Urban carbon emission statistical system based on multi-source data fusion

Abstract

The invention discloses an urban carbon emission statistical system based on multi-source data fusion. The system comprises a data acquisition module, a data processing module, a carbon emission accounting model construction module, a high-resolution spatialization module and a visualization and management module. The data acquisition module is used for acquiring multi-source data including statistical yearbook data, energy consumption data, night light data, point of interest (POI) data and land utilization / coverage data. And the data processing module performs cleaning, alignment and dimension reduction processing on the data. The carbon emission accounting model construction module constructs a sub-industry carbon emission spatialization model in combination with night light data and POI data. The system uses an improved clustering algorithm to accurately classify different urban areas, and combines multi-source data weight distribution to solve the problems of one-sided statistics, data missing, poor timeliness and low spatial resolution in the prior art. According to the invention, high-precision, real-time and dynamic monitoring and zoning statistics of urban carbon emission can be realized.

Description

Technical Field

[0001] This invention relates to the field of environmental monitoring and big data processing technology, specifically to an urban carbon emission statistics system based on multi-source data fusion. Background Technology

[0002] Carbon emissions generally refer to greenhouse gas emissions. Greenhouse gas emissions cause the greenhouse effect, leading to a rise in global temperatures. With the acceleration of industrialization and urbanization, cities have become one of the main sources of greenhouse gas emissions. Currently, my country's urbanization rate is approximately 65%, and in the future, population and economic activities will continue to concentrate in cities, foreseeably further increasing the proportion of carbon emissions from Chinese cities. Energy consumption, industrial production, transportation, and residential activities in cities generate large amounts of greenhouse gases such as carbon dioxide, significantly impacting global climate change.

[0003] Scientific accounting of urban carbon emissions is the core and foundation of urban carbon emission management. Current urban carbon emission accounting primarily employs manual statistical methods, with various industries reporting carbon emission data from the bottom up, ultimately resulting in a comprehensive carbon emission estimate. While this method, based on IPCC (Intergovernmental Panel on Climate Change) guidelines and provincial inventories, offers high accuracy, it also has significant limitations:

[0004] (1) Low efficiency: It involves many industries, and the types of fuels and processes are complex. It requires specialized personnel to compile and calculate the list, resulting in a large time lag.

[0005] (2) Statistical bias and data gaps: It is easily affected by the completeness of data reporting and is difficult to cover all emission sources;

[0006] (3) Low spatial resolution: Existing statistical methods are usually based on administrative divisions, lacking spatial information within cities, making it difficult to reflect the detailed carbon emissions of different areas and industries within cities, which makes it difficult to estimate carbon dioxide emissions of functional zones with high spatial resolution.

[0007] (4) Static estimation, lack of real-time performance: Traditional methods rely heavily on macro-statistical data, which cannot reflect the changing trends of urban carbon emissions and oxygen consumption in real time, and are difficult to adapt to the real-time monitoring needs of rapid urban development and smart city management.

[0008] (5) Neglecting the role of ecosystems: Most estimation methods focus on carbon emissions from human activities and do not adequately consider the carbon sequestration and oxygen release role of urban ecosystems (such as forests, grasslands, wetlands, etc.), resulting in incomplete estimation results.

[0009] Furthermore, due to the different urban functional zones (industry, transportation, residential life, service industry, etc.) located in different areas, carbon emissions exhibit significant spatial variations within cities. Although some studies have used satellite-observed nighttime light intensity (NTL) data for carbon dioxide emission spatialization, there are relatively few studies that integrate satellite observation data and point-of-interest density data to establish high-resolution carbon emission data for different sectors. Meanwhile, in terms of urban feature classification, traditional clustering methods (such as K-means) are prone to getting trapped in local optima when dealing with nonlinear distributions and mixed feature parameters, which may lead to biases in the clustering of urban functional zones, thus affecting the spatialization accuracy of carbon emissions.

