Green garden energy-saving design simulation model construction method based on fusion digital twinning

Abstract

The invention discloses a green garden energy-saving design simulation model construction method based on fusion digital twinning, and relates to the technical field of gardens, and the method comprises the steps: S1, data collection and demand analysis; s2, building and enhancing a BIM model; s3, multi-physics field simulation integration is carried out; s4, digital twinning integration; and S5, verifying and optimizing. According to the green garden energy-saving design simulation model construction method based on fusion of digital twinning, multi-software collaborative modeling is realized by constructing a green garden design basic database and developing a customized plug-in, various aspects of garden design, including vegetation ecological benefits, soil hydrological processes and the like, can be comprehensively covered, and the method has the advantages of being high in practicability and high in practicability. The capability of simulating natural systems such as specific vegetation ecological benefits and soil hydrological processes of gardens is high, so that an accurate simulation result is provided; meanwhile, the advantages of digital twinning are fused, energy consumption simulation, hydrological simulation and ecological simulation are integrated, comprehensive simulation of multiple physical fields is achieved, and the performance of the garden system can be comprehensively evaluated.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of gardens, in particular to a green garden energy-saving design simulation model construction method based on fusion digital twinning. BACKGROUND

[0002] The green garden energy-saving design simulation model is an important tool for modern landscape architectural design and ecological sustainable development. It simulates and predicts key indicators such as energy consumption, water resource utilization, and vegetation ecological benefits of the garden system through digital means. With the advancement of the "double carbon" goal, the construction industry is undergoing profound changes from traditional modes to green, low-carbon, and intelligent directions, and green building standards have become a "must-have" rather than a "optional". The most mainstream simulation model construction method is based on BIM digital modeling.

[0003] The core of the BIM-based digital modeling method is to integrate various related information of the engineering project with three-dimensional digital technology to form a data-rich building information model, and to realize energy-saving performance simulation through professional analysis tools. In green garden design, common BIM energy consumption simulation tools include Ecotect and other software. These tools can integrate building envelope insulation system data, natural ventilation design parameters, and rainwater reuse system information to form a comprehensive energy-saving analysis model.

[0004] However, in actual operation, although the BIM integrated modeling method has many advantages, there are still obvious limitations in the application of green garden energy-saving design. The primary problem is that the professional coverage is not complete. Due to the focus of software developers on building professional needs and the inherent complexity of vegetation ecological process simulation, existing BIM software is mostly developed for architectural design and structural engineering, and has weak modeling capabilities for natural systems such as vegetation ecological benefits and soil hydrological processes unique to landscape design. For example, animal habitat design, ecological corridor planning, and food source plant configuration, which need to be considered in landscape design, lack specialized modeling tools and analysis modules in mainstream BIM software, resulting in poor model results. SUMMARY

[0005] The purpose of the present application is to provide a green garden energy-saving design simulation model construction method based on fusion digital twinning to solve the problems raised in the background art.

