Urban form multi-objective intelligent optimization decision-making method based on energy efficiency driving
By utilizing multidimensional databases, extreme gradient boosting algorithms, and Lasso regression methods, combined with UAV oblique photography, satellite image recognition, and computer vision technologies, the systemic mechanism of urban morphology and energy efficiency was revealed and efficiently regulated. This enabled efficient coupling analysis and optimal solution generation of urban morphology and energy efficiency, reduced computational costs, and improved the efficiency and stability of spatial search.
Patent Information
- Application Number
- CN202511284373.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-09-09
- Publication Date
- 2025-12-09
AI Technical Summary
Existing technologies are insufficient to reveal the systematic mechanisms and efficiently regulate urban morphology and energy efficiency. Traditional methods are computationally expensive and slow to iterate, making it difficult to meet the needs for rapid decision-making in large-scale complex parameter spaces.
This paper establishes a patent-based technical solution by constructing a multidimensional database, using the Extreme Gradient Boosting (XGBoost) algorithm for single-objective prediction models, and employing the Lasso regression method, combined with UAV oblique photogrammetry, satellite image recognition, and computer vision technologies. By introducing these technologies, a multi-objective optimization problem is solved. Furthermore, by incorporating data acquisition, the Extreme Gradient Boosting algorithm, and computer vision technologies, a research path based on multi-objective optimization algorithms and the Lasso regression method is established, enabling efficient coupling analysis and optimal solution generation of urban morphology and energy efficiency.
This approach achieves efficient coupling analysis and optimal solution generation between urban morphology and energy efficiency, reduces the computational cost of multi-objective optimization, and improves the stability and generalization ability of the solution model.
Smart Images
Figure CN121093784A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The application relates to an energy-efficiency-driven multi-objective intelligent optimization decision method for urban morphology, and belongs to the technical field of intelligent algorithms and urban building group design, BACKGROUND
[0002] Energy is an important material basis for promoting the normal operation of urban functions, and is the lifeblood of urban production and life, and to a large extent, maintains the benign and healthy development of social economy. With the rapid increase of the global urbanization rate, the city scale is continuously expanding, and the urban energy consumption is also growing at an alarming rate.
[0003] The city is a complex giant system. There is a possibility of mutual influence and even conflict between urban morphology and different energy efficiency objectives, so under the premise of considering multi-objective collaborative optimization, how to realize the optimal scheme of urban morphology has become the focus of current research. According to the literature retrieval at home and abroad, most of the previous urban building energy consumption models are concentrated on the correlation analysis and strategy optimization research of single factor and single target or multiple factors and single target. Although in recent years, research on comprehensive energy efficiency has started, but the systematic mechanism has not been revealed, and the regulation method and technology platform have not been developed.
[0004] At the same time, with the rapid development of big data analysis, data mining, and machine learning, deep learning and other technologies, the traditional research method relying on experience judgment and qualitative analysis is gradually being replaced by quantitative methods based on simulation and intelligent algorithms. For the multi-objective optimization problem of the coupling relationship between urban morphology and energy efficiency, considering the characteristics of large amount of data required, high analysis and mining difficulty, this project proposes a research path based on multi-objective optimization algorithm and Lasso regression method, in order to obtain a better solution set and design scheme. SUMMARY
[0005] The application provides an energy-efficiency-driven multi-objective intelligent optimization decision method and device for urban morphology, which aims to realize efficient coupling analysis and optimal scheme generation between urban morphology and energy efficiency by establishing a multi-dimensional database and constructing a prediction and optimization model.
[0006] In order to achieve the above purpose, the technical scheme of the application is as follows: an energy-efficiency-driven multi-objective intelligent optimization decision method for urban morphology, comprising the following steps:
[0007] Step 1: Establishing a multi-dimensional urban basic database related to sample urban morphology and energy efficiency;
[0008] Step 2: Constructing a single-target prediction model for urban morphology and energy efficiency based on the extreme gradient boosting algorithm (XGBoost);
[0009] Step 3: Reveal the relevant key urban form factors and their coupling degrees by using Lasso regression method;
[0010] Step 4: Establish a multi-objective optimization-based "urban form-energy efficiency" simulation model and carry out automatic optimization to reveal the coupling mechanism and form a high-performance urban form design pattern;
[0011] Step 5: Build a multi-objective optimization method for "urban form-energy efficiency" and build a high-performance urban form optimization design management platform.
