Multi-scale cooperative urban residential district carbon reduction and sink increase implementation method, device and medium

Abstract

The application relates to the computer technical field, in particular to a multi-scale cooperative urban residential area carbon reduction and sink increase implementation method, equipment and medium, the method comprises the following steps: for each spatial form scale in a plurality of spatial form scales, energy consumption status evaluation is carried out; for each spatial form scale, optimization is carried out based on the energy consumption status evaluation, and a single-scale optimization scheme is generated; based on the correlation between different spatial form scales, a cross-scale correlation coupling optimization scheme is generated; based on the basic carbon reduction amount and the sink increase amount corresponding to each single-scale optimization scheme respectively and the coupling carbon reduction and sink increase amount corresponding to each cross-scale correlation coupling optimization scheme respectively, a total carbon reduction and sink increase amount is obtained; and target matching is carried out with a pre-set multi-layer hierarchical target. Multi-scale system coverage: the spatial form scales such as buildings, traffic, energy and green land are uniformly included in the evaluation and optimization framework, the traditional single-scale limitation is broken, and full-factor systematic carbon reduction and sink increase design of the residential area is realized.

Description

Technical Field

[0001] This invention relates to a method, equipment, and medium for achieving carbon reduction and sequestration in urban residential areas based on multi-scale collaborative approaches, and belongs to the field of computer technology. Background Technology

[0002] Traditional residential renewal projects often limit carbon reduction design to a single scale (such as building energy-saving retrofits or the addition of photovoltaics), lacking systematic multi-factor coordination. This results in the carbon reduction potential not being fully realized, specifically manifested in the following ways: 1. Limited Scale: Traditional solutions often focus on individual buildings or independent facilities, failing to incorporate transportation, pipelines, green spaces, and energy systems into a unified design framework, making it difficult to achieve full-chain carbon emission reduction; 2. Lack of coupling effect: The synergistic carbon reduction potential between elements of different scales (such as photovoltaic power generation and charging facilities, green space layout and ventilation corridors) has not been quantitatively explored, and the carbon reduction benefits have a ceiling; 3. Vague accounting system: There is no unified accounting model for carbon emission reduction and carbon sink increase before and after the update, making it difficult to accurately assess changes in carbon density and support scientific decision-making; 4. Inadequate target matching: No quantitative correspondence has been established between carbon reduction effects and low-carbon, near-zero-carbon, and zero-carbon targets, and the designed strategies lack target orientation and differentiation.

[0003] Therefore, there is an urgent need for a design method and system for carbon reduction and sequestration enhancement in residential area-level renewal that features multi-scale collaboration, quantifiable coupling increments, scientific accounting, and hierarchical matching of objectives, in order to fill the gaps in traditional solutions. Summary of the Invention

[0004] This invention addresses the shortcomings of existing technologies by providing a method for achieving carbon reduction and sequestration in urban residential areas based on multi-scale collaborative approaches, comprising: For each spatial scale across multiple spatial scales, an assessment of the current energy consumption status is conducted using the corresponding geospatial data and residential energy consumption environmental time-series data collected. For each spatial scale, optimization is performed based on the energy consumption status assessment to generate a single-scale optimization scheme. Based on the correlation between different spatial scales, a cross-scale correlation coupling optimization scheme is generated on the basis of the single-scale optimization scheme. Based on the basic carbon reduction and carbon sequestration amounts corresponding to each single-scale optimization scheme, and the coupled carbon reduction increments corresponding to each cross-scale coupled optimization scheme, the total carbon reduction and carbon sequestration amount is obtained. Based on the total amount of carbon reduction, target matching is performed with pre-set multi-level hierarchical targets.

[0005] In one example, an assessment of the current energy consumption status is conducted using the collected geospatial data and residential energy consumption environmental time-series data, specifically including: The drones acquire the corresponding remote sensing image data and digital surface models, and the sensors collect the corresponding residential area energy time series data and environmental data. By using point cloud registration and timestamp synchronization, the remote sensing image data, the digital surface model, the energy time series data, and the environmental data are spatially registered and temporally aligned to obtain residential area digital twin data. Based on the digital twin data of the residential area, the corresponding core features are extracted; the core features include the year of construction of the residential area, the slope of the terrain, and the degree of functional mixing. A scene classifier is constructed based on the random forest algorithm. The core features are used as input to determine the corresponding residential area type and match the corresponding evaluation model. The evaluation model contains different evaluation dimensions and / or the evaluation weights of each evaluation dimension for different residential area types. The energy consumption status is assessed using the assessment model based on the digital twin data of the residential area to obtain energy efficiency assessment data. The energy efficiency assessment data and on-site sampling data are then sampled and verified using the Monte Carlo simulation method. When the data variation coefficient reaches a preset value, the energy efficiency assessment data is reassessed and / or the on-site sampling data is supplemented.

[0006] In one example, the spatial morphological scale includes point scale, line scale, surface scale, and volume scale; The assessment scope at the point scale corresponds to the building itself and independent facilities and equipment in urban residential areas; The assessment scope at the linear scale corresponds to transportation facilities and pipeline systems in urban residential areas; The assessment scope at the surface scale corresponds to ecological green spaces and public spaces in urban residential areas; The volumetric scale assessment range corresponds to the overall collaborative system of the residential area obtained by integrating the assessment ranges corresponding to the point scale, the line scale, and the surface scale.

[0007] In one example, based on the correlation between different spatial morphological scales, a cross-scale correlation coupling optimization scheme is generated on the basis of the single-scale optimization scheme, specifically including: Based on the digital twin data of the residential area, a graph neural network is used to construct a full-element correlation map of the residential area. In the full-element correlation map of the residential area, the scale elements corresponding to each spatial morphology scale are used as nodes, and the correlation relationships corresponding to each scale element are used as edges. The correlation relationships include material flow relationships, energy flow relationships, and spatial relationships. Based on the community's full-element correlation map, strong correlations between nodes are identified, and the synergistic carbon reduction sensitivity of each strong correlation is determined using response surface methodology. The carbon reduction increment per unit renovation cost is obtained based on the synergistic carbon reduction sensitivity, and the nodes are prioritized. Based on priority, select the nodes corresponding to the highest number of strong correlations, and generate corresponding cross-scale correlation coupling optimization schemes through a genetic algorithm based on a pre-set coupling technology measure library. The cross-scale correlation and coupling optimization scheme is simulated using the digital twin data of the residential area, and the final cross-scale correlation and coupling optimization scheme is output.

[0008] In one example, based on a pre-defined library of coupling techniques, a genetic algorithm is used to generate a corresponding cross-scale correlation coupling optimization scheme, specifically including: For the strong point-line correlation corresponding to point and line scales, a long short-term memory network is selected from a preset coupling technology library. The time series data of the residential area's energy consumption environment is used as training data to train a load-output prediction model. The load-output prediction model is then used to predict the dynamic characteristics of energy supply and demand in the residential area during specific time periods. Based on the dynamic characteristics of energy supply and demand during specific time periods, corresponding dynamic energy supply and demand scheduling strategies are added to the cross-scale correlation coupling optimization scheme generated by the genetic algorithm. To address the strong line-area correlation between line and area scales, the U-Net semantic segmentation model and Frechet distance algorithm are selected from a pre-defined coupling technology library. Multi-source raster image preprocessing is performed on the geospatial data, and pixel segmentation is performed using the U-Net semantic segmentation model to extract corridor elements. Furthermore, spatial overlay is applied to various types of corridor elements, and the spatial overlap of corridor centerlines for each type of corridor element is calculated using the Frechet distance algorithm to obtain composite utilization areas. Based on these composite utilization areas, spatial conflict detection and conflict avoidance are performed during the cross-scale correlation coupling optimization process generated by a genetic algorithm. To address the strong point-to-area correlation between point and area scales, the U-Net semantic segmentation model, building information model parsing, and minimum spanning tree algorithm are selected from a pre-defined coupling technology library. For the residential area's digital twin data, pixel segmentation is performed using the U-Net semantic segmentation model, and model parsing is performed using the building information model parsing to identify carbon sink nodes corresponding to the point-scale assessment range. Based on the minimum spanning tree algorithm, using each carbon sink node as an endpoint, a strip-shaped space as a wiring constraint, and aiming to minimize the total path length and the spatial modification of the residential area, an ecological corridor path connecting each carbon sink node is planned and generated. The connectivity index of the three-dimensional carbon sink network corresponding to the volumetric scale assessment range is obtained using the connectivity index of landscape ecology. Finally, a cross-scale correlation coupling optimization scheme for the photovoltaic panel layout is generated using a genetic algorithm, with the objective functions of maximizing photovoltaic installed capacity, maximizing carbon sink volume, and minimizing transmission distance.

[0009] In one example, the cross-scale correlation and coupling optimization scheme is simulated using the digital twin data of the residential area, and the final cross-scale correlation and coupling optimization scheme is output, specifically including: The cross-scale correlation and coupling optimization scheme is simulated using the digital twin data of the residential area; If the simulation results do not reach the verification threshold, the core influencing factors of the deviation are located by Pearson correlation coefficient analysis. The identification dimensions include: photovoltaic installed capacity, energy storage equipment capacity, charging pile layout, transmission distance, scheduling strategy parameters, and pump station operating power. Based on the core influencing factor of deviation, the particle swarm optimization algorithm is used to automatically iterate and optimize the core influencing factor of deviation with the objective function of maximizing photovoltaic absorption rate and minimizing grid electricity consumption, and with the verification threshold as the constraint condition. The optimized parameter combination of the identification dimension is then output until the verification threshold is reached or the scheme reconstruction is triggered.

[0010] In one example, based on the basic carbon reduction and carbon sequestration amounts corresponding to each single-scale optimization scheme, and the coupled carbon reduction increments corresponding to each cross-scale coupled optimization scheme, the total carbon reduction increment is obtained, specifically including: Calculate the basic carbon reduction corresponding to each single-scale optimization scheme separately, and sum them to obtain the total basic carbon reduction; Calculate the coupled carbon reduction increment corresponding to each cross-scale correlation and coupling optimization scheme, and sum them to obtain the total coupled carbon reduction increment; Calculate the increase in sink volume corresponding to each single-scale optimization scheme separately, and sum them to obtain the total increase volume; The total carbon reduction is obtained by summing the basic total carbon reduction, the coupled incremental carbon reduction, and the total increase.

[0011] In one example, the total carbon reduction increase is matched with pre-set multi-level targets, specifically including: Based on the total carbon reduction increase, the total carbon emissions before the optimization plan update, and the carbon density before the optimization plan update, determine the corresponding carbon density reduction ratio. Based on the carbon density reduction ratio, it is compared with the ratio range of a pre-set multi-level hierarchical target to perform target matching.

