Smart city planning method and system based on digital twinning

Abstract

The invention discloses a smart city planning method and system based on digital twinning, relates to the technical field of city planning, and solves the problems that the scheme is not feasible due to dependence on manual interpretation of regulation texts, lack of automatic generation and matching of index constraint lists and possible neglect of key constraint conditions in planning suggestions, and meanwhile, the scheme is high in subjectivity and low in cost. And a unified quantitative model is lacked. Through the NLP technology, planning suggestions and regulation terms are compared in real time, conflict indexes are automatically marked, planning conditions and preliminary index calculation results are combined, the generated project planning suggestions strictly follow a policy framework, and manual interpretation errors or violation risks are avoided. A collaborative quantitative evaluation model of the single building and the regional landscape is established, and the landscape effect of the planning suggestion is evaluated from the dimensions of spatial layout, visual vestibule, skyline, color coordination and the like. The fusion degree of the project and the surrounding environment is visually reflected through quantitative scoring, and the deviation of subjective evaluation is avoided.

Description

A Smart City Planning Method and System Based on Digital Twins Technical Field

[0001] This invention belongs to the field of urban planning, specifically a smart city planning method and system based on digital twins. Background Technology

[0002] Digital twin is a technological system that uses technologies such as the Internet of Things, big data, cloud computing, and artificial intelligence to construct virtual mapping models of physical entities (such as cities, buildings, and transportation systems), enabling two-way interaction between the virtual and real worlds, dynamic simulation, and intelligent optimization. The deep integration of digital twin technology with urban planning concepts is driving the transformation of planning models from "traditional experience-based" to "modern intelligent," necessitating the development of smart city planning solutions based on digital twins.

[0003] Traditional planning relies on manual interpretation of legal texts and lacks automated generation and matching of indicator constraint lists. Planning recommendations may overlook key constraints, leading to infeasible solutions. It also relies on expert experience or public perception, resulting in highly subjective solutions. Furthermore, it lacks a unified quantitative model, making it inefficient and prone to errors. Summary of the Invention

[0004] This invention aims to solve at least one of the technical problems existing in the prior art; to this end, this invention proposes a smart city planning method and system based on digital twins to solve the technical problems of relying on manual interpretation of regulatory texts, lack of automated generation and matching of indicator constraint lists, planning suggestions may ignore key constraints, resulting in infeasible solutions, and strong subjectivity of solutions, lack of a unified quantitative model.

[0005] To address the aforementioned issues, the first aspect of this invention provides a smart city planning method and system based on digital twins, comprising the following steps: parsing basic information of urban planning projects using NLP technology, matching corresponding statutory and non-statutory planning texts from a planning basis database, extracting binding and guiding indicators, and generating an indicator constraint list; the system automatically calls project surveying data and operational performance data, and generates preliminary indicators by analyzing core indicators of planning conditions; based on the indicator constraint list and the calculation results of preliminary indicators of planning conditions, generating project planning recommendations; analyzing the matching degree between project planning recommendations and planning basis under multi-objective optimization constraints, and optimizing and verifying project planning recommendations; establishing a landscape coordination quantitative evaluation model by analyzing the unity of individual units and regions, evaluating the landscape coordination of project planning recommendations, and setting landscape indicator adjustment formulas to optimize the indicators of project planning recommendations; during project construction, dividing the project into blocks, obtaining actual landscape data of completed blocks, comparing them with the actual implemented project planning recommendations, and analyzing the landscape implementation fit to adjust the parameters of the landscape indicator adjustment formula.

[0006] Optionally, in one example of the above aspects, NLP technology is used to analyze the basic information of urban planning projects, match corresponding statutory and non-statutory planning texts from the planning basis database, extract binding and guiding indicators, and generate an indicator constraint list. This includes the following steps: For statutory planning texts, integrate territorial spatial planning texts and control detailed planning texts, extract land use location, nature, and scale fields, and construct a structured planning database; For non-statutory planning texts, incorporate special planning texts and industrial planning texts, mark their planned land use location, target development nature, and target development scale, and add them to the structured planning database; Use NLP natural language processing tools to segment statutory and non-statutory planning texts in the structured planning database and tag them with parts of speech; Set binding and guiding indicators as key entity information and train... The Named Entity Recognition (NER) model identifies and marks key entity information in both statutory and non-statutory planning texts within the structured planning database, associating the corresponding values ​​or ranges of these key entities. It then performs spatial overlay analysis between the project site coordinates or address and the land parcel boundaries in the structured planning database to determine the planning scope to which the project site belongs. From the statutory planning texts within the planning scope of the structured planning database, it retrieves the mandatory clauses for the corresponding land parcels and their marked key entity information, labeling them as hard constraints. From the non-statutory planning texts within the planning scope of the structured planning database, it retrieves the guiding requirements for the corresponding land parcels and their marked key entity information, labeling them as soft constraints. Statutory planning texts specifically include: territorial spatial planning and control planning. The model extracts hard and soft constraints, associates them with the land location in the structured planning database, and generates a list of indicator constraints.