[0012] Further, the step 1 specific steps are:
[0013] Step 1.1: Call the API interface of the open map data source OpenStreetMap (OSM) to obtain OSM data, and import it into the ArcGIS platform to complete the establishment of the city three-dimensional space model;
[0014] Step 1.2: For missing data, combine unmanned aerial photography, satellite image recognition and computer vision technology to extract missing three-dimensional space information and data and complete the supplementary modeling. For example, using the computer vision processing library OpenCV (mainly using the functions cv2.findContours and cv2.drawContours) of Python, the contours of buildings, roads, water systems and green lands are extracted and generated from the high-resolution images taken by WorldView-2 satellite after preprocessing (radiation correction and atmospheric correction based on ENVI remote sensing image digital processing software), and then converted into GIS data format;
[0015] Step 1.3: Based on the ArcGIS toolkit, describe and calculate the urban form (building type, urban density, urban texture, blue-green system, etc.) and building form data (roof area, window-wall ratio, shape coefficient, age, height, etc.). Including but not limited to the urban form indicators listed in Table 1. The above indicators correspond to the urban form and building feature parameters listed in Table 1, providing quantitative basis for subsequent urban space analysis, environmental simulation and planning decision-making.
[0016] Table 1 Urban form indicators
[0017]
[0018]
[0019] Step 1.4: Obtain typical meteorological year (TMY) data from the EnergyPlus website (www.energyplus.net), and select reference points for microclimate data measurement and comparison. Microclimate measurement mainly records four basic parameters of air temperature, relative humidity, wind speed and direction, and solar radiation.Figure 1 Microclimate data measuring instruments used in this project are demonstrated.
[0020] Step 1.5: Obtain part of the actual energy consumption data of urban buildings in cooperation with the power department;
[0021] Step 1.6: Based on the ArcGIS platform, clean and organize the data, and perform unified standardized processing (normalization, discretization, etc.), complete data integration.
[0022] Combining open source data, unmanned aerial vehicle oblique photography, satellite image recognition and computer vision technology, etc. to obtain urban form, climate data and energy consumption data can reduce the uncertainty in performance simulation and reduce decision-making errors.
[0023] Further, the specific steps of step 2 are:
[0024] Step 2.1: Divide and slice the study area, then import it into the Rhino&Grasshopper parameterized modeling platform, and input the corresponding modeling preset parameters (urban environment, urban form, building form, etc.), complete the parameterized modeling of the sample street form;
[0025] Step 2.2: Use the Ladybug Tools series of environmental performance evaluation plug-ins to build a single-target parameterized "urban form-energy efficiency" simulation model for building energy consumption, solar potential (photovoltaic, photothermal) and outdoor human comfort simulation;
[0026] Step 2.3: Calibrate and optimize the simulation results with the actual energy efficiency values in the database repeatedly to ensure the accuracy of the simulation results;
[0027] Step 2.4: Visit the sample street in the three simulation models to obtain single-target "urban form-energy efficiency" simulation data;
[0028] Step 2.5: Use simulation data as training and verification data (ratio 8:2) to establish a single-target "urban form-energy efficiency" prediction model based on the extreme gradient boosting algorithm (XGBoost). The model takes urban form data as input and energy efficiency (building energy consumption or solar potential or human comfort) data as output.
[0029] Although the traditional multi-objective optimization method based on physical simulation is accurate, it has high computational cost and slow iteration, which is difficult to meet the rapid decision-making demand in large-scale and complex parameter space, to some extent, limiting the application and popularization of the technology in the actual urban design process. This method replaces high-frequency simulation operation by building an extreme gradient boosting algorithm (XGBoost) proxy model, which takes about 1 second for single prediction, which is 420 times faster than traditional simulation.