[0006] To achieve the above purpose, the present application provides the following technical solution: a green garden energy-saving design simulation model construction method based on fusion digital twinning, comprising the following steps: S1, Data collection and demand analysis: Collect data to build a green garden design database, which is used for subsequent comprehensive simulation of the operation of the garden system, and to determine the goals and needs of green garden energy-saving design, as well as the key indicators that the simulation model needs to simulate, including energy consumption simulation, water resource utilization, and vegetation ecological benefit evaluation, etc. S2, BIM model construction and enhancement: Evaluate the functions of existing BIM software, match the software according to design requirements, and develop custom plug-ins to realize multi-software collaborative modeling and output three-dimensional BIM models. S3, Multi-physical field simulation integration: S31, Energy consumption simulation (EnergyPlus calculates air conditioning load): EnergyPlus calculates air conditioning load through thermodynamic models, combined with BIM geometric parameters (such as wall thermal conductivity, window shading coefficient) and meteorological data, to simulate energy consumption demand in different seasons or time periods. S32, Hydrological simulation (SWMM simulates stormwater runoff): SWMM (Storm Water Management Model) simulates surface runoff, pipe drainage, and other processes, combined with BIM terrain data to identify areas prone to water accumulation, and assess the storage effect of sponge city facilities (such as rain gardens), while digital twins can dynamically update rainfall prediction data to achieve real-time flood control scheduling. S33, Ecological simulation (i-Tree evaluates vegetation carbon sequestration): i-Tree calculates the carbon sequestration capacity of vegetation through biomass models, combined with BIM plant spatial distribution, to analyze the carbon sink potential of different configuration schemes, while digital twins simulate tree growth cycles to predict long-term carbon absorption trends. S4, Digital twin integration: Integrate energy consumption simulation, hydrological simulation, and ecological simulation in a unified digital twin platform, achieve data interconnection through DLMS five-layer architecture (physical layer, data layer, model layer, function layer, interaction layer), and use IoT sensors to update real-time meteorological, soil moisture, and other data, adjust simulation parameters, and form a "monitoring-simulation-optimization" closed loop. S5, Verification and optimization: S51, Simulation result comparison: Compare the air conditioning load calculated by EnergyPlus with the electricity meter data to calculate the root mean square error; Compare the runoff volume simulated by SWMM with the actual drainage monitoring point data to verify the Nash efficiency coefficient, NSE>0.7 is qualified; S52, Vegetation carbon sequestration verification: Scan the vegetation canopy volume using LiDAR and perform regression analysis (R²≥0.8) with the carbon sink volume estimated by i-Tree; S53, Multi-objective optimization: NSGA-II optimization algorithm is adopted to balance multiple objectives such as energy saving (lowest energy consumption), ecology (highest carbon sink), cost (lowest cost of LID facilities), etc., and Pareto frontier analysis is performed to output a non-inferior solution set for decision makers to weigh and select.

[0007] Further, in step S1, an automatic network crawler system is used for data collection. Specifically, a distributed crawler cluster based on Python Scrapy framework is deployed to perform targeted collection on target websites, including official websites of natural resource departments (land use / vegetation cover data), historical data platforms of meteorological bureaus (near 30 years of meteorological data), monitoring sites of environmental protection departments (air quality / water quality data), and urban planning display networks (surrounding building / road network data). The crawler system is configured with an automatic cleaning pipeline to perform the following preprocessing operations on the collected raw data: Outlier filtering: automatically delete abnormal records, such as temperature > 50℃ or <-20℃; Temporal and spatial alignment: unified conversion to WGS84 coordinate system and UTC time zone; Missing data interpolation: Kriging spatial interpolation method is used.

[0008] Further, in step S1, the green garden design basic database includes the following data: 1) Geospatial data: terrain data, boundary data, land use data, etc., wherein the accuracy requirement of terrain data is plane error ≤0.1m and height error ≤0.05m; 2) Climate and meteorological data: Conventional meteorology: temperature, humidity, precipitation, wind speed and direction, etc.; Special parameters: evaporation, solar radiation, relative humidity, etc.: Extreme events: number of heat waves (≥35℃), heavy rain return period data (50 years / 100 years), etc.; 3) Ecological background data: Vegetation data: species list, growth parameters, community structure, etc., updated every year during the growing season; Soil data: physical properties (volume, permeability coefficient), chemical properties (pH value, organic matter content, heavy metal content), etc.; 4) Hydrogeological data: surface water (river / lake water level fluctuation, water quality indicators), groundwater (depth, flow direction, mineralization), rainwater runoff (runoff coefficient, peak flow), etc.; 5) Social and economic data: population density (determine garden service pressure), traffic flow (noise pollution assessment), cultural protection points (avoid design conflicts such as ancient trees), etc.

[0009] Further, in step S1, the key indicators that the simulation model needs to simulate are as follows: Energy consumption simulation: assess the energy consumption of the garden system, including lighting, irrigation, equipment operation, etc. Water resource utilization: simulate the water resource demand and supply of the garden system, including rainwater collection, irrigation efficiency, drainage system, etc. Vegetation ecological benefit evaluation: evaluate the improvement of vegetation on the environment, including air purification, temperature regulation, biodiversity protection, etc.

[0010] Further, in step S2, the matching of BIM software and design requirements is as follows: For building / terrain modeling, use Autodesk Revit; For vegetation design, use Vectorworks Landmark; For hydrological engineering, use Bentley OpenBuildings Designer; Multi-software collaborative process: Use Revit to establish building / terrain model, Vectorworks Landmark to design vegetation, and Bentley OpenBuildings Designer to design rainwater runoff.

[0011] Further, in step S2, the customized plug-ins include: Plug-in 1, Revit ecological parameter extension: inject Vectorworks plant data into Revit model; Plug-in 2, cross-platform data verifier: ensure that the coordinate system / unit of the three software models is consistent.