[0007] Optionally, in one example of the above aspects, the system automatically retrieves project surveying data and operational performance data, and generates preliminary indicators by analyzing the core indicators of planning conditions, including the following steps: the system automatically retrieves project surveying data and operational performance data, whereby the project surveying data includes: topography, buildings, and transportation; and the operational performance data includes: the load of surrounding public services and traffic efficiency; and generates preliminary indicators by analyzing the core indicators of planning conditions.

[0008] Optionally, in one example of the above aspects, based on the indicator constraint list and the preliminary indicator calculation results of the planning conditions, a project planning suggestion is generated, including the following steps: importing the preliminary indicators of the project planning conditions, performing correlation screening with the values ​​in the indicator constraint list, marking data where the preliminary indicators of the planning conditions do not meet the numerical requirements in the indicator constraint list, and sending a manual modification prompt; using GIS tools to spatially overlay the project land boundary with the plot range and surrounding facilities in the structured planning database to clarify the locational relationship of the project land boundary; marking the existing surrounding facilities at the project land boundary, generating alternative schemes for the project planning suggestion, including: a compliance scheme, which includes: planning the project land using the screened preliminary indicators of the project planning conditions, including the surrounding facilities marked at the project land boundary in the project planning surrounding facilities, with the surrounding facilities required to meet the minimum standards in the indicator constraint list; and a quality scheme. This includes: planning the project land use based on the preliminary indicators of the screened project planning conditions, excluding surrounding facilities marked at the project land boundary from the planned surrounding facilities, and requiring surrounding facilities to meet the minimum standards in the indicator constraint list; a hybrid optimization scheme, including: planning the project land use based on the preliminary indicators of the screened project planning conditions, excluding surrounding facilities marked at the project land boundary from the planned surrounding facilities, and requiring surrounding facilities to meet the minimum standards in the indicator constraint list; simultaneously, under the premise of meeting the numerical requirements in the indicator constraint list, selecting several indicators from the preliminary indicators of the project planning conditions to set key guiding optimization indicators, optimizing by 10% based on the preliminary indicator values ​​of the original project planning conditions; and extracting the land use location, nature, and scale from the text of the project planning proposal using the NER model, comparing it with the planning nature and scale of the corresponding plots in the structured planning database to verify consistency.

[0009] Optionally, in one example of the above aspects, the land location, nature, and scale are extracted from the text of the project planning proposal using a NER model, and compared with the planning nature and scale of the corresponding plots in the structured planning database to verify consistency. This includes: extracting the constraint relationship between key entity information and binding indicators through dependency parsing; comparing the project land location, nature, and scale with the land location, planning nature, and scale of the corresponding plots in the structured planning database based on the constraint relationship between key entity information and binding indicators; calculating the cosine similarity of the corresponding location, nature, and scale texts respectively; and weighting and summing the results to obtain a consistency coefficient. Data with a consistency coefficient lower than a threshold is marked with an early warning and a manual modification prompt is sent.

[0010] Optionally, in one example of the above aspects, by analyzing the matching degree between the project planning recommendations and the planning basis under multi-objective optimization constraints, and optimizing and verifying the project planning recommendations, the following steps are included: setting the optimization constraint range for each indicator in the alternative solutions of the project planning recommendations; calculating the compliance verification coefficient by analyzing the multi-objective optimization constraints; and analyzing the matching degree between the project planning recommendations and the planning basis. Where: Com represents the optimization verification result of the indicators in the alternative solutions proposed in the project planning. Let Com be the optimization constraint interval for the k-th indicator, k∈(1,2,…,m), where m is the total number of indicators; perform abnormal warnings and corrections. If Com=0, the system automatically marks abnormal indicators, pushes conflict evidence, and provides corrective solutions for compliance, quality, and hybrid optimization based on historical cases through a large voice model, and sends a manual review prompt; otherwise, it determines that the alternative solutions do not need to be modified.

[0011] Optionally, in one example of the above aspects, by analyzing the unity of individual units and regions, a landscape synergy quantitative evaluation model is established to assess the landscape coordination of project planning recommendations, and a landscape index adjustment formula is set to optimize the indicators of project planning recommendations. This includes the following steps: collecting landscape parameters of the project area and surrounding areas; establishing a landscape synergy quantitative evaluation model by analyzing the unity of individual units and regions: L=ωv*Lv+ωe*Le+ωf*Lf; where L is the overall landscape synergy coordination degree; Lv is the visual coordination score; Le is the ecological integration score; Lf is the functional adaptability score; ωv is the evaluation dimension weight, ωe is the ecological integration weight, and ωf is the functional adaptability weight, set according to urban planning priorities; by establishing the landscape synergy quantitative evaluation model, the overall landscape synergy coordination degree is calculated, the landscape coordination of project planning recommendations is assessed, and early warnings are issued for project planning recommendations with an overall landscape synergy coordination degree below a threshold.

[0012] Optionally, in one example of the above aspects, setting a landscape indicator adjustment formula to optimize the indicators of the project planning recommendations includes the following steps: for project planning recommendations that have passed the landscape harmony assessment, optimizing the indicators of the project planning recommendations by setting a landscape indicator adjustment formula.