[0030] Furthermore, step 3 specifically involves the following steps:
[0031] Step 3.1: Using the trained "urban morphology-energy efficiency" single-objective prediction model based on the extreme gradient boosting algorithm, the corresponding energy efficiency prediction data is generated in batches and quickly by inputting large-scale random morphology data within a reasonable range;
[0032] Step 3.2: Perform outlier filtering and standardization on the predicted data;
[0033] Step 3.3: Using the processed data, conduct a comprehensive analysis of the correlation between building energy consumption, solar energy potential, human comfort and urban morphology factors. Based on the analysis results, remove urban morphology factors that are not related to energy efficiency goals.
[0034] Step 3.4: Use the Lasso regression method to fit the above three energy efficiency targets with urban morphology factors that have a high degree of correlation, obtain the magnitude and ranking of the influence of urban morphology factors, and thus reveal the key urban morphology factors that affect a single energy efficiency target and their coupling degree.
[0035] The objective function of Lasso regression is a least-squares function with an L1 regularization term, i.e.:
[0036]
[0037] in, λ represents the deviation of the i-th component, and λ≥0 is the regularization parameter used to control the influence of the L1 regularization term. A larger λ indicates stronger model stability, meaning fewer features are selected. λ can be determined using cross-validation, selecting the λ value that minimizes the cross-validation error. Finally, the model is refitted using all the data according to the obtained λ value. This represents the standard Lasso estimator. The L1 norm penalty leads to sparsity of solutions, hence Lasso is widely used for variable selection in high-dimensional data.
[0038] The advantage of this step in the overall method is that it can automatically perform feature selection, thereby eliminating some redundant or useless features, reducing model complexity, and improving model stability and generalization ability. Furthermore, this step has few restrictions on data type and can accept data of any type.
[0039] Furthermore, step 4 specifically involves the following steps:
[0040] Step 4.1: By analyzing the characteristics of typical case cities, such as climate, morphology, texture, and building types, typical urban blocks with different functions are extracted as research objects.
[0041] Step 4.2: Based on the Rhino & Grasshopper platform, complete the setting and parameterized modeling of typical block site development constraints (land red line, constructable area boundary, etc.), surrounding environment information (adjacent buildings, roads, greenery, etc.), block form design variables, etc. Realize the generation of parameter-controlled block form;
[0042] Step 4.3: Call the multi-objective optimization algorithm RBFMOpt through the Opossum plug-in to automatically iterate and optimize the generated block form, set optimization objectives including building energy consumption minimization, outdoor human comfort optimization, and solar potential maximization. Each automatically generated scheme will be evaluated in terms of the above three aspects of energy efficiency by the simulation model, and the energy efficiency value will be input into the algorithm for decision evaluation;
[0043] Step 4.4: After optimization, analyze the Pareto optimal solution selected by the algorithm, draw the three-dimensional and two-dimensional distribution graph of the Pareto frontier, and reveal the mutual constraint relationship between different objectives combined with the correlation analysis result;
[0044] Step 4.5: Cluster analysis of the Pareto optimal solution set, and explore the form cause and principle combined with parallel coordinate analysis chart, so as to provide decision basis for scheme optimization;
[0045] Step 4.6: Analyze the "urban form-energy efficiency" coupling mechanism based on multi-objective optimization and form high-performance urban form mode based on the key factor conclusion of the above research that affects single energy efficiency target.
[0046] Further, the specific steps of step 5 are:
[0047] Step 5.1: According to the urban design target orientation of different plots, formulate corresponding control and guiding different level form control index and design guide;
[0048] Step 5.2: Build a "urban form-energy efficiency" multi-objective optimization control method;
[0049] Step 5.3: Based on the Rhino & Grasshopper platform, integrate the optimization workflow, complete the page integration and UI human-computer interaction interface design (User Interface Design) of environment information, form variable, and performance preset parameter setting, realize the development of multi-objective optimization urban form management platform.
[0050] An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor executes the program to realize the energy efficiency driven urban form generative design optimization method.
[0051] A computer readable storage medium, which stores computer instructions, the computer instructions are executed by a processor to implement the energy efficiency driven urban form generative design optimization method.