[0012] Further, in step S2, the three-dimensional BIM model is exported in IFC / gbXML format data for subsequent simulation, with precision requirement LOD350.

[0013] Further, in step S4, the digital twin platform building process: based on Unity3D / Unreal engine to develop visual interface, and configure Azure IoT Hub to receive real-time sensor data, and develop data dashboard (such as PowerBI) to display dynamic indicators such as energy consumption / water level, finally output the virtual-real interactive digital twin system.

[0014] Further, in step S4, the five-layer architecture of DLMS is as follows: Physical layer: As the bottom layer, it is responsible for connecting physical devices (such as sensors, smart meters) with digital systems, using standardized communication interfaces (such as RS-485, PLC power carrier or wireless LoRa), and following the physical layer specification of DLMS / COSEM (such as HDLC protocol) to realize the collection and transmission of raw data. In the green garden scene, it includes temperature and humidity sensors, soil monitoring devices, etc. Data layer: It cleans, stores and manages the raw data collected by the physical layer, and provides a unified data access interface for the upper layer. Through the COSEM object model defined by the DLMS protocol (such as "association SN / LN object"), it converts heterogeneous data (energy consumption, hydrology, ecology) into a structured format. The storage architecture adopts a hierarchical design, and the raw data is stored in a time series database, and the processed data enters the data warehouse for analysis. Model layer: It encapsulates multi-physical field simulation logic (such as the algorithms of EnergyPlus, SWMM, i-Tree), builds the core computing capability of digital twin, and integrates different simulation tools through FMI (functional Mock-up interface). Function layer: It defines business logic and automation rules, coordinates the operation of multiple models, supports decision optimization, exposes APIs in the form of microservices for simulation engines, and integrates DLMS encryption and authentication (such as AES-128 encryption, two-way identity verification) to ensure data interaction security. Interaction layer: It provides human-computer interaction interface and real-time control interface, forms a "monitoring-simulation-optimization" closed loop, displays three-dimensional simulation results of multi-physical field (such as heat map, runoff animation), and supports dynamic parameter adjustment. Specifically, it uses IoT feedback control to issue optimization instructions to physical devices (such as adjusting irrigation strategy) through the "request-response" mechanism of DLMS / COSEM.

[0015] The present application provides a green garden energy-saving design simulation model construction method based on fusion digital twin, which has the following beneficial effects: 1、The present application can comprehensively cover all aspects of garden design, including vegetation ecological benefits, soil hydrological processes, etc., by constructing a green garden design database and developing custom plug-ins to realize multi-software collaborative modeling, and has strong modeling capability for natural systems such as vegetation ecological benefits and soil hydrological processes unique to gardens, thereby providing accurate simulation results.

[0016] 2. This invention integrates multiphysics simulation results through a digital twin platform and uses IoT sensors to update data in real time, which can dynamically adjust simulation parameters to make the model more closely resemble the actual situation. Combined with a verification mechanism, it significantly improves the accuracy and reliability of the simulation and solves the problem that traditional BIM models have relatively fixed data after construction, making it difficult to update and adjust them in real time according to the actual situation. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating the steps of the method for constructing a green garden energy-saving design simulation model based on integrated digital twins according to the present invention. Figure 2 This is an example diagram of ecological parameters for the method of constructing a green garden energy-saving design simulation model based on integrated digital twins according to the present invention; Figure 3 This is a schematic diagram of the cross-platform data validator operation logic of the green garden energy-saving design simulation model construction method based on integrated digital twins of the present invention; Figure 4 This table shows the simulation results verification of the simulation model construction method for green garden energy-saving design based on integrated digital twins of the present invention. Detailed Implementation

[0018] The embodiments of the present invention will be described in further detail below with reference to the accompanying drawings and examples. The following examples are for illustrative purposes only and should not be construed as limiting the scope of the invention.

