern area, analyzes the change sequence of each climate factor, identifies the response delay relationship between multiple climate factors, identifies the risk propagation trigger sequence, and generates risk trend activation path. The risk level assignment module calls the risk trend triggering path, and based on the frequency and amplitude of data changes, it analyzes the intensity of climate data changes in multiple regions, combines the spatial continuity between risk levels in adjacent regions, calculates and adjusts the risk levels of multiple regions, and generates spatial distribution values ​​of risk levels. The data reconstruction module includes: The difference analysis submodule acquires multi-source meteorological monitoring data, calculates the density value of observation points from multiple data sources based on the number and spatial distribution of observation points from multiple data sources, analyzes the observation interval difference of data sources, calculates the data numerical deviation of each data source, and generates a data difference coefficient. The data source rating submodule calls the data difference coefficient, and based on the data difference coefficient, grades and evaluates the credibility of each data source, identifying the credibility level value of each data source. The data fusion submodule calls the credibility level value of each data source, calculates the weight coefficient of multiple data sources based on the credibility level value, and combines the location information corresponding to the meteorological monitoring data to perform data fusion on the multi-source meteorological monitoring data and establish a spatial reconstruction dataset. The specific formula for data fusion of multi-source meteorological monitoring data is as follows: ； Calculate the fused data values; in, For spatial location The fused data value at the location, For the first Data sources in location The original monitoring values, For the first The trust level value of each data source. For the first Monitoring points of each data source to spatial location Spatial distance value, For the first The trust level value of each data source. For the first Monitoring points of each data source to spatial location Spatial distance value, The total number of all data sources participating in spatial data fusion. For the index number of each data source in the weighted calculation, This refers to the index number of each data source in the weight normalization calculation. This refers to the number of the location point to be calculated within the spatial fusion region.

2. The urban climate risk monitoring system based on multi-source data fusion according to claim 1, characterized in that, The spatial reconstruction dataset includes grid location units, fused observations, and confidence level coefficients. The precursor disturbance trend interval includes disturbance time period, disturbance factor type, and risk concern area number. The risk trend triggering path includes factor triggering sequence, response delay time difference, and trend path number. The risk level spatial distribution value specifically includes level numerical unit, level change amplitude, and level continuity index.

3. The urban climate risk monitoring system based on multi-source data fusion according to claim 1, characterized in that, The precursor identification module includes: The trend extraction submodule calls the spatial reconstruction dataset to obtain time series data of each climate factor in multiple areas of the city and establishes a data change trend curve. The consistency analysis submodule calls the data change trend curve, extracts the fluctuation direction and duration of each climate factor trend curve, analyzes the consistency of the fluctuation direction and duration of multiple data trend curves, and generates a trend consistency coefficient. The regional marking submodule calls the trend consistency coefficient to mark risk-concern areas and establish precursor disturbance trend intervals based on the consistency of the changing trends of each meteorological monitoring data.

4. The urban climate risk monitoring system based on multi-source data fusion according to claim 1, characterized in that, The trend chain extraction module includes: The sequence extraction submodule calls the precursor disturbance trend interval, extracts real-time monitoring sequence values ​​of various climate factors in the region based on the coordinates of the risk concern area, including temperature, humidity, precipitation, and wind speed, and arranges the data according to the timestamp to establish factor monitoring sequence data; The delay determination submodule calls the factor monitoring sequence data, calculates the correlation between the numerical changes of each climate factor based on the numerical changes of each data sequence, analyzes the response relationship of multiple climate factors, and obtains the response delay coefficient. The triggering sorting submodule calls the response delay coefficient to analyze the triggering order of various climate factors in risk propagation and establish a risk trend activation path.

5. The urban climate risk monitoring system based on multi-source data fusion according to claim 1, characterized in that, The level assignment module includes: The intensity calculation submodule calls the risk trend activation path, analyzes the frequency and amplitude of changes in climate data values ​​in each region, calculates the intensity of change, and obtains data change intensity information. The continuity analysis submodule calls the data change intensity information to detect the boundary relationship between adjacent areas. By comparing the change intensity between adjacent areas, it analyzes the spatial relationship of regional risk levels and obtains the spatial continuity coefficient. The risk level adjustment submodule calls the spatial continuity coefficient, calculates the risk level benchmark value for multiple regions based on the change intensity value, adjusts the risk level in combination with the spatial relationship of regional risk levels, and establishes the spatial distribution value of risk level.

6. The urban climate risk monitoring system based on multi-source data fusion according to claim 1, characterized in that, The system also includes: The risk diffusion module calls the spatial distribution value of the risk level, obtains the surface material type, building density and topographic elevation difference of each region, analyzes the spatial impedance characteristics between regions, calculates the spatial diffusion direction and path of urban climate risk, marks the diffusion risk area, and generates climate risk prediction results. The climate risk prediction results specifically refer to the diffusion direction and path, risk impact boundary, and spatial propagation intensity.

7. The urban climate risk monitoring system based on multi-source data fusion according to claim 6, characterized in that, The risk diffusion module includes: The impedance calculation submodule calls the risk level spatial distribution value to obtain the surface material type, building density, and terrain elevation difference of each area. Based on the surface material friction coefficient, building density barrier value, and terrain elevation difference value, it calculates the spatial impedance value and establishes the spatial impedance coefficient. The specific formula for calculating the spatial impedance value is as follows: ； Calculate the overall spatial impedance value; in, Representative area Spatial composite impedance value, Representative area The average coefficient of friction of the surface material in the interior, Representative area Internal building density barrier value, Representative area The highest topographic elevation, Representative area The lowest terrain elevation, This represents the average elevation difference across all regions. The weight representing the coefficient of friction of surface materials. The weight representing the building density barrier value, The weights representing the differences in terrain elevation. For indexing a region; The directional analysis submodule calls the spatial impedance coefficient, analyzes the spatial diffusion direction of climate risk based on the obstruction strength of diffusion risk propagation between multiple regions, identifies diffusion paths, and generates risk diffusion prediction results. The path marking submodule calls the risk diffusion prediction results, marks the diffusion risk areas by predicting the spatial diffusion trajectory of urban climate risk, and establishes climate risk prediction results.

8. An application for urban climate risk monitoring and early warning based on multi-source data fusion, characterized in that: The application is used to carry the urban climate risk monitoring system based on multi-source data fusion according to any one of claims 1-7.

Images