[0052] The advantages of the present application are as follows: compared with the prior art, the accuracy of urban form multi-objective intelligent optimization is further improved by open source data collection, unmanned aerial vehicle oblique photography, satellite image recognition and computer vision technology; the computational cost of multi-objective problems in high-dimensional form parameter space is reduced by introducing an extreme gradient boosting algorithm (XGBoost) proxy model, and the efficiency and stability of the solution space search are improved; the method can eliminate redundant or useless features, reduce the complexity of the model, and improve the stability and generalization ability of the model; BRIEF DESCRIPTION OF DRAWINGS
[0053] Figure 1 is the overall flowchart of the present application;
[0054] Figure 2 Multi-objective optimization model flowchart; DETAILED DESCRIPTION
[0055] In order to deepen the understanding and understanding of the present application, the following will introduce the scheme in detail in combination with examples.
[0056] Embodiment:
[0057] Reference Figure 1 , Figure 2 A multi-objective intelligent optimization decision method based on energy efficiency driven urban form, comprising the following steps:
[0058] Step 1: Establish a multi-dimensional urban basic database related to sample urban form and energy efficiency;
[0059] Step 2: Build a "city form-energy efficiency" single-objective prediction model based on the extreme gradient boosting algorithm (XGBoost);
[0060] Step 3: Use Lasso regression method to reveal the relevant key urban form factors and their coupling degree;
[0061] Step 4: Establish a "city form-energy efficiency" simulation model based on multi-objective optimization and expand automatic optimization, reveal the coupling mechanism and form a high-performance urban form design pattern;
[0062] Step 5: Build a "city form-energy efficiency" multi-objective optimization control method, and build a high-performance urban form optimization design management platform.
[0063] Further, the step 1 is specifically as follows:
[0064] Step 1.1: Call the API interface of the open map data source OpenStreetMap (OSM) to obtain OSM data, and import it into the ArcGIS platform to complete the establishment of the city's three-dimensional spatial model;
[0065] Step 1.2: For missing data, combine unmanned aerial photography, satellite image recognition, and computer vision technology to extract missing three-dimensional spatial information and data and complete the supplementary modeling. For example, using the computer vision processing library OpenCV (mainly using the functions cv2.findContours and cv2.drawContours) of Python, the contours of buildings, roads, water systems, and green spaces are extracted and generated from high-resolution images taken by WorldView-2 satellite after preprocessing (radiation correction and atmospheric correction based on ENVI remote sensing image digital processing software), and converted into GIS data format;
[0066] Step 1.3: Based on the ArcGIS toolkit, describe and calculate the city form (building type, city density, city texture, blue-green system, etc.) and building form data (roof area, window-wall ratio, volume coefficient, age, height, etc.). Including but not limited to the city form indicators listed in Table 1.
[0067] Table 1 City Form Indicators
[0068]
[0069]
[0070] Step 1.4: Obtain typical meteorological year (TMY) data from the EnergyPlus website (www.energyplus.net), and select reference points for microclimate data measurement and comparison. Microclimate measurement mainly records four basic parameters: air temperature, relative humidity, wind speed and direction, and solar radiation.
[0071] Step 1.5: Cooperate with the power department to obtain part of the actual energy consumption data of urban buildings;
[0072] Step 1.6: Based on the ArcGIS platform, clean and organize the data, and perform unified normalization processing (normalization, discretization, etc.), complete data integration.
[0073] Further, the specific steps of Step 2 are as follows:
[0074] Step 2.1: Divide and slice the study area, then import it into the Rhino & Grasshopper parametric modeling platform, and input the corresponding modeling preset parameters (city environment, city form, building form, etc.), complete the parametric modeling of the sample street form;
[0075] Step 2.2: Use the Ladybug Tools series of environmental performance assessment plugins to build a single-objective parameterized "urban form-energy efficiency" simulation model for building energy consumption, solar potential (photovoltaic, photothermal), and outdoor human comfort simulation;
[0076] Step 2.3: Calibrate and optimize the simulation results with actual energy efficiency values in the database to ensure the accuracy of the simulation model simulation results;
[0077] Step 2.4: Visit the sample block in the three simulation models to obtain single-objective "urban form-energy efficiency" simulation data;
[0078] Step 2.5: Use simulation data as training and verification data (ratio 8:2) to establish a "urban form-energy efficiency" single-objective prediction model based on the extreme gradient boosting algorithm (XGBoost). The model takes urban form data as input and energy efficiency (building energy consumption or solar potential or human comfort) data as output.