[0019] like Figures 1-4 As shown, the method for constructing a green garden energy-saving design simulation model based on integrated digital twins includes the following steps: S1. Data Acquisition and Requirements Analysis: Data is collected to build a basic database for green landscape design, which will be used for subsequent comprehensive simulation of the operation of the landscape system, clarify the goals and requirements of green landscape energy-saving design, and determine the key indicators that the simulation model needs to simulate, including energy consumption simulation (assessing the energy consumption of the landscape system, including energy consumption for lighting, irrigation, equipment operation, etc.), water resource utilization (simulating the water resource demand and supply of the landscape system, including rainwater harvesting, irrigation efficiency, drainage system, etc.), and vegetation ecological benefit assessment (assessing the environmental improvement effect of vegetation, including air purification, temperature regulation, biodiversity protection, etc.).

[0020] In this embodiment, an automated web crawler system is used for data collection. Specifically, a distributed crawler cluster based on the Python Scrapy framework is deployed to collect data from targeted websites, including the official website of the natural resources department (land use / vegetation cover data), the historical data platform of the meteorological bureau (near 30 years of meteorological data), the monitoring site of the environmental protection department (air quality / water quality data), and the city planning display network (surrounding building / road network data). The crawler system is configured with an automatic cleaning pipeline to perform the following preprocessing operations on the collected raw data: Outlier filtering: automatically delete abnormal records, such as temperature > 50℃ or <-20℃; Temporal and spatial alignment: unified conversion to WGS84 coordinate system and UTC time zone; Missing data interpolation: Kriging spatial interpolation method is used.

[0021] In this embodiment, the green garden design foundation database includes the following data: 1) Geospatial data: topographic data, boundary data, land use data, etc., where the accuracy requirement for topographic data is plane error ≤ 0.1m and elevation error ≤ 0.05m.

[0022] 2) Climate and weather data: Conventional weather: temperature, humidity, precipitation, wind speed and direction, etc.

[0023] Special parameters: evaporation, solar radiation, relative humidity, etc.

[0024] Extreme events: number of heat waves (≥35℃), heavy rain return period data (50 years / 100 years), etc.

[0025] 3) Ecological background data: Vegetation data: species list, growth parameters, community structure, etc., updated every year during the growing season.

[0026] Soil data: physical properties (volume, permeability coefficient), chemical properties (pH value, organic matter content, heavy metal content), etc.

[0027] 4) Hydrogeological data: surface water (river / lake water level fluctuation, water quality indicators), groundwater (depth, flow direction, mineralization), rainwater runoff (runoff coefficient, peak flow), etc.

[0028] 5) Social and economic data: population density (determine garden service pressure), traffic flow (noise pollution assessment), cultural protection points (avoid design conflicts such as ancient trees), etc.

[0029] S2, BIM model construction and enhancement: Evaluate the functions of existing BIM software, match the software according to the design requirements, use Revit to establish the building / terrain model, Vectorworks Landmark to design vegetation, Bentley OpenBuildings Designer to design rainwater runoff, and develop custom plug-ins to realize multi-software collaborative modeling. Output three-dimensional BIM model, and the model exports IFC / gbXML format data, with precision requirement LOD350.

[0030] In this embodiment, the multi-software collaborative modeling process is as follows: Operation 1, Revit basic modeling: Use "Dynamo" script to generate terrain surface (import survey point CSV), create building model with pre-embedded device pipeline wall bushing, output: IFC with geographic coordinates ExportIFC(revit_model, geolocation=(121.47,31.23)).

[0031] Operation 2, Vectorworks vegetation configuration: Link Revit terrain as a base map, use "Plant Tool" to arrange native tree species, and associate ecological parameters. For example, as shown in the figure, the software automatically calculates the planting spacing and generates a 3D preparation model. Figure 2

[0032] Operation 3, Bentley hydrological analysis: Import Revit terrain and Vectorworks vegetation data, use "Sustain" module to calculate runoff coefficient: C=∑(Areai×RunoffCoeffi) / TotalArea, output: pipe network slope adjustment PDF+3D DGN; Use "Flow Reduction Calculator" to calculate runoff coefficient: C=0.15*(green area)+0.9*(hardened area), output: rainwater pipe network optimization scheme (DGN format).

[0033] In this embodiment, the custom plug-ins include: Plug-in 1, Revit ecological parameter expansion: inject Vectorworks plant data into Revit model.

[0034] Plug-in 2, cross-platform data verifier: as shown in the figure, ensure that the coordinate systems / units of the three software models are consistent. Figure 2

[0035] S3, multi-physical field simulation integration: ​​S31, Energy consumption simulation (EnergyPlus calculates air conditioning load) 1) Input data: BIM geometric model: Provides basic data such as building structure, spatial layout, material properties, etc., ensuring consistency between virtual models and physical entities.

