Intelligent prediction method for dynamic evolution of human habitation type heritage system

Abstract

The invention provides an intelligent prediction method for dynamic evolution of a living type heritage system. The intelligent prediction method comprises the following steps: establishing a heritage material spatial form evolution model, a community multi-subject dynamic interaction model and a culture factor propagation and diffusion model; coupling the heritage material spatial form evolution model, the community multi-subject dynamic interaction model and the culture factor propagation and diffusion model to form a multi-modal coupling model; and finally, predicting a future contextual model based on the multi-modal coupling model. According to the method, time sequence modeling and trend prediction of core elements such as a space structure, a population structure and cultural activities in a human settlement system are realized.

Description

Technical Field

[0001] The present invention belongs to the technical field of human settlement heritage, and particularly relates to an intelligent prediction method for the dynamic evolution of a human settlement heritage system. Background Art

[0002] Human settlement heritage refers to settlements, blocks or building groups that have gradually formed during the historical evolution process, still continue to carry the living function to this day, and maintain the inheritance of living culture. Such heritage has unique dual attributes: it is both a heritage carrying historical memories and cultural values, and an important spatial carrier for the daily life of contemporary residents. Its core characteristics are reflected in the dynamic symbiotic relationship among "people - land - culture", showing the organic integration of historical inheritance and real life. Analyzed from the perspective of systems theory, human settlement heritage can be regarded as a complex adaptive system composed of three core elements: physical space, social groups and cultural traditions. This system not only has the complexity of structure, that is, it covers multi-level, multi-element, multi-scale spatial and social networks; but also shows significant adaptability, and can flexibly adjust its functions and structure under the change of external environment to maintain the continuous survival and cultural continuity of the community. At the same time, this system is also a continuously evolving dynamic system, and under the action of multiple factors such as population replacement, function transformation, policy intervention, etc., it continuously undergoes updates and reconstructions. In recent years, China has continuously increased the intensity of human settlement heritage protection at the policy level, promoting the coordinated advancement of heritage protection and community development. However, in the process of practical operation, it still faces prominent challenges in two dimensions: it is necessary to strictly maintain the authenticity and integrity of the heritage ontology. Currently, the traditional methods for the protection and renewal of human settlement heritage have the following limitations: (1) The evolution process lacks quantitative modeling and prediction mechanisms. Existing research mostly stays at qualitative analysis, lacking systematic modeling methods that can depict the dynamic change process of the human settlement system, and also lacking data-driven evolution trend prediction means, making it difficult to provide effective support for the scientific formulation of protection strategies.

[0003] (2) The ability of data integration and analysis is weak. The data of the human settlement heritage system often has complex characteristics such as cross-space (spatial distribution), cross-time (historical evolution), and cross-dimension (population, industry, culture, etc.). Problems such as fragmented data sources, heterogeneous structures, and inconsistent semantics seriously hinder the development of unified modeling and sequential analysis.

[0004] (3) The evolution driving mechanism is complex and the coupling relationship is unclear. The evolution of the human settlement system is affected by the interweaving of multiple factors, such as policy orientation, population migration, economic structure changes, cultural identity, etc. Traditional analysis methods are difficult to reveal the underlying non-linear coupling relationship and deep driving mechanism.

[0005] (4) The update intervention lacks foresight and scientific support. In the practice of urban renewal and heritage conservation, the lack of predictive analysis of future evolution trends has led to frequent problems such as unreasonable function substitution, resident loss, and spatial landscape distortion, seriously affecting the sustainable vitality of human settlement heritage and the true continuation of cultural value. Summary of the Invention

[0006] The purpose of the present invention is to provide an intelligent prediction method for the dynamic evolution of a human settlement heritage system, aiming to achieve time-series modeling and trend prediction of core elements such as spatial structure, population structure, and cultural activities in the human settlement system in a multi-source heterogeneous data environment. The technical solution adopted is as follows: An intelligent prediction method for the dynamic evolution of a human settlement heritage system includes the following steps: Establish a model for the evolution of the physical space form of the heritage, a dynamic interaction model of multiple agents in the community, and a model for the spread and diffusion of cultural factors; Subsequently, couple the model for the evolution of the physical space form of the heritage, the dynamic interaction model of multiple agents in the community, and the model for the spread and diffusion of cultural factors to form a multi-modal coupling model; Finally, based on the multi-modal coupling model, predict the future scenario pattern.

[0007] Preferably, the specific steps for establishing the model for the evolution of the physical space form of the heritage include: Step 1A: Collect multi-modal data of human settlement heritage; Step 1B: Analyze the physical space form of human settlement heritage; Step 1C: Construct a model for the evolution of the physical space form of the heritage.

[0008] Preferably, Step 1B specifically includes the following steps: Step 1B1: Analyze the macroscopic external space form; Step 1B2: Analyze the microscopic internal space form.

[0009] Preferably, Step 1C specifically includes the following steps: Step 1C1: Establish a macroscopic space form evolution prediction model; Step 1C2: Establish a microscopic space form evolution prediction model; Step 1C3: Couple the macroscopic space form evolution prediction model and the microscopic space form evolution prediction model to form a model for the evolution of the physical space form of the heritage.

[0010] Preferably, the steps for establishing the dynamic interaction model of multiple agents in the community include: Step 2A: Design a specific index system; Step 2B: Collect and fuse multi-source heterogeneous data; Step 2C: Construct and model a dynamic relationship network.

[0011] Preferably, the indicators included in step 2A are: Neighborhood Density Index (NDI), Intangible Cultural Heritage Inheritance Chain Length (TCL), and Local Employment Ratio (LER).

[0012] Preferably, the specific steps for establishing the cultural factor propagation and diffusion model include: Step 3A: Node semantic encoding; Step 3B: Constructing a multi - cultural propagation network; Step 3C: Based on the multi - cultural propagation network, establishing a cultural factor propagation and diffusion model.

[0013] Preferably, the cultural propagation network in step 3B includes: master - apprentice inheritance network, festival participation network, and language diffusion network.

[0014] Preferably, the steps for forming a multi - modal coupling model include: Step 4A: Coupling mechanism design; Step 4B: Multi - modal coupling model structure design; Step 4C: Cross - modal interaction mechanism design.

[0015] Compared with the prior art, the advantages of the present invention are: The human - settlement - type heritage system is a complex adaptive system composed of three major elements: "physical space - social group - cultural activities". The three elements not only have their own independent evolution laws but also jointly shape the overall evolution path of the system through dynamic interactions. Therefore, the present invention first independently models the three subsystems to capture their respective dynamic characteristics and evolution trends; then, through coupling mechanism modeling, it identifies the interaction relationships among the three, thereby constructing a dynamic evolution prediction model for the overall system, that is, a multi - modal coupling model.