[0013] Optionally, in one example of the above aspects, during the project construction process, the project is divided into blocks, and the actual landscape data of the completed blocks is obtained. This data is then compared with the actual implemented project planning proposal. By analyzing the fit between the landscape implementation and the actual implementation, the parameters of the landscape indicator adjustment formula are adjusted accordingly. This includes the following steps: During the project construction process, the project is divided into blocks, and the actual landscape data of the completed blocks is obtained. The target value of the landscape synergy and coordination degree is incorporated into the block-based formula for block-based analysis. Under the constraints, the range of the target value of the landscape synergy and coordination degree for each block in the project area is calculated. The minimum value within the range is averaged to obtain the optimized target value of the landscape synergy and coordination degree, which replaces the corresponding parameter in the original landscape indicator adjustment formula.

[0014] According to another aspect of this disclosure, a smart city planning system based on digital twins is provided, which adopts a smart city planning method based on digital twins as described above to realize smart city planning based on digital twins.

[0015] Compared with existing technologies, the beneficial effects of this invention are as follows: By integrating data in a digital twin environment to generate a list of indicator constraints or preliminary indicators, this invention forms a closed loop of "design-simulation-optimization," reducing repeated modifications and resource waste in the physical world. Quantitative analysis based on surveying data and operational performance provides objective evidence for planning recommendations, avoiding subjective decision-making biases. Real-time comparison of planning recommendations with regulatory clauses using NLP technology automatically identifies conflicting indicators, such as exceeding the floor area ratio or failing to meet green space requirements, reducing legal risks.

[0016] This invention utilizes NLP technology to analyze a planning basis database, extracting binding and guiding indicators from statutory and non-statutory planning texts to form an indicator constraint list. Combined with the preliminary indicator calculation results of the planning conditions, the generated project planning recommendations strictly adhere to the policy framework, avoiding errors in human interpretation or the risk of violations. A collaborative quantitative evaluation model for individual buildings and the regional landscape is established, assessing the landscape effect of the planning recommendations from dimensions such as spatial layout, visual corridors, skyline, and color harmony. Quantitative scoring intuitively reflects the integration of the project with its surrounding environment, avoiding biases in subjective evaluation. During project construction, actual landscape data is obtained through block division and compared with the planning recommendations to analyze implementation fit. Based on feedback on deviations, the landscape indicator formula parameters are adjusted, forming a closed loop of "planning-construction-feedback-optimization" to ensure the effective implementation of the plan. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 is a schematic diagram of the method flow of the present invention; Figure 2 is a schematic diagram of the system framework of the present invention. Detailed Implementation

[0019] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] Please refer to Figures 1 and 2. The first aspect of this invention provides a smart city planning method and system based on digital twins, comprising the following steps: using NLP technology to analyze core information such as land location, nature, and scale of urban planning projects; matching corresponding statutory and non-statutory planning texts from the planning basis database; extracting binding and guiding indicators; generating an indicator constraint list; the system automatically calls project surveying data and operational performance data; generating preliminary indicators by analyzing core indicators of planning conditions; generating project planning suggestions based on the indicator constraint list and the calculation results of preliminary indicators of planning conditions; analyzing the matching degree between the project planning suggestions and the planning basis under multi-objective optimization constraints; optimizing and verifying the project planning suggestions; establishing a landscape coordination quantitative evaluation model by analyzing the unity of individual units and regions; evaluating the landscape coordination of the project planning suggestions; and setting landscape indicator adjustment formulas to optimize the indicators of the project planning suggestions; during project construction, dividing the project into blocks, obtaining actual landscape data of the completed blocks, comparing it with the actual implemented project planning suggestion scheme, and adjusting the parameters of the landscape indicator adjustment formula by analyzing the landscape implementation fit.

[0021] Specifically, in this embodiment, core information such as land location, nature, and scale is automatically extracted using natural language processing technology. It then matches statutory and non-statutory planning texts from the planning basis database to ensure that planning recommendations strictly adhere to the policy framework. For example, binding indicators such as plot ratio and green space ratio are extracted to avoid human interpretation errors. The system calls upon surveying and mapping data and operational performance data for cross-validation with the indicator constraint list, generating preliminary indicators that better reflect actual conditions and reducing the risk of a disconnect between planning and implementation.

[0022] The entire process, from information analysis to indicator generation and recommendation output, is automated, replacing traditional manual surveys, text comparisons, and indicator calculations, significantly shortening the planning time. For example, Xiong'an New Area uses a digital twin platform to record the entire lifecycle of its planning schemes, improving efficiency. The digital twin model can dynamically access urban operational data, enabling planning recommendations to quickly respond to urban changes.

[0023] In a digital twin environment, data integration generates a list of indicator constraints or preliminary indicators, forming a closed loop of "design-simulation-optimization," reducing repeated modifications and resource waste in the physical world. Quantitative analysis based on surveying data and operational performance provides objective evidence for planning recommendations, avoiding subjective decision-making biases. 3D visualization technology can be used to present planning schemes, allowing managers to intuitively compare the advantages and disadvantages of different options. By analyzing data such as population distribution and traffic flow, the layout of public facilities can be optimized, reducing resource idleness or over-concentration. Digital twin models can simulate extreme situations, allowing for advance planning of emergency resource reserves and evacuation routes, enhancing urban resilience.