[0036] Meteorological data: Including temperature, humidity, solar radiation, etc., used to simulate the impact of dynamic environment on building energy consumption.

[0037] 2) Operation process: EnergyPlus calculates air conditioning load through thermodynamic models, combined with BIM geometric parameters (such as wall thermal conductivity, window shading coefficient) and meteorological data, to simulate energy consumption demand in different seasons or time periods.

[0038] 3) Output results: Generate energy consumption reports, including hourly electricity consumption, peak load, etc., for optimizing equipment selection or energy-saving strategies (such as night pre-cooling).

[0039] S32, Hydrological simulation (SWMM simulates stormwater runoff) 1) Input data: BIM terrain model: Terrain elevation, slope, permeability, etc., affecting stormwater runoff paths; Rainfall intensity formula: Based on historical or design rainfall data (such as return period), determine the spatio-temporal distribution of rainfall.

[0040] 2) Operation process: SWMM (Storm Water Management Model) simulates surface runoff, pipe drainage, etc., combined with BIM terrain data to identify areas prone to waterlogging, and assess the detention effect of sponge city facilities (such as rain gardens), while digital twins can dynamically update rainfall prediction data to achieve real-time flood control scheduling.

[0041] 3) Output results: Provide runoff coefficient (ratio of surface runoff to rainfall), quantify rainwater management efficiency, and assist in designing low-impact development (LID) measures.

[0042] S33, Ecological simulation (i-Tree evaluates vegetation carbon sequestration) 1) Input data: BIM plant list: Tree species, age, canopy size, etc., determine photosynthetic efficiency.

[0043] 2) Operation process: i-Tree calculates the carbon sequestration capacity of vegetation through biomass models, combined with the spatial distribution of plants in BIM, analyzes the carbon sink potential of different configuration schemes, and digital twins simulate tree growth cycles to predict long-term carbon absorption trends.

[0044] 3) Output results: Generate carbon sink quantitative values (such as tons of CO2 per year), support ecological benefit evaluation and carbon neutralization target planning.

[0045] S4, Digital Twin Integration: Integrate energy consumption simulation, hydrological simulation, and ecological simulation in a unified digital twin platform. Achieve data interconnection through the DLMS five-layer architecture (physical layer, data layer, model layer, function layer, and interaction layer). Real-time update weather, soil moisture, and other data using IoT sensors, adjust simulation parameters, and form a "monitoring-simulation-optimization" closed loop.

[0046] In this embodiment, the digital twin platform construction process: develop a visual interface based on Unity3D / Unreal Engine, configure Azure IoT Hub to receive real-time sensor data, and develop data dashboards (such as Power BI) to display dynamic indicators such as energy consumption and water level. Finally, output a virtual-real interactive digital twin system. The DLMS five-layer architecture of the digital twin platform is as follows: Physical layer: As the bottom layer, it is responsible for connecting physical devices (such as sensors, smart meters) and digital systems. It uses standardized communication interfaces (such as RS-485, PLC power carrier, or wireless LoRa) and follows the physical layer specifications of DLMS / COSEM (such as HDLC protocol) to achieve raw data collection and transmission. In the green garden scenario, it includes temperature and humidity sensors, soil monitoring devices, etc.

[0047] Data layer: Clean, store, and manage the raw data collected by the physical layer, providing a unified data access interface for the upper layer. Through the COSEM object model defined by the DLMS protocol (such as "associated SN / LN object"), heterogeneous data (energy consumption, hydrology, ecology) is converted into a structured format. The storage architecture uses a hierarchical design, with raw data stored in a time series database and processed data stored in a data warehouse for analysis.

[0048] Model layer: Encapsulate multi-physical field simulation logic (such as EnergyPlus, SWMM, i-Tree algorithms), build the core computing capability of digital twin, and integrate different simulation tools through FMI (functional Mock-up interface).

[0049] Function layer: Define business logic and automation rules, coordinate the operation of multiple models, support decision optimization, expose API for simulation engines in the form of microservices, and integrate DLMS encryption and authentication (such as AES-128 encryption, two-way identity verification) to ensure data interaction security.