[0079] Further, the specific steps of step 3 are:
[0080] Step 3.1: Use the trained "urban form-energy efficiency" single-objective prediction model based on the extreme gradient boosting algorithm to quickly generate corresponding energy efficiency prediction data by inputting large-scale random form data within a reasonable range;
[0081] Step 3.2: Perform outlier filtering and standardization on the prediction data;
[0082] Step 3.3: Use the processed data to comprehensively analyze the correlation between building energy consumption, solar potential, human comfort, and urban form factors. According to the analysis results, remove urban form factors that are not related to energy efficiency targets;
[0083] Step 3.4: Use Lasso regression to fit the three energy efficiency targets and urban form factors with high correlation to obtain the degree of influence and ranking of urban form factors, thereby revealing the key urban form factors affecting a single energy efficiency target and their coupling degree.
[0084] The objective function of Lasso regression is a least squares function with an L1 regularization term, that is:
[0085]
[0086] where, is the bias of the i-th component, and λ > 0 is a regularization parameter to control the degree of influence of the L1 regularization term. When λ is larger, the stability of the model is stronger, i.e. fewer features are selected. The determination of λ can be made by cross-validation method, and the λ value with the minimum cross-validation error is selected. Finally, according to the obtained λ value, the model is refitted with all the data. represents the conventional Lasso estimator. This L1 norm penalty leads to the sparsity of the solution, so Lasso is widely used for variable selection in high-dimensional data. The advantage of Lasso regression is that it can automatically implement feature selection, thereby eliminating some redundant or useless features, reducing the complexity of the model, and improving the stability and generalization ability of the model. In addition, Lasso has little restriction on data types and can accept any type of data.
[0087] Further, the specific steps of step 4 are as follows:
[0088] Step 4.1: By analyzing the characteristics of climate, morphology, texture and building types of typical case cities, typical urban blocks with different functions are extracted as research objects
[0089] Step 4.2: Based on the Rhino & Grasshopper platform, the setting and parametric modeling of the site development constraints (red line, constructable area boundary, etc.), surrounding environment information (adjacent buildings, roads, greenery, etc.), and street block morphology design variables of the typical block are completed, and the parameter-controlled block morphology generation is realized;
[0090] Step 4.3: The generated block morphology is automatically iteratively optimized by calling the multi-objective optimization algorithm RBFMOpt through the Opossum plug-in, and optimization objectives including building energy minimization, outdoor human comfort optimization and solar potential maximization are set. Each automatically generated scheme will be evaluated in terms of the above three aspects of energy efficiency by the simulation model, and the energy efficiency value will be input into the algorithm for decision evaluation;
[0091] Step 4.4: After optimization, the Pareto optimal solution selected by the algorithm is analyzed, and the three-dimensional and two-dimensional distribution graphs of the Pareto frontier are drawn, combined with the correlation analysis results, to reveal the mutual constraint relationship between different objectives;
[0092] Step 4.5: The Pareto optimal solution set is subjected to cluster analysis, and the parallel coordinate analysis chart is used to explore the morphology causes and principles, thereby providing decision basis for scheme optimization;
[0093] Step 4.6: Based on the above research findings and the key factor conclusions affecting a single energy efficiency target, the coupling mechanism of "urban morphology-energy efficiency" based on multi-objective optimization is analyzed and a high-performance urban morphology mode is formed.
[0094] Further, the specific steps of step 5 are as follows:
[0095] Step 5.1: According to the urban design target orientation of different plots, corresponding control and guiding different level form control indicators and design guidelines are formulated;
[0096] Step 5.2: A multi-objective optimization regulation method of "urban form-energy efficiency" is constructed.
[0097] Step 5.3: Based on the Rhino&Grasshopper platform, the optimization workflow is integrated, the page integration and UI human-computer interaction interface design (User Interface Design) of environmental information, form variables and performance preset parameter settings are completed, and the development of the multi-objective optimization urban form management platform is realized.
[0098] It should be noted that the above embodiments are not used to limit the protection scope of the present application, and equivalent transformations or substitutions made on the basis of the above technical solutions all fall within the scope of protection of the claims of the present application.