[0012] On the other hand, this application also proposes a device for achieving carbon reduction and sequestration in urban residential areas based on multi-scale collaborative methods, comprising: At least one processor; and, A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, which enables the at least one processor to perform the multi-scale collaborative urban residential carbon reduction and sequestration method described in the above example.

[0013] On the other hand, this application also proposes a non-volatile computer storage medium storing computer-executable instructions configured to implement the multi-scale collaborative urban residential carbon reduction and sequestration method described in the above example.

[0014] The beneficial effects of this invention are: 1. Multi-scale system coverage: Integrating spatial forms such as buildings, transportation, energy, and green spaces into the assessment and optimization framework, breaking the limitations of traditional single scales, and realizing a systematic carbon reduction and sequestration design for all elements of the residential area.

[0015] 2. Coupling Mechanism Quantification: Based on the inter-scale correlation, cross-scale coupling optimization schemes are generated, the carbon reduction increment of coupling is clarified, the additional emission reduction benefits of scale synergy are explored, and a complete path from single-scale optimization to system integration is formed.

[0016] 3. Scientific accounting model: Based on the basic carbon reduction and carbon sequestration at each single scale, the total amount of carbon reduction increment is calculated by superimposing and coupling, which supports the itemized traceability and accurate evaluation of carbon reduction effect and overcomes the fuzzy problem of traditional accounting.

[0017] 4. Target-based matching: By matching the total amount with pre-set multi-level targets, quantitative correspondence between different targets such as low carbon, near-zero carbon, and zero carbon can be achieved, thereby enhancing the differentiation and guidance of strategy formulation.

[0018] 5. Applicability and Intelligent Support: The method is streamlined and can be adapted to different residential characteristics. It also supports the access of multi-source geospatial and energy consumption time series data, providing efficient tool support for scientific decision-making. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of the method for achieving carbon reduction and carbon sequestration in urban residential areas based on multi-scale collaborative approaches in this application. Figure 2 This is a schematic diagram illustrating the calculation of the total carbon reduction gain under one scenario in the embodiments of this application; Figure 3 This is a schematic diagram illustrating target matching in one scenario of an embodiment of this application; Figure 4 This is a schematic diagram of a device for achieving carbon reduction and carbon sequestration in urban residential areas based on multi-scale collaborative methods, as described in this application. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0021] The technical solutions provided by the various embodiments of this application are described in detail below with reference to the accompanying drawings.

[0022] like Figure 1 As shown in the embodiments of this application, a method for achieving carbon reduction and carbon sequestration in urban residential areas based on multi-scale collaborative methods is provided, including: S101: For each spatial scale among multiple spatial scales, conduct an assessment of the current energy consumption status by collecting the corresponding geospatial data and residential energy consumption environmental time series data.

[0023] A comprehensive evaluation system was constructed, encompassing automated data acquisition, multi-source data fusion, scenario-adaptive assessment, and dynamic error verification. This system utilizes remote sensing image analysis, IoT sensor monitoring, energy consumption simulation software, and on-site surveys to comprehensively evaluate residential areas across multiple spatial scales. Furthermore, specific evaluation correction models were developed for certain special residential scenarios (e.g., older residential areas built ≥30 years ago, mountainous residential areas with a slope ≥8%, and mixed-use commercial and residential blocks) to address the bias issues in evaluation results from conventional solutions under these special circumstances.

[0024] Specifically, corresponding remote sensing image data and a Digital Surface Model (DSM) are obtained via an unmanned aerial vehicle as geospatial data, and corresponding residential energy time-series data and environmental data are collected through sensors (e.g., existing smart electricity meters, water meters and environmental sensors in residential areas) as residential energy consumption and environmental time-series data. Automated data collection replaces the basic data collection link of manual census, thereby improving data collection coverage and collection efficiency.

[0025] Through point cloud registration and timestamp synchronization, spatial registration and time alignment are performed on remote sensing image data, digital surface models, energy time-series data and environmental data to obtain residential digital twin data (also referred to as a residential digital twin base). The Iterative Closest Point (ICP) algorithm can be used as the point cloud registration algorithm, and spatial registration and time alignment are realized through timestamp synchronization and an interpolation algorithm for time-series data. The residential digital twin data not only includes static physical spatial information of the residential area, such as building contours, heights, green space distribution, road networks, etc., but also integrates dynamic energy consumption data (e.g., hourly power load, gas consumption) and environmental parameters (e.g., temperature, humidity, PM2.5 concentration, light intensity), forming a digital mirror with a unified data spatial coordinate system and time dimension that can reflect the actual operation status of the residential area. Of course, for some parameters that are difficult to collect directly, such as the construction year of the residential area, parameters for which no relevant smart sensors are installed, and structural parameters inside buildings, they can also be obtained by importing corresponding attribute libraries, or improved and supplemented manually.

[0026] Corresponding core features are extracted based on the residential digital twin data; the core features include the construction year of the residential area, terrain slope, and functional mixing degree. Among them, the terrain slope can be directly calculated through the digital surface model, the construction year is obtained through association with an attribute library, and the functional mixing degree is obtained through analysis of building contours and energy consumption time series. For example, the residential area is first divided into independent building units based on building contour vector data, and the energy time-series data corresponding to each unit (including hourly energy consumption time-series curves of electricity, water and gas) is extracted; then a K-means clustering algorithm is used to perform pattern recognition on the energy consumption curves, and automatically classify buildings into functional labels such as residential, office and commercial; finally, the diversity of the proportion of various functional building areas is calculated based on the Shannon entropy index, where a higher entropy value indicates a higher functional mixing degree.

[0027] A scene classifier is constructed based on the random forest algorithm. Core features are used as input to determine the corresponding residential area type and match it with the corresponding evaluation model. The evaluation dimensions and / or the evaluation weights for each dimension differ depending on the residential area type. When the core features are all within the normal range, the evaluation dimensions and weights in the evaluation model are standard values. However, for special cases such as old residential areas, mountainous residential areas, and mixed-use commercial and residential blocks, corresponding evaluation dimensions and weights are used to achieve scene-adaptive adjustment of the evaluation model. For example, for old residential areas, evaluation weights are added for dimensions such as damaged building envelope and aging pipe networks; for mountainous residential areas, new evaluation dimensions such as wind and heat environment and site drainage are added; and for mixed-use commercial and residential blocks, evaluation weights are added for traffic carbon emissions and energy consumption of public services and supporting facilities.

[0028] The assessment model evaluates the current energy consumption status based on the digital twin data of the residential area, yielding energy efficiency assessment data. The assessment model uses the digital twin data of the residential area as input, and employs building energy consumption simulation software such as Energy Plus and DeST to perform hourly dynamic simulations of building heat transfer and equipment operation, or uses the IPCC activity level method to linearly calculate transportation and pipeline energy consumption, aggregating and outputting carbon emissions and carbon density distributions at various scales.

[0029] Based on the Monte Carlo simulation method, energy efficiency assessment data and field sampling data are sampled and verified. When the coefficient of variation of the data reaches a preset value, the reassessment of the energy efficiency assessment data and / or supplementary sampling of the field sampling data are triggered. For example, using the Monte Carlo simulation method, 1000 random sampling verifications are performed on energy efficiency assessment data and field sampling data (not only for verification, but also for supplementing geospatial data and residential energy consumption environmental time series data). When the coefficient of variation of the data is >10%, supplementary sampling and simulation recalculation are automatically triggered to ensure that the deviation of the assessment data is ≤5%.

[0030] Furthermore, the spatial morphological scale includes point scale, line scale, surface scale, and volume scale. The different scales increase sequentially in spatial morphology. The evaluation scope of the point scale corresponds to the building body and independent facilities and equipment in the urban residential area; the evaluation scope of the line scale corresponds to the transportation facilities and pipeline system in the urban residential area; the evaluation scope of the surface scale corresponds to the ecological green space and public space in the urban residential area; and the evaluation scope of the volume scale corresponds to the overall collaborative system of the residential area obtained by integrating the evaluation scopes corresponding to the point scale, line scale, and surface scale.

[0031] At each spatial scale, the entire process of "automated data acquisition - multi-source data fusion - scene adaptive evaluation - dynamic error verification" is executed sequentially. In addition, for each scale, an independent process can be set up to execute the relevant actions at that scale.

[0032] This section primarily provides an exemplary description of specific implementation methods for automated data acquisition operations in process operations at various spatial scales.

[0033] For the point scale, it mainly includes three dimensions: thermal performance evaluation of building envelope, energy efficiency ratio evaluation of equipment, and distributed energy potential evaluation. An example is given to illustrate the automated data collection for these three dimensions.

[0034] In the thermal performance evaluation of building envelopes, in addition to the data that needs to be collected regularly, a comprehensive survey of relevant data can be carried out. For example, on-site verification of the building age and the construction of the envelope (materials and thickness of exterior walls / roofs / windows), recording problems such as damage and leakage, and filling in the "Current Status Survey Form of Building Envelope" (with photos of typical parts and structural diagrams). At the same time, select 3 to 5 typical buildings, take 2 exterior wall measuring points and 1 roof measuring point for each building, and entrust a third-party organization to test the heat transfer coefficient K value according to the corresponding standards to obtain the design optimization benchmark data and enter it into the attribute database.

[0035] In the equipment energy efficiency ratio assessment dimension, the model, service life, and operating status of existing air conditioners, water pumps, fans, and other equipment can be checked, and the rated power and COP value can be calculated. High-energy-consuming old equipment (service life ≥ 10 years, COP value more than 15% lower than the current standard) can be marked.

[0036] In the dimension of distributed energy potential assessment, solar irradiance data for the past 5 years is obtained through PVGIS tools and local weather stations. On-site measurements of roof area, slope, and orientation are taken, and roof obstructions (such as water tanks and antennas) are identified to assess the area where photovoltaics can be installed. Simultaneously, the existing hot water consumption in the residential area (water meter data for the past year) is statistically analyzed to determine the demand for solar thermal systems. Furthermore, the electricity load curve of the residential area is plotted (combining residents' electricity consumption habits and equipment operating patterns), and the hourly output curve of the photovoltaic system is simulated to analyze the matching degree between the two, thereby determining the photovoltaic installed capacity (it is recommended to configure it according to 80% of the installable area) and the energy storage equipment capacity (configured according to 30% of the daily photovoltaic power generation).

[0037] For the linear scale, it mainly includes three dimensions: traffic organization efficiency assessment, pipeline transmission loss assessment, and corridor greening status assessment. An example is provided to illustrate the automated data collection for these three dimensions.

[0038] In the dimension of traffic organization efficiency assessment, traffic flow counters can be set up at the entrances and exits of residential areas and major nodes to monitor the distribution of vehicle and pedestrian traffic over 7 days (including weekdays and weekends); at the same time, issues such as congestion nodes, parking gaps, and pedestrian walkway damage can be recorded to create a "Residential Traffic Status Analysis Map".