[0010] In summary, there is an urgent need for an urban carbon emission statistics system that can integrate multi-source data, ensuring accuracy while achieving high spatial resolution and real-time dynamic monitoring. To this end, we propose a novel urban carbon emission statistics system based on multi-source data fusion. Summary of the Invention

[0011] The purpose of this invention is to provide an urban carbon emission statistics system based on multi-source data fusion. This system integrates multi-source information such as statistical data, nighttime light data, and POI data to achieve high-precision spatialization, real-time accounting, and zonal statistics of urban carbon emissions, providing scientific support for urban low-carbon development planning.

[0012] To achieve the above objectives, the present invention provides the following technical solution: an urban carbon emission statistics system based on multi-source data fusion, comprising:

[0013] The data acquisition module is used to collect multi-source data related to urban carbon emissions. The multi-source data includes statistical yearbook data, energy consumption inventory data, satellite-observed nighttime light data, point of interest (POI) data, and urban land use data.

[0014] The data preprocessing module is used to clean and normalize the multi-source data, and to unify the data from different sources in terms of time and space.

[0015] The urban functional zone identification module is used to identify the functional zones of the urban area based on the POI data and land use data, using a clustering algorithm to divide the city into industrial, transportation, residential, service and ecological functional zones.

[0016] The carbon emission spatialization model building module is used to build a carbon emission spatialization model. Based on the characteristics of different functional zones, it assigns carbon emission weights to each zone and combines nighttime light data and energy consumption data to map statistical carbon emission data into a high spatial resolution grid.

[0017] The carbon emission dynamic accounting module is used to perform real-time accounting and statistics on carbon emissions of different functional zones within the city, based on the aforementioned carbon emission spatialization model and combined with real-time updated energy flow data and IoT monitoring data.

[0018] The visualization and management module is used to spatially visualize the carbon emission statistics obtained from the accounting, and provides interfaces for data query, trend analysis and emission reduction policy evaluation.

[0019] Preferably, the clustering algorithm used in the urban functional area identification module is an improved clustering algorithm, which includes:

[0020] Kernel principal component analysis was used to reduce the dimensionality of POI density data and land use characteristics;

[0021] Clustering algorithms based on density peaks or optimized K-means algorithms are used to cluster the dimensionality-reduced data to determine the functional type of each grid.

[0022] Preferably, the carbon emission spatialization model construction module is specifically used for:

[0023] Establish a regression relationship between total carbon emissions and nighttime light intensity;

[0024] By introducing POI density as an auxiliary variable, the spatial distribution of carbon emissions in different functional zones such as industry, transportation, and service industries is corrected.

[0025] By utilizing ecological land use information from land use data and combining it with vegetation carbon sequestration models, the carbon sequestration of urban ecosystems is calculated and deducted from total carbon emissions statistics or listed separately.

[0026] Preferably, the data preprocessing module includes:

[0027] Time alignment units are used to match annually published statistical data with high-frequency satellite remote sensing data and IoT real-time data in terms of time scale.

[0028] Spatial alignment cells are used to map statistical data at the administrative region scale to a spatial grid consistent with satellite data through spatial interpolation or resampling.

[0029] Preferably, the visualization and management module includes:

[0030] The carbon emission heat map generation unit is used to display the carbon emission intensity of different regions and industries on the city map using different color depths;

[0031] The time trend analysis unit is used to display the curve of urban carbon emissions over time and to mark the impact of major events or policy implementation nodes on carbon emissions.

[0032] Preferably, the system also includes a data verification module, which compares the carbon emission results estimated based on the model with the measured data from ground monitoring stations, and adjusts the parameters of the carbon emission spatialization model based on error feedback.

[0033] Preferably, when spatializing carbon emissions using nighttime light data and POI data, the following logic is adopted:

[0034] For industrial and commercial service areas, a higher weight is assigned to POI density to reflect high-energy-consuming activities;

[0035] For residential areas, nighttime light data is given a higher weight to reflect energy consumption.