[0016] This multi - modal coupling model comprehensively applies multi - source data fusion, spatio - temporal modeling, complex network analysis, and intelligent prediction algorithms such as deep learning and system dynamics, and can simulate the state changes of the human - settlement - type heritage system under specific interventions or natural evolution, providing a scientific basis for policy formulation and intervention. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a technical roadmap for the intelligent prediction method of the dynamic evolution of the human - settlement - type heritage system: Figure 2 It is a flow chart for the intelligent prediction of the evolution of the heritage physical space; Figure 3 It is a schematic diagram of the dynamic relationship network; Figure 4 It is a schematic diagram of parameter optimization; Figure 5 It is a schematic diagram of the multi - modal coupling model design; Figure 6Schematic diagram of the hierarchical modeling strategy for the coupling structure Figure 7 Schematic diagram of the regional grid divided based on GIS Figure 8 Schematic diagram of extracting the regional boundary length in QGIS Figure 9 Schematic diagram of the integrated network module Detailed implementation manners

[0018] The intelligent prediction method for the dynamic evolution of the human settlement heritage system of the present invention will be described in more detail below with reference to the schematic diagrams, in which the preferred embodiments of the present invention are shown. It should be understood that those skilled in the art can modify the present invention described herein while still achieving the advantageous effects of the present invention. Therefore, the following description should be understood as a broad guidance for those skilled in the art and not as a limitation to the present invention.

[0019] An intelligent prediction method for the dynamic evolution of the human settlement heritage system proposed by the present invention, the overall technical framework of which is as Figure 1 shown.

[0020] Based on the human settlement heritage theory, this method constructs a technical system of "four-dimensional analysis - dynamic coupling", including the following core modules: analysis of the evolution of the physical space form of the heritage, analysis of the evolution of the community network structure of the heritage, analysis of the spread and diffusion of cultural factors, and a multi-modal dynamic coupling model.

[0021] Each module is organically integrated through a spatio-temporal graph neural network, and finally forms a dynamic prediction system supporting multi-scenario simulation.

[0022] In the physical space evolution analysis module, a "space - land - history" trinity data acquisition strategy is adopted to construct a systematic spatio-temporal database. This module includes two parts of models: one is to construct a long short-term memory network for prediction aiming at the trends of physical space structure characteristics such as integration degree and connectivity changing with the tourism population pressure; the other is to construct a cellular automaton model to simulate the intensity and speed of its spatial expansion for the external form expansion process.

[0023] In the heritage community network structure evolution analysis module, multi-modal data such as questionnaire survey data and mobile phone signaling are fused, and a quantitative index system conforming to the characteristics of the heritage community is proposed to quantitatively model the community network characteristics. Further, an agent-based modeling method is introduced to simulate the influence mechanism of individual and group behaviors on the evolution of the community structure.

[0024] In the cultural factor dissemination analysis module, combined with the CultureVerse multimodal cultural understanding benchmark dataset, the voice transcription and sentiment analysis technology of oral history is used to encode and express cultural genes; in the dissemination mechanism modeling, a Bass diffusion model considering the spatial decay effect is introduced to depict the spatial diffusion process of cultural factors.

[0025] Finally, a multimodal dynamic coupling model integrating the three subsystems of physical space, community structure, and cultural dissemination is constructed. Through spatio-temporal graph neural networks, deep fusion and dynamic collaboration of cross-domain features are achieved, supporting the simulation and prediction of various future scenario patterns.

[0026] Step 1: Analysis of the morphological evolution of the heritage physical space.

[0027] The morphological evolution of the heritage physical space is a key component of the dynamic evolution of the human settlement heritage system, mainly reflected in changes in building density and form, migration and transformation of spatial functional areas, evolution of the block spatial pattern, and retention, reconstruction, or disappearance of the landscape structure. By introducing space syntax theory and systematically analyzing the spatial accessibility and connectivity of the block, the feedback mechanism of human activities on the spatial structure can be effectively revealed, and key spatial structure indicators such as integration and connectivity can be quantitatively integrated, thus reflecting the vitality level and dynamic evolution characteristics of the heritage space.

[0028] Step 1A: Collection of multimodal data of human settlement heritage.

[0029] When conducting the analysis of the morphological evolution of the heritage physical space in the present invention, a dual analysis perspective combining the macroscopic external space form and the microscopic internal space form is adopted. To meet this analysis requirement, a "space-ground-history" trinity data acquisition scheme is proposed, and modern drones, lidar scanning, high-precision thermal infrared imaging and other technologies are comprehensively used to achieve comprehensive and refined acquisition of the heritage space.

[0030] Step 1A1: Collection of macroscopic external space form data.

[0031] At the macroscopic level, it focuses on the large-scale building form and spatial pattern of the heritage area. The present invention selects the DJI M300 RTK drone equipped with a five-lens oblique photography system as the acquisition device. This device can achieve a ground resolution of 2.5 cm at a flight altitude of 150 meters, and uses the PPK post-differential positioning technology to control the plane precision error within 3 cm. By collecting multi-angle high-resolution images to generate a three-dimensional model, macroscopic spatial information such as the built-up area, average annual expansion area, base period area, and grid side length of the heritage area (macroscopic external space form data) can be obtained, providing high-quality data support for subsequent modeling analysis. The regional grid divided based on GIS is as Figure 7 shown.

[0032] Specifically, the macroscopic external spatial form data includes: the built-up area in the year, time interval average annual expansion area base period area grid side length minimum number of grids covering the heritage boundary at this scale regional planar projected area regional boundary length .

[0033] The extraction of the regional boundary length based on QGIS is shown as Figure 8 shown.

[0034] Step 1A2, collection of microscopic internal spatial form data.

[0035] At the microscopic level, the present invention combines a handheld lidar with an 8K global camera to achieve refined modeling of the interior space of a building. The handheld lidar has a scanning rate of up to 650,000 points / second, capable of efficiently obtaining high-density point cloud data to meet the three-dimensional modeling requirements of complex indoor structures. At the same time, in cooperation with an 8K resolution global camera and a thermal infrared imager with a thermal sensitivity of 0.03°C, it can not only accurately collect building material texture images but also detect thermal anomalies in the building structure to assist in identifying potential structural defects. Through the collaborative collection of multi-source data, a comprehensive reconstruction of the internal space of the heritage area is achieved, forming a high-precision and semantically rich three-dimensional space model (microscopic internal spatial form data).

[0036] Among them, the microscopic internal spatial form data includes: roof form, dougong structure, door and window styles, wall materials, wall damaged areas, building image texture data, three-dimensional point cloud data, historical building image datasets, noise vectors, weathering and lighting condition parameters.

[0037] Step 1A3, data integration and spatio-temporal database construction.