[0039] In the dimension of pipeline transmission loss assessment, acoustic detectors are used to detect leaks in water supply networks, and smart water meters are installed in sections to monitor leakage rates; for heating networks, the condition of insulation layer damage and inlet / outlet temperature difference are detected, and aging and corrosion problems of the network are recorded to form a "Network Status Inspection Report".

[0040] In the assessment of the current status of corridor greening, the width, length, and distance to surrounding buildings of the traffic / pipeline corridor are measured on-site, the existing vegetation types, health status, and green coverage are recorded, idle spaces and damaged areas are marked, and a "Current Status Map of Corridor Greening" is drawn.

[0041] For the area scale, it mainly includes three dimensions: green coverage assessment, vegetation carbon sequestration capacity assessment, and public space thermal environment assessment. An example is given to illustrate the automated data collection for these three dimensions.

[0042] In the green space coverage assessment dimension, high-resolution remote sensing images of residential areas are acquired, green space boundaries are extracted using ArcGIS software, and the current green space coverage is calculated; on-site verification is conducted to correct misjudged areas and to identify areas such as idle plots and hardened surfaces that can be transformed into green spaces.

[0043] In the assessment dimension of vegetation carbon sequestration capacity, the existing vegetation species, quantity, diameter at breast height / tree height are investigated in the field, the current carbon sequestration is calculated using the allometric growth equation, and vegetation with low carbon sequestration and poor growth (such as weeds and dead trees) is marked.

[0044] In the assessment dimension of thermal environment in public spaces, 5 to 8 measuring points are set up in public spaces on typical summer days (≥35℃) to monitor air temperature, relative humidity, and wind speed, calculate heat island intensity, and mark high-temperature nodes (such as hardened plaza surfaces and unshaded areas).

[0045] For the overall scale, the main dimensions include energy self-sufficiency assessment, carbon flow cycle efficiency assessment, and system synergy assessment. An example is provided for the automated data collection of these three dimensions.

[0046] In the energy self-sufficiency assessment dimension, the current energy consumption of the residential area (building electricity, transportation energy, and public facility energy) is summarized, the power generation / heat supply of existing distributed energy (if any) is calculated, and the current energy self-sufficiency rate is calculated.

[0047] In the carbon flow cycle efficiency assessment dimension, identify the carbon input items (electricity supply from the grid, gas input) and output items (building emissions, traffic emissions, waste emissions) of the residential area, establish a current carbon flow inventory, and calculate the carbon flow cycle efficiency.

[0048] In the system synergy evaluation dimension, the synergy evaluation index system is optimized, and new design-specific indicators are added (such as multi-disciplinary design synergy rate and equipment and building layout adaptability), with the weights determined using the analytic hierarchy process.

[0049] S102: For each spatial scale, optimize based on the energy consumption status assessment to generate a single-scale optimization scheme.

[0050] A single-scale optimization scheme refers to optimizing energy consumption based on an assessment of the current energy consumption status, taking only the current spatial form and scale into account. This optimization involves optimizing the existing conditions (including equipment optimization, building structure optimization, wiring optimization, and material optimization).

[0051] Develop independent technical solutions for each scale. For example, at the point scale, you can carry out structural reforms, replace high-efficiency equipment, implement distributed photovoltaic / solar thermal systems, and rooftop / vertical greening; at the line scale, you can optimize traffic organization, improve slow-traffic systems, upgrade pipeline networks for energy conservation, and green corridors; at the area scale, you can optimize green space layout, configure high-carbon-sequestering vegetation, and implement permeable paving and sponge city facilities; at the volume scale, you can build microgrids, deploy intelligent energy management systems, and establish a global carbon sink network.

[0052] In generating single-scale optimization schemes, an automated approach can be adopted. For example, a multimodal large model and related automated tools (including image generation models such as Stable Diffusion, energy consumption simulation software such as DeST and Energy Plus, and traffic simulation software such as VISSIM) can be used, combined with corresponding manual actions based on requirements (such as manual screening based on multiple output results, inputting relevant instructions into the multimodal large model, and using automated tools) to achieve semi-automated and rapid generation of single-scale optimization schemes.

[0053] For some schemes that are strongly related to human activities (such as meeting planning and travel guidance), corresponding scheme plans can be generated through large language models.

[0054] To elaborate, regarding point scale: In the thermal performance evaluation of building envelopes, during scenario-adaptive assessment, energy consumption simulation software such as DeST / Energy Plus / PKPM can be used to establish a current status model and 3-5 renovation scheme models (such as thickening external wall insulation, replacing windows with Low-E glass, etc.). Local meteorological parameters are input, and the annual heat transfer consumption and air conditioning load of each scheme are simulated and compared, outputting a "Comparison Table of Optimized Envelope Schemes". The current status model can be directly used for the current status assessment in step S101, while the renovation scheme model can be used for the single-scale optimization scheme in step S102. At this point, the optimal scheme is determined based on the simulation results, specifying the type and thickness of insulation materials and construction details (such as the finishing of external wall insulation and door and window joints) to ensure compliance with local residential building energy-saving design standards.

[0055] In the equipment energy efficiency ratio assessment dimension, based on the building load requirements after the renovation, 3-4 suitable equipment models (such as first-level energy efficiency inverter air conditioners and high-efficiency centrifugal pumps) are selected. Rated parameters provided by manufacturers are collected, and the annual operating energy consumption, investment cost, and payback period of different equipment are calculated to form an "Equipment Selection Comparison Analysis Table". The equipment model and installation location are determined, and space is reserved for equipment room and pipeline routing. Intelligent monitoring interfaces (such as power monitoring module installation points) are designed simultaneously to ensure that equipment operation data can be collected and controlled. At the same time, based on the set benchmark verification and referring to relevant standards, the energy efficiency level of the selected equipment is verified to ensure that the average energy efficiency ratio of the equipment after the renovation is improved by more than 20%.

[0056] In the dimension of distributed energy potential assessment, layout design is carried out, and layout diagrams of photovoltaic panels and solar thermal collectors are drawn, specifying the installation spacing (photovoltaic panel spacing ≥ 1.5 times the panel length), fixing method (roof bracket selection), and avoiding roof maintenance passages, parapet walls, etc.; the routing of transmission cables is designed simultaneously to shorten the distance to the power load center and reduce line loss; and benefit calculation is carried out: the annual power generation / heat supply of distributed energy, investment payback period (usually 5~8 years), and annual carbon reduction are calculated, and a "Distributed Energy Design Scheme Benefit Table" is output as the basis for the feasibility of the transformation scheme.

[0057] For line scale: In the traffic organization efficiency assessment dimension, based on the renovation and upgrading needs, 2-3 traffic optimization schemes are proposed, such as adding one-way lanes, planning dedicated non-motorized vehicle lanes, adding multi-level parking garages (or converting ground parking spaces), and optimizing entrance and exit locations. A master plan of the schemes is drawn. Using VISSIM traffic simulation software, the average vehicle travel speed, congestion duration, and parking turnover rate of each scheme are simulated, and the optimization effects are compared. The number of parking spaces is calculated simultaneously (based on a standard of 1.2 spaces per household) to ensure that the demand is met. The optimal scheme is determined, specifying the lane width (motorized vehicle lane ≥ 4m, non-motorized vehicle lane ≥ 2.5m), parking space dimensions, traffic sign locations, speed bump / zebra crossing layout, etc., and simultaneously reserving charging pile installation points (more than 15% of the total number of parking spaces).

[0058] In the dimension of pipeline transmission loss assessment, for leakage / loss issues, renovation solutions are proposed, such as replacing old pipes (PPR pipes instead of cast iron pipes), adding pipe insulation layers (rock wool insulation layer thickness ≥50mm), and optimizing pipeline routing (shortening transmission distance), etc., and calculating the loss rate and investment cost of each solution; drawing pipeline renovation construction drawings, specifying pipe diameter, material, and insulation methods, designing the installation points of segmented meters and leakage detection interfaces, and optimizing pipeline bends and diameter change nodes (reducing resistance loss); referring to relevant standards, verifying that the leakage rate of the water supply network after renovation is ≤8% and the heat transfer loss rate of the heating network is ≤10%.

[0059] In the assessment of the current status of corridor greening, a multi-layered greening scheme of "trees + shrubs + herbs" is designed in combination with the corridor functions (traffic buffer, ecological protection). Pollution-resistant, easy-to-maintain, and high carbon-fixing plants (such as camphor, cypress, and liriope) are selected, and the planting spacing and seedling specifications are specified. Greening design drawings are drawn. The spatial relationship between the greening layout and pipelines and cables is coordinated to ensure that the distance between the planting of vegetation and pipelines is ≥1.5m (to avoid root damage to pipelines). A green buffer zone with a width of 3-5m is set on both sides of the traffic corridor, and a pedestrian walkway (width ≥1.2m) is designed simultaneously to achieve the synergy of "greenway + walkway". In accordance with the relevant standards, the green coverage rate is specified to be ≥30% and the seedling survival rate is ≥95%, and the "Corridor Greening Renovation Design Description" is output.

[0060] At the surface level, in the green space coverage rate assessment dimension, combined with relevant standards, determine the green space coverage rate target after renovation (≥30% for old area renovation, ≥35% for new area construction), and plan the location and area of ​​centralized green spaces, pocket parks, and sunken green spaces; draw an optimized green space layout map to ensure that the area of ​​centralized green spaces is ≥0.5㎡ / person, the service radius of pocket parks is ≤300m, and sunken green spaces account for more than 30% of the total green space area (for rainwater storage), specifying the terrain slope (1%-3%) and vegetation configuration; output a "Comparison Table of Green Space Before and After Renovation", marking the green space area, coverage rate, and proportion of each type of green space to ensure that the planning indicators are met.

[0061] In the vegetation carbon sequestration capacity assessment dimension, based on local climate and soil conditions, high carbon sequestration and easy-to-maintain vegetation (trees: camphor, osmanthus; shrubs: red photinia, arborvitae; herbs: liriope, zoysia) are selected to form a "Vegetation Selection List"; the annual carbon sequestration after transformation is calculated according to "area × carbon sequestration coefficient", and the improvement rate is compared with the current situation (target improvement of more than 20%); the vegetation community structure is optimized (multi-layered community accounts for ≥60%) to improve carbon sequestration efficiency; the planting location, specifications, and spacing of vegetation are specified, a planting map is drawn, the core green areas with high carbon sequestration are marked, and the "Green Space Carbon Sequestration Optimization Design Description" is output.