[0036] For traffic functional zones, a joint weight allocation is performed by combining road network density and traffic facility data in POIs.

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

[0038] 1. This invention achieves a refined mapping of carbon emission data from the administrative scale to the high-resolution grid scale by combining nighttime light data and POI data. It can identify emission hotspots in different functional zones within the city, improve spatial resolution, and not only consider carbon emissions from human activities, but also assess the carbon sequestration role of the urban ecosystem by combining land use data, providing a more comprehensive perspective on net carbon emissions and improving statistical comprehensiveness.

[0039] 2. This invention introduces IoT data and real-time energy flow data, combined with high-frequency satellite remote sensing data, to overcome the problem of poor timeliness of traditional statistical methods, realize dynamic monitoring of urban carbon emissions, and enhance timeliness. At the same time, it adopts an improved clustering algorithm to solve the bias problem of traditional algorithms in urban feature classification, lays the foundation for accurate estimation of carbon emissions by type, and optimizes functional area identification. Attached Figure Description

[0040] Figure 1 This is a logical diagram of the system in an embodiment of the present invention;

[0041] Figure 2 This is a schematic diagram of the workflow in an embodiment of the present invention. Detailed Implementation

[0042] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0043] Example

[0044] Please see Figure 1 - Figure 2 This invention provides an urban carbon emission statistics system based on multi-source data fusion, including a data acquisition module, a data preprocessing module, an urban functional area identification module, a carbon emission spatialization model construction module, a carbon emission dynamic accounting module, a visualization and management module, and a data verification module.

[0045] 1. Data Acquisition Module

[0046] This module is used to collect multi-source heterogeneous data related to urban carbon emissions required for model building.

[0047] (1) Statistical data: including statistical yearbooks, energy balance sheets (energy consumption inventory data), industrial energy consumption data, etc., mainly used to determine the total carbon emission benchmark of the city;

[0048] (2) Nighttime light data (NTL): Remote sensing data from satellites such as NPP-VIIRS are used to reflect the intensity of human activities and the distribution of energy consumption at night;

[0049] (3) Point of Interest (POI) data: collect geographic coordinate data including factories, shopping malls, schools, hospitals, gas stations, road intersections, etc., to identify functional areas and assist in spatialization;

[0050] (4) Urban land use / cover data; derived from remote sensing interpretation or national land planning data, distinguishing between construction land, cultivated land, forest land, grassland, water area, etc., for ecosystem carbon sequestration calculation and functional zone delineation.

[0051] 2. Data Preprocessing Module

[0052] Because various types of data have different formats and spatiotemporal scales, preprocessing is required. This module is used to clean and normalize multi-source data, and to unify data from different sources in terms of time and spatial scale.

[0053] The data preprocessing module includes a time alignment unit and a spatial alignment unit: the time alignment unit is used to match the annually published statistical data with high-frequency satellite remote sensing data and IoT real-time data in terms of time scale (e.g., using time series interpolation technology); the spatial alignment unit is used to map the statistical data at the administrative region scale to a spatial grid consistent with the satellite data (e.g., a 1km×1km or 500m×500m grid) through spatial interpolation or resampling.

[0054] Specifically, including:

[0055] (1) Data cleaning: Remove duplicate and outlier points from POI data; repair background noise in nighttime light data.

[0056] (2) Spatial gridding: Establish a unified spatial grid system (e.g., 1km×1km), convert vector data (POI, administrative boundaries) into raster data, and calculate the POI density, total light value, etc. in each grid.

[0057] (3) Time alignment: For annual statistical data, it is allocated to the months by interpolation; for satellite data, it is matched with the statistical period by synthesizing monthly products.

[0058] 3. Urban Functional Zone Identification Module

[0059] Based on the POI data and land use data, a clustering algorithm is used to identify functional zones in urban areas, dividing the city into industrial, transportation, residential, service, and ecological functional zones.

