Urban renewal project generation and simulation deduction method based on ai and digital twinning

Abstract

The application relates to the technical field of digital twinning, in particular to an urban renewal project generation and simulation deduction method based on AI and digital twinning, which comprises the following steps: constructing a digital twinning body and generating an update scheme library through multi-source data fusion, providing accurate models and various choices for urban renewal projects, and laying a solid foundation; avoiding invalid deduction and improving the subsequent deduction efficiency and the feasibility of the scheme by identifying potential contradictions of the scheme in advance and screening out a set of schemes to be deduced; comprehensively understanding the influence of the scheme and providing key basis for quantifying contradictions by multi-dimensional deep deduction and extracting contradiction transmission relationships among dimensions; eliminating unfeasible schemes by quantifying the comprehensive contradiction degree of each scheme, making the screening more scientific, and improving the success rate of urban renewal project implementation; forming a closed loop by feeding back the contradiction signal to optimize the generation strategy, continuously improving the quality of the scheme, and ensuring that the update scheme is more in line with actual demands.

Description

Technical Field

[0001] This invention relates to the field of digital twin technology, and in particular to a method for generating and simulating urban renewal projects based on AI and digital twins. Background Technology

[0002] In the field of urban renewal, traditional project generation and planning methods often rely on human experience and local data analysis. Typically, planners first develop several renewal plans based on a preliminary understanding of the urban area and basic data, such as simple geographical information and building distribution. These plans are then initially screened and refined through expert discussions and on-site surveys.

[0003] However, this traditional method has many limitations. On the one hand, data acquisition is incomplete and lacks dynamism. Relying solely on limited basic data makes it difficult to fully grasp the complex realities of urban renewal areas, such as dynamic changes in population flow and the real-time operational status of infrastructure, resulting in generated plans that are difficult to accurately meet actual needs. On the other hand, plan evaluation lacks systematicity. Static indicator analysis cannot deeply consider the complex interactions between various elements during plan implementation, such as the comprehensive impact of the renewal plan on traffic flow, social carrying capacity, and other dimensions, as well as the potential contradictions between different plans in terms of space, time, and resources. When faced with large-scale and complex urban renewal projects, traditional methods struggle to effectively coordinate numerous factors, easily leading to unforeseen problems during implementation, such as increased traffic congestion, insufficient public service facilities, and environmental degradation, thereby affecting the overall benefits and sustainable development of urban renewal projects and failing to meet the requirements of modern urban renewal for scientific rigor, precision, and efficiency. To address this, this invention proposes a method for generating and simulating urban renewal projects based on AI and digital twins. Summary of the Invention

[0004] The purpose of this invention is to solve the problems in the background art and to propose a method for generating and simulating urban renewal projects based on AI and digital twins.

[0005] To overcome the aforementioned shortcomings of existing technologies and to achieve the above objectives, this invention provides a method for generating and simulating urban renewal projects based on AI and digital twins, comprising: S1. Obtain multi-source data of the area to be updated, construct a digital twin of the area to be updated, and obtain an update scheme library generated based on the update requirements, wherein the update scheme library contains multiple alternative update schemes; S2. Identify and judge the potential conflict relationships between each candidate update scheme in the update scheme library, and select candidate update schemes whose conflict relationships between schemes meet the preset level threshold to form a set of schemes to be deduced. S3. Load each scheme in the set of schemes to be simulated into the updated digital twin of the area in sequence for multi-dimensional in-depth simulation, and identify and extract the contradiction transmission relationship between dimensions from the simulation results; S4. Based on the contradiction transmission relationship between dimensions, the degree of comprehensive contradiction exposed by each scheme in the simulation is quantified, and the schemes with a comprehensive contradiction degree exceeding the preset contradiction threshold are eliminated, and the remaining schemes are retained to form a candidate scheme set. S5. The contradiction transmission relationship between dimensions is used as a deduced contradiction signal and fed back to the update scheme library. The generation strategy of the update scheme is optimized based on the deduced contradiction signal.

[0006] Further, S1 includes: Acquire multi-source data for the area to be updated, namely, current geographic information data, existing building BIM data, infrastructure pipeline data, population distribution data, and traffic flow data; among which, the area to be updated is the urban renewal geographic area targeted for acquiring multi-source data; By fusing multi-source data, an updated digital twin of the area is constructed, which includes the area's physical space and dynamic operational status. Obtain the update scheme library generated based on update requirements and update constraints.

[0007] Furthermore, in S1, each candidate update scheme in the update scheme library and the digital twin of the update area maintain a unified reference in the spatial coordinate system, and the spatial range information of each candidate update scheme is recorded in the form of spatial coordinate boundaries.

[0008] Further, S2 includes: Analyze the spatial range information of each alternative update scheme, and identify scheme pairs with overlapping or adjacent spatial ranges within the physical space of the digital twin of the update area; For the identified scheme pairs, based on the execution time sequence, analyze whether there are overlapping periods in the execution time of the scheme pairs; at the same time, based on the expected types and quantities of resources to be invested, analyze whether there are competing relationships in terms of resource occupancy types of the scheme pairs; based on the analysis process, obtain the spatial overlap degree, the length of the time overlap period, and the number of resource competition types of the scheme pairs as the judgment factors characterizing the conflict level. Based on the judgment factors that characterize the level of contradiction, the potential contradiction relationship level between the scheme pairs is comprehensively determined. The alternative update schemes with potential conflict levels below a preset threshold are selected and retained to form a set of schemes to be deduced.

[0009] Furthermore, in S2, the method for determining the level of potential conflict relationships includes: Obtain the percentage of overlapping area between the two schemes as the degree of spatial overlap; Obtain the number of overlapping days in the execution time sequence between the scheme pairs, and use it as the length of the time sequence overlap period; The number of identical resource types among the expected resource types to be invested in the two schemes is used as the number of resource competition types. By combining the degree of spatial overlap, the length of time of temporal overlap, and the number of resource competition types with preset corresponding weights, a potential conflict relationship level between scheme pairs is generated.