[0062] In the assessment of the thermal environment of public spaces, the ENVI-met software is used to establish models of the current situation and renovation plans (such as adding shading facilities, planting large trees, and laying permeable paving) to simulate the hourly thermal environment distribution and compare the cooling effects of each plan (the target heat island intensity is reduced by 1-2℃). The optimal plan is determined, specifying the location and size of the shading canopy, the planting points of large trees (spacing 5-8m), and the area of ​​permeable paving (occupying ≥60% of the public space ground). Spray cooling facilities are designed simultaneously (in densely populated areas). The thermal environment comfort of the public space after renovation is checked with reference to the human comfort index standard (70~80 is suitable), and the "Public Space Thermal Environment Optimization Plan Report" is output.

[0063] At the scale of scale, in the energy self-sufficiency rate assessment dimension, combined with the low-carbon goals of urban renewal, the post-renovation energy self-sufficiency rate target is determined (≥15% for old residential areas, ≥25% for new residential areas), and the configuration scale of facilities such as photovoltaics, solar thermal, and energy storage is planned; an integrated energy system design diagram is drawn to realize the coordinated linkage of distributed energy, power grid, and energy storage equipment, and the charging and discharging strategy of energy storage equipment is clarified (peak-hour charging of photovoltaics, peak-hour discharging of loads); the interface of the intelligent energy management platform is designed simultaneously to realize real-time monitoring of energy supply and demand; the post-renovation energy self-sufficiency rate, annual electricity savings, and annual carbon reduction are calculated, and the "Energy System Renovation Benefit Analysis Table" is output to ensure that the preset goals are achieved.

[0064] In the carbon flow cycle efficiency assessment dimension, optimization solutions are proposed to address the bottleneck of "high input-low cycle," such as increasing distributed energy (reducing grid power input), promoting electric vehicles (reducing traffic emissions), and adding waste sorting and recycling points (improving waste recycling rate). An "energy-carbon sink-waste" cycle system is designed, such as photovoltaic power generation replacing grid electricity, green space carbon sequestration offsetting some emissions, and resource utilization of kitchen waste (biogas production), and a carbon flow cycle diagram is drawn. The carbon flow cycle efficiency after the transformation is calculated (target improvement of more than 30%), and a "Carbon Flow Cycle Optimization Design Report" is output, clarifying the division of responsibilities for carbon reduction / carbon sequestration in each link.

[0065] In the system synergy assessment dimension, a multi-disciplinary collaborative design mechanism involving architecture, HVAC, landscape, and electrical engineering should be established, and regular collaborative meetings should be held to ensure the synergy and compatibility of distributed energy with building layout, pipeline network with green space, and intelligent systems with facilities and equipment. A 1-5 point scale should be used to score synergy indicators, and a weighted total score should be calculated (target ≥ 3.5 points, indicating good synergy). For indicators with low scores (such as management synergy), supplementary design optimization measures should be implemented (such as clarifying the responsible party for intelligent system operation and maintenance). A "System Synergy Transformation Design Specification" should be output, clarifying the collaborative nodes, division of responsibilities, and operation and maintenance requirements of each system.

[0066] S103: Based on the correlation between different spatial morphological scales, a cross-scale correlation coupling optimization scheme is generated on the basis of the single-scale optimization scheme.

[0067] In the above text, a single-scale optimization scheme was proposed based on the current situation assessment for a single spatial morphological scale. However, this optimization is limited to a single scale and does not consider the mutual influence and synergistic effects between different scales, which may lead to limited optimization results or even conflicts.

[0068] Based on this, an intelligent coupling design system is constructed, which consists of "intelligent identification of coupling relationships, quantitative ranking of synergistic potential, automated generation of coupling schemes, and dynamic simulation of benefits". This system identifies and designs pairwise synergistic relationships between the three element scales of "point, line, and surface", explores additional carbon reduction increments through their interaction, and finally achieves system integration optimization at the "volume" scale.

[0069] Specifically, based on the digital twin data of the residential area, a comprehensive relational map of all elements of the residential area is constructed using a graph neural network. In this map, scale elements corresponding to each spatial morphology scale are used as nodes, and the relationships between these scale elements are used as edges. These relationships include material flow relationships, energy flow relationships, and spatial relationships. Based on the constructed digital twin data of the residential area, a graph neural network (GNN) is used to construct the comprehensive relational map of all elements of the residential area. Using "points, lines, and surfaces" as scale elements as graph nodes, and material flow, energy flow, and spatial relationships between elements as edges, the map automatically identifies strong relationships between nodes (e.g., a correlation threshold ≥ 0.7) and outputs a list of potential coupling pairs, replacing the manual identification of coupling relationships.

[0070] Scale elements refer to physical entities or functional units that can be independently identified at each scale, possess independent attributes, and participate in the interaction of material / energy flows. For example, point-scale elements may include individual residential / public buildings, rooftop distributed photovoltaic arrays, outdoor air conditioning units / heat pump units, energy storage battery cabinets, and individual charging piles. Line-scale elements may include traffic segments (sections of motor vehicle lanes, non-motor vehicle lanes, and pedestrian walkways), water supply network segments, heating network segments, power cable corridors, and roadside green belts. Area-scale elements may include concentrated green space patches, pocket parks, sunken rain gardens, public squares, and rooftop green areas. Edges may include material flows (water, electricity, and heat transmission), energy flows (photovoltaic power generation transmission and distribution), and spatial relationships (adjacency, inclusion, and shading) between nodes.

[0071] Based on the comprehensive correlation map of the residential area, strong correlations between nodes are identified. A correlation threshold (e.g., ≥0.7) can be pre-set to filter out potential coupling pairs. For example, at the point scale, "rooftop distributed photovoltaic arrays" and at the line scale, "power cable corridors" have a significant energy flow correlation, with a correlation degree potentially reaching 0.85; at the surface scale, "sunken rain gardens" and at the point scale, "residential building rainwater downpipes" have a material flow (rainwater collection) correlation, with a correlation degree potentially reaching 0.78; at the line scale, "roadside green belts" and at the point scale, "residential building windows" have a spatial correlation (summer shading, winter lighting), with a correlation degree potentially reaching 0.72.

[0072] The synergistic carbon reduction sensitivity of each strongly correlated pair is determined using response surface methodology (RSM). Based on this sensitivity, the carbon reduction increment per unit retrofit cost is obtained and prioritized. RSM is an optimization method combining mathematical modeling and statistical analysis. Its core function is to construct an approximate functional relationship between input variables and output response using limited experimental or simulation data, thereby quickly finding the optimal parameter combination. First, a finite number of simulations are performed on key design parameters for each strongly correlated pair to fit an approximate functional surface between the carbon reduction increment and the parameters. The partial derivatives of each parameter are then calculated as the synergistic carbon reduction sensitivity. Next, the carbon reduction potential corresponding to the sensitivity is divided by the corresponding retrofit cost to obtain the carbon reduction increment per unit cost. Finally, these values ​​are sorted from highest to lowest, and the top 30% are prioritized for design.

[0073] Based on priority, nodes corresponding to the highest number of strong correlations are selected. Using a pre-defined coupling technology library (containing multiple pre-defined optional technologies), a genetic algorithm generates corresponding cross-scale correlation coupling optimization schemes. These schemes include core design elements such as spatial layout, equipment parameters, and scheduling logic. Based on the priority ranking, the top-ranked (e.g., top 30%) strong correlation node pairs are selected, and corresponding optional technology combinations are retrieved from the pre-defined coupling technology library. Using spatial layout, equipment parameters, and scheduling logic as optimization variables, and maximizing carbon reduction increment and minimizing retrofit costs as objective functions, a genetic algorithm performs multi-objective optimization, automatically generating a set of Pareto optimal coupling schemes. The final output is a design scheme that includes spatial arrangement, equipment selection parameters, and operational scheduling strategies.

[0074] Using digital twin data of residential areas, simulations of cross-scale correlation and coupling optimization schemes are performed, and the final cross-scale correlation and coupling optimization scheme is output. The generated cross-scale correlation and coupling optimization schemes are imported into the digital twin data of residential areas and simulated hourly for 8760 hours throughout the year. The incremental carbon reduction due to coupling is quantitatively calculated, invalid schemes with negative incremental reductions are eliminated, and the optimal coupling scheme set is output.

[0075] Furthermore, in the process of generating cross-scale correlation and coupling optimization schemes, corresponding scheme designs can be carried out during the generation process for different types of cross-scale correlation and coupling optimization schemes.

[0076] Specifically, the strong point-line correlation, also known as point-line coupling, corresponds to the collaboration between buildings / facilities and linear facilities. Examples of measures include the intelligent linkage between building-distributed photovoltaic systems and transportation charging pile networks (prioritizing local consumption of photovoltaic power generation); and the collaboration between building-intelligent water terminals and regional water supply network scheduling systems (smoothing peak water usage and reducing pumping energy consumption). The quantifiable indicator is: the incremental carbon reduction from point-line coupling. .

[0077] The core logic of the execution process is to establish a closed loop of "data linkage - load matching - intelligent scheduling" between buildings / facilities and linear facilities, realize the coordinated optimization of energy flow and material flow, and adapt to the integrated layout of "photovoltaic-charging pile-pipeline pump station" in the renovation design.

[0078] In the specific implementation steps: First, interface design is carried out. In the design of building rooftop photovoltaics, charging piles, and pipeline pumping stations, intelligent monitoring interfaces (RS485 / 5G interfaces) are reserved, and the data collection content (power generation, charging load, pumping station energy consumption) is clearly defined to ensure data interconnection.

[0079] Secondly, a coordinated layout design was implemented, drawing a layout coordination diagram of "photovoltaic-charging pile-pipeline pumping station". Charging piles were centrally arranged under the photovoltaic roof (such as the entrance and exit of underground garages and ground parking lots) to shorten the power transmission distance; pipeline pumping stations were located close to energy storage equipment to facilitate load matching.

[0080] Next, the scheduling strategy is designed and embedded. In the design of the intelligent energy management platform, a dynamic adaptive scheduling logic based on Long Short-Term Memory (LSTM) networks is embedded to replace the fixed-time scheduling mode. In this process, LSTM networks are selected from a pre-set library of coupling technology measures, and time-series data of residential energy consumption environment are used as training data. The load-output prediction model is trained through the LSTM network. For example, hourly photovoltaic power output data, charging pile charging load data, pipeline water load data, local hourly meteorological data, and grid peak-valley electricity price data of the residential area over the past year are collected to construct a training dataset. The load-output prediction model is trained through the LSTM algorithm, and the model prediction accuracy is ≥90%.