[0060] The clustering algorithm used in the urban functional zone identification module is an improved clustering algorithm. This system utilizes kernel principal component analysis (KPCA) to reduce the dimensionality of POI density data and land use characteristics, extracting nonlinear features. Then, it employs a density peak-based clustering algorithm or an optimized K-means algorithm (such as K-means++ initialization) to cluster the dimensionality-reduced data, determining the functional type of each grid, thereby improving the accuracy of functional zone identification.

[0061] Specifically, this module aims to address the issue of significant spatial differences in carbon emissions within cities by first dividing cities into functional zones.

[0062] (1) Feature construction: For each grid, the density of POI type (industrial POI density, living service POI density, transportation POI density) and the proportion of land use type are statistically analyzed.

[0063] (2) Clustering analysis: Given the shortcomings of the K-means algorithm, which is sensitive to initial values ​​and prone to getting trapped in local optima, this system adopts an optimized clustering strategy. First, the high-dimensional features are reduced using kernel principal component analysis (KPCA) to extract the main feature components; then, the cluster centers are initialized using the K-means++ algorithm, or the density peak clustering algorithm (CLARANS) is used to divide the urban grid into industrial areas, commercial service areas, residential areas, transportation hub areas and ecological areas.

[0064] 4. Carbon Emission Spatialization Model Construction Module

[0065] This module is the core of the system, used to build a spatial model of carbon emissions. Based on the characteristics of different functional zones, it assigns carbon emission weights to each zone and combines nighttime light data and energy consumption data to map statistical carbon emission data into a high spatial resolution grid, so as to realize the decomposition of total carbon emission data into grid data from the top down.

[0066] Specifically, based on the IPCC guidelines, the carbon emission spatialization model construction module includes:

[0067] (1) Basic regression model: Establish a linear or nonlinear regression relationship between the total emissions of carbon cities or administrative regions and the nighttime light intensity (total light value) of the region;

[0068] (2) Multi-source data fusion correction: Since nighttime light may have a saturation effect in industrial and residential areas or weak signal in sparse areas (this is because simple nighttime light data can only reflect the intensity of human activities and it is difficult to distinguish specific industries), POI density is introduced as an auxiliary variable to correct the spatial distribution of carbon emissions in different functional zones such as industry, transportation, and service industries.

[0069] Set grid The estimated carbon emissions are The model is then represented as:

[0070] in, For nighttime light intensity, , , These are density indices for industrial, service, and residential POI categories (which need to be standardized). , , , is the regression coefficient.

[0071] By introducing POI density, we can more accurately depict the distribution of factories in industrial zones, the distribution of service industries in commercial zones, and the distribution of transportation networks, thereby improving the accuracy of spatialization.

[0072] (3) Zoning weight adjustment: In conjunction with the urban functional areas identified in step 3, when using nighttime light data and POI data to spatialize carbon emissions, the following differentiated weight allocation logic is adopted to zonally constrain the above model: For industrial areas and commercial service areas, POI density is given a higher weight to reflect the concentration of point sources and high-energy-consuming activities; for residential areas, nighttime light data is given a higher weight to reflect the distribution of residential energy consumption in an area; for transportation functional areas, joint weight allocation is carried out by combining road network density and transportation facility data in POI.

[0073] For example, in grids identified as "industrial zones", the weight of industrial POI density is increased. In "residential areas", increase the weight of nighttime lighting. Or the weight of the POI where the resident resides.

[0074] (4) Ecological carbon sequestration calculation: For grids identified as ecological zones, the net primary productivity (NPP) of the grid is calculated using a light energy utilization model (such as the CASA model), combined with vegetation type (NDVI data) and meteorological data. The NPP is then converted into carbon sequestration as a deduction item for carbon emissions. It is deducted in the total carbon emissions statistics or listed separately to form net carbon emissions statistics to reflect the city's net carbon emissions.