[0010] Further, S3 includes: Each scheme in the set of schemes to be simulated is loaded into the updated digital twin of the area in sequence, and in-depth simulations are carried out in the dimensions of spatial efficiency, social carrying capacity, and environmental carrying capacity, generating a sequence of simulation indicators for each dimension. Anomaly detection is performed on the inference index sequence of the same scheme in different dimensions, and the indexes that deviate abnormally and their respective dimensions are marked. By analyzing the temporal correlation of abnormal deviation indicators between different dimensions, when there is a causal relationship between the occurrence of abnormal deviation indicators in one dimension and the occurrence of abnormal deviation indicators in another dimension in time, the causal relationship is identified as a contradiction transmission relationship between dimensions. Extract the dimension categories, transmission directions, and transmission strengths involved in the contradiction transmission relationships between dimensions.

[0011] Furthermore, in S3, the spatial efficiency dimension simulation includes simulations of the spatial utilization rate, traffic accessibility, and functional mixing of the area after the implementation of the renewal plan; the social carrying capacity dimension simulation includes simulations of population carrying capacity density, public service facility coverage, and community vitality index; and the environmental carrying capacity dimension simulation includes simulations of carbon emission intensity, microclimate comfort, and green space accessibility.

[0012] Further, S4 includes: Obtain the contradiction transmission relationships between all dimensions identified during the simulation of each scheme; The total number of contradictory transmission relationships between the dimensions involved in each solution, and the transmission strength of each contradictory transmission relationship, are counted. The total number and intensity of contradiction transmission relationships are comprehensively quantified to generate the overall contradiction degree of the scheme; The overall degree of contradiction is compared with the preset contradiction threshold, and the schemes whose overall degree of contradiction exceeds the preset contradiction threshold are eliminated from the schemes to be simulated. The remaining solutions after elimination form a candidate solution set, and the comprehensive contradiction degree of each solution in the candidate solution set is sorted from low to high and output.

[0013] Further, S5 includes: The contradiction transmission relationship between dimensions exposed by each candidate scheme in the simulation, as well as the spatial location, temporal stage and resource type involved, are extracted in a structured manner as the contradiction signal of the simulation. The deduced contradiction signals are fed back to the generation strategy module of the updated solution library, and the constraint parameters of the generation strategy of the updated solution are adjusted. Among them, the adjustment of constraint parameters includes: modifying the boundary generation rules of the scheme spatial range to avoid spatial overlap areas, and optimizing the arrangement rules of the scheme execution sequence to stagger the resource competition sequence. New alternative updated schemes are generated based on the generation strategy after the constraint parameters are adjusted, and added to the updated scheme library, forming a closed loop of two-way iteration between scheme generation and simulation deduction.

[0014] Furthermore, in S5, the closed loop of bidirectional iteration stops iterating when any of the following termination conditions are met: there is a scheme in the candidate scheme set whose comprehensive contradiction degree is lower than the preset excellent threshold; or the number of iteration rounds reaches the preset maximum number of iterations; or the distribution of the comprehensive contradiction degree of the candidate scheme set tends to converge in two adjacent iteration rounds.

[0015] Compared with existing technologies, the beneficial effects of this invention in providing a method for generating and simulating urban renewal projects based on AI and digital twins are as follows: 1) By acquiring multi-source data of the area to be updated and constructing a digital twin, an update solution library is generated. The fusion of multi-source data enables the digital twin to fully reflect the physical and operational status of the area, providing a data foundation for subsequent simulations. The update solution library provides a variety of alternative solutions, increasing the diversity of project selection and meeting different update needs. 2) By identifying potential conflicts between alternative update schemes, schemes that meet preset level thresholds are selected to form a set of schemes to be simulated. This allows for the early detection of potential conflicts between schemes, avoiding wasted resources in subsequent simulations, improving simulation efficiency, and ensuring that the schemes entering the simulation stage have high feasibility. By loading the schemes into a digital twin for multi-dimensional in-depth simulation, the contradiction transmission relationship between dimensions is identified and extracted. Multi-dimensional simulation comprehensively evaluates the impact of the schemes. Identifying the contradiction transmission relationship between dimensions can provide a deeper understanding of the interaction between various factors during the implementation of the schemes, providing a basis for subsequent quantification of the degree of contradiction. 3) By quantifying the comprehensive degree of contradiction among various schemes based on the contradiction transmission relationship between dimensions, schemes exceeding the preset contradiction threshold are eliminated, and the candidate scheme set is retained. Quantitative evaluation makes the scheme selection more scientific and objective, effectively reduces infeasible schemes, and improves the success rate of project implementation. By feeding back the contradiction transmission relationship to the updated scheme library, the generation strategy is optimized. A two-way iterative closed loop of scheme generation and simulation is formed to continuously improve the quality of the schemes, making the final determined renewal schemes more reasonable and feasible, and better adapting to the needs of urban renewal. Attached Figure Description

[0016] Figure 1 This is a flowchart of the method for generating and simulating urban renewal projects based on AI and digital twins proposed in this invention. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] Please see Figure 1 This invention provides a method for generating and simulating urban renewal projects based on AI and digital twins, including: S1. Obtain multi-source data of the area to be updated, construct a digital twin of the area to be updated, and obtain an update scheme library generated based on the update requirements, wherein the update scheme library contains multiple alternative update schemes; S2. Identify and judge the potential conflict relationships between each candidate update scheme in the update scheme library, and select candidate update schemes whose conflict relationships between schemes meet the preset level threshold to form a set of schemes to be deduced. S3. Load each scheme in the set of schemes to be simulated into the updated digital twin of the area in sequence for multi-dimensional in-depth simulation, and identify and extract the contradiction transmission relationship between dimensions from the simulation results; S4. Based on the contradiction transmission relationship between dimensions, the degree of comprehensive contradiction exposed by each scheme in the simulation is quantified, and the schemes with a comprehensive contradiction degree exceeding the preset contradiction threshold are eliminated, and the remaining schemes are retained to form a candidate scheme set. S5. The contradiction transmission relationship between dimensions is used as a deduced contradiction signal and fed back to the update scheme library. The generation strategy of the update scheme is optimized based on the deduced contradiction signal.