Claims
1. A multi-objective intelligent optimization decision-making method for urban morphology based on energy efficiency, characterized in that, Includes the following steps: Step 1: Establish a multi-dimensional urban basic database related to the morphological characteristics and energy efficiency of the sample cities; Step 2: Construct a single-objective prediction model of "urban morphology-energy efficiency" based on the Extreme Gradient Boosting algorithm (XGBoost) to realize the predictive mapping between urban morphology factors and energy efficiency indicators; Step 3: Use Lasso regression to reveal key urban morphological factors with significant influence and quantify their coupling relationships; Step 4: Establish a simulation model of "urban form-energy efficiency" based on multi-objective optimization, use intelligent optimization algorithms to achieve automatic optimization, reveal the coupling mechanism between form factors and energy efficiency, and form a high-performance urban form design mode. Step 5: Construct a multi-objective optimization and control method for "urban form-energy efficiency" and build a high-performance urban form optimization and design management platform to realize intelligent decision support for urban planning and design.
2. The energy efficiency-driven multi-objective intelligent optimization decision-making method for urban morphology according to claim 1, characterized in that: Step 1 specifically involves: Step 1.1: Obtain the raw geospatial data by calling the API interface of the OpenStreetMap (OSM) data source, and import it into the ArcGIS platform to complete the construction of the basic three-dimensional spatial model of the city; Step 1.2: To address the missing parts in the OSM data, we combine UAV oblique photography, satellite image recognition, and computer vision technologies to extract and model the missing 3D spatial information and data. Using the Python computer vision processing library OpenCV, we call the cv2.findContours and cv2.drawContours functions to extract the outlines of target objects such as buildings, roads, water systems, and green spaces from the preprocessed remote sensing images. The extracted vector outline information is converted into standard GIS data format and fused with the existing 3D model to achieve complete modeling of urban spatial information. Step 1.3: After the model is built, the ArcGIS toolkit is used to describe and quantify the macro-morphology of the city and the micro-morphology of the buildings in multiple dimensions. The city morphology indicators include, but are not limited to, building type, city density, urban fabric, and blue-green system, which are used to depict the overall spatial pattern. Building morphology indicators encompass roof area, window-to-wall ratio, shape coefficient, building age, and building height, used to characterize individual buildings. Step 1.4: Download typical meteorological year (TMY) data from the EnergyPlus website and select reference points for actual measurement and comparison of microclimate data. Microclimate measurements mainly record four basic parameters: temperature, relative humidity, wind speed and direction, and solar radiation. Step 1.5: In cooperation with the power sector, obtain actual energy consumption data for some urban buildings; this data serves as an important reference for calibrating and validating the simulation model, ensuring the high reliability and practicality of subsequent results. Step 1.6: Based on the ArcGIS platform, clean and unify the multi-source heterogeneous data, including: eliminating dimensional differences, performing unified standardization processing (normalization, discretization, etc.), and realizing unified and standardized management of remote sensing images, GIS data, energy consumption data and meteorological data to form a complete data system.
3. The energy efficiency-driven multi-objective intelligent optimization decision-making method for urban morphology according to claim 1, characterized in that: Step 2 specifically involves: Step 2.1: Divide the study area into spatial segments and grids, and then import them into the Rhino & Grasshopper parametric modeling platform. At the same time, input the corresponding preset modeling parameters, including urban environment, urban form and building form, to complete the parametric modeling of the sample street morphology. Step 2.2: Use Ladybug Tools series environmental performance assessment plugins to build a single-objective parametric "urban morphology-energy efficiency" simulation model for simulating building energy consumption, solar energy utilization potential, and outdoor human comfort. Step 2.3: Repeatedly calibrate and optimize the above simulation results with the actual energy efficiency values in the database to ensure the accuracy and reliability of the simulation model in terms of building energy consumption, solar energy potential and human comfort. Step 2.4: Visit the model street in the three simulation models to obtain single-objective "urban form-energy efficiency" simulation data; Step 2.5: Using simulated data as training and validation data at a ratio of 8:2, establish a single-objective prediction model of "urban morphology-energy efficiency" based on the Extreme Gradient Boosting (XGBoost) algorithm. This model takes urban morphology data as input and energy efficiency (building energy consumption, solar energy potential, or human comfort) data as output.