[0075] 5. Dynamic carbon emission accounting module

[0076] To address the issue of poor timeliness in traditional IPCC methods, this module introduces a dynamic update mechanism. This mechanism is used to perform real-time calculation and statistics of carbon emissions in different functional zones within a city, based on the aforementioned carbon emission spatialization model and combined with real-time updated energy flow data and IoT monitoring data.

[0077] Specifically:

[0078] (1) Use Internet of Things (IoT) technology to obtain real-time energy consumption data, traffic flow data, etc. of key enterprises.

[0079] (2) Use these high-frequency real-time data as driving variables and input them into the above spatialization model to update the carbon emission estimates of the corresponding grid.

[0080] (3) The system can output carbon emission statistics results at different time granularities such as hour, day, week, and month, so as to realize real-time monitoring.

[0081] 6. Visualization and Management Module

[0082] It is used to spatially visualize the carbon emission statistics obtained from the accounting, and provides interfaces for data query, trend analysis and emission reduction policy evaluation.

[0083] The visualization and management module includes a carbon emission heat map generation unit and a time trend analysis unit: the carbon emission heat map generation unit is used to display the carbon emission intensity of different regions and industries on the city map with different color depths, and intuitively identify high emission areas; the time trend analysis unit is used to display the change curve of urban carbon emissions over time, and mark the impact of major events or policy implementation nodes on carbon emissions, and evaluate the policy effect.

[0084] Specifically:

[0085] (1) Statistical reports: Automatically generate carbon emission statistical reports at the city, district and street levels, and display them by industry (industry, transportation, residents, etc.).

[0086] (2) Spatial distribution map: Display carbon emission heat map on the city GIS map, clearly showing high emission areas (hot spots) and low emission areas.

[0087] (3) Trend analysis: Generate time series curves to show the changing trend of carbon emission intensity and help evaluate the effectiveness of emission reduction policies. For example, analyze whether the carbon emissions in an industrial zone have decreased significantly after the implementation of production restriction policies.

[0088] 7. Data Validation Module

[0089] To ensure accuracy, the system incorporates a verification mechanism. This mechanism compares the carbon emission estimates based on the model with actual data from ground monitoring stations, and adjusts the parameters of the carbon emission spatialization model based on error feedback, thereby enabling the model to self-optimize.

[0090] Specifically:

[0091] (1) Collect a small amount of data from ground-based high-precision CO2 concentration monitoring stations (or flux tower data) in the city.

[0092] (2) Input the grid carbon emission data estimated by the model into the atmospheric diffusion model to simulate the concentration at the monitoring stations, or directly compare the regional total trend.

[0093] (3) Calculate the error between the simulated value and the measured value. If the error exceeds the threshold, use machine learning algorithms (such as neural networks) to adjust the parameter weights of the spatialization model in reverse.

[0094] Through the above system, the present invention effectively integrates multi-source data, overcomes the limitations of a single data source, and not only realizes the total carbon emission statistics, but also achieves high spatiotemporal resolution and refined statistics, providing a powerful technical means for low-carbon management of smart cities.

[0095] Workflow:

[0096] Phase 1: Multi-source data acquisition and preprocessing

[0097] 1. Data Acquisition:

[0098] Macroeconomic statistics: Obtain basic data such as annual / monthly total energy consumption and industrial output of cities and administrative regions from statistical yearbooks and energy balance sheets (as a benchmark for total carbon emissions).

[0099] Remote sensing data: acquire satellite-observed nighttime light (NTL) data to reflect the intensity of nighttime human activity; acquire land cover / use data (LUCC) to identify surface vegetation types.

[0100] Geographic information data: Collecting city point of interest (POI) data (such as the location of factories, shopping malls, residences, and gas stations).