[0081] The load-output forecasting model is used to predict the dynamic characteristics of energy supply and demand in residential areas over time periods. For example, based on the forecasting model, the hourly photovoltaic output forecast curve, charging pile load forecast curve, and pipeline water load forecast curve are output 24 hours in advance for the next day. The model automatically identifies the top 30% of photovoltaic output as the peak photovoltaic period and the top 20% of pipeline water load as the peak water consumption period, replacing fixed time settings.

[0082] Based on the dynamic characteristics of energy supply and demand over time periods, a corresponding dynamic energy supply and demand scheduling strategy is added during the cross-scale correlation and coupling optimization process generated by the genetic algorithm. At this point, the dynamic scheduling strategy is set as follows: during peak photovoltaic periods, photovoltaic output is prioritized for local supply to charging piles, with excess electricity stored in energy storage devices, and the curtailment rate control threshold is ≤2%; one hour before the peak water consumption period, the charging and discharging plans of the energy storage devices are automatically matched, and the energy storage devices are activated to supply power to the pipeline pumping stations, smoothing the peak load on the power grid during peak water consumption and reducing the proportion of peak electricity consumption on the power grid.

[0083] It also allows for dynamic strategy iteration, with incremental training of the prediction model every 7 days. The model parameters are optimized based on actual operating data, and the scheduling strategy is updated synchronously to adapt to seasonal and long-term changes in residential energy consumption habits.

[0084] The strong line-surface correlation corresponding to the line and surface scales, also known as line-surface coupling, corresponds to the synergy between linear facilities and spatial planes. Examples of measures include the spatial integration of transportation corridors with ecological ventilation corridors and green space systems (synergistically reducing traffic emissions and mitigating the urban heat island effect); and the combination of pipeline corridors with vegetated buffer zones and sunken green spaces (enhancing landscape carbon sequestration capacity and reducing the environmental load on the pipeline network). The quantifiable indicator is the incremental carbon reduction from line-surface coupling. .

[0085] The core logic of the implementation process is to achieve a dual benefit of reducing carbon emissions from linear facilities and enhancing carbon sequestration at the area scale through the synergy of spatial composite utilization and ecological functions, adapting to the composite transformation design of "corridors-green spaces". A fully automated technical solution is constructed, consisting of "multi-source raster image processing - automatic extraction of corridor elements - spatial overlay analysis - intelligent determination of composite areas - automatic generation of layout maps", replacing the manual GIS operation mode.

[0086] In the specific implementation steps: The U-Net semantic segmentation model and Frechet distance algorithm were selected from a pre-defined library of coupling techniques. Multi-source raster image preprocessing was performed on geospatial data, and pixel segmentation was performed using the U-Net semantic segmentation model to extract corridor elements. High-resolution remote sensing images of residential areas, underground pipe network CAD vector maps, and control detailed planning vector maps were acquired. Gaussian filtering was used to denoise the remote sensing images, and coordinate registration was performed to unify all vector / raster data to the 2000 National Geodetic Coordinate System, with a registration error ≤ 0.5 pixels. The U-Net semantic segmentation model was used to perform pixel-level segmentation on the preprocessed remote sensing images, automatically extracting the spatial boundaries and centerlines of traffic corridors, green spaces, and ventilation corridors. Vector data parsing algorithms were used to automatically extract the orientation, centerline, and spatial boundaries of underground pipe network corridors, outputting vector datasets for various types of corridors.

[0087] Further spatial overlay is applied to various types of corridor elements, and the spatial overlap of the corridor centerlines is calculated using the Frechet distance algorithm to obtain the composite utilization area. Based on the GIS spatial analysis engine, various types of corridor vector data are spatially overlaid, and the spatial overlap of the centerlines of two types of corridors is calculated using the Frechet distance algorithm, as shown in Formula 1: Formula 1; Wherein, P and Q are the center lines of the two corridors, d is the Euclidean distance, and α and β are parametric mappings; the overlap of the center lines is calculated based on the Frechet distance, and when the overlap is ≥80% and the composite width after the corridors are superimposed is ≥5m, it is automatically determined to be a composite utilization area.

[0088] Based on the mixed-use areas, during the process of generating cross-scale correlation and coupling optimization schemes using genetic algorithms, spatial conflict detection and conflict avoidance are performed on the mixed-use areas. For the identified mixed-use areas, spatial conflicts with the main building, underground structures, and fire lanes are automatically detected. If conflicts exist, the corridor boundaries are automatically optimized to avoid conflicting areas while meeting overlap and width requirements. Based on the finally confirmed mixed-use areas, a spatial composite layout map is automatically generated, marking the corridor centerline, composite area boundaries, width, and overlap parameters, and outputting design drawings conforming to CAD / GIS format.

[0089] In addition, ecological configurations can be refined based on needs. For example, a 3-5m wide and high carbon-fixing vegetation strip (camphor tree + cypress + liriope) can be planted on both sides of the transportation corridor, and a planting cross-section diagram can be drawn (tree spacing 2-3m, shrub spacing 1-1.5m); a sunken green space (elevation 5-10cm lower than the surrounding road surface) can be laid above the sewage pipe network of the pipeline corridor, a topographic cross-section diagram can be drawn, and moisture-tolerant plants (calamus, reeds) can be selected, and the rainwater storage capacity can be determined; in the combined section of ventilation corridor and transportation corridor, a "tree + shrub + herb" approach can be adopted. The multi-layered structure optimizes vegetation height (6-8m for trees and 1-2m for shrubs) to improve ventilation efficiency. Monitoring points (CO2 concentration, PM2.5 concentration, and heat island intensity) are set up in both composite and non-composite areas, with reserved installation points for monitoring equipment (such as weather station brackets and concentration monitoring instrument interfaces). The design scheme specifies the vegetation maintenance cycle (pruning once per quarter and replanting once per year) and the dredging cycle of sunken green spaces (twice a year) to ensure long-term stability of synergistic effects and realize dynamic optimization plans.

[0090] The strong point-to-surface correlation, also known as point-to-surface coupling, refers to the coordination between buildings / facilities and spatial planes, corresponding to the relationship between point scale and surface scale. Examples of measures include: coordination between building layout, form, and optimized design of residential wind and heat environment (passive energy saving); ecological connectivity between vertical greening of building facades, roof gardens, and concentrated green spaces on the site (building a continuous three-dimensional carbon sink network); and matching the layout of distributed energy facilities with the energy demand of public spaces. The quantifiable indicator is: carbon reduction increment of point-to-surface coupling. .

[0091] The core logic of the execution process is to achieve passive energy conservation and carbon sink network enhancement through the coordinated design of building form and surface-scale spatial environment and the connection of ecological elements, and to adapt to the integrated transformation design of "building-green space".

[0092] First, CFD software (Phoenics) was used to simulate the wind and heat environment of the residential area, optimize the building layout (mainly north-south orientation, with a spacing ≥ 1.5 times the building height) and window ratio (0.3-0.4 for south-facing and ≤ 0.2 for north-facing), and draw up the design drawing of the cross ventilation corridor.

[0093] Next, the vertical greening design of the building facade will be carried out, specifying that the vertical greening coverage rate is ≥30%, selecting climbing plants such as Virginia creeper and ivy, designing the location of the planting trough (1.5m height on the exterior wall of the building) and the irrigation method (drip irrigation), and drawing the facade greening construction drawings.

[0094] Next, a collaborative automated design of the carbon sink network is implemented, constructing a full-process design system encompassing "automatic carbon sink node identification, intelligent network topology optimization, dynamic connectivity correction, and automated layout map generation." In this process, the U-Net semantic segmentation model, Building Information Modeling (BIM) parsing, and minimum spanning tree algorithm are selected from a pre-defined library of coupled technologies. For the residential area's digital twin data, pixel segmentation is performed using the U-Net semantic segmentation model, and model parsing is performed using BIM parsing to identify carbon sink nodes corresponding to the point-scale assessment range. Based on the residential area's digital twin data, remote sensing image semantic segmentation using BIM model parsing and the U-Net semantic segmentation model automatically identifies areas such as building rooftops, building facades, concentrated green spaces, and vacant land that can serve as carbon sink nodes. The greenable area and carbon sequestration potential of each node are calculated, and a list of carbon sink nodes and their spatial coordinates are output.

[0095] Based on the minimum spanning tree algorithm, the optimal path of a 2m wide ecological corridor is automatically planned, with each carbon sink node (e.g., the building itself and independent facilities and equipment included in the point-scale assessment scope) as the endpoint and the strip space (including building fire lanes, residential paths, and idle strip spaces) as the wiring constraint. The goal is to generate ecological corridor paths connecting each carbon sink node with the shortest total path length and the least amount of space modification to the residential area.

[0096] The connectivity index of the three-dimensional carbon sink network corresponding to the volumetric assessment range is obtained using the connectivity index of landscape ecology. Based on the connectivity index (PC) calculation formula of landscape ecology, the connectivity index of the three-dimensional carbon sink network is automatically calculated, as shown in Formula 2: Formula 2; in, , Let i be the area of ​​carbon sink nodes i and j. Let be the maximum connectivity probability between nodes i and j. The total land area of ​​the residential area; ensure that the connectivity index of the calculated result is ≥0.7. If the standard is not met, automatically optimize the corridor path and carbon sink node layout until the threshold requirement is met.

[0097] Using a genetic algorithm, with the objective functions of maximizing photovoltaic (PV) installed capacity, maximizing carbon sequestration, and minimizing transmission distance, a cross-scale, interconnected optimization scheme for PV panel layout is generated. A multi-objective genetic algorithm is employed, with the objective functions of maximizing PV installed capacity, maximizing carbon sequestration, and minimizing transmission distance, to automatically optimize PV panel layout, prioritizing their placement on building rooftops and parking lot canopies, and automatically avoiding core areas of concentrated green spaces. For the lower edge areas of PV panels, a list of shade-tolerant vegetation is automatically matched to complete the "light-green" collaborative layout design. Based on the optimized carbon sequestration network and collaborative layout scheme, a two-dimensional plan layout diagram and a three-dimensional BIM model are automatically generated, annotating carbon sequestration nodes, ecological corridors, PV panel layout, and core parameters of carbon sequestration, outputting drawing files that conform to design specifications.

[0098] Furthermore, in point-to-area coupling, collaborative automated design of energy demand can be carried out to construct a full-process design system of "energy load spatial distribution simulation - intelligent equipment location optimization - automated benefit verification". In this process, the energy load of public spaces is first simulated automatically: the location, area, and pedestrian density distribution of public spaces such as residential squares, fitness areas, and walkways are identified through a 3D point cloud model. Combined with the simulation results of local sunshine duration and building shadows, the hourly lighting and equipment energy load of public spaces are calculated, and an energy load heat map is output.