4. The energy efficiency-driven multi-objective intelligent optimization decision-making method for urban morphology according to claim 1, characterized in that: Step 3 specifically involves: Step 3.1: Using the trained "urban morphology-energy efficiency" single-objective prediction model based on the extreme gradient boosting algorithm, the corresponding energy efficiency prediction data is generated in batches and quickly by inputting large-scale random morphology data within a reasonable range; Step 3.2: Perform outlier filtering and standardization on the predicted data; Step 3.3: Using the processed data, conduct a comprehensive analysis of the correlation between building energy consumption, solar energy potential, human comfort and urban morphology factors. Based on the analysis results, remove urban morphology factors that are not related to energy efficiency goals. Step 3.4: The Lasso regression method is used to fit the three energy efficiency targets with highly correlated urban morphology factors to obtain the magnitude and ranking of the influence of urban morphology factors, thereby revealing the key urban morphology factors affecting a single energy efficiency target and their coupling degree. The objective function of Lasso regression is a least-squares function with an L1 regularization term, i.e.: in, λ represents the deviation of the i-th component, and λ≥0 is the regularization parameter used to control the influence of the L1 regularization term. A larger λ indicates stronger model stability, meaning fewer features are selected. λ is determined using cross-validation, selecting the λ value with the smallest cross-validation error. Finally, the model is refitted using all data according to the obtained λ value. Let L1 be the standard Lasso estimator, which is penalized by the L1 norm, resulting in sparsity of solutions.
5. The energy efficiency-driven multi-objective intelligent optimization decision-making method for urban morphology according to claim 1, characterized in that: Step 4 specifically involves: Step 4.1: By analyzing the climate, morphology, texture and architectural characteristics of typical case cities, typical urban blocks with different functions are extracted as research objects; Step 4.2: Based on the Rhino & Grasshopper platform, complete the setting and parametric modeling of site development constraints, surrounding environment information, and street form design variable elements for typical blocks, and realize parameter-controlled block form generation; Step 4.3: The multi-objective optimization algorithm RBFMOpt is called through the Opossum plugin to automatically iterate and optimize the generated street block form. Optimization objectives include minimizing building energy consumption, optimizing outdoor human comfort, and maximizing solar energy potential. Each automatically generated scheme will be evaluated for energy efficiency in the above three aspects through a simulation model, and the energy efficiency value will be input into the algorithm for decision evaluation. Step 4.4: After optimization, the Pareto optimal solution selected by the algorithm is analyzed, and the three-dimensional and two-dimensional distribution maps of the Pareto front are drawn. Combined with the correlation analysis results, the mutual constraints between different objectives are revealed. Step 4.5: Perform cluster analysis on the Pareto optimal solution set, and explore the causes and principles of the morphology in conjunction with the parallel coordinate analysis diagram, so as to provide a decision basis for the optimal solution; Step 4.6: Based on the above research findings and the conclusions of key factors affecting a single energy efficiency target, analyze the coupling mechanism of "urban form-energy efficiency" based on multi-objective optimization and form a high-performance urban form model.
6. The energy efficiency-driven urban form generative design optimization method according to claim 1, characterized in that: Step 5 specifically involves: Step 5.1: Based on the urban design objectives of different plots, formulate corresponding control indicators and design guidelines at different levels of form, both controlling and guiding. Step 5.2: Construct a multi-objective optimization control method for "urban form-energy efficiency"; Step 5.3: Based on the Rhino & Grasshopper platform, integrate and optimize the workflow, complete the page integration and UI human-computer interaction interface design for setting environmental information, morphological variables, and performance preset parameters, and realize the development of a multi-objective optimized urban morphology management platform.
7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements an energy efficiency-driven urban form generative design optimization method as described in any one of claims 1 to 6.
8. A computer-readable storage medium storing computer instructions thereon, characterized in that, When executed by the processor, the computer instructions implement an energy efficiency-driven urban morphology generative design optimization method as described in any one of claims 1-6.