[0101] Real-time monitoring data: High-frequency data such as real-time energy consumption and traffic flow of key energy-consuming enterprises are acquired through IoT sensors.

[0102] 2. Data cleaning and standardization:

[0103] Remove duplicate points and outliers from the POI data; remove background noise and temporary light source interference such as fire points from the nighttime light data. Normalize the data for different units.

[0104] 3. Spatiotemporal alignment:

[0105] Spatial gridding: Establish a unified spatial grid system (e.g., a 1km×1km grid) to map data such as administrative boundaries, POI points, and light pixels to the same grid coordinates.

[0106] Time alignment: Low-frequency annual statistics are broken down into monthly or quarterly data using an interpolation algorithm and aligned with high-frequency satellite and IoT data on the timeline.

[0107] Phase Two: Intelligent Identification of Urban Functional Zones

[0108] 1. Feature Extraction: Calculate the density of various POI types (e.g., industrial POI density, catering POI density, transportation POI density) within each grid. Combine this with the proportion of various land uses (construction land, forest land, water area, etc.) in the land use data.

[0109] 2. Dimensionality reduction: Kernel principal component analysis (KPCA) is used to reduce the dimensionality of high-dimensional eigenvectors and extract nonlinear principal components that can characterize the functional region attributes.

[0110] 3. Clustering Partitioning: The grid is clustered using improved clustering algorithms (such as optimized K-means or density kurtosis-based clustering algorithms).

[0111] Output results: The city area is automatically divided into industrial zones, residential areas, commercial service zones, transportation hub zones, ecological functional zones, etc.

[0112] Phase 3: Construction and Calculation of Spatial Model for Carbon Emissions

[0113] 1. Establish a baseline relationship: Based on the total energy carbon emissions in the statistical yearbook and the corresponding total nighttime light emissions in the region, establish a preliminary regression relationship.

[0114] 2. Multi-source data fusion and weight allocation: POI density is introduced as a spatialization auxiliary variable. Based on the functional area types identified in the second stage, differentiated weights are assigned to different grids:

[0115] Industrial Zones: Industrial POI density is given a higher weight to correct for underestimation that may be caused by light saturation.

[0116] Residential areas: Nighttime lighting data is given higher weight to reflect daily energy consumption.

[0117] Commercial / Transportation Zone: Weighting is based on a combination of commercial POIs and road network density.

[0118] 3. Grid carbon emission calculation:

[0119] Using a fusion model, the total carbon emissions of the administrative region are decomposed and allocated to each grid in a top-down manner, resulting in a preliminary grid carbon emission distribution map.

[0120] 4. Ecosystem carbon sequestration and restoration:

[0121] For the grid of "ecological functional zones", the amount of carbon sequestration in the area is calculated using a vegetation carbon sequestration model (combining NDVI data and meteorological parameters).

[0122] Calculate net carbon emissions: Net carbon emissions = carbon emissions from human activities - carbon sequestration by ecosystems.

[0123] Phase 4: Dynamic Accounting and Data Verification

[0124] 1. Real-time dynamic updates: Input the real-time energy consumption data collected by the Internet of Things into the model and dynamically adjust the carbon emission estimates of the corresponding grid, thereby realizing the transformation from "annual statistics" to "real-time / near real-time monitoring".

[0125] 2. Data Validation and Feedback: The model calculation results are input into the atmospheric diffusion model to simulate the CO2 concentration at ground monitoring stations. The results are compared with the measured concentrations at the ground monitoring stations to calculate the error.

[0126] Model optimization: If the error exceeds the threshold, the parameter weights in the spatialized model are automatically adjusted through a feedback mechanism to continuously improve the accuracy.

[0127] Phase 5: Visual Analysis and Management Decision-Making

[0128] 1. Statistical report generation: Automatically summarizes carbon emission statistical reports at the city, district, and street levels, as well as by industry (industry, transportation, services, etc.).