[0019] In this embodiment of the invention, the detailed process of S1—acquiring multi-source data of the area to be updated, constructing a digital twin of the updated area, and obtaining a library of update schemes generated based on update requirements—includes: S11. Obtain multi-source data for the area to be updated, namely, current geographic information data, existing building BIM data, infrastructure pipeline data, population distribution data, and traffic flow data; among which, the area to be updated is the urban renewal geographic area targeted for obtaining multi-source data. In this embodiment, multiple data acquisition channels are executed simultaneously to obtain multi-source data for the area to be updated: First, high-precision current geographic information data of the area is retrieved from the data platform of the natural resources and planning authority. This data includes topographic elevation models, cadastral boundary layers, and land cover classification information. Second, the housing and urban-rural construction archive system is connected to extract the as-built BIM model data of existing buildings in the area. For older buildings lacking BIM data, a 3D laser scanning reverse modeling method is used to complete the data. Next, underground pipeline detection results and pipeline topology data are obtained from the municipal public utility management unit to form an infrastructure pipeline data set. At the same time, anonymized residential population distribution data and time-segmented population activity heatmap data are obtained from the public security household registration management platform and the telecommunications operator signaling platform, and cross-sectional traffic flow and pedestrian flow monitoring data of major road sections are accessed from the urban traffic management platform. It is understood that after obtaining multi-source data, it is uniformly converted to the same geographic coordinate system, and the data format and spatiotemporal granularity are standardized and preprocessed to form a basic data resource pool for the construction of digital twins.

[0020] S12. Construct an updated digital twin of the area, including the area's physical space and dynamic operational status, through multi-source data fusion; Understandably, the specific process of building and updating the digital twin of the area is divided into two stages: static physical space modeling and dynamic operational status mapping. The first stage of static physical space modeling uses existing geographic information data and existing building BIM data as core inputs. A spatial semantic fusion engine is used to overlay and geometrically register the terrain, building outline, interior space structure and underground pipelines layer by layer to generate a three-dimensional solid model of the area covering the entire space above and below ground. Each building component, pipeline segment and plot unit in the three-dimensional solid model of the area is assigned a unique spatial identifier code to establish a one-to-one correspondence between spatial entities and attribute information. The second phase, dynamic operation status, involves embedding a dynamic data perception layer into the static physical space model. Population distribution data is mapped to corresponding building volumes according to community grid units, forming a population carrying capacity density distribution layer. Traffic flow data is mapped to the road network model according to road segments, forming a dynamic traffic situation layer. Infrastructure pipeline data is correlated with real-time monitoring values ​​from IoT sensors, forming a pipeline network operation status layer. Subsequently, through a time-series data stream access engine, the population carrying capacity density distribution layer, dynamic traffic situation layer, and pipeline network operation status layer contained in the dynamic data perception layer are aligned with the static physical space model along the time axis. This allows the digital twin to continuously reflect the synchronous changes in the physical spatial form and operational status of personnel activities, traffic flow, and facility operation within the area over time. This results in the creation of an updated digital twin of the area that describes spatial geometric features and depicts dynamic operational patterns. It should be noted that in the process of constructing the digital twin of the updated area, the spatial coordinate system benchmark, multi-source data fusion semantic standard, and dynamic layer access protocol adopted all refer to the Zhongbang Digital Twin Construction Technical Specifications. Among them, in the spatial semantic fusion of existing geographic information data and existing building BIM data, the urban information model data exchange format provided by the Zhongbang CIM basic platform is used for intermediate data conversion to ensure semantic consistency of heterogeneous data from different sources in geometric registration and attribute attachment. In addition, the mapping process of population distribution data and traffic flow data to the dynamic layer of the digital twin follows the Zhongbang Urban Operation Vital Signs Data Access Specifications, enabling the constructed digital twin of the updated area to achieve seamless connection and data interoperability with the municipal urban operation management platform.

[0021] S13. Obtain an update scheme library generated based on update requirements and update constraints, wherein update requirements and update constraints include at least one of the following: area functional positioning requirements, planning indicator restrictions, investment budget constraints, and historical protection restrictions. Specifically, the planning unit or project decision-making body shall input the functional positioning requirements and planning indicator restrictions for the area to be updated. For example, the core function of the area shall be determined as residential improvement, industrial upgrading, or public space supplementation. The planning control values ​​such as the upper limit of plot ratio, building height control, and lower limit of green space ratio shall be input. At the same time, the total amount of investment budget constraints and annual allocation plan shall be input, as well as the list of protected buildings and style control boundaries involved in historical protection restrictions. The system invokes a pre-built update scheme generation engine, which includes a library of scheme templates and parametric components for different update goals. Based on the input requirements and constraints, the engine automatically generates multiple alternative update schemes by combining different building renovation methods, spatial layout adjustment strategies, and functional business format replacement schemes. Each alternative update scheme is stored in a structured manner and output to the update scheme library in a standardized data format. The update scheme library serves as a unified input source for subsequent analysis and deduction, supporting the addition, deletion, version management, and related retrieval of alternative schemes. It should be noted that each alternative renewal plan includes the spatial scope covered by the plan, the implementation schedule, the expected types and quantities of resources to be invested, and information on the functional adjustments involved. Specifically, the spatial scope covered by the plan refers to the geographical boundary of the alternative renewal plan within the renewal area; the implementation schedule indicates the time plan and sequence of various engineering activities of the alternative renewal plan; the expected types and quantities of resources to be invested indicate the types and corresponding scale of human resources, machinery, materials, funds, etc. required for the implementation of the alternative renewal plan; and the information on the functional adjustments involved refers to the changes in the use function and business layout of existing buildings or spaces caused by the alternative renewal plan. In the update scheme library, each candidate update scheme and the digital twin of the update area maintain a unified benchmark in the spatial coordinate system, and the spatial range information of each candidate update scheme is recorded in the form of spatial coordinate boundaries.

[0022] In this embodiment of the invention, the detailed implementation steps of S2, which identifies and judges the potential conflicting relationships between the candidate update schemes in the update scheme library and filters out candidate update schemes whose conflicting relationships meet a preset level threshold, to form a set of schemes to be deduced, include: S21. Analyze the spatial extent information of each candidate update scheme. Within the physical space of the digital twin of the update area, identify scheme pairs with overlapping or adjacent spatial extents, including: The system sequentially reads the spatial extent information of each candidate update scheme from the update scheme library, which is recorded in the form of spatial coordinate boundaries. It then converts the spatial extent of each candidate update scheme into a polygonal geometric object and loads it into the unified spatial coordinate system of the digital twin of the update area. A spatial overlay analysis is performed, calculating the spatial relationship between the polygonal geometric objects corresponding to any two candidate update schemes: if the two polygons have overlapping areas, they are considered spatially overlapping; if the boundaries of the two polygons are adjacent and the distance is less than a preset proximity tolerance, they are considered spatially adjacent. The system iterates through all pairwise combinations of candidate update schemes in the update scheme library, outputting the combinations that are determined to have spatially overlapping or spatially adjacent relationships as a scheme pair. Each scheme pair is accompanied by a type label recording its spatial relationship and a description of the spatial location of the intersecting area or adjacent boundary.