[0099] Next, intelligent optimization of photovoltaic panel locations is carried out: based on the energy load heat map, a simulated annealing algorithm is used with the constraints of "transmission distance ≤ 50m and minimum line loss" to automatically optimize the layout and installed capacity of photovoltaic panels on the rooftops of buildings around the public space, ensuring that the matching degree between photovoltaic output and the energy load of the public space is ≥ 85%.

[0100] Then, the lighting system layout is automatically optimized: based on the simulation results of building reflected light in public spaces and the distribution of pedestrian density, the Cuckoo Search algorithm is used to automatically optimize the installation points, spacing and power of light-sensing + human body sensing dual-control lamps. Under the standard requirements of road surface average illuminance ≥10lx and uniformity ≥0.4, the number of lamps is minimized, reducing the number of lamps by 15%~20% compared with conventional layout, and automatically generating lighting layout diagram.

[0101] Finally, automated effect verification is performed: the optimized design scheme is imported into Energy Plus and ENVI-met software, automatically simulating the building's annual air conditioning energy consumption and green space carbon sink after the renovation, automatically comparing the data before and after the renovation, ensuring that air conditioning energy consumption is reduced by ≥15% and carbon sink is increased by ≥20%; if the standards are not met, the building facade greening, photovoltaic layout, and lighting design parameters are automatically iterated and optimized until the threshold requirements are met.

[0102] Furthermore, during the simulation verification process, an automated closed-loop process of "simulation verification - deviation diagnosis - automatic parameter iteration - scheme recalculation and optimization" is constructed to replace the manual verification and adjustment mode.

[0103] A cross-scale correlation and coupling optimization scheme was simulated using digital twin data of residential areas. Based on the digital twin data of residential areas, the hourly operation of the coupling scheme was simulated for 365 consecutive days under typical meteorological conditions. The power grid consumption and photovoltaic absorption rate of charging piles and pump stations before and after coupling were compared. The verification thresholds were set as follows: photovoltaic absorption rate increased by ≥15% and power grid consumption decreased by ≥10%.

[0104] If the simulation results do not reach the verification threshold, the core influencing factors of the deviation are located by Pearson correlation coefficient analysis. The identification dimensions include: photovoltaic installed capacity, energy storage equipment capacity, charging pile layout, transmission distance, scheduling strategy parameters, and pump station operating power.

[0105] Based on the core influencing factors of deviation, a particle swarm optimization algorithm is used to automatically iteratively optimize the core influencing factors of deviation, with the objective functions of maximizing photovoltaic absorption rate and minimizing grid power consumption, and with the verification threshold as a constraint. The optimized parameter combination for the identification dimensions is output until the verification threshold is reached or a scheme reconstruction is triggered. The upper limit of the number of iterations can be set to 500. The optimized parameter combination is then re-imported into the simulation model for recalculation and verification until the result meets the verification threshold. If the threshold is not met even after iteration to the upper limit, a coupling scheme reconstruction is automatically triggered, adjusting the spatial layout and coordination mode of photovoltaic-charging pile-pump station, re-executing the entire process verification, and finally outputting the optimal coupling scheme that meets the threshold requirements.

[0106] In addition, regarding the point-to-line coupling carbon reduction increment The calculation is shown in Formula 3: Formula 3; in, The power grid electricity consumed (kWh / a) of the pre-coupling linear facilities (charging piles + pipeline pumping stations) is measured using the pre-renovation monitoring data or comparative data from similar residential areas. The grid power consumption (kWh / a) of the coupled linear facility is simulated and calculated using Energy Plus software (input photovoltaic installed capacity and dispatch strategy parameters). The average carbon emission factor (tCO2 / kWh) for the regional power grid is based on the latest data released by the relevant local authorities. The charge / discharge loss of the energy storage device (kWh / a) is determined according to the parameters provided by the energy storage device manufacturer (usually 5% to 8% of the total charge / discharge).

[0107] at this time, The calculation results, as a core indicator of the carbon reduction effect of the coupled scheme, must be included in the "Low-Carbon Retrofit Design Scheme Demonstration Report". If the value is less than 0, adjustments to the photovoltaic installed capacity, dispatch strategy, or energy storage equipment parameters are required.

[0108] For line-to-surface coupling carbon reduction increment The calculation is shown in Formula 4: Formula 4; in, To reduce carbon emissions from transportation, For the increase of carbon sequestration, The figures represent the average daily traffic flow (vehicles / day) of the traffic corridors before and after coupling. Take the current monitoring value, Take the optimized design value (determined through traffic simulation); These are the carbon emission factors per unit vehicle km before and after coupling, respectively. Adopt the recommended values ​​from relevant guidelines (e.g., provincial greenhouse gas inventory compilation guidelines). Calculated based on a 5% to 10% reduction in heat island intensity; The length of the traffic corridor (km) is taken from the actual measured length according to the design drawings; The green space area (m²) for the composite area and the single area are respectively calculated based on the design drawings. The values ​​for carbon sequestration per unit area of ​​composite / single green space are respectively adopted, using the recommended values ​​from relevant guidelines (such as the guidelines for carbon sequestration measurement and monitoring in afforestation projects), and adjusted in combination with the designed vegetation type. The total carbon reduction amount must be included in the design plan to ensure that the annual carbon reduction of the composite renovation plan is ≥ This serves as a core indicator of the feasibility of the plan.

[0109] For point-to-surface coupled carbon increment The calculation is shown in Formula 5: Formula 5; in, For passive energy conservation and carbon reduction in buildings, To increase the volume of three-dimensional carbon sequestration, The building air conditioning energy consumption (kWh / a) before and after coupling is calculated using Energy Plus software (input building layout, window ratio, and greening parameters). These are the carbon sequestration per unit area of ​​three-dimensional / two-dimensional carbon sink networks, respectively, and are weighted and calculated based on the proportion of rooftop, vertical, and centralized green space areas in the design. The total green area of ​​the residential area (m²) is calculated according to the design drawings (roof greening is calculated based on the projected area, and vertical greening is calculated based on 30% of the unfolded area).

[0110] As a core carbon reduction indicator for integrated "building-green space" renovation, the calculation process and results must be clearly defined in the design plan to ensure that the annual carbon reduction after the renovation is ≥ .

[0111] The benefits of the coupling among the above three types of elements are achieved through unified scheduling, integration and enhancement of the community-level microgrid, intelligent energy management system and carbon flow monitoring and optimization platform. Ultimately, these benefits are reflected in system-level carbon reduction and carbon sequestration results such as improved energy self-sufficiency rate and optimized carbon cycle efficiency in the community as a whole (scale).

[0112] S104: Based on the basic carbon reduction and carbon sequestration amounts corresponding to each single-scale optimization scheme and the coupled carbon reduction increments corresponding to each cross-scale coupled optimization scheme, the total carbon reduction and carbon sequestration amount is obtained.

[0113] Construct a closed-loop accounting system that integrates "dynamic parameter traceability, full automated accounting, adaptive result correction, and full data traceability" and establish an accounting model.

[0114] In the closed-loop accounting system for carbon reduction and total increase, the first step is to dynamically trace the source of parameters. A unique identifier is established for all accounting parameters, such as carbon emission factors, carbon sequestration coefficients, and energy consumption parameters. The source documents, test reports, standard documents, and release dates of the parameters are linked to automatically verify the timeliness and regional adaptability of the parameters. When the local housing and construction department releases the latest power grid carbon emission factors and vegetation carbon sequestration coefficients, the parameters are automatically updated, replacing the manual parameter entry mode.

[0115] Next, full-scale automated accounting is performed. Based on the digital twin data of the residential area and the cross-scale correlation and coupling optimization scheme, the basic data such as the amount of work, equipment parameters, area, and duration required for accounting are automatically extracted, and the preset formula is called to complete the full-scale accounting. The accounting details are output, and the accounting granularity can be refined to each building, each section of corridor, and each piece of green space.

[0116] Then, adaptive correction of the results is performed by introducing correction coefficients for interannual meteorological fluctuations, occupancy rates, and equipment aging and degradation. The calculation results are dynamically corrected by introducing correction coefficients for interannual meteorological fluctuations, occupancy rates, and equipment aging and degradation. The correction coefficients are automatically generated by fitting local meteorological data and operational data of similar residential areas over the past 10 years through a linear regression algorithm, which solves the problem of deviation between static calculation and actual operation.

[0117] Finally, the data is made fully traceable, and the entire accounting process automatically retains the data source, calculation formula, correction coefficient, and simulation log, generating a traceable accounting report that supports third-party institutions to review the accounting results throughout the entire process.

[0118] like Figure 2As shown in Part 1, the basic carbon reduction corresponding to each single-scale optimization scheme is calculated, and the total basic carbon reduction is obtained by summing them up, as shown in Formula 6: Formula Six; in, Based on the total reduction of carbon emissions, , , , These represent the basic carbon reduction amounts corresponding to point scale, line scale, surface scale, and volume support, respectively.

[0119] To elaborate, Defined as the carbon reduction resulting from the renovation of the building itself and its independent facilities ( In its calculation process, the carbon loss due to heat transfer in the building envelope is calculated, as shown in Formula 7: Formula 7; in, Carbon loss due to heat transfer in building envelope. Take the current DeST simulation value, Take the local building energy consumption carbon emission factor.

[0120] The carbon consumption of computing equipment is shown in Formula 8: Formula 8; in, Carbon consumption during equipment operation Take the existing equipment nameplate parameters respectively. Take the average operating time of similar equipment in the local area; Calculate point-scale carbon emissions before modification As shown in Formula Nine: Formula Nine; in, To improve carbon emissions at the point scale.

[0121] After the building envelope was modified, a new simulation was performed based on the heat transfer coefficient K value from the design scheme. After equipment replacement, the new equipment selected according to the design should be used. calculate .

[0122] The newly added distributed energy calculation is shown in Formula 10: Formula 10; in, To add distributed energy, Take the annual photovoltaic power generation of the design scheme.

[0123] Calculate point-scale carbon emissions after the renovation As shown in Formula 11: Formula 11; in, This is for point-scale carbon emissions after the transformation.

[0124] at this time, The calculation is shown in Formula Twelve: Formula twelve.

[0125] A calculation process table must be attached, clearly indicating the source of each parameter (such as design drawings, simulation reports), as the core basis for design scheme review.

[0126] Defined as the carbon reduction resulting from the transformation of transportation and pipeline systems ( In its calculation process, carbon reduction of the transportation system is calculated, as shown in Formula Thirteen: Formula Thirteen; in, To reduce carbon emissions in the transportation system, Take the optimized values ​​from the traffic simulation. Calculated based on the number of charging piles designed (≥15% of parking spaces).

[0127] Carbon reduction in the pipeline system is calculated using Formula Fourteen: Formula Fourteen; in, To reduce carbon emissions in the pipeline system, Take the pipeline length and unit energy consumption of the design scheme.