[0024] By using NLP technology to compare planning recommendations with regulatory clauses in real time, conflicting indicators such as exceeding the floor area ratio or failing to meet the green space ratio are automatically marked, reducing legal risks. Construction plans are tested in a virtual environment to identify the risk of project rework in advance.

[0025] By analyzing the planning basis database using NLP technology, binding and guiding indicators are extracted from statutory and non-statutory planning documents to form an indicator constraint list. Combined with the preliminary indicator calculation results of the planning conditions, the generated project planning recommendations strictly adhere to the policy framework, avoiding errors in human interpretation or the risk of violations. For example, in historical preservation areas, the system can automatically identify building height restrictions to ensure that the planning recommendations meet protection requirements.

[0026] A collaborative quantitative evaluation model for individual buildings and the surrounding landscape is established to assess the landscape effects of planning recommendations from dimensions such as spatial layout, visual corridors, skyline, and color harmony. Quantitative scoring intuitively reflects the degree of integration between the project and its surrounding environment, avoiding biases from subjective evaluations. For example, the model can analyze the impact of new buildings on the skyline of historic districts and propose height adjustment suggestions.

[0027] During the planning recommendation generation phase, the system uses a multi-objective optimization algorithm based on a list of indicator constraints to balance economic, social, and environmental goals. For example, in commercial district planning, it can simultaneously optimize floor area ratio and traffic flow to avoid congestion caused by high-density development.

[0028] During project construction, actual landscape data is obtained through block division and compared with the proposed planning scheme to analyze its suitability for implementation. Based on feedback on deviations, the landscape indicator formula parameters are adjusted, forming a closed loop of "planning-construction-feedback-optimization" to ensure the effective implementation of the plan. For example, if the actual greening rate is lower than the planned value, the system can automatically increase the greening indicator requirements for subsequent blocks.

[0029] Digital twin technology maps project construction progress in real time, enabling phased evaluation through block division. If a construction deviation is detected in a certain block, the system can immediately trigger an alert and generate adjustment plans to prevent problems from accumulating. The optimization and verification of planning recommendations not only target the current project but also consider future urban development needs. For example, by simulating population growth and traffic flow changes, the flexibility of planning recommendations is assessed to ensure the long-term feasibility of the plan.

[0030] In one embodiment of the present invention, NLP technology is used to analyze the basic information of urban planning projects, match the corresponding statutory planning texts and non-statutory planning texts from the planning basis database, extract binding indicators and guiding indicators, and generate an indicator constraint list. The steps include: for statutory planning texts, integrating territorial spatial planning texts and control detailed planning texts, extracting land location, nature and scale fields, and constructing a structured planning database; in this embodiment, the correspondence between "plot number-land use nature-plot ratio" in the control plan is stored in a relational database.

[0031] For non-statutory planning documents, they are incorporated into special planning documents and industrial planning documents, with their planned land use location, target development nature, and target development scale marked, and added to the structured planning database. NLP (Natural Language Processing) tools are used to segment and tag the statutory and non-statutory planning documents in the structured planning database. Binding and guiding indicators are set as key entity information, and a NER (Named Entity Recognition) model is trained to identify and tag key entity information in both statutory and non-statutory planning documents in the structured planning database, associating the corresponding values ​​or ranges of the key entity information. The project land coordinates are... Alternatively, spatial overlay analysis can be performed between the address and the land parcel boundaries in the structured planning database to determine the planning scope to which the project land belongs; mandatory clauses and their marked key entity information for the corresponding land parcels can be retrieved from the statutory planning texts of the planning scope to which the structured planning database belongs, and these are marked as hard constraints; guiding requirements and their marked key entity information for the corresponding land parcels can be retrieved from the non-statutory planning texts of the planning scope to which the structured planning database belongs, and these are marked as soft constraints; statutory planning texts specifically include: territorial spatial planning and control planning; hard and soft constraints will be extracted and linked to the land location in the structured planning database to generate an indicator constraint list.

[0032] In this embodiment, the generation of the indicator list includes: generating a table by extracting hard and soft constraints according to indicator type (binding / guiding), indicator name and value / range, planning basis, and matching plot format; comparing the indicators of the same plot in different plans (such as the control plan requiring a green space ratio of 30%, and the special plan requiring 35%), marking conflicting items for manual review; and using charts or maps to mark and assist decision-making.

[0033] In one embodiment of the present invention, the system automatically retrieves project surveying data and operational performance data, and generates preliminary indicators by analyzing core indicators of planning conditions, including the following steps: The system automatically retrieves project surveying data and operational performance data. The project surveying data includes: topography, buildings, and transportation; the operational performance data includes: surrounding public service load and traffic efficiency; and preliminary indicators are generated by analyzing core indicators of planning conditions. Where: Inn is the preliminary planning condition indicator (such as plot ratio and building height); Inga is the upper limit of the indicator specified by the statutory planning; β is the operation performance correction coefficient, with a value of 0.1~0.3, and the coefficient is smaller for better performance; Pera is the actual operation performance of the surrounding area, including: public service coverage and traffic efficiency; Pert is the operation performance target value; γ is the current density correction coefficient, with a value of 0.2~0.4, and the coefficient is larger for higher density; Deac is the current building density of the surrounding area; and Dest is the planned standard building density.