[0023] S22. For the identified scheme pair, based on the scheme execution sequence, further analyze whether there is an overlap in the execution time of the scheme pair; at the same time, based on the expected type and quantity of resources to be invested, analyze whether there is a competitive relationship in the resource occupation type of the scheme pair; based on the analysis process, obtain the spatial overlap degree, the length of the time overlap period and the number of resource competition types of the scheme pair as the judgment factors representing the contradiction level. Specifically, for each identified pair of schemes, the execution time sequence of each scheme is extracted; the execution time sequence is represented by a Gantt chart to show the planned start and end times of each engineering activity; the execution time intervals of the two schemes are aligned on a unified time axis to determine whether there is a time overlap between the overall execution period of the two schemes or any sub-activity period; if there is a time overlap, the start and end times of the overlap interval are recorded and the number of overlapping days is calculated; if there is no overlap, it is marked as no time overlap. During the synchronous execution of the time-series analysis, the expected resource types and quantities for each of the two plans are extracted. The expected resource types and quantities are stored in the form of a resource list, in which resource types are at least divided into four categories: manpower teams, construction machinery, special materials, and temporary land use. The resource lists of the two plans are compared to identify resource types that appear simultaneously in both plans. For each resource type that appears simultaneously, it is further determined whether the sum of the demand for that resource by the two plans exceeds the supply limit of that resource in the corresponding time period. If it exceeds the supply limit, the resource type is marked as having a competitive relationship. The resource types with competitive relationships are recorded one by one, and the total number of resource types with competitive relationships is counted.

[0024] S23. Based on the determination factors representing the level of contradiction, comprehensively determine the level of potential contradiction relationship between the scheme pairs, specifically as follows: The overlapping area ratio between scheme pairs is obtained as the degree of spatial overlap. Specifically, after completing spatial range analysis and scheme pair identification, spatial overlay analysis results for each scheme pair are generated. These results record the two polygonal geometric objects of each scheme pair and their intersection relationship data. The intersection area of ​​the polygonal geometric objects of the two schemes is extracted from the spatial overlay analysis results. At the same time, the footprint of each polygonal geometric object of the two schemes is obtained. The intersection area is divided by the smaller value of the footprint of the two schemes to obtain the value of the overlapping area ratio. The value of this ratio ranges from zero to one. The closer the value is to one, the closer the two schemes are to complete overlap in space, and the higher the degree of spatial competition. This overlapping area ratio is recorded as the spatial overlap degree judgment value of the scheme pair.

[0025] Obtain the number of overlapping days between the execution times of the scheme pairs, as the length of the overlapping time period, including: Read the execution time sequence data of each of the two schemes; perform an intersection operation on the continuous time axis by calculating the execution intervals defined by the planned start time and planned end time of the two schemes; if there is an intersection time interval, calculate the number of calendar days of the intersection interval, which is the overlap number; if the two schemes have no overlapping intervals on the time axis, the overlap number is recorded as zero; record the overlap number as the determination value of the time sequence overlap period length of the scheme pair.

[0026] The number of identical resource types among the expected resource types to be invested in the two schemes is used as the number of resource competition types. Understandably, the expected resource type lists for both schemes are read; the resource type names are used as comparison keys to match and search the two lists; the number of resource type names that appear in both lists is counted. For example, if both schemes require the use of tower cranes and concrete mixer trucks, the number of identical resource types is counted as two; the number of identical resource types is recorded as the resource competition type quantity judgment value for the scheme pair.

[0027] By combining the degree of spatial overlap, the length of temporal overlap, and the number of resource competition types with preset corresponding weights, a potential conflict relationship level between scheme pairs is generated. Specifically, a set of corresponding weights is preset, consisting of three weight coefficients: spatial dimension weight, temporal dimension weight, and resource dimension weight. The sum of the three weight coefficients is a fixed value. The spatial overlap value is multiplied by the spatial dimension weight to obtain the spatial conflict contribution component. The temporal overlap period length value is multiplied by the temporal dimension weight to obtain the temporal conflict contribution component. The number of resource competition types is multiplied by the resource dimension weight to obtain the resource conflict contribution component. The three contribution components are summed to obtain the comprehensive conflict score. Based on the preset numerical range in which the comprehensive conflict score falls, the scheme pairs are classified into corresponding levels of potential conflict relationships. The numerical range is a continuous sub-range into which the possible values ​​of the comprehensive conflict score are divided, and each sub-range corresponds to a potential conflict relationship level. This range division method is completed in the initialization phase based on historical project data. It can be understood that the preset corresponding weights are pre-calibrated according to the actual situation of the area to be updated, that is, based on the degree of conflict between schemes in each dimension, to reflect the relative importance of the conflict in different dimensions to the overall feasibility of the scheme. After completing the three analytical tasks of spatial overlap extraction, temporal overlap period length calculation, and resource competition type quantity statistics, the specific values ​​of the three judgment factors are obtained. Spatial overlap is represented by the ratio of the overlapping area to the smaller of the areas of the two schemes. The temporal overlap period length is represented by the number of days in the overlapping period of the execution time of the two schemes. At the same time, the number of resource competition types with competition relationship is used to represent the quantity of resource competition types. The values ​​of the three judgment factors are input into a pre-constructed conflict level assessment mapping table. The specific implementation of the conflict level assessment mapping table can adopt a pre-trained decision tree classification model. This decision tree model uses spatial overlap, temporal overlap period length, and the quantity of resource competition types as input features. At each internal node of the tree, the model selects path branches based on the comparison results of feature values ​​and preset split thresholds. Finally, the corresponding potential conflict relationship level is output at the leaf node. The tree structure and split threshold of the decision tree are obtained by supervised learning training using actual conflict case data of scheme pairs in historical urban renewal projects, which can adapt to the nonlinear boundary requirements of conflict level judgment under different area characteristics.