[0038] To achieve efficient management and continuous update of heritage physical space data, the present invention relies on local professional resources, accesses the existing three-dimensional spatial database system, integrates historical materials and real-time collected data, and constructs a unified full spatio-temporal data management platform. This platform supports the storage, retrieval, and analysis of multi-source heterogeneous data, providing a solid data foundation for subsequent dynamic evolution modeling and predictive analysis.

[0039] Step 1B, analysis of the physical space form of human settlement heritage.

[0040] For the systematic processing and analysis of the collected multi-modal data, the present invention constructs a quantitative index system from two levels, namely macroscopic and microscopic, to reveal the dynamic evolution law of the material spatial form of human settlement heritage. At the macroscopic level, it focuses on the spatial expansion characteristics at the regional scale and adopts a four-dimensional index system: expansion speed, expansion intensity, contour fractal dimension, and shape compactness; at the microscopic level, it focuses on the internal characteristics of buildings and uses architectural style and texture features for description.

[0041] Step 1B1, Macroscopic external spatial form analysis. That is, obtain the output data set for model training.

[0042] (1) The land use expansion speed is used to measure the average annual growth rate of the built-up area of the heritage area, reflecting the overall scale and development trend of spatial expansion. Its calculation formula is: (1) Among them, represents the expansion speed, represents the built-up area in the th year, represents the time interval (year); (2) The land use expansion intensity reflects the relative change degree of the average annual expansion amount per unit base period area, depicting the intensity level of the spatial expansion of the heritage area. The calculation formula is: (2) Among them, represents the expansion intensity, represents the average annual expansion area, represents the base period area; (3) The contour fractal dimension is based on the fractal geometry theory to measure the complexity and spatial fragmentation of the boundary form of the heritage area. The grid counting method is often used for estimation, and the specific formula is as follows: (3) Among them, represents the contour fractal dimension, represents the grid side length, represents the minimum number of grids covering the heritage boundary at this scale. The grid side length is not a single fixed value, but a dynamic value to cover multiple scales from coarse to fine.

[0043] (4) The shape compactness index is used to evaluate the geometric compactness of the shape of the heritage area, usually reflecting the spatial utilization efficiency and development stage characteristics. Its expression is: (4) Among them represents the compactness index, represents the area of the regional plane projection, Indicates the length of the regional boundary. In the actual process, it can be read from Figure 8 the area and length of the regional boundary.

[0044] Step 1B2, Microscopic internal space form analysis. That is, obtaining the output dataset for model training.

[0045] (1) Architectural style analysis: Extract typical components and architectural style elements within the heritage area through the roof form, dougong structure, window and door styles, wall materials, and damaged wall areas, and identify their style characteristics. Combining deep learning and image recognition technologies such as the CNN network can achieve automatic recognition of architectural styles and prediction of evolution trends.

[0046] (2) Architectural texture analysis: Use the gray-level co-occurrence matrix (GLCM) to extract texture feature parameters such as contrast, energy, and entropy from the architectural image texture data, and combine machine learning algorithms such as random forest and SVM to achieve quantitative expression of the architectural surface material texture and time-series evolution modeling.

[0047] The output dataset for training the microscopic space form evolution prediction model includes architectural style features and architectural texture features.

[0048] Step 1C, Construction of the heritage material space form evolution model.

[0049] Through the quantitative analysis of the heritage material space form, construct a multi-level prediction model to achieve accurate prediction of the future space form evolution trend, providing a scientific basis and technical support for heritage protection and renewal.

[0050] Step 1C1, Macroscopic space form evolution prediction. That is, establishing a macroscopic space form evolution prediction model.

[0051] The spatial evolution at the macroscopic level usually manifests as the dynamic changes of time-series indicators such as spatial expansion speed, expansion intensity, contour fractal dimension, and shape compactness. These indicators have obvious time correlation and non-linear change characteristics, and are suitable for using time series modeling methods.

[0052] The long short-term memory network has good long-term dependence modeling ability, and can effectively capture the historical trajectory and trend characteristics of the heritage space form evolution. At the same time, introducing the attention mechanism can automatically identify and focus on the historical moment that is most critical to the current prediction, thereby enhancing the model's perception ability of key evolution nodes and overcoming the "equal weight distribution" problem in the traditional LSTM's historical information processing.

[0053] Based on this, adopting the LSTM-ATT model that combines LSTM and Attention mechanisms can accurately predict the core indicators such as the future spatial expansion speed, expansion intensity, and morphological complexity of the heritage area.

[0054] In summary, the macroscopic spatial form evolution prediction model, that is, the trained LSTM-ATT model, takes the macroscopic external spatial form data collected in step 1A1 as its input, and its output is the expansion speed, expansion intensity, contour fractal dimension, and shape compactness.

[0055] Step 1C2, Microscopic spatial form evolution prediction. That is, establish a microscopic spatial form evolution prediction model.

[0056] (1) Architectural style evolution prediction. That is, establish an architectural style evolution prediction model.

[0057] As a visual and culturally recognizable spatial feature, the evolution of architectural style is affected by various factors such as historical periods, cultural contexts, material aging, and climate factors. Style prediction is essentially a comprehensive problem of image style generation and classification.

[0058] This model integrates three network modules: StyleGAN2, StyleNet3D, and EfficientNetB7. This model is also called an integrated network module. The integrated network module is as Figure 9 shown.

[0059] StyleGAN2 is used to generate architectural images with temporal evolution features, simulating natural or human impacts such as weathering and lighting changes; the input is a historical architectural image dataset, a noise vector, and weathering and lighting condition parameters, and the output is an architectural image with temporal evolution features.

[0060] StyleNet3D is used to extract architectural style features from the 3D point cloud data in step 1A2 and model its evolution process; the input is the 3D point cloud data, component segmentation labels such as roofs and brackets. The output is a 3D style feature vector.

[0061] EfficientNetB7 is used to perform fine-tuning of top-level parameters based on image feature transfer learning to further improve the accuracy of style classification and evolution prediction. The input is the architectural image with temporal evolution features generated by StyleGAN2 and the 3D style feature vector generated by StyleNet3D, and the output is the fineness style classification result.

[0062] Through the above multi-model integration strategy, the model's perception and prediction ability of the architectural style evolution trend are significantly enhanced.

[0063] The architectural style evolution prediction model is the trained integrated network module.

[0064] (2) Architectural texture evolution prediction. That is, establish an architectural texture evolution prediction model.

[0065] The evolution of building surface textures is usually driven by multiple factors such as natural weathering, artificial restoration, and functional reuse, presenting significant temporal and structural change characteristics. A series of quantifiable texture indices such as contrast, homogeneity, energy, and entropy can be extracted through the Gray-Level Co-Occurrence Matrix (GLCM) to reflect the detailed changes in the building surface materials.