[0034] In one embodiment of the present invention, a project planning suggestion is generated based on the indicator constraint list and the preliminary indicator calculation results of the planning conditions. This includes the following steps: importing the preliminary indicators of the project planning conditions and screening them against the values ​​in the indicator constraint list; marking data where the preliminary indicators of the planning conditions do not meet the numerical requirements in the indicator constraint list and sending a manual modification prompt; using GIS tools to spatially overlay the project land boundary with the plot area and surrounding facilities in the structured planning database to clarify the locational relationship of the project land boundary; marking the existing surrounding facilities at the project land boundary and generating alternative plans for the project planning suggestion, including: a compliance plan, which involves planning the project land using the screened preliminary indicators of the project planning conditions, including the marked surrounding facilities at the project land boundary in the project planning's surrounding facilities, with the surrounding facilities required to meet the minimum standards in the indicator constraint list; and a quality plan. This includes: planning the project land use based on the preliminary indicators of the screened project planning conditions, excluding surrounding facilities already marked at the project land boundary from the planned surrounding facilities, and requiring surrounding facilities to meet the minimum standards in the indicator constraint list; a hybrid optimization scheme, including: planning the project land use based on the preliminary indicators of the screened project planning conditions, excluding surrounding facilities already marked at the project land boundary from the planned surrounding facilities, and requiring surrounding facilities to meet the minimum standards in the indicator constraint list; simultaneously, under the premise of meeting the numerical requirements in the indicator constraint list, selecting several indicators from the preliminary indicators of the project planning conditions to set key guiding optimization indicators, optimizing by 10% based on the preliminary indicator values ​​of the original project planning conditions; and extracting the land use location, nature, and scale from the project planning suggestion text using the NER model, comparing it with the planning nature and scale of the corresponding plots in the structured planning database to verify consistency.

[0035] In one embodiment of the present invention, the location, nature, and scale of land use are extracted from the text of the project planning proposal using a NER model, and compared with the planning nature and scale of the corresponding plots in the structured planning database to verify consistency. This includes: extracting the constraint relationship between key entity information and binding indicators through dependency parsing; for example, residential land must meet the requirement of a green space ratio of ≥30%; based on the constraint relationship between key entity information and binding indicators, the location, nature, and scale of the project land use are compared with the location, planning nature, and scale of the corresponding plots in the structured planning database, and the cosine similarity of the corresponding location, nature, and scale texts is calculated respectively. The results are then weighted and summed to obtain a consistency coefficient. Data with a consistency coefficient lower than a threshold is marked with an early warning and a manual modification prompt is sent.

[0036] In one embodiment of the present invention, the matching degree between project planning suggestions and planning basis is analyzed under multi-objective optimization constraints, and the project planning suggestions are optimized and verified. This includes the following steps: setting optimization constraint intervals for various indicators in the alternative solutions of the project planning suggestions; calculating compliance verification coefficients under multi-objective optimization constraints; and analyzing the matching degree between project planning suggestions and planning basis. Where: Com represents the optimization verification result of the indicators in the alternative solutions proposed in the project planning (1 for compliance, 0 for non-compliance). Let be the optimization constraint interval (legal requirements + multi-objective optimization constraints) for the k-th indicator, k∈(1,2,…,m), where m is the total number of indicators; perform abnormal warnings and corrections. If Com=0, the system automatically marks abnormal indicators, pushes conflict evidence, and provides correction schemes for compliance, quality and hybrid optimization based on historical cases through a large voice model, and sends a manual review prompt.

[0037] In one embodiment of the present invention, a landscape synergy quantitative evaluation model is established by analyzing the unity of individual units and regions to assess the landscape coordination of project planning recommendations, and a landscape index adjustment formula is set to optimize the indicators of project planning recommendations. This includes the following steps: collecting landscape parameters of the project area and surrounding areas; establishing a landscape synergy quantitative evaluation model by analyzing the unity of individual units and regions: L = ωv*Lv + ωe*Le + ωf*Lf; where L is the overall landscape synergy coordination degree; and Lv is the visual coordination score. Le represents the score for ecological integration. Lf represents the functional adaptability score. ωv represents the evaluation dimension weight, ωe represents the ecological integration weight, and ωf represents the functional adaptation weight, set according to the city planning priorities. In this embodiment, the visual coordination weight ωv = 0.4, the ecological integration weight ωe = 0.3, and the functional adaptation weight ωf = 0.3. Hpr represents the building height of the project, and Hav represents the average building height of the surrounding area. Sts represents the similarity between the project's architectural style and its surroundings. In this embodiment, the similarity Sts is obtained by calculating the cosine similarity between the key semantic features of the actual project's architectural style and the key semantic features of the surrounding building styles. Coc represents the contrast between the project's building color and its surroundings. A drone is used to take 360-degree panoramic photos of the project building and its surroundings at a preset distance. Multiple photos are stitched together into a complete 360-degree panoramic image using PTGui software. The color histograms of the project building and its surrounding areas are extracted. Based on the selected contrast calculation method, in this embodiment, the standard can be selected. The color contrast between the project building and surrounding buildings is calculated using the standard deviation method, variance method, or CIEDE2000 color difference method; Grec is the regional green space connectivity index, calculated using Conefor software to determine the overall connectivity index (IIC) of green space patches, with a value ranging from 0 to 1; Grep is the project's vegetation coverage rate, and Grea is the average vegetation coverage rate of the area surrounding the project; Fum is the project's land use function's fit with the regional planning. In this embodiment, the fit is obtained by calculating the cosine similarity between the semantic features of the actual project's land use function and the semantic features of the regional planned land use function; Spac is the connectivity length between the project's public space and the regional public space, and Spat is the total length of the regional public space between the project's public space and the regional public space; by establishing a landscape synergy quantitative evaluation model, the comprehensive coordination degree of landscape synergy is calculated to assess the landscape coordination of the project planning recommendations, and warnings are issued for project planning recommendations with a comprehensive coordination degree of landscape synergy below the threshold.