[0028] S24. Select and retain alternative update schemes whose potential conflict relationship level is lower than the preset level threshold to form a set of schemes to be deduced. Specifically, a level threshold is set, which corresponds to a specific numerical node in the comprehensive contradiction score delineation. The level threshold is a criterion used to distinguish whether the potential contradiction risk between scheme pairs is acceptable. The level threshold is pre-set based on the actual management tolerance of the area to be updated, the sufficiency of resource supply, and the urgency of the project implementation cycle. For example, in update projects with tight resource supply and high sensitivity to overlapping construction periods, the pre-set level threshold can be lowered accordingly to filter high contradiction scheme pairs with stricter standards; conversely, the threshold can be appropriately raised to retain more alternative schemes for subsequent simulation. The purpose of setting the level threshold is to eliminate multi-scheme combinations with significant internal contradictions before investing computing resources in in-depth simulation, so as to avoid the ineffective consumption of simulation resources. For each generated scheme pair, a threshold comparison analysis is performed to assess the potential conflict level. If the overall conflict score of a scheme pair is higher than or equal to the preset level threshold, the scheme pair is determined to have an inherent conflict risk that cannot be ignored, and both schemes constituting the scheme pair will be marked as pending screening. After traversing all scheme pairs, the number of times each candidate update scheme is marked as pending screening is counted. If the number of times a scheme is marked exceeds the preset tolerance limit, the scheme is removed from the update scheme library. The tolerance limit refers to the maximum number of times a single candidate update scheme is judged as a high-conflict party by different scheme pairs. This limit is preset according to the expected size of the scheme set to be deduced, and is used to prevent the mistaken removal of schemes with overall feasibility due to a few marginal conflicts in the conflict pre-screening stage. If a scheme is marked zero times or less than the tolerance limit, the scheme is retained. All retained candidate update schemes together constitute the scheme set to be deduced, which only contains candidate schemes with controllable conflict risks between schemes, and serves as the input object for subsequent in-depth deduction.

[0029] In this embodiment of the invention, the specific implementation method of S3, which sequentially loads each scheme in the set of schemes to be simulated into the updated digital twin of the region for multi-dimensional in-depth simulation, and identifies and extracts the contradiction transmission relationship between dimensions from the simulation results, is as follows: S31. Load each scheme in the set of schemes to be simulated into the updated digital twin of the area in sequence, and conduct in-depth simulations of spatial efficiency, social carrying capacity, and environmental carrying capacity dimensions respectively. The method for generating the simulation index sequence corresponding to each dimension is as follows: After loading a single scheme from the set of schemes to be simulated into the digital twin, three parallel simulation calculation channels are started, corresponding to the spatial efficiency dimension, social carrying capacity dimension, and environmental carrying capacity dimension, respectively. Among them, the spatial efficiency dimension simulation includes simulation of the spatial utilization rate, traffic accessibility and functional mixing of the area after the implementation of the renewal plan; the social carrying capacity dimension simulation includes simulation of population carrying density, public service facility coverage and community vitality index; the environmental carrying capacity dimension simulation includes simulation of carbon emission intensity, microclimate comfort and green space accessibility. It should be noted that space utilization rate refers to the ratio of the actual building area in use to the total usable building area in the area after the implementation of the renovation plan. It is calculated through the building volume model and usage status label layer in the digital twin. Traffic accessibility reflects the convenience of any location in the area to the nearest public transportation station or main road entrance. It is specifically measured by the average time required and is accurately assessed by the joint calculation of the road network model and traffic flow simulation model. Functional mixing degree describes the degree of spatial mixing of various functional formats such as residential, commercial, office, and public services in the area. Its calculation is based on the information entropy of the building function label layer, which can comprehensively reflect the diversity and mixing degree of functions in the area. Population carrying capacity density, an important indicator for measuring the degree of population concentration in an area, refers to the number of people actually carried per unit area of ​​construction land. This indicator can be obtained by combining population distribution data with the zoning statistics of land area. Public service facility coverage rate reflects the proportion of residents in an area who can easily access public service facilities such as basic education, medical care, and culture within walking distance. It is obtained by overlaying service radius buffer analysis with population grids. Community vitality index is a comprehensive indicator for measuring the intensity and diversity of human activity during non-commuting hours in an area. This index can comprehensively reflect the vitality of the community by analyzing mobile phone signaling trajectory dwell points and combining statistical data on the frequency of visits to points of interest. Carbon emission intensity, as a key indicator for measuring the environmental performance of an area, refers to the amount of carbon dioxide emitted per unit of building area or per unit of output value. Its calculation process involves the conversion between energy consumption simulation models and carbon emission factors. Microclimate comfort refers to the human comfort evaluation level based on factors such as temperature, humidity, wind speed, and solar radiation. It is obtained by coupling computational fluid dynamics simulation with outdoor thermal comfort evaluation standards. Green space accessibility reflects the ease with which residents in the area can reach the nearest park or open space. It is specifically measured by walking distance. This indicator can be calculated by combining green space patch layers with road network shortest path analysis. Each simulation channel progresses at a set simulation time step, from the start time to the end time. At each simulation time step node, the key indicator values ​​for that dimension are calculated. For example, in the spatial efficiency dimension channel, space utilization rate, traffic accessibility index, and functional mixing index are calculated every simulation time step, and the results are arranged in chronological order. In the social carrying capacity dimension channel, population carrying capacity density, public service facility coverage rate, and community vitality index are calculated every corresponding time step. In the environmental carrying capacity dimension channel, carbon emission intensity, microclimate comfort, and green space accessibility are calculated every corresponding time step. The set of indicator values ​​output by each simulation calculation channel in chronological order constitutes the simulation indicator sequence for that dimension. The simulation indicator sequences for the three dimensions—spatial efficiency, social carrying capacity, and environmental carrying capacity—maintain synchronization on the time axis.

[0030] S32. Detect abnormal deviations in the inference index sequences of the same scheme across different dimensions, and mark the indicators exhibiting abnormal deviations and their respective dimensions, including: A normal fluctuation baseline is pre-set for the inferred indicator sequence of each dimension. This normal fluctuation baseline is generated based on the statistical analysis of historical inferred data from similar updated areas, or it is converted from the recommended value range in the planning specifications. Each value in the inferred indicator sequence is compared point by point with the normal fluctuation baseline. When a value deviates from the normal fluctuation baseline by more than the preset allowable deviation range, the inferred indicator item corresponding to that value is marked as an abnormal deviation indicator, and the simulation time step node where the abnormal deviation indicator appears and its corresponding dimension category are recorded. The allowable deviation range is a numerical range determined based on the recommended value range in the planning specifications or the fluctuation range of historical inferred statistical data from similar areas, and is used to define the normal fluctuation boundary of the inferred indicators. All inferred indicators of all dimensions are continuously traversed throughout the entire inferred time interval to form an abnormal deviation indicator event set.