[0066] Based on the multi-temporal building image texture data in step 1A2, this paper constructs a GLCM-RF (Gray-Level Co-Occurrence Matrix - Random Forest) prediction model to deeply explore the evolution law of building surface textures, and then realizes the regression prediction of texture features in future time periods. The prediction results of this model can be further used to simulate the aging process of building materials, the change of structural wear, and the decline trend of visual appearance, providing technical support for the reconstruction of heritage appearance and visual restoration.

[0067] The building texture evolution prediction model, that is, the trained GLCM-RF prediction model, takes the building image texture data as the input and the building texture features in future time periods as the output.

[0068] In summary, the micro spatial form evolution prediction model includes the building style evolution prediction model and the building texture evolution prediction model.

[0069] Step 1C3, the implementation plan of the comprehensive prediction system. That is, coupling the macro spatial form evolution prediction model and the micro spatial form evolution prediction model to form a heritage physical space form evolution model.

[0070] This system adopts a "macro-micro" collaborative modeling strategy to construct an integrated heritage physical space evolution prediction system. The macro model focuses on the analysis of the spatial structure and form evolution trend, and the micro model focuses on the dynamic change modeling of the building appearance and material details. The two are unified and integrated through a coupling module to achieve the global collaborative prediction of spatial scale and appearance level.

[0071] This system has good interpretability, predictability, and visualization capabilities, and can be widely applied to scenarios such as the formulation of heritage protection strategies, the study of evolution trends, and simulation restoration, providing technical support for intelligent and scientific heritage management decisions. The overall framework of the intelligent prediction process of heritage physical space evolution is as Figure 2As shown below. First, input data and perform preprocessing. Spatiotemporal alignment is used to align the preprocessed data in terms of time and space, ensuring that data from different sources can be accurately matched and correlated in terms of time and geographical location. Feature engineering extracts, selects, transforms, and creates new features from the aligned data to improve the prediction ability and performance of the model. The macro spatial form prediction module uses the data processed by feature engineering to predict the spatial form at the macro level. The micro spatial form prediction runs in parallel with the macro prediction and also uses the data processed by feature engineering to predict the spatial form at the micro level. The results of the macro spatial form prediction and the micro spatial form prediction are integrated and fused to obtain a more comprehensive and accurate final prediction result. Finally, the fused prediction result is presented in the form of 3D visualization for easy understanding and analysis by users.

[0072] The overall system adopts a modular + microservices architecture, encapsulating various prediction models into independent microservice modules to achieve high scalability, easy deployment, and flexible upgradability of the system.

[0073] At the data layer, the system stores raw data such as images, point clouds, and gray-level co-occurrence matrices based on distributed file systems such as HDFS and MinIO. At the same time, it uses a spatio-temporal composite database such as PostGIS + TimescaleDB to manage historical and real-time dynamic data, ensuring the timeliness and spatial consistency of data access.

[0074] In the model service layer, each prediction model is encapsulated in the form of a microservice, supporting service calls through RESTful or gRPC interfaces, specifically including: Microservice A is an LSTM-ATT time series prediction module. It inputs macro external spatial form data, uses the LSTM network to model the long-term evolution trend, and combines the attention mechanism to identify key time nodes, outputting the predicted values of parameters such as the contour fractal dimension and compactness index for several future time periods.

[0075] Microservice B is a building style generation and classification module. It uses StyleNet3D + StyleGAN2 to model the historical building style features and generate images, simulating style evolution and weathering effects; and combines the EfficientNet network to identify the style and classify the weathering stage of the generated images, improving the visualization ability and accuracy of the model. The inputs are historical building images, 3D point cloud data, control parameters such as weathering degree and lighting angle, 3D point cloud data, and component segmentation labels such as roofs and brackets. The output is the classification result of the building style, such as Qing Dynasty official style.

[0076] The microservice C is a texture evolution regression and synthesis module that uses GLCM to extract building surface texture features (such as contrast, entropy, energy, etc.), performs regression prediction on the future evolution trend of texture features based on a random forest model, and synthesizes aging effect images at the same time to achieve the time-series simulation of the visual decline of building materials. The input is the high-resolution surface images collected from the historical heritage community and the time series of historical texture parameters, and the output is the synthesized aging effect images and texture evolution prediction values Finally, the material space features are formed by integrating the macroscopic form prediction results with the evolution trends of microscopic style and texture, and the unified space situation simulation map is constructed based on the material space features

[0077] The system uses the output of all microservices, such as the macroscopic prediction of microservice A, the style probability of microservice B, and the texture parameters of microservice C, as inputs to model using the graph neural network GNN. The nodes are morphological indicators, style labels, and texture features; the edges are dynamic association strengths, such as the probability of compactness decline or texture entropy increase, and the output is the heritage space situation simulation map, mining the dynamic association path between "form-style-texture" and revealing its implicit evolution logic chain Step 2: Analysis of the evolution of the heritage community network structure

[0078] Based on the theory of Complex Adaptive Systems (CAS), the present invention proposes a three-dimensional analysis framework of "environment-individual-interaction" to systematically model the dynamic evolution process of multiple agents in the heritage community at the levels of social relations, cultural inheritance, and economic activities. Through the integrated modeling of multi-source heterogeneous data, the evolution mechanism of the community network structure is revealed, providing a scientific basis for understanding and intervening in group behaviors in cultural heritage protection

[0079] Step 2A: Design of specific index system

[0080] To quantitatively characterize the evolution characteristics of the heritage community network structure, three categories of core index systems are constructed, which are developed from three dimensions: community structure, cultural inheritance, and economic resilience

[0081] (1) The Neighborhood Density Index (NDI) is used to quantify the spatial social density of residents in the heritage community and reflect the support ability of the physical space layout for social interaction. Its definition is as follows (5) Among them represents the social connection strength between residents and The social connection strength between represents the residential space distance of residents represents the distance decay function represents the effective social interaction area

[0082] (2) The length of the intangible cultural heritage inheritance chain (TCL) is used to measure the length of the intergenerational transmission experienced by a certain intangible cultural heritage from the source inheritor to the contemporary practitioner. The greater the intergenerational span, the deeper the cultural accumulation. Its definition is as follows: (6) Among them, represents the maximum number of generations traced back, represents the inheritance effectiveness of the nth generation, represents the weight of this generation.

[0083] (3) The proportion of local employment (LER) reflects the degree of living continuation of the traditional production and lifestyle in the heritage area. The definition is as follows: (7) Among them, represents the number of local residents engaged in traditional-related work, represents the total employed population within the heritage protection area, which is obtained from the data collected through the following social questionnaire surveys and other methods.

[0084] The neighborhood density index is used to calculate the real-time spatial interaction distance in step 2C and dynamically update the weights of the adjacency matrix.

[0085] The length of the intangible cultural heritage inheritance chain is used to construct the inheritance relationship map in step 2C.

[0086] The proportion of local employment is used for the social relationship statement in step 2C to quantify the strength of social connections.