[0038] In one embodiment of the present invention, setting a landscape index adjustment formula to optimize the indicators of project planning recommendations includes the following steps: For project planning recommendations that have passed the landscape harmony assessment, the indicators of the project planning recommendations are optimized by setting a landscape index adjustment formula: Where: Ina is the optimized landscape-related index, Lta is the target value of landscape synergy and coordination (preset to 80 points in this embodiment); Lcu is the landscape synergy and coordination degree of the current scheme; Int is the initial planning index.

[0039] In one embodiment of the present invention, during the project construction process, the project is divided into blocks, and the actual landscape data of the completed blocks is obtained. This data is then compared with the actual implemented project planning proposal. By analyzing the landscape implementation fit, the parameters of the landscape indicator adjustment formula are adjusted accordingly. This includes the following steps: During the project construction process, the project is divided into blocks, and the actual landscape data of the completed blocks is obtained. The landscape synergy and coordination target value Lta is incorporated into the block-based formula for block analysis. For each block in the project area, establish the following landscape coordination constraints: stLta≥80; for each block in the project area, establish the following target constraints: stFit≥90%; where Lactual is the actual landscape coordination degree of each block in the project area, and Fit is the optimization target coefficient of each block in the project area; calculate the range of the target value Lta of the landscape coordination degree of each block in the project area under the constraints, average the minimum value within the range to obtain the optimized target value Lta of the landscape coordination degree, and replace the corresponding parameters of the original landscape index adjustment formula. Through the adjusted landscape index adjustment formula, recalculate the optimized landscape-related indicators for the undeveloped blocks.

[0040] In one embodiment of the present invention, the system employs a digital twin-based smart city planning method as described above to realize digital twin-based smart city planning, including: a constraint establishment module: using NLP technology to analyze core information such as land location, nature, and scale of urban planning projects, matching corresponding statutory and non-statutory planning texts from the planning basis database, extracting binding and guiding indicators, and generating an indicator constraint list; an association calculation module: the system automatically calls project surveying data and operational performance data, and generates preliminary indicators by analyzing core indicators of planning conditions; and a planning generation and verification module, including: a planning suggestion generation unit: based on the indicator constraint list and planning conditions... The preliminary indicator calculation results generate project planning recommendations; the compliance verification unit analyzes the matching degree between the project planning recommendations and the planning basis under multi-objective optimization constraints, and optimizes and verifies the project planning recommendations; the landscape coordination calculation module establishes a landscape coordination quantitative evaluation model by analyzing the unity of individual units and regions, evaluates the landscape coordination of the project planning recommendations, and sets landscape indicator adjustment formulas to optimize the indicators of the project planning recommendations; the landscape effect tracking and feedback module divides the project into blocks during the project construction process, obtains the actual landscape data of the completed blocks, compares it with the actual implemented project planning recommendations, and analyzes the fit of landscape implementation to feed back into the parameter adjustment of the landscape indicator adjustment formula.

[0041] The above embodiments are only used to illustrate the technical methods of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical methods of the present invention without departing from the spirit and scope of the technical methods of the present invention.

Claims

1. A smart city planning method based on digital twins, characterized in that, Includes the following steps: By using NLP technology to analyze the basic information of urban planning projects, the system matches corresponding statutory and non-statutory planning texts from the planning basis database, extracts binding and guiding indicators, and generates an indicator constraint list. The system automatically calls project surveying data and operational performance data, and generates preliminary indicators by analyzing the core indicators of planning conditions. Based on the indicator constraint list and the calculation results of the preliminary indicators of planning conditions, the system generates project planning recommendations. By analyzing the matching degree between the project planning recommendations and the planning basis under multi-objective optimization constraints, the system optimizes and verifies the project planning recommendations. By analyzing the unity of individual units and regions, a quantitative evaluation model for landscape synergy is established to assess the landscape coordination of project planning recommendations, and a landscape indicator adjustment formula is set to optimize the indicators of project planning recommendations. During the project construction process, the project is divided into blocks, and the actual landscape data of the completed blocks is obtained. This data is then compared with the actual project planning proposals. By analyzing the fit between the landscape implementation and the proposed solutions, the parameters of the landscape indicator adjustment formula are adjusted accordingly.