[0031] S33. Analyze the temporal correlation of abnormal deviation indicators between different dimensions. When there is a causal relationship between the occurrence of abnormal deviation indicators in one dimension and the occurrence of abnormal deviation indicators in another dimension in time, the causal relationship is identified as a contradiction transmission relationship between dimensions. Specifically, the events in the set of abnormal deviation indicator events are sorted by their occurrence time. A sliding time window method is used to check whether abnormal deviation indicator events belonging to two different dimensions occur within a preset time window. If an abnormal deviation indicator event of the first dimension occurs first and an abnormal deviation indicator event of the second dimension occurs later within the time window, the causal relationship discrimination rule base is further invoked. This causal relationship discrimination rule base pre-stores causal relationship templates between dimensions derived from domain knowledge. For example, an abnormally high space utilization rate in the spatial efficiency dimension followed by an abnormally high population carrying capacity in the social carrying capacity dimension within the extrapolation period is identified as a valid causal chain. If the dimension to which the abnormal deviation indicator event occurs first and the dimension to which the abnormal deviation indicator event occurs later have corresponding causal chain templates in the rule base, then a causal relationship is determined to exist between the two, and this causal relationship is identified as a set of contradictory transmission relationships between dimensions.

[0032] S34. The process of extracting the dimension categories, transmission directions, and transmission strengths involved in the contradiction transmission relationships between dimensions is as follows: For each identified set of contradictory transmission relationships between dimensions, the dimension in which the abnormal deviation index occurs first is recorded as the transmission source dimension, and the dimension in which the abnormal deviation index occurs later is recorded as the transmission target dimension. Together, they constitute the dimension category information. The transmission direction is from the transmission source dimension to the transmission target dimension. The transmission strength is quantified based on the relative magnitude of the deviation of the abnormal deviation index value of the target dimension and the deviation of the abnormal deviation index value of the source dimension. The dimension category, transmission direction, and transmission strength are structured and encapsulated into contradictory transmission relationship data units.

[0033] In this embodiment of the invention, step S4 quantifies the degree of comprehensive contradiction exposed in the deduction of each scheme based on the contradiction transmission relationship between dimensions, eliminates schemes whose comprehensive contradiction degree exceeds a preset contradiction threshold, and retains the remaining schemes to form a candidate scheme set. The detailed implementation steps include: S41. The process of obtaining the contradiction transmission relationships between all dimensions identified in the simulation of each scheme is as follows: After the simulation process is completed, a record table of contradiction transmission relationships is established for the current simulation scheme. All inter-dimensional contradiction transmission relationship events identified by the inter-dimensional causal relationship discrimination rule base within the simulation period are traversed. For each marked contradiction transmission relationship event, its complete attribute field information is extracted, including the name of the transmission source dimension, the name of the transmission target dimension, the transmission direction identifier, the transmission intensity value, and the simulation time step interval corresponding to the occurrence of the contradiction transmission relationship. Each extracted contradiction transmission relationship event is appended to the record table as an independent data entry. After traversing all inter-dimensional causal relationship judgment results for the scheme throughout the entire simulation period, the set of entries summarized in the record table represents all inter-dimensional contradiction transmission relationships identified by the scheme.

[0034] S42. Count the total number of contradictory transmission relationships between the dimensions involved in each scheme, and the transmission strength of each contradictory transmission relationship; Specifically, the contradiction transmission relationship record table is read, and the total number of data entries in the table is counted. The count result is the total number of contradiction transmission relationships between dimensions involved in the scheme. At the same time, the transmission intensity value field stored in each entry in the record table is read in turn, and the transmission intensity value of each entry is retained separately and bound to the scheme for storage, forming a list of transmission intensity values ​​corresponding to the scheme.

[0035] S43. Quantify the total number and intensity of contradictory transmission relationships to generate the overall degree of contradiction for the scheme, specifically: The total number of contradiction transmission relationships and the corresponding transmission intensity value list of the scheme are used as input; the sum of all values ​​in the transmission intensity value list is calculated to obtain the total transmission intensity value; the total number of contradiction transmission relationships and the total transmission intensity value are converted into dimensionless relative scores through a preset normalization mapping curve; the relative scores of the quantity dimension and the relative scores of the intensity dimension are weighted and summed, where the weights of the quantity dimension and the intensity dimension are preset constants; the result of the weighted summation is the comprehensive contradiction degree quantification value of the scheme; the higher the comprehensive contradiction degree quantification value, the more significant the inherent contradictions between dimensions exposed in the deduction of the scheme.

[0036] S44. Compare the overall degree of contradiction with a preset contradiction threshold, and eliminate schemes whose overall degree of contradiction exceeds the preset contradiction threshold from the pool of schemes to be simulated. The contradiction threshold is a pre-set quantitative reference value for the overall degree of contradiction. The contradiction threshold is used to define the dividing line between acceptable and unacceptable contradiction levels among multiple alternative schemes. When the quantitative value of the overall degree of contradiction of a certain scheme exceeds the preset contradiction threshold, it indicates that the inherent contradictions exposed by the scheme in the in-depth simulation have exceeded the limits that can be tolerated in project management. Continuing to retain the scheme will increase the risk in the subsequent implementation stage. Conversely, if the overall degree of contradiction is lower than the contradiction threshold, the scheme can be regarded as a candidate scheme with controllable contradictions in the simulation and enter the next processing stage. S45. After eliminating all the remaining schemes, form a candidate scheme set, and sort and output the comprehensive contradiction degree of each scheme in the candidate scheme set from low to high.

[0037] In this embodiment of the invention, step S5 uses the inter-dimensional contradiction transmission relationship as a deduced contradiction signal and feeds it back to the update scheme library. The specific implementation of the generation strategy for optimizing the update scheme based on the deduced contradiction signal is as follows: S51. The interdimensional contradiction transmission relationship exposed by each candidate scheme in the simulation, as well as the spatial location, temporal stage and resource type involved, are extracted in a structured manner as the contradiction signal of the simulation. Specifically, for each retained scheme in the candidate scheme set, the contradiction transmission relationship record table generated during the simulation phase is analyzed a second time. For each contradiction transmission relationship event in the record table, the state snapshot of the updated area's digital twin is traced back to when the contradiction occurred. Spatial entities associated with the contradiction transmission source dimension and transmission target dimension are located from the state snapshot, and the geographical coordinate range or building unit number corresponding to the spatial entity is extracted as the spatial location information involved in the contradiction. At the same time, the simulation time step interval of the contradiction transmission relationship event is extracted and converted into the specific stage name in the scheme execution sequence arrangement as the temporal stage information involved in the contradiction. The types of resources consumed or competed by both dimensions when contradiction transmission occurs in the two dimensions within the temporal stage are further analyzed, and the resource type names are recorded as the resource type information involved in the contradiction. The contradiction transmission relationship between each set of dimensions, along with its attached spatial location information, temporal stage information, and resource type information, are encapsulated into a structured simulation contradiction signal data unit. All simulation contradiction signal data units of all candidate schemes constitute the simulation contradiction signal set.