[0129] 2. Visualization: Generate carbon emission heat maps: Display the carbon emission intensity of each grid on the city map with different colors, intuitively locating high-emission hotspots.

[0130] Generate a time-series trend chart: showing the curve of carbon emissions changing over time.

[0131] The above are merely preferred embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A city carbon emission statistics system based on multi-source data fusion, characterized in that, include: The data acquisition module is used to collect multi-source data related to urban carbon emissions. The multi-source data includes statistical yearbook data, energy consumption inventory data, satellite-observed nighttime light data, point of interest (POI) data, and urban land use data. The data preprocessing module is used to clean and normalize the multi-source data, and to unify the data from different sources in terms of time and space. The urban functional zone identification module is used to identify the functional zones of the urban area based on the POI data and land use data, using a clustering algorithm to divide the city into industrial, transportation, residential, service and ecological functional zones. The carbon emission spatialization model building module is used to build a carbon emission spatialization model. Based on the characteristics of different functional zones, it assigns carbon emission weights to each zone and combines nighttime light data and energy consumption data to map statistical carbon emission data into a high spatial resolution grid. The carbon emission dynamic accounting module is used to perform real-time accounting and statistics on carbon emissions of different functional zones within the city, based on the aforementioned carbon emission spatialization model and combined with real-time updated energy flow data and IoT monitoring data. The visualization and management module is used to spatially visualize the carbon emission statistics obtained from the accounting, and provides interfaces for data query, trend analysis and emission reduction policy evaluation.

2. The urban carbon emission statistics system based on multi-source data fusion according to claim 1, characterized in that, The clustering algorithm used in the urban functional area identification module is an improved clustering algorithm, which includes: Kernel principal component analysis was used to reduce the dimensionality of POI density data and land use characteristics; Clustering algorithms based on density peaks or optimized K-means algorithms are used to cluster the dimensionality-reduced data to determine the functional type of each grid.

3. The urban carbon emission statistics system based on multi-source data fusion according to claim 1, characterized in that, The carbon emission spatialization model construction module is specifically used for: Establish a regression relationship between total carbon emissions and nighttime light intensity; By introducing POI density as an auxiliary variable, the spatial distribution of carbon emissions in different functional zones such as industry, transportation, and service industries is corrected. By utilizing ecological land use information from land use data and combining it with vegetation carbon sequestration models, the carbon sequestration of urban ecosystems is calculated and deducted from total carbon emissions statistics or listed separately.

4. The urban carbon emission statistics system based on multi-source data fusion according to claim 1, characterized in that, The data preprocessing module includes: Time alignment units are used to match annually published statistical data with high-frequency satellite remote sensing data and IoT real-time data in terms of time scale. Spatial alignment cells are used to map statistical data at the administrative region scale to a spatial grid consistent with satellite data through spatial interpolation or resampling.

5. The urban carbon emission statistics system based on multi-source data fusion according to claim 1, characterized in that, The visualization and management module includes: The carbon emission heat map generation unit is used to display the carbon emission intensity of different regions and industries on the city map using different color depths; The time trend analysis unit is used to display the curve of urban carbon emissions over time and to mark the impact of major events or policy implementation nodes on carbon emissions.

6. The urban carbon emission statistics system based on multi-source data fusion according to claim 1, characterized in that, The system also includes a data verification module, which compares the carbon emission estimates based on the model with the measured data from ground monitoring stations and adjusts the parameters of the carbon emission spatialization model based on error feedback.

7. The urban carbon emission statistics system based on multi-source data fusion according to claim 1, characterized in that, When spatializing carbon emissions using nighttime light data and POI data, the following logic is adopted: For industrial and commercial service areas, a higher weight is assigned to POI density to reflect high-energy-consuming activities; For residential areas, nighttime light data is given a higher weight to reflect energy consumption. For traffic functional zones, a joint weight allocation is performed by combining road network density and traffic facility data in POIs.

Images