2. The smart city planning method based on digital twins according to claim 1, characterized in that, This study uses Natural Language Processing (NLP) technology to analyze basic information of urban planning projects. It matches corresponding statutory and non-statutory planning texts from a planning basis database, extracts binding and guiding indicators, and generates an indicator constraint list. The steps include: For statutory planning texts, integrating territorial spatial planning texts and regulatory detailed planning texts, extracting land use location, nature, and scale fields, and constructing a structured planning database; For non-statutory planning texts, incorporating special planning texts and industrial planning texts, marking their planned land use location, target development nature, and target development scale, and adding them to the structured planning database; Using NLP tools, segmenting and tagging the statutory and non-statutory planning texts in the structured planning database; and setting binding and guiding indicators as key entity information to train NER (Named Entity Recognition) software. The model identifies and marks key entity information in both statutory and non-statutory planning texts within the structured planning database, associating the corresponding values ​​or ranges of these key entity information. It then performs spatial overlay analysis between the project land coordinates or address and the land parcel boundaries in the structured planning database to determine the planning scope to which the project land belongs. From the statutory planning texts within the planning scope of the structured planning database, it retrieves the mandatory clauses for the corresponding land parcels and their marked key entity information, labeling them as hard constraints. From the non-statutory planning texts within the planning scope of the structured planning database, it retrieves the guiding requirements for the corresponding land parcels and their marked key entity information, labeling them as soft constraints. Statutory planning texts specifically include: territorial spatial planning and control planning. The model extracts hard and soft constraints, associates them with the land location in the structured planning database, and generates a list of indicator constraints.

3. The smart city planning method based on digital twins according to claim 1, characterized in that, The system automatically retrieves project surveying data and operational performance data. By analyzing core indicators of the planning conditions, it generates preliminary indicators, including the following steps: The system automatically retrieves project surveying data and operational performance data. Project surveying data includes topography, buildings, and transportation; operational performance data includes the load of surrounding public services and traffic efficiency; by analyzing core indicators of the planning conditions, it generates preliminary indicators: Where: Inn is the preliminary planning condition indicator (such as plot ratio and building height), Inga is the upper limit of the indicator specified by the statutory planning, β is the operation performance correction coefficient, Pera is the actual operation performance of the surrounding area, Pert is the operation performance target value, γ is the current density correction coefficient, Deac is the current building density of the surrounding area, and Dest is the planned standard building density.

4. The smart city planning method based on digital twins according to claim 1, characterized in that, Based on the indicator constraint list and the preliminary indicator calculation results of the planning conditions, project planning suggestions are generated, including the following steps: Importing the preliminary indicators of the project planning conditions and screening them against the values ​​in the indicator constraint list; marking data where the preliminary indicators of the planning conditions do not meet the numerical requirements in the indicator constraint list and sending manual modification prompts; using GIS tools to spatially overlay the project land boundary with the plot area and surrounding facilities in the structured planning database to clarify the locational relationship of the project land boundary; marking existing surrounding facilities at the project land boundary and generating alternative plans for the project planning suggestions, including: a compliance plan, which includes planning the project land according to the screened preliminary indicators of the project planning conditions, including the surrounding facilities already marked at the project land boundary in the project planning surrounding facilities, with the surrounding facilities required to meet the minimum standards in the indicator constraint list; and a quality plan, which includes... The preliminary indicators of the project planning conditions are used to plan the project land use. Surrounding facilities already marked at the project land boundary are not included in the project planning's surrounding facilities list, and the surrounding facilities are required to meet the minimum standards in the indicator constraint list. A hybrid optimization scheme includes: planning the project land use based on the preliminary indicators of the screened project planning conditions, excluding surrounding facilities already marked at the project land boundary from the project planning's surrounding facilities list, and requiring surrounding facilities to meet the minimum standards in the indicator constraint list; simultaneously, while meeting the numerical requirements in the indicator constraint list, selecting several indicators from the preliminary indicators of the project planning conditions to set key guiding optimization indicators, optimizing them by 10% based on the original preliminary indicator values ​​of the project planning conditions; and using a NER model to extract the land use location, nature, and scale from the project planning suggestion text, comparing it with the planning nature and scale of the corresponding plots in the structured planning database to verify consistency.

5. A smart city planning method based on digital twins according to claim 1, characterized in that, The NER model is used to extract land location, nature, and scale from the text of project planning proposals. This information is then compared with the planning nature and scale of corresponding plots in the structured planning database to verify consistency. This includes: extracting the constraint relationships between key entity information and binding indicators through dependency parsing; comparing the project land location, nature, and scale with the corresponding land location, planning nature, and scale in the structured planning database based on these constraints; calculating the cosine similarity of the corresponding location, nature, and scale texts; and weighting and summing the results to obtain a consistency coefficient. Data with consistency coefficients below a threshold are marked with an alert and a manual modification prompt is sent.