[0038] S52. Feedback the deduced contradiction signals to the generation strategy module of the updated solution library, and adjust the constraint parameters of the generation strategy for the updated solution, including: The set of inferred conflict signals generated in the previous step is sent to the generation strategy module associated with the update solution library. The generation strategy module is the core functional component associated with the update solution library. Internally, it encapsulates rules for generating spatial boundaries of solutions, temporal arrangement rules, and resource requirement estimation logic. It automatically generates new alternative update solutions after receiving update requirements and constraints. After receiving the set of inferred conflict signals, the generation strategy module performs statistical analysis on the signal content, identifying frequently occurring conflict spatial regions, frequently conflicting temporal phase types, and frequently contested resource types. Based on the statistical analysis results, the generation strategy module adjusts the constraint parameters in its internally maintained update solution generation strategy. For example, it increases the exclusivity weight of solutions in high-incidence conflict spatial regions, shortens the upper limit of the number of activities allowed to be executed in parallel within high-conflict temporal phases, or increases the resource buffer coefficient for highly contested resource types. After the constraint parameters are adjusted, the subsequent solution generation behavior of the generation strategy module will be limited by the new set of constraint parameters. Among them, the adjustment of constraint parameters includes: modifying the boundary generation rules of the scheme spatial range to avoid spatial overlap areas, and optimizing the arrangement rules of the scheme execution sequence to stagger the resource competition sequence. The specific operation is as follows: the boundary generation rules of the scheme's spatial range are modified; the generation strategy module reads the high-incidence area of ​​spatial overlap in the inference contradiction signal and adds avoidance polygon constraints to the boundary generation rules of the scheme's spatial range; when new alternative updated schemes are generated in the future, the spatial range boundary of the scheme will actively bypass the spatial area covered by the avoidance polygon constraints during the automatic division process, thereby reducing the degree of spatial overlap between different schemes at the source of scheme design. The execution sequence arrangement rules of the scheme are optimized; the strategy generation module adjusts the parallelism determination logic of each activity in the sequence arrangement rules based on the distribution of high resource competition time sequence stages statistically analyzed in the inference conflict signals. That is, for resource types that frequently appear in the competition relationship, the conditions for simultaneous execution of multiple activities involving the resource in the arrangement rules are tightened, and they are forced to be staggered on the time axis to reduce the length of overlapping time periods and the possibility of resource competition conflicts in the execution sequence of subsequent schemes.

[0039] S53. Generate new alternative update schemes based on the generation strategy after the constraint parameters are adjusted, add them to the update scheme library, and form a closed loop of two-way iteration between scheme generation and simulation deduction. Understandably, the closed loop of bidirectional iteration stops iterating when any of the following termination conditions are met: there is a solution in the candidate solution set whose overall contradiction degree is lower than the preset excellent threshold; or the number of iteration rounds reaches the preset maximum number of iterations; or the distribution of the overall contradiction degree of the candidate solution set tends to converge in two adjacent iteration rounds. In the process of this method, scheme generation and simulation deduction constitute a cyclical iterative framework. Each iteration includes a complete set of steps such as generating a batch of new candidate schemes using the current generation strategy, pre-screening the contradictions between schemes, performing in-depth deduction and calculating the comprehensive contradiction degree of each scheme, eliminating schemes according to the contradiction threshold and outputting the candidate scheme set, and extracting the deduction contradiction signal to adjust the generation strategy. At the end of each iteration, the termination condition is checked. The first termination condition is that there is a solution in the candidate solution set whose overall contradiction level is lower than the preset excellent threshold. The preset excellent threshold is a reference value for the overall contradiction level. When the overall contradiction level of a solution is lower than the excellent threshold, it indicates that the number of contradiction transmission relationships between dimensions of the solution is very small and the transmission intensity is weak. The solution has high overall internal consistency and has reached the maturity level that can be recommended for implementation. At this time, the iteration goal has been achieved, the iteration stops and the solution is output as the preferred solution. The second termination condition is that the number of iterations reaches the preset maximum number of iterations; the preset maximum number of iterations is a pre-set positive integer used to prevent the iteration process from running indefinitely; when the number of iterations reaches this value and no solution that meets the excellent threshold has yet appeared, the solution with the lowest degree of comprehensive contradiction in the current candidate solution set is output as the second best solution and the iteration stops. The third termination condition is that the distribution of the overall contradiction degree of the candidate solution set tends to converge in two adjacent iterations; compare the statistical distribution characteristics of the overall contradiction degree of each solution in the current round of candidate solutions with the statistical distribution characteristics of the previous round; if the difference measure value is lower than the preset convergence judgment limit for several consecutive rounds, it indicates that it is difficult to further reduce the overall contradiction degree of the solution in subsequent iterations, the iteration has entered the convergence plateau period, and the iteration stops and the current optimal solution is output; the above three termination conditions are jointly controlled by logical OR relation to automatically end the iteration process.

[0040] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

[0041] It should be noted that all formulas in this manual are calculated by removing dimensions and taking their numerical values. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters and thresholds in the formulas are set by those skilled in the art according to the actual situation.

[0042] Although embodiments of the invention have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.