[0087] Step 2B: Multi-source heterogeneous data collection and fusion.

[0088] The key subjects in the heritage community include community residents and tourists. There are differences and tensions between the two in terms of economic interests, values, and development demands, constituting a complex and dynamically evolving social network system. To support the modeling and evolutionary analysis of the above characteristic indicators, it is necessary to construct a multi-scale and multi-level population behavior data system, including the following two core parts: Resident data collection: Obtain the social network structure of residents through structured methods (such as social relationship questionnaire surveys) for modeling the social relationships and interaction patterns among residents.

[0089] Tourist dataset: Use means such as scenic spot ticket scanning records and camera video analysis (such as pedestrian flow statistics and heat maps) to capture the spatial movement behavior and aggregation dynamics of tourists at the heritage site, reflecting the visiting patterns and potential impacts of tourists on the heritage site.

[0090] Semantic data integration: Integrate semantic attributes such as the cultural identity tags, intangible heritage participation, and spatial cognitive maps of residents in the resident data to construct a social relationship database with dual dimensions of time and space. This database serves as the core support for the modeling and evolution prediction of the heritage community network, enabling the analysis of the evolution process from the "individual - relationship - group" level.

[0091] The data in Step 2B, such as the resident dataset, tourist dataset, etc., are used to capture the spatio-temporal trajectories and face-to-face contact records in the dynamic relationship network of Step 2C.

[0092] Step 2C, construction and modeling of the dynamic relationship network. That is, construct a dynamic interaction model for multiple agents in the community.

[0093] 1. Composition of the model framework This invention is based on agent-based modeling (ABM) technology to realize the simulation of the dynamic interaction process of multiple types of agents in the heritage community.

[0094] The model regards key participants such as residents, tourists, intangible heritage inheritors, and policymakers as "intelligent agents" with the ability of autonomous decision-making. They make decisions and interact in the heritage space according to their respective behavior rules, such as economic incentives, cultural preferences, and social willingness, forming an evolvable dynamic relationship network.

[0095] The modeling framework mainly consists of three parts: the environment layer, the individual layer, and the interaction layer.

[0096] The environment layer describes the spatial structure basis of the heritage site, covering spatial topological relationships, building function distributions, traffic accessibility networks, layouts of intangible heritage inheritance points, and spatio-temporal semantic layers of historical evolution, etc., to construct the spatial context for agent activities.

[0097] The individual layer mainly includes three types of agents: community residents, tourists, and intangible heritage inheritors. Each type of agent has autonomous attributes and behavior strategies, and their behaviors are driven by multiple factors such as cultural identity, social bonds, and economic returns.

[0098] The interaction layer simulates the social relationship network among agents, the propagation path of cultural elements, and the response behaviors of individuals to the spatial environment, capturing the process of social network evolution and cultural diffusion. The structure of this dynamic network is as Figure 3As shown. Signaling location mainly relies on various signaling data generated by the mobile communication network when users use their mobile phones. These data were originally used for network management and communication, but through specific technologies and algorithms, the geographical location information of users can be extracted from them to generate or update spatio-temporal trajectories. These trajectories represent the movement paths of entities in time and space. Then this spatio-temporal trajectory information will be transmitted and integrated into the relational graph. Questionnaire data is mainly collected by designing a series of questions for respondents. After being processed, the questionnaire data is used to generate social relationship statements. Sensor data refers to the real-time or quasi-real-time perception of the environment or object state through physical sensor devices. For example, through smart bracelets and GPS trajectory recorders, the activity patterns and spatial behaviors of residents are captured in real time, and are used to identify and record face-to-face contact events between entities (face-to-face contact records), and these records reveal the close physical interactions. After the above spatio-temporal trajectories, social relationship statements, and face-to-face contact records are generated, they jointly act on dynamically updating the adjacency matrix, and the adjacency matrix will be continuously adjusted and optimized according to the received real-time information.

[0099] (1) Spatial interaction distance It is constructed based on the geographical distance and neighborhood density index between residents, and the specific formula is defined as follows:

[0100] Where represents resident and The geographical distance between them, represents the neighborhood density index (2) Inheritance relationship graph It is constructed based on the length of the intangible cultural heritage inheritance chain, and the specific formula is defined as follows:

[0101] Where, represents the length of the intangible cultural heritage inheritance chain (3) Social relationship statement It is constructed based on the local employment ratio, and the specific formula is defined as follows:

[0102] Where, represents the local employment ratio of entity i.

[0103] (4) Social connection strength It is constructed based on the social relationship statement, and the specific formula is defined as follows:

[0104] 2. Parameter optimization and model calibration To improve the authenticity of simulation and prediction accuracy, the present invention introduces a parameter optimization and learning mechanism. Through historical survey data and behavioral trajectory data, a prior distribution of behavioral rule parameters is constructed and continuously iteratively optimized during the simulation process. The specific implementation includes: Introduce intelligent algorithms such as particle swarm optimization (PSO) or genetic algorithm (GA) to adjust the key parameters in the behavioral rules; implement multiple rounds of "simulation - verification - correction" mechanism, compare the simulation results with the actual community evolution history, and continuously adjust the model to improve its fitting degree and interpretability; support parameter sensitivity analysis and uncertainty quantification to ensure that the model has a stable response ability when facing different scenarios and policy interventions. The schematic diagram of parameter optimization is as Figure 4 shown. First, the input historical data set is used to drive subsequent parameter search and model optimization. Then, a set of parameters is selected from a predefined parameter space for evaluation, providing an initial exploration range for Bayesian optimization. Bayesian optimization is an efficient global optimization algorithm that constructs a Gaussian process to fit the objective function and uses an acquisition function to determine the next evaluation point, aiming to find the optimal solution with as few iterations as possible. After a set of parameters is generated by Bayesian optimization, these parameters are input into a simulator or model to generate corresponding simulation results. The simulation results are compared and evaluated with the actually observed "true community structure", and the root mean square error (RMSE) is used to measure the deviation between the model prediction value and the true value. When the RMSE is less than 0.1, it indicates that the matching degree between the simulation result and the true community structure is high enough to meet the requirements. When the result of "comparing with the true community structure" meets the condition of RMSE < 0.1, the current parameter set is identified as the optimal parameter set.

[0105] Step 3: Analysis of the spread and diffusion of cultural factors.

[0106] The present invention regards the cultural dissemination process in the human settlement heritage system as a complex adaptive system. Multiple individuals in the system, such as residents, families, masters and apprentices, tourists, etc., interact based on factors such as geographical location, social network, cognitive differences, and cultural absorption willingness, showing system characteristics such as nonlinearity, emergence, and self-organizing evolution.

[0107] Combined with classical cultural dissemination theory and diffusion dynamics model, a three-dimensional modeling method of "structure - semantics - behavior" is constructed. The propagation path is simulated at the network structure layer, the cultural factors are encoded at the semantic layer, and the propagation probability function is defined at the behavior layer, so as to realize the spatio-temporal diffusion modeling of cultural elements and the prediction of future trends.