6. A smart city planning method based on digital twins according to claim 5, characterized in that, By analyzing the matching degree between project planning recommendations and planning basis under multi-objective optimization constraints, the project planning recommendations are optimized and verified, including the following steps: setting optimization constraint ranges for various indicators in the alternative solutions of the project planning recommendations; calculating compliance verification coefficients by analyzing the multi-objective optimization constraints; and analyzing the matching degree between project planning recommendations and planning basis. Where: Com represents the optimization verification result of the indicators in the alternative solutions proposed in the project planning. The optimization constraint interval for the k-th indicator is k∈(1,2,…,m), where m is the total number of indicators. Anomaly warnings and corrections are performed. If Com=0, the system automatically marks the abnormal indicator, pushes the conflict basis, and provides correction schemes for compliance, quality and hybrid optimization schemes based on historical cases through a large voice model, and sends a manual review prompt; otherwise, it is determined that the alternative scheme does not need to be modified.

7. A smart city planning method based on digital twins according to claim 1, characterized in that, By analyzing the unity of individual projects and the overall landscape, a quantitative evaluation model for landscape synergy is established to assess the landscape coordination of project planning recommendations. A landscape index adjustment formula is then set to optimize the indicators of the project planning recommendations. The process includes the following steps: collecting landscape parameters from the project area and surrounding areas; and establishing a quantitative evaluation model for landscape synergy by analyzing the unity of individual projects and the overall landscape. The model is: L = ωv*Lv + ωe*Le + ωf*Lf; where L represents the overall landscape synergy degree; and Lv represents the visual coordination score. Le represents the score for ecological integration. Lf represents the functional adaptability score. ωv represents the evaluation dimension weight, ωe represents the ecological integration weight, and ωf represents the functional adaptation weight, set according to the city's planning priorities; Hpr represents the project's building height, and Hav represents the average building height in the surrounding area; Sts represents the similarity between the project's architectural style and its surroundings; Coc represents the contrast between the project's building color and its surroundings; Grec represents the regional green space connectivity index; Grep represents the project's vegetation coverage rate, and Grea represents the average vegetation coverage rate in the surrounding area; Fum represents the fit between the project's land use function and the regional planning; Spac represents the connection length between the project's public space and the regional public space, and Spat represents the total length of the regional public space between the project's public space and the regional public space; By establishing a landscape synergy quantitative evaluation model, the comprehensive coordination degree of landscape synergy is calculated, the landscape coordination of the project planning recommendations is assessed, and warnings are issued for project planning recommendations with a comprehensive coordination degree of landscape synergy below the threshold.

8. A smart city planning method based on digital twins according to claim 1, characterized in that, Setting up landscape indicator adjustment formulas to optimize project planning recommendations involves the following steps: For project planning recommendations that have passed the landscape harmony assessment, the indicators of the project planning recommendations are optimized by setting up landscape indicator adjustment formulas: Where: Ina is the optimized landscape-related index, Lta is the target value of landscape synergy and coordination, Lcu is the landscape synergy and coordination degree of the current scheme, and Int is the initial planning index.

9. A smart city planning method based on digital twins according to claim 8, characterized in that, During project construction, the project is divided into blocks, and actual landscape data of the completed blocks is obtained. This data is then compared with the actual implemented project planning proposals. By analyzing the fit between the landscape implementation and the proposed solutions, the parameters of the landscape indicator adjustment formula are adjusted accordingly. This process includes the following steps: During project construction, the project is divided into blocks, and actual landscape data of the completed blocks is obtained. The landscape synergy and coordination target value Lta is incorporated into the block-based formula for block analysis. For each block in the project area, establish the following landscape coordination constraints: stLta≥80; for each block in the project area, establish the following target constraints: stFit≥90%; where Lactual is the actual landscape coordination degree of each block in the project area, and Fit is the optimization target coefficient of each block in the project area; calculate the range of the target value Lta of the landscape coordination degree of each block in the project area under the constraints, average the minimum value within the range to obtain the optimized target value Lta of the landscape coordination degree, and replace the corresponding parameter in the original landscape index adjustment formula.

10. A smart city planning system based on digital twins, characterized in that, This system employs a digital twin-based smart city planning method as described in any one of claims 1-9 to achieve digital twin-based smart city planning, comprising: a constraint establishment module: using NLP technology to parse basic information of urban planning projects, matching corresponding statutory and non-statutory planning texts from the planning basis database, extracting binding and guiding indicators, and generating an indicator constraint list; an association calculation module: the system automatically calls project surveying data and operational performance data, and generates preliminary indicators by analyzing core indicators of planning conditions; and a planning generation and verification module, comprising: a planning suggestion generation unit: based on the indicator constraint list and the calculation results of preliminary indicators of planning conditions, ... The system generates project planning recommendations; a compliance verification unit analyzes the matching degree between project planning recommendations and planning basis under multi-objective optimization constraints, and optimizes and verifies the project planning recommendations; a landscape coordination calculation module establishes a landscape coordination quantitative evaluation model by analyzing the unity of individual units and regions, evaluates the landscape coordination of project planning recommendations, and sets landscape indicator adjustment formulas to optimize the indicators of project planning recommendations; and a landscape effect tracking and feedback module divides the project into blocks during project construction, obtains the actual landscape data of the completed blocks, compares it with the actual implemented project planning recommendations, analyzes the fit of landscape implementation, and feeds back the parameter adjustments of the landscape indicator adjustment formulas.

Images