Claims

1. A method for generating and simulating urban renewal projects based on AI and digital twins, characterized in that, include: S1. Obtain multi-source data of the area to be updated, construct a digital twin of the area to be updated, and obtain an update scheme library generated based on the update requirements, wherein the update scheme library contains multiple alternative update schemes; S2. Identify and judge the potential conflict relationships between each candidate update scheme in the update scheme library, and select candidate update schemes whose conflict relationships between schemes meet the preset level threshold to form a set of schemes to be deduced. S3. Load each scheme in the set of schemes to be simulated into the updated digital twin of the area in sequence for multi-dimensional in-depth simulation, and identify and extract the contradiction transmission relationship between dimensions from the simulation results; S4. Based on the contradiction transmission relationship between dimensions, the degree of comprehensive contradiction exposed by each scheme in the simulation is quantified, and the schemes with a comprehensive contradiction degree exceeding the preset contradiction threshold are eliminated, and the remaining schemes are retained to form a candidate scheme set. S5. The contradiction transmission relationship between dimensions is used as a deduced contradiction signal and fed back to the update scheme library. The generation strategy of the update scheme is optimized based on the deduced contradiction signal.

2. The method for generating and simulating urban renewal projects based on AI and digital twins according to claim 1, characterized in that, S1 includes: Acquire multi-source data for the area to be updated, namely current geographic information data, existing building BIM data, infrastructure pipeline data, population distribution data, and traffic flow data; among which, the area to be updated is the urban renewal geographic area targeted for acquiring multi-source data; By fusing multi-source data, an updated digital twin of the area is constructed, which includes the area's physical space and dynamic operational status. Obtain the update scheme library generated based on update requirements and update constraints.

3. The method for generating and simulating urban renewal projects based on AI and digital twins according to claim 2, characterized in that, In the update scheme library, each candidate update scheme and the digital twin of the update area maintain a unified benchmark in the spatial coordinate system, and the spatial range information of each candidate update scheme is recorded in the form of spatial coordinate boundaries.

4. The method for generating and simulating urban renewal projects based on AI and digital twins according to claim 1, characterized in that, S2 includes: Analyze the spatial range information of each alternative update scheme, and identify scheme pairs with overlapping or adjacent spatial ranges within the physical space of the digital twin of the update area; For the identified scheme pairs, based on the execution time sequence, analyze whether there are overlapping periods in the execution time of the scheme pairs; at the same time, based on the expected types and quantities of resources to be invested, analyze whether there are competing relationships in terms of resource occupancy types of the scheme pairs; based on the analysis process, obtain the spatial overlap degree, the length of the time overlap period, and the number of resource competition types of the scheme pairs as the judgment factors characterizing the conflict level. Based on the judgment factors that characterize the level of contradiction, the potential contradiction relationship level between the scheme pairs is comprehensively determined. The alternative update schemes with potential conflict levels below a preset threshold are selected and retained to form a set of schemes to be deduced.

5. The method for generating and simulating urban renewal projects based on AI and digital twins according to claim 4, characterized in that, The methods for determining the level of potential conflict relationships include: Obtain the percentage of overlapping area between the two schemes as the degree of spatial overlap; Obtain the number of overlapping days in the execution time sequence between the scheme pairs, and use it as the length of the time sequence overlap period; The number of identical resource types among the expected resource types to be invested in the two schemes is used as the number of resource competition types. By combining the degree of spatial overlap, the length of time of temporal overlap, and the number of resource competition types with preset corresponding weights, a potential conflict relationship level between scheme pairs is generated.

6. The method for generating and simulating urban renewal projects based on AI and digital twins according to claim 1, characterized in that, S3 includes: Each scheme in the set of schemes to be simulated is loaded into the updated digital twin of the area in sequence, and in-depth simulations are carried out in the dimensions of spatial efficiency, social carrying capacity, and environmental carrying capacity, generating a sequence of simulation indicators for each dimension. Anomaly detection is performed on the inference index sequence of the same scheme in different dimensions, and the indexes that deviate abnormally and their respective dimensions are marked. By analyzing the temporal correlation of abnormal deviation indicators between different dimensions, when there is a causal relationship between the occurrence of abnormal deviation indicators in one dimension and the occurrence of abnormal deviation indicators in another dimension in time, the causal relationship is identified as a contradiction transmission relationship between dimensions. Extract the dimension categories, transmission directions, and transmission strengths involved in the contradiction transmission relationships between dimensions.

7. The method for generating and simulating urban renewal projects based on AI and digital twins according to claim 6, characterized in that, The spatial efficiency dimension simulation includes simulations of the area's spatial utilization rate, traffic accessibility, and functional mix after the implementation of the renewal plan; the social carrying capacity dimension simulation includes simulations of population carrying density, public service facility coverage, and community vitality index; and the environmental carrying capacity dimension simulation includes simulations of carbon emission intensity, microclimate comfort, and green space accessibility.

8. The method for generating and simulating urban renewal projects based on AI and digital twins according to claim 1, characterized in that, S4 includes: Obtain the contradiction transmission relationships between all dimensions identified during the simulation of each scheme; The total number of contradictory transmission relationships between the dimensions involved in each solution, and the transmission strength of each contradictory transmission relationship, are counted. The total number and intensity of contradiction transmission relationships are comprehensively quantified to generate the overall contradiction degree of the scheme; The overall degree of contradiction is compared with the preset contradiction threshold, and the schemes whose overall degree of contradiction exceeds the preset contradiction threshold are eliminated from the schemes to be simulated. The remaining solutions after elimination form a candidate solution set, and the comprehensive contradiction degree of each solution in the candidate solution set is sorted from low to high and output.

9. The method for generating and simulating urban renewal projects based on AI and digital twins according to claim 1, characterized in that, S5 includes: The contradiction transmission relationship between dimensions exposed by each candidate scheme in the simulation, as well as the spatial location, temporal stage and resource type involved, are extracted in a structured manner as the contradiction signal of the simulation. The deduced contradiction signals are fed back to the generation strategy module of the updated solution library, and the constraint parameters of the generation strategy of the updated solution are adjusted. Among them, the adjustment of constraint parameters includes: modifying the boundary generation rules of the scheme spatial range to avoid spatial overlap areas, and optimizing the arrangement rules of the scheme execution sequence to stagger the resource competition sequence. New alternative updated schemes are generated based on the generation strategy after the constraint parameters are adjusted, and added to the updated scheme library, forming a closed loop of two-way iteration between scheme generation and simulation deduction.

10. The method for generating and simulating urban renewal projects based on AI and digital twins according to claim 9, characterized in that, The closed loop of the bidirectional iteration stops iterating when any of the following termination conditions are met: there is a solution in the candidate solution set whose comprehensive contradiction degree is lower than the preset excellent threshold; or the number of iteration rounds reaches the preset maximum number of iterations. Or the distribution of the overall contradiction degree of the candidate solution set tends to converge in two adjacent iterations.

Images