[0108] This model includes three core construction levels: The node semantic encoding layer is used to structurally and semantically vector-encode cultural factors and their dissemination carriers; The multi - channel dissemination network construction layer is used to construct multiple types of dissemination paths such as "mentor - apprentice relationship network", "festival participation network", "language diffusion network", etc. The diffusion and evolution simulation layer introduces an improved Bass diffusion model to simulate the evolution process of cultural factors in the network, considering dynamic factors such as imitation dissemination and spatial decay.

[0109] Step 3A: Node semantic encoding.

[0110] This model classifies the cultural factors of the heritage community into two categories: explicit factors and implicit factors. Among them, the explicit factors include traditional crafts and festival ceremonies. The implicit factors include values and oral traditions.

[0111] For the traditional crafts among the explicit factors, a "process flow knowledge graph" can be constructed. The nodes represent the process operation steps, the edges represent the operation dependency paths, and the node attributes include the materials used, tools, process difficulty, etc. For the festival ceremonies, a "spatiotemporal behavior pattern matrix" is constructed to encode the participants, time nodes, spatial locations, and behavior sequences.

[0112] For the values and oral traditions among the implicit factors, a dual - coding method of "semantic embedding + pragmatic labels" is adopted. First, BERT is used to perform semantic vector embedding on the text, and then cultural pragmatic attribute labels are introduced to enhance its context expression ability.

[0113] Step 3B: Multi - cultural dissemination network construction.

[0114] To comprehensively depict the cultural dissemination paths and mechanisms in the human - inhabited heritage system, the present invention constructs the following three types of cultural dissemination network layers based on the heterogeneity of cultural subjects and dissemination media: 1. Mentor - apprentice inheritance network.

[0115] The mentor - apprentice inheritance network mainly simulates the "point - to - point" inter - generational inheritance path of intangible cultures such as traditional crafts, operas, and folk skills in the mentor - apprentice relationship.

[0116] Each node represents an individual with an inheritance relationship, and the attributes include the level of the inheritor (national - level, provincial - level, etc.), the type of skill, the active years, and the geographical location. The edge represents the existence of a skill - teaching behavior. The edge weight is defined as the teaching intensity index, and the definition is as follows: (8) Among them, represents the teaching years, represents the teaching frequency, , represents the adjustment parameter.

[0117] 2. Festival participation network.

[0118] The festival participation network mainly depicts the collective participation behaviors of community residents and festival activities, highlighting their emotional connection and sharing nature.

[0119] The nodes are families, community organizations, or festival organizers. The edges indicate that two nodes have a record of jointly participating in a festival, and the edge weights are defined as follows: (9) Among them, K represents the number of festival types, represents the family 、 's number of joint participations in k festivals. represents the festival culture weight.

[0120] 3. Language diffusion network.

[0121] It is used to simulate the spread and evolution of language cultures such as dialects, slang, and oral traditions in geographical space.

[0122] The nodes represent language geographical units such as townships, streets, and dialect areas, and the attributes include population size, language types, and usage frequencies. The edges represent the spread relationships. The edge weights are defined as follows: (10) Among them, represents the lexical similarity rate, which can be calculated through the Jaccard coefficient or edit distance, represents the population flow ratio, , represents the adjustment parameter.

[0123] Step 3C. Cultural diffusion simulation and model enhancement strategy.

[0124] Based on the above-mentioned spread network, an improved Bass diffusion model is used to simulate the evolutionary diffusion process of cultural factors: (11) Among them, represents the number of adopters of cultural factors up to time ; represents the maximum size of the potential adopter group; represents the innovation adoption coefficient; represents the imitation spread coefficient, represents the spatial spread attenuation factor; represents the geographical distance, represents the simulation space or social network spread resistance.

[0125] Formula (11) is the cultural factor propagation and diffusion model.

[0126] The master-apprentice inheritance network (Formula 8) provides the and in Formula 11. The longer the teaching years, the more systematic the skills, and the higher the imitation and transmission coefficient q. The faster the teaching frequency, the larger the maximum scale of the adopting group will be; Here, the and relationship provided by (Formula 8) in Formula 11 is given:

[0127]

[0128] represents the maximum value of all edge weights in the network, represents the basic scale (statistically obtained from historical data, such as the peak number of people covered by the past spread of the skill) The festival participation network (Formula 9) provides the in Formula 11. The more times of joint celebration, the easier it is to imitate each other, and the imitation and transmission coefficient will be higher.

[0129] Here, the relationship provided by (Formula 9) in Formula 11 is given:

[0130] represents the festival influence adjustment parameter, and the default setting is 0.3.

[0131] The language diffusion network (Formula 10) provides the and in Formula 11. The higher the lexical similarity, the smaller the cultural gap, the weaker the social network transmission resistance, and when the population flow is frequent, the innovation adoption coefficient is high.

[0132] Here, the relationship provided by (Formula 10) in Formula 11 is given:

[0133] To make the performance of the model better, the present invention performs the following three aspects of model enhancement strategies: (1) Introduce the "cultural absorption willingness function" : reflecting the adoption tendency of an individual towards a certain cultural factor at time (2)Construct a heterogeneous propagation kernel function : Dynamically adjust the propagation path and intensity according to the type of cultural factor; (3)Model the cultural loss mechanism: Considering the actual loss mechanisms such as the break in the inheritance line and the migration of inheritors, dynamically update and adjust the propagation scale and the propagation path diagram.

[0134] Step 4: Establish a multimodal coupling model.

[0135] In the above three steps, a model for the evolution of the physical space form of the heritage, a model for the evolution of the community network (a dynamic interaction model of multiple community agents), and a model for the propagation and diffusion of cultural factors are constructed respectively. However, in the human settlement heritage system, these three types of subsystems are interconnected and interact with each other, showing significant coupling. Therefore, the present invention proposes a multimodal coupling modeling method to predict the overall dynamic evolution of the system and can be used to carry out multi-scenario simulation. The schematic diagram of the multimodal coupling model design is as Figure 5 shown. First, input the model for the evolution of the physical space form of the heritage, the dynamic interaction model of multiple community agents, and the cultural propagation and diffusion model. These are coupled through the ST-GNN multimodal spatio-temporal network to effectively associate and integrate these different modalities and spatio-temporal information to understand their interactions and influences. One output of the coupling model is used to dynamically predict the "human settlement heritage system". This model can comprehensively consider physical, social, and cultural factors and predict the future state and trends of the heritage in the human living environment. Another output is to support the "intervention simulation platform simulation". This model can serve as the basis for the simulation platform to test the effects of different intervention measures such as policy adjustments and protection strategies on the dynamic evolution of the heritage system, thereby providing support for decision-making.