[0128] at this time, The calculation is shown in Formula 15: Formula 15.

[0129] In the design and application process, a comparison table of energy consumption before and after pipeline renovation and a calculation table of carbon reduction through traffic optimization must be attached to ensure that the data can be verified.

[0130] Defined as the carbon reduction resulting from the transformation of green spaces and public spaces ( In its calculation process, carbon reduction of public space lighting / facilities is calculated, as shown in Formula Sixteen: Formula Sixteen; in, Reduce carbon emissions for public space lighting / facilities. Take the total power of the luminaires after the design optimization.

[0131] The carbon reduction from green space maintenance is calculated as shown in Formula 17: Formula 17; in, To reduce carbon emissions for green space maintenance The value should be taken according to the designed ecological maintenance plan (reducing fertilizer use by more than 50%).

[0132] at this time, The calculation is shown in Formula 18: Formula 18.

[0133] In design and application, a comparison table of lighting equipment selection and a description of green space maintenance plan should be attached to clearly define the basis for carbon reduction.

[0134] Defined as the amount of carbon reduction resulting from the overall optimization of the residential system ( In its calculation process, the energy system is optimized, as shown in Formula 19: Formula 19; in, For energy system optimization, Take the energy transmission loss rate of the design scheme (3%-5%).

[0135] The carbon flow cycle optimization is calculated as shown in Equation 20: Formula 20; in, For carbon flow cycle optimization, The value is determined according to the designed waste sorting and recycling plan (recycling rate increased by more than 30%).

[0136] at this time, The calculation is shown in Formula 21: Formula 21.

[0137] In design and application, energy system optimization plans and waste sorting and recycling design specifications must be attached to support the calculation of carbon reduction.

[0138] Part 2 calculates the coupled carbon reduction increment corresponding to each cross-scale correlation and coupling optimization scheme, and sums them to obtain the total coupled carbon reduction increment, which can be shown in Formula 22: Formula 22; in, To couple the total increase in carbon reduction, , , These represent the total incremental carbon reduction from point to line, line to surface, and point to surface, respectively.

[0139] and , , The specific calculations have been described above and will not be repeated here.

[0140] Part 3 calculates the increase in carbon sink corresponding to each single-scale optimization scheme and sums them to obtain the total increase, as shown in Formula 23: Formula 23; in, To increase the total amount, , , , These represent the incremental amounts corresponding to point scale, line scale, surface scale, and volume support, respectively.

[0141] (Corresponding point-scale carbon sequestration) is defined as the amount of carbon sequestration brought about by building rooftops / vertical greening. In its calculation process, it is obtained through formula twenty-four: Formula 24; in, The value is taken from the roof greening projection area in the design drawings; Values ​​are taken according to the designed roof vegetation type (herbaceous plants 0.005-0.008, shrubs 0.01-0.015). The value is determined based on the vertical greening area as shown in the design drawings; Values ​​are taken according to the designed vertical vegetation type (0.004-0.006 for Virginia creeper).

[0142] (Corresponding linear scale carbon sequestration) is defined as the amount of carbon sequestration brought about by the new / optimized greening of the corridor. In its calculation process, it is obtained through formula twenty-five: Formula 25; in, The value is taken from the corridor greening area based on the current situation survey / design drawings; The value is taken as (0.012-0.018) based on the designed corridor vegetation community (multi-layered trees and shrubs).

[0143] (Corresponding surface-scale carbon sequestration) is defined as the increase in carbon sequestration brought about by the optimization of green space layout and the configuration of high carbon sequestration vegetation. It can be obtained through formula twenty-six: Formula 26; in, The values ​​are taken from the green space area according to the current situation survey / design drawings; The average carbon sequestration coefficient of the existing vegetation is taken as (0.006-0.009). The value is taken as 0.015-0.02 based on the designed high carbon sequestration vegetation (camphor tree, osmanthus).

[0144] (Corresponding to the volumetric scale increase in carbon sequestration) is defined as the incremental carbon sequestration brought about by the synergy of the entire carbon sequestration network. It can be obtained through formula twenty-seven: Formula 27; in, The value is taken from the average carbon sequestration coefficient when the carbon sinks at each scale operate independently; The carbon sequestration coefficient of the designed carbon sink network (after connection) is taken as (increased by 15%-25%). The value is taken from the total carbon sink area of ​​the residential area as shown in the design drawings.

[0145] At this point, the total carbon reduction increment is obtained by summing the basic total carbon reduction, the coupled carbon reduction increment, and the total increment, as shown in Formula 28: Formula 28; in, To reduce carbon emissions and increase total output.

[0146] S105: Based on the total amount of carbon reduction, perform target matching with the pre-set multi-level hierarchical targets.

[0147] Based on the total carbon reduction increase, the total carbon emissions before the optimization plan was updated, and the carbon density before the optimization plan was updated, the corresponding carbon density reduction ratio was determined.

[0148] The percentage decrease in carbon density is shown in Formula 29: Formula 29; in, To reduce the proportion of carbon density, ; To optimize the carbon density before the scheme update, it refers to the annual carbon emissions per unit land area of ​​the residential area before the update (unit: / (Ten thousand Carbon emission intensity (CPI) is a core benchmark indicator for measuring the carbon emission intensity of residential areas and is used to calculate the transformation target (carbon reduction rate).

[0149] The method for determining the value should prioritize the actual calculated value, as shown in Formula 30: Formula 30; in, The total annual carbon emissions of the residential area before the update ( ), which is the sum of carbon emissions at the point, line, surface, and volume scales ( The values ​​of each item are the same as those in the previous text. - The carbon emission accounting data is consistent with that before the renovation.

[0150] Residential land area (10,000 square meters) The calculation is based on the land boundary in the design drawings, deducting the area of ​​non-residential exclusive land such as urban roads and public green spaces, and retaining two decimal places.

[0151] Here is a calculation example: If an old residential area was before renovation... land area ,but .

[0152] For scenarios where existing data is missing or difficult to calculate, alternative value selection methods can be used. .

[0153] For example, using analogy with similar residential areas: Select 3-5 residential areas with similar construction years (±5 years), plot ratios (±0.3), and unit types (consistent proportion of multi-story / high-rise buildings) to this residential area that have already had their carbon density calculated; collect comparable residential areas. The data was calculated using a weighted average method (weights were assigned based on similarity: the highest similarity score had a weight of 0.4, decreasing by 0.1 for each subsequent score), and this average was used as the data for this residential area. Values.

[0154] Here is an example: 3 analog residential areas They are 38, 42, and 39 respectively. / (Ten thousand •a) If the similarity weights are 0.4, 0.3, and 0.3 respectively, then the original residence .

[0155] For example, the standard value method (referencing authoritative benchmark data): Referencing the "Urban Renewal Carbon Emission Reduction Benchmark Guidelines" issued by the relevant local authorities, values ​​are taken based on the type of residential area (old residential area / newly built residential area / mixed residential area). If there is no local standard, the data in the appendix of the corresponding standard can be referenced and adjusted according to the climate zone of the residential area (increase by 10% to 15% in severely cold / cold regions, and decrease by 5% to 10% in hot summer and warm winter regions).

[0156] Here is a method for validating the value: When using actual calculated values, carbon emission accounting tables for each scale must be attached, and the data authenticity must be verified by a third-party organization or design review group. When using the alternative value method, a selection explanation of comparable residential areas and the standard reference basis must be submitted to ensure that the value deviation is ≤10%. If the value deviation exceeds 10%, current monitoring data must be supplemented (such as energy consumption monitoring of 1-2 new buildings), and the calculation must be recalculated. .

[0157] Based on the carbon density reduction ratio, the ratio range of the pre-set multi-level hierarchical targets is compared to perform target matching.

[0158] like Figure 3 As shown, based on regional development plans, resource endowments, and policy requirements, three levels of targets and their corresponding carbon density reduction thresholds are set in advance:

[0159] Level 1, Low-carbon target: The carbon density after the update is reduced by 30% to 50% (excluding 50%) compared to before the update. The second level, near-zero carbon target: reduce carbon emissions by 50% to 80% (excluding 80%). Level 3: Reduce emissions by ≥80%, or achieve net-zero emissions through carbon sequestration and carbon offsetting.

[0160] At this point, the target is matched according to the carbon density reduction ratio (η): if (30%≤η<50%), the low carbon target is met; if (50%≤η<80%), the near-zero carbon target is met; if (η≥80%), the zero carbon target is met.

[0161] Based on the thresholds of the three-tiered carbon reduction targets, the "threshold edge" is defined as the range from the first 5 percentage points (inclusive) of the target's upper limit to the upper limit itself, specifically as follows: For the edge of the low-carbon target: 45% ≤ η < 50%.

[0162] This indicates that while this range meets the low-carbon target (30% ≤ η < 50%), it is close to the lower limit of the near-zero carbon target. This can be mitigated through slight optimization (such as increasing rooftop photovoltaic area by 10%) or purchasing a small amount of carbon credits (≤ 5%). ) Strive for near-zero carbon targets; if optimization costs are too high, low carbon targets can be maintained.

[0163] For the near-zero carbon target edge: 75%≤η<80% (excluding 80%).

[0164] This range indicates that it represents the upper limit of the near-zero carbon target. The gap between this range and the zero carbon target (η≥80%) is relatively small. Priority should be given to achieving the zero carbon target through optimized coupling strategies (such as increasing the photovoltaic absorption rate by 5%~8%). If site conditions are limited, carbon offsetting mechanisms (such as purchasing green electricity or forestry carbon sinks) can be introduced, with the offset amount required to be ≥2%~5% of the annual carbon emissions of the residential area.

[0165] For the edge of the zero-carbon target (special case): 78%≤η<80%.

[0166] This indicates that the range does not meet the η≥80% requirement of the zero carbon target, but is close to the threshold. Net zero emissions can be achieved through a combination of "technology optimization + carbon offsetting": for example, through technology optimization, increase carbon sink (such as adding 5%~10% of high carbon sequestration green space); through carbon offsetting, purchase carbon sink ≥ (80%-η)×ρ0×A to ensure that the final net carbon emissions are 0.

[0167] Of course, the determination of the threshold edge needs to be combined with the economics of the project. If the optimization cost accounts for more than 15% of the total investment, we can give up on striving for higher goals and maintain the current compliance status.

[0168] 1. Multi-scale system coverage: Integrating spatial forms such as buildings, transportation, energy, and green spaces into the assessment and optimization framework, breaking the limitations of traditional single scales, and realizing a systematic carbon reduction and sequestration design for all elements of the residential area.

[0169] 2. Coupling Mechanism Quantification: Based on the inter-scale correlation, cross-scale coupling optimization schemes are generated, the carbon reduction increment of coupling is clarified, the additional emission reduction benefits of scale synergy are explored, and a complete path from single-scale optimization to system integration is formed.