[0050] Interaction layer: Provide human-computer interaction interface and real-time control interface, form a "monitoring-simulation-optimization" closed loop, three-dimensional display of multi-physical field simulation results (such as heat map, runoff animation), and support parameter dynamic adjustment, specifically using IoT feedback control, through the "request-response" mechanism of DLMS / COSEM, the optimization instruction is issued to the physical device (such as adjusting the irrigation strategy).

[0051] S5, verification and optimization: S51, simulation result comparison: Compare the air conditioning load calculated by EnergyPlus with the electricity meter data, and calculate the root mean square error.

[0052] Compare the runoff simulated by SWMM with the actual drainage monitoring point data, verify the Nash efficiency coefficient, and NSE>0.7 is qualified.

[0053] S52, verification of carbon sequestration of vegetation: The volume of the vegetation canopy is scanned by LiDAR, and the carbon sink estimated by i-Tree is regressed (R²≥0.8). The verification result of the embodiment is shown in Figure 4 .

[0054] S53, multi-objective optimization: Adopt NSGA-II optimization algorithm, balance energy saving (lowest energy consumption), ecology (highest carbon sink), cost (lowest LID facility cost) and other multi-objectives, and Pareto frontier analysis, output non-inferior solution set, for decision makers to weigh and choose.

[0055] Embodiments of the present application are given for the purpose of example and description, and are not exhaustive or limit the present application to the disclosed form. Many modifications and variations will be apparent to those of ordinary skill in the art. The embodiments are chosen and described in order to better illustrate the principles and practical application of the present application, and to enable those of ordinary skill in the art to understand the present application in order to design various embodiments with various modifications for specific use.

Claims

1. A method for constructing a green garden energy-saving design simulation model based on fusion digital twinning, characterized in that, Includes the following steps: S1. Data Collection and Needs Analysis: Collect data to build a basic database for green landscape design, clarify the goals and needs of green landscape energy-saving design, and determine the key indicators that the simulation model needs to simulate, including energy consumption simulation, water resource utilization, and vegetation ecological benefit assessment. S2, BIM Model Construction and Enhancement: Evaluate the functionality of existing BIM software, match software according to design requirements, and develop customized plugins to achieve collaborative modeling of multiple software programs and output a 3D BIM model; S3, Multiphysics Simulation Integration: S31. Energy consumption simulation: EnergyPlus calculates air conditioning load through a thermodynamic model and combines BIM geometric parameters and meteorological data to simulate energy consumption demand in different seasons or time periods. S32. Hydrological Simulation: SWMM simulates surface runoff and pipeline drainage processes, combines BIM topographic data to identify areas prone to water accumulation, and evaluates the retention effect of sponge city facilities. S33. Ecological Simulation: i-Tree calculates the carbon sequestration capacity of vegetation through a biomass model and analyzes the carbon sink potential of different configuration schemes by combining the spatial distribution of plants in BIM. S4. Digital Twin Integration: Integrate energy consumption simulation, hydrological simulation, and ecological simulation in a unified digital twin platform. Achieve data interoperability through the DLMS five-layer architecture and utilize IoT sensors to update data in real time and adjust simulation parameters.

2. The green garden energy-saving design simulation model construction method based on fusion digital twinning according to claim 1, characterized in that, In step S1, an automated web crawler system is used for data collection. Specifically, a distributed crawler cluster based on the Python Scrapy framework is deployed to collect data from target websites, including the official websites of natural resources departments, historical data platforms of meteorological bureaus, monitoring stations of environmental protection departments, and urban planning display websites. The crawler system is configured with an automated cleaning pipeline to perform the following preprocessing operations on the collected raw data: Outlier filtering: Automatically deletes outlier records; Spatiotemporal alignment: uniformly convert to WGS84 coordinate system and UTC time zone; Missing data imputation: Kriging spatial interpolation method is used.

3. The green garden energy-saving design simulation model construction method based on fusion digital twinning according to claim 1, characterized in that, In step S1, the green landscape design basic database includes the following data: Geospatial data: topographic data, boundary data, and land use data. The accuracy requirements for topographic data are: horizontal error ≤ 0.1m and elevation error ≤ 0.05m. Climate and meteorological data: Standard meteorological data: temperature, humidity, precipitation, wind speed and direction; Special parameters: evaporation rate, solar radiation, relative humidity: Extreme events: heatwave days, heavy rain recurrence interval data; Ecological baseline data: Vegetation data: species list, growth parameters, community structure, updated annually during the growing season; Soil data: physical properties, chemical properties; Hydrogeological data: surface water, groundwater, and stormwater runoff; Socioeconomic data: population density, traffic flow, cultural preservation sites.