[0136] Step 4A: Design the coupling mechanism.

[0137] The present invention constructs a "three-dimensional coupling mechanism". By analyzing the interaction relationships between subsystems, the dynamic coupling paths are clarified: (1)Physical space - social network coupling: The building layout and spatial form directly affect the spatial accessibility and interaction frequency between individuals, thereby driving the evolution of the social network structure within the community. Therefore, a building form - spatial accessibility - social interaction - network structure evolution can be established.

[0138] (2)Social network - cultural propagation coupling: The building layout and spatial form directly affect the spatial accessibility and interaction frequency between individuals, thereby driving the evolution of the social network structure within the community. Therefore, a network centrality - information circulation efficiency - intangible cultural heritage memory transmission - cultural factor diffusion can be established (3) Cultural dissemination - physical space coupling: The practice activities of certain cultural crafts put forward specific requirements for the spatial form, which in turn promotes the adjustment of building functions and layouts, resulting in the reshaping of the physical space. Its action path is as follows. Therefore, a process activity - space use - building transformation - space evolution can be constructed.

[0139] Step 4B. Design of the multi-modal coupling model structure.

[0140] That is, couple the mutual relationships between the three models. For example, if the physical space may be affected by the social network, then a relationship between the physical space and the social network needs to be established. If the social network may be affected by the cultural dissemination network, then a relationship between the social network and the cultural dissemination network needs to be established. The same is true for the relationship between cultural dissemination and the physical space.

[0141] Step 4B1. Modeling framework: ST-GNN multi-modal spatio-temporal network The present invention selects the spatio-temporal graph neural network ST-GNN as the coupling backbone network for the following reasons: Use the graph structure to model irregular spatial units such as blocks and nodes, and support topological modeling to achieve spatial heterogeneity processing; The time series learning LSTM gating mechanism captures the historical evolution trends of multi-modal indicators.

[0142] Finally, the graph attention mechanism (GAT) realizes the dynamic weighting and coupling modeling of the features of different subsystems, enhancing the feature fusion ability.

[0143] Compared with models such as system dynamics (SD) and cellular automata (CA), ST-GNN has advantages in terms of spatial resolution, data fusion ability, and prediction accuracy.

[0144] Step 4B2. Setting of the coupling equation system.

[0145] Regarding the physical space M, the social group structure S, and the cultural dissemination state C as dynamic state variables respectively, construct the following multi-modal coupling evolution equations, namely formula (12):

[0146]

[0147]

[0148] Among them, represents the reaction force effect of the community network model and the cultural dissemination factor diffusion model on the physical space evolution network; represents the diffusion coefficient of the physical space form, which is a set value; Indicates the reconstruction effect of the material space evolution network model and the cultural factor diffusion model on the community network. Indicates the social network adjacency matrix, calculated from formula (13); Indicates the support effect of the material space evolution network model and the community network model on the cultural factor diffusion model.

[0149] Indicates the Laplacian operator of the cultural factor, which is a set value.

[0150] What formula 12 outputs is the future material space layout, social network topology, and cultural dissemination scope.

[0151] The dynamic state variables M, S, and C here respectively refer to the heritage material space evolution model obtained in step 1C, the dynamic interaction model of community multi-agents obtained in step 2C, and the cultural factor diffusion model obtained in step 3C.

[0152] The ST-GNN multimodal spatio-temporal network in step 4B1 is to fit this coupled equation, that is, to fit formula 12, and find the optimal solution, that is 、 、 。

[0153] Step 4C, Cross-modal Interaction Mechanism Design.

[0154] 1. Material Space - Social Network: Use the spatial accessibility matrix to set the initial social network edge weights and initialize the social network adjacency matrix : (13) Among them, Indicates the geographical distance between nodes, is the attenuation coefficient.

[0155] 2. Social Network - Cultural Dissemination: Adjust the cultural dissemination rate based on the network centrality of nodes , used to adjust the function in the cultural dissemination equation: (14) Among them, is the centrality of the central node 。

[0156] 3. Cultural Dissemination - Material Space: Increase the probability of building update based on the activity of cultural crafts , For adjusting the function (15) Wherein, represents the level of technological inheritance, is the fitness of building functions.

[0157] Step 4D. Hierarchical modeling strategy for the coupling structure To clarify the functional positioning and coupling direction among subsystems, the present invention introduces a three-layer coupling modeling idea to clarify the functional positioning of each subsystem. The underlying physical space provides a physical carrier for cultural activities and a spatial platform for interaction with residents. The middle-layer social network forms a conduction channel for the flow of culture and behavior and has the ability of dynamic evolution. The upper-layer cultural genes dominate heritage identity, spatial use intention and system evolution direction. Specifically as Figure 6 shown. Among them, the physical space includes geographical location and building type, which constitute the physical basis and environmental constraints for the operation of the system. The spatial indicators of the physical layer act on the social layer through "spatial constraints". This means that physical attributes such as physical space layout, accessibility, and geographical barriers will limit or affect the formation and development of social relations, thereby shaping the structure of the social network. The network structure in the social layer acts as a "transmission carrier" to transmit information, ideas, practices or "cultural genes" from one node to another. Social relations and network structures are channels and media for cultural transmission. The "gene distribution" in the cultural layer will generate "demand feedback", which in turn affects the spatial indicators of the physical layer. This means that specific cultural preferences, lifestyles or values (i.e., the distribution of cultural genes) will generate demands for the physical space, which may lead to adjustments in spatial form, construction or renovation of infrastructure, and changes in land use patterns.

[0158] The input of the coupling model is the output features of three models, namely physical space features such as the expansion speed of the heritage community, fractal dimension of the contour, synthetic aging effect image, and predicted value of texture evolution, etc., community network features (such as neighborhood density index, length of the intangible cultural heritage inheritance chain, proportion of local employment, etc.), and cultural factor transmission features (such as the number of adopters of cultural factors representing the cut-off time and innovation adoption coefficient, etc.).

[0159] Figure 1 In

[0160] Scenario simulation outputs the evolutionary paths of the heritage system under different scenarios after inputting different scenario variables (such as the growth of the tourism population, policy intervention, infrastructure construction, etc.). For example, it predicts the trend of spatial function reconstruction in the core area within the next decade under the condition of "high tourism pressure + low governance response".

[0161] Step 5: Prediction and scenario simulation of the heritage system of human settlements.

[0162] To achieve the intelligent prediction and regulation simulation of the heritage system of human settlements, based on the multi-modal coupling model, the present invention designs an "integration mechanism of prediction - feedback - visualization", constructs three major functional modules including a scenario generation engine, an ABM multi-agent behavior simulation system, and a three-dimensional visualization platform, and collaboratively supports the analysis of the evolutionary trend of the complex heritage system and the assessment of the situational response ability.