[0170] 3. Scientific accounting model: Based on the basic carbon reduction and carbon sequestration at each single scale, the total amount of carbon reduction increment is calculated by superimposing and coupling, which supports the itemized traceability and accurate evaluation of carbon reduction effect and overcomes the fuzzy problem of traditional accounting.

[0171] 4. Target-based matching: By matching the total amount with pre-set multi-level targets, quantitative correspondence between different targets such as low carbon, near-zero carbon, and zero carbon can be achieved, thereby enhancing the differentiation and guidance of strategy formulation.

[0172] 5. Applicability and Intelligent Support: The method is streamlined and can be adapted to different residential characteristics. It also supports the access of multi-source geospatial and energy consumption time series data, providing efficient tool support for scientific decision-making.

[0173] like Figure 4 As shown in the embodiments of this application, a device for achieving carbon reduction and sequestration in urban residential areas based on multi-scale collaborative methods is also proposed, comprising: At least one processor; and, A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, which enables the at least one processor to perform the multi-scale collaborative urban residential carbon reduction and sequestration method as described in any of the above embodiments.

[0174] This application also proposes a non-volatile computer storage medium storing computer-executable instructions, which are configured to implement the multi-scale collaborative urban residential carbon reduction and sequestration method as described in any of the above embodiments.

[0175] The various embodiments in this application are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the device and medium embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the description of the method embodiments.

[0176] The devices and media provided in this application are one-to-one with the methods. Therefore, the devices and media also have similar beneficial technical effects as their corresponding methods. Since the beneficial technical effects of the methods have been described in detail above, the beneficial technical effects of the devices and media will not be repeated here.

[0177] The above description is merely an embodiment of this application and is not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A method for achieving carbon reduction and sequestration in urban residential areas based on multi-scale collaborative approaches, characterized in that, include: For each spatial scale across multiple spatial scales, an assessment of the current energy consumption status is conducted using the corresponding geospatial data and residential energy consumption environmental time-series data collected. For each spatial scale, optimization is performed based on the energy consumption status assessment to generate a single-scale optimization scheme. Based on the correlation between different spatial scales, a cross-scale correlation coupling optimization scheme is generated on the basis of the single-scale optimization scheme. Specifically, this includes: constructing a full-element correlation map of the residential area using graph neural networks based on the digital twin data of the residential area; in the full-element correlation map, scale elements corresponding to each spatial scale are used as nodes, and the correlation relationships corresponding to each scale element are used as edges; the correlation relationships include material flow relationships, energy flow relationships, and spatial relationships; based on the full-element correlation map of the residential area, strong correlation relationships between nodes are identified, and the synergistic carbon reduction sensitivity of each strong correlation relationship is determined using response surface methodology. The carbon reduction increment per unit renovation cost is obtained based on the synergistic carbon reduction sensitivity, and priority is given to... The algorithm first ranks nodes according to their priority, then selects nodes corresponding to the highest-ranking strong correlations. Based on a pre-defined library of coupling techniques, it generates corresponding cross-scale correlation coupling optimization schemes using a genetic algorithm. The algorithm then simulates the cross-scale correlation coupling optimization schemes using the digital twin data of the residential area, outputting the final cross-scale correlation coupling optimization scheme. Specifically, generating the corresponding cross-scale correlation coupling optimization schemes using a genetic algorithm based on the pre-defined library of coupling techniques includes: for point-line strong correlations corresponding to point and line scales, selecting a Long Short-Term Memory (LSTM) network from the pre-defined library of coupling techniques, using the time-series data of the residential area's energy consumption environment as training data, and training the LTM network to obtain the load... - Output prediction model; and predict the dynamic characteristics of energy supply and demand in the residential area during the time period through the load-output prediction model; according to the dynamic characteristics of energy supply and demand during the time period, add corresponding dynamic scheduling strategies for energy supply and demand in the process of generating cross-scale correlation coupling optimization scheme through genetic algorithm; for the strong correlation between line and surface scales, select U-Net semantic segmentation model and Frechet distance algorithm from the preset coupling technology measure library, perform multi-source raster image preprocessing on the geospatial data, and perform pixel segmentation through U-Net semantic segmentation model to extract corridor elements; and perform spatial overlay on various types of corridor elements through Frechet distance algorithm. The t-distance algorithm calculates the spatial overlap of the corridor centerlines of various corridor elements to obtain the composite utilization area. Based on the composite utilization area, during the process of generating a cross-scale correlation coupling optimization scheme through a genetic algorithm, spatial conflict detection and conflict avoidance are performed on the composite utilization area. For the point-area strong correlation relationship corresponding to the point scale and the area scale, the U-Net semantic segmentation model, building information model parsing and minimum spanning tree algorithm are selected from the preset coupling technology measures library. For the digital twin data of the residential area, pixel segmentation is performed through the U-Net semantic segmentation model and model parsing is performed through the building information model parsing to identify the carbon sink nodes corresponding to the evaluation range of the point scale.Based on the minimum spanning tree algorithm, with each carbon sink node as an endpoint, the strip space as a wiring constraint, and the goal of minimizing the total path length and the spatial modification of the residential area, an ecological corridor path connecting each carbon sink node is planned and generated. The connectivity index of the three-dimensional carbon sink network corresponding to the volume scale assessment range is obtained through the connectivity index of landscape ecology. Through the genetic algorithm, with the objective functions of maximizing photovoltaic installed capacity, maximizing carbon sink volume, and minimizing transmission distance, a cross-scale correlation and coupling optimization scheme corresponding to the photovoltaic panel layout is generated. Based on the basic carbon reduction and carbon sequestration amounts corresponding to each single-scale optimization scheme, and the coupled carbon reduction increments corresponding to each cross-scale coupled optimization scheme, the total carbon reduction and carbon sequestration amount is obtained. Based on the total amount of carbon reduction, target matching is performed with pre-set multi-level hierarchical targets.

2. The method for achieving carbon reduction and sequestration in urban residential areas based on multi-scale collaborative methods according to claim 1, characterized in that, An assessment of the current energy consumption status is conducted using the collected geospatial data and residential energy consumption environmental time-series data, specifically including: The drones acquire the corresponding remote sensing image data and digital surface models, and the sensors collect the corresponding residential area energy time series data and environmental data. By using point cloud registration and timestamp synchronization, the remote sensing image data, the digital surface model, the energy time series data, and the environmental data are spatially registered and temporally aligned to obtain residential area digital twin data. Based on the digital twin data of the residential area, the corresponding core features are extracted; the core features include the year of construction of the residential area, the slope of the terrain, and the degree of functional mixing. A scene classifier is constructed based on the random forest algorithm. The core features are used as input to determine the corresponding residential area type and match the corresponding evaluation model. The evaluation model contains different evaluation dimensions and / or the evaluation weights of each evaluation dimension for different residential area types. The energy consumption status is assessed using the assessment model based on the digital twin data of the residential area to obtain energy efficiency assessment data. The energy efficiency assessment data and on-site sampling data are then sampled and verified using the Monte Carlo simulation method. When the data variation coefficient reaches a preset value, the energy efficiency assessment data is reassessed and / or the on-site sampling data is supplemented.

3. The method for achieving carbon reduction and sequestration in urban residential areas based on multi-scale collaborative methods according to claim 2, characterized in that, The spatial morphological scales include point scale, line scale, surface scale, and volume scale; The assessment scope at the point scale corresponds to the building itself and independent facilities and equipment in urban residential areas; The assessment scope at the linear scale corresponds to transportation facilities and pipeline systems in urban residential areas; The assessment scope at the surface scale corresponds to ecological green spaces and public spaces in urban residential areas; The volumetric scale assessment range corresponds to the overall collaborative system of the residential area obtained by integrating the assessment ranges corresponding to the point scale, the line scale, and the surface scale.

4. The method for achieving carbon reduction and sequestration in urban residential areas based on multi-scale collaborative methods according to claim 1, characterized in that, Using the digital twin data of the residential area, the cross-scale correlation and coupling optimization scheme is simulated, and the final cross-scale correlation and coupling optimization scheme is output, specifically including: The cross-scale correlation and coupling optimization scheme is simulated using the digital twin data of the residential area; If the simulation results do not reach the verification threshold, the core influencing factors of the deviation are located by Pearson correlation coefficient analysis. The identification dimensions include: photovoltaic installed capacity, energy storage equipment capacity, charging pile layout, transmission distance, scheduling strategy parameters, and pump station operating power. Based on the core influencing factor of deviation, the particle swarm optimization algorithm is used to automatically iterate and optimize the core influencing factor of deviation with the objective function of maximizing photovoltaic absorption rate and minimizing grid electricity consumption, and with the verification threshold as the constraint condition. The optimized parameter combination of the identification dimension is then output until the verification threshold is reached or the scheme reconstruction is triggered.

5. The method for achieving carbon reduction and sequestration in urban residential areas based on multi-scale collaborative methods according to claim 1, characterized in that, Based on the basic carbon reduction and carbon sequestration amounts corresponding to each single-scale optimization scheme, and the coupled carbon reduction increments corresponding to each cross-scale coupled optimization scheme, the total carbon reduction increment is obtained, which specifically includes: Calculate the basic carbon reduction corresponding to each single-scale optimization scheme separately, and sum them to obtain the total basic carbon reduction; Calculate the coupled carbon reduction increment corresponding to each cross-scale correlation and coupling optimization scheme, and sum them to obtain the total coupled carbon reduction increment; Calculate the increase in sink volume corresponding to each single-scale optimization scheme separately, and sum them to obtain the total increase volume; The total carbon reduction is obtained by summing the basic total carbon reduction, the coupled incremental carbon reduction, and the total increase.

6. The method for achieving carbon reduction and sequestration in urban residential areas based on multi-scale collaborative methods according to claim 1, characterized in that, Based on the total carbon reduction increase, target matching is performed with pre-set multi-level hierarchical targets, specifically including: Based on the total carbon reduction increase, the total carbon emissions before the optimization plan update, and the carbon density before the optimization plan update, determine the corresponding carbon density reduction ratio. Based on the carbon density reduction ratio, it is compared with the ratio range of a pre-set multi-level hierarchical target to perform target matching.

7. A device for achieving carbon reduction and sequestration in urban residential areas based on multi-scale collaborative methods, characterized in that, include: At least one processor; as well as, A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, which enables the at least one processor to perform the multi-scale collaborative urban residential carbon reduction and sequestration method as described in any one of claims 1 to 6.

8. A non-volatile computer storage medium storing computer-executable instructions, characterized in that, The computer-executable instructions are configured to implement the multi-scale collaborative urban residential carbon reduction and sequestration method as described in any one of claims 1 to 6.

Images