4. The method for constructing a green garden energy-saving design simulation model based on fused digital twins according to claim 1, characterized in that, In step S1, the key indicators that the simulation model needs to simulate are as follows: Energy consumption simulation: Evaluate the energy consumption of the garden system, including energy consumption for lighting, irrigation, and equipment operation; Water resource utilization: Simulate the water resource demand and supply of the garden system, including rainwater harvesting, irrigation efficiency, and drainage system; Vegetation ecological benefits assessment: assessing the environmental improvement effects of vegetation, including air purification, temperature regulation, and biodiversity conservation.

5. The method for constructing a green garden energy-saving design simulation model based on fused digital twins according to claim 1, characterized in that, In step S2, the matching status between the BIM software and the design requirements is as follows: For building / terrain modeling, use Autodesk Revit; For vegetation design, use Vectorworks Landmark; For hydrological engineering, use Bentley OpenBuildings Designer; Multi-software collaboration process: Use Revit to create building / terrain models, Vectorworks Landmark to design vegetation, and BentleyOpenBuildings Designer to design rainwater runoff.

6. The method for constructing a green garden energy-saving design simulation model based on fused digital twins according to claim 1, characterized in that, In step S2, the customized plugin includes: Plugin 1, Revit Ecological Parameter Extension: Injects Vectorworks plant data into the Revit model; Plugin 2, Cross-Platform Data Validator: Ensures that the coordinate systems / units of the three software models are consistent.

7. The method for constructing a green garden energy-saving design simulation model based on fused digital twins according to claim 1, characterized in that, In step S2, the 3D BIM model is exported as IFC / gbXML format data with a required accuracy of LOD350.

8. The method for constructing a green garden energy-saving design simulation model based on fused digital twins according to claim 1, characterized in that, In step S4, the digital twin platform construction process involves: developing a visual interface based on the Unity3D / Unreal engine, configuring Azure IoT Hub to receive real-time sensor data, developing a data dashboard to display dynamic indicators of energy consumption / water level, and finally outputting a digital twin system that allows for virtual-real interaction.

9. The method for constructing a green garden energy-saving design simulation model based on fused digital twins according to claim 1, characterized in that, In step S4, the DLMS five-layer architecture is as follows: Physical layer: As the lowest layer, it is responsible for connecting physical devices and digital systems. Specifically, it adopts standardized communication interfaces and follows the physical layer specifications of DLMS / COSEM to realize the acquisition and transmission of raw data. Data layer: Cleans, stores and manages the raw data collected from the physical layer, and provides a unified data access interface for the upper layer. Specifically, it uses the COSEM object model defined by the DLMS protocol to convert heterogeneous data into a structured format. The storage architecture adopts a layered design, with raw data stored in a time-series database and processed data entering a data warehouse for analysis. Model layer: encapsulates multiphysics simulation logic, builds the core computing capabilities of digital twins, and integrates different simulation tools through FMI; Functional layer: Defines business logic and automation rules, coordinates the collaborative operation of multiple models, supports decision optimization, exposes the simulation engine as a microservice API, and integrates DLMS encryption and authentication. Interaction Layer: Provides human-computer interaction interface and real-time control interface, displays multi-physics simulation results in 3D, and supports dynamic parameter adjustment. Specifically, it adopts IoT feedback control and sends optimization commands to physical devices through DLMS / COSEM.

10. The method for constructing a green garden energy-saving design simulation model based on fused digital twins according to claim 1, characterized in that, The construction method further includes the following steps: S5. Verification and Optimization: S51. Comparison of simulation results: Compare the air conditioning load calculated by EnergyPlus with the electricity meter data, and calculate the root mean square error. The Nash efficiency coefficient was verified by comparing the runoff simulated by SWMM with the actual drainage monitoring data. NSE > 0.7 is considered qualified. S52. Verification of vegetation carbon sequestration: The volume of vegetation canopy was scanned by LiDAR and regression analysis was performed with the carbon sink estimated by i-Tree. S53. Multi-objective optimization: The NSGA-II optimization algorithm is adopted to balance energy saving, ecological and cost objectives, and Pareto front analysis is performed to output a non-dominated solution set for decision-makers to weigh and choose.

Images