[0163] (I) Scenario generation engine This module is used to construct a system perturbation scenario under multiple external input conditions and realize the prediction of the regulatory effect of macro variables such as policy intervention and environmental changes on the system evolutionary path. The specific design is as follows: 1. Multi-strategy parameterized configuration: Support users to customize input variables, including but not limited to the following scenarios: (1) Adjustment of the intensity of heritage protection policies, such as building renovation restrictions and the scale of investment in cultural relics protection funds; (2) Changes in the resident structure, such as the out-migration rate and changes in the age structure, etc.; (3) Tourism development pressure, such as the annual growth rate of tourists and changes in the frequency of cultural activities, etc.; (4) Promotion of cultural policies, such as the popularization rate of intangible cultural heritage project education.

[0164] 2. Perturbation path injection mechanism: By dynamically adjusting the key parameters in the coupling model, such as ( ), simulate the influence of external input on the internal evolutionary mechanism of the system and drive the multi-path evolution of the system under different constraint conditions.

[0165] (II) ABM multi-agent behavior simulation linkage module This module is used to simulate the behavior decision-making and interaction process of various micro-individuals (residents, tourists, intangible cultural heritage inheritors, policy executors, etc.) in the heritage system and reveal the social feedback mechanism and system emergence characteristics.

[0166] 1. Agent type classification: (1) Resident Agent: Has state variables such as spatial living location, cultural identity level, social circle, etc.; (2) Intangible cultural heritage inheritor Agent: Includes inheritance level, influence, and activity level; (3) Policy implementer Agent: can conduct regulatory intervention, such as capital injection and functional adjustment (4) Tourist Agent: Adjusts behavior trajectory based on the cultural attraction heat map.

[0167] 2. Coupling model empowerment mechanism: The state variables of each agent (such as migration probability, cultural adoption willingness, and dissemination ability) are dynamically output by the ST-GNN model; the social network structure affects the behavior diffusion path, and the cultural heat field affects behavioral preferences; agent behavior in turn acts on the model input, forming a closed-loop feedback system.

[0168] 3. Example of rules of conduct: (1) Residents' willingness to move out: influenced by both living comfort and cultural participation; (2) The inheritor agent selects the recipient of the art: the neighbor nodes with high cultural acceptance are given priority; (3) Tourist Agent Path Selection: It is guided by both the frequency of intangible cultural heritage activities and the popularity of spatial aggregation.

[0169] (3) 3D visualization platform To enable intuitive presentation and interactive analysis of system evolution results, this paper constructs a web-based 3D GIS visualization platform. This platform supports interactive operations such as timeline scrolling, perspective rotation, and switching between multidimensional data layers. The backend, connected to the coupled model, refreshes the simulation scene in real time using GeoJSON format. Key features include:

[0170] 1. System evolution path animation: Dynamic demonstration process showing changes in building density, changes in social network structure and the propagation trajectory of cultural factors.

[0171] 2. Heat map: Calculate the "cultural attraction field" based on the density of cultural factors. High-heat areas are highlighted in red, and low-heat areas are faded. Supports simultaneous comparison with policy intervention events to facilitate analysis of response effects.

[0172] 3. Scenario comparison chart: Supports the "dual-window simultaneous playback" function, presenting the similarities and differences in system evolution under two strategy settings. It can dynamically compare indicators such as resident migration rate, cultural activity, and the spatial distribution density of intangible cultural heritage inheritors over time.

[0173] The above description is merely a preferred embodiment of the present invention and does not limit the present invention in any way. Any person skilled in the art who, without departing from the scope of the present invention, makes any equivalent substitution, modification, or other changes to the technical solution and technical content disclosed in the present invention shall be deemed to be within the scope of the present invention and still fall within the scope of protection of the present invention.

Claims

1. An intelligent prediction method for the dynamic evolution of a human settlement type heritage system, characterized in that, It includes the following steps: Establish an evolution model of the material spatial form of the heritage, a dynamic interaction model of multiple agents in the community, and a cultural factor dissemination and diffusion model; Subsequently, couple the evolution model of the material spatial form of the heritage, the dynamic interaction model of multiple agents in the community, and the cultural factor dissemination and diffusion model to form a multi-modal coupling model; Finally, based on the multi-modal coupling model, predict the future scenario pattern.

2. The intelligent prediction method for the dynamic evolution of the human settlement heritage system according to claim 1, wherein The specific steps for establishing the evolution model of the material spatial form of the heritage include: Step 1A: Collect multi-modal data of the human settlement heritage; Step 1B: Analyze the material spatial form of the human settlement heritage; Step 1C: Construct an evolution model of the material spatial form of the heritage.

3. The intelligent prediction method for the dynamic evolution of the human settlement heritage system according to claim 2, wherein Step 1B specifically includes the following steps: Step 1B1: Analyze the macroscopic external spatial form; Step 1B2: Analyze the microscopic internal spatial form.

4. The intelligent prediction method for the dynamic evolution of the human settlement heritage system according to claim 3, wherein Step 1C specifically includes the following steps: Step 1C1: Establish a macroscopic spatial form evolution prediction model; Step 1C2: Establish a microscopic spatial form evolution prediction model; Step 1C3: Couple the macroscopic spatial form evolution prediction model and the microscopic spatial form evolution prediction model to form an evolution model of the material spatial form of the heritage.

5. The intelligent prediction method for the dynamic evolution of the human settlement heritage system according to claim 1, characterized in that, The steps for establishing a dynamic interaction model of multiple agents in the community include: Step 2A: Design a specific index system; Step 2B: Collect and fuse multi-source heterogeneous data; Step 2C: Construct and model a dynamic relationship network.

6. The intelligent prediction method for the dynamic evolution of the human settlement heritage system according to claim 5, characterized in that The indicators included in Step 2A are: Neighborhood Density Index (NDI), Traditional Cultural Heritage Inheritance Chain Length (TCL), and Local Employment Ratio (LER).

7. The intelligent prediction method for the dynamic evolution of the human settlement heritage system according to claim 1, wherein The specific steps for establishing a cultural factor dissemination and diffusion model include: Step 3A: Node semantic encoding; Step 3B: Construct a multi-cultural dissemination network; Step 3C: Based on the multi-cultural dissemination network, establish a cultural factor dissemination and diffusion model.

8. The intelligent prediction method for the dynamic evolution of the human settlement heritage system according to claim 7, wherein The cultural dissemination networks in Step 3B include: master-apprentice inheritance network, festival participation network, and language diffusion network.

9. The intelligent prediction method for the dynamic evolution of the human settlement heritage system according to claim 7, characterized in that, The steps for forming a multi-modal coupling model include: Step 4A: Design a coupling mechanism; Step 4B: Design the structure of the multi-modal coupling model; Step 4C: Design a cross-modal interaction mechanism.

Images