A digital-narrative-based evaluation system for adaptive protection of vernacular architectural heritage

Abstract

This invention discloses an adaptive conservation assessment system for vernacular architectural heritage based on digital narrative. The system first constructs a digital narrative knowledge graph integrating spatiotemporal attributes, linking intangible cultural information such as oral history and craft memories with a high-precision 3D model. Then, through dynamic binding of multi-source sensor data, it achieves continuous integration and updating of architectural physical state, environmental parameters, and social usage information. Based on this, a risk assessment model coupling physical risk and the degree of damage to the integrity of the cultural narrative is established, outputting a dual risk index of structural and cultural value. Finally, digital twin technology is used to simulate the long-term effects of alternative conservation strategies, supporting adaptive decision-making. This invention realizes a transformation from "static recording" to "dynamic assessment," and from "material protection" to "cultural continuity," enhancing the scientific rigor, predictability, and public participation in the conservation process.

Description

Technical Field

[0001] This invention relates to the field of vernacular architectural heritage protection technology, specifically to a digital narrative-based adaptive protection assessment system for vernacular architectural heritage. Background Technology

[0002] Existing digital recording and modeling technologies widely employ techniques such as 3D laser scanning and oblique photogrammetry, which can efficiently acquire high-precision geometric information of architectural heritage and generate realistic 3D models or Building Information Models (BIM). These technologies focus on the precise "replication" of physical forms. However, existing digital technologies primarily focus on the geometric form and physical state of buildings, while the intangible cultural connotations carried by buildings (such as oral history and community memory) often exist in the form of independent texts and images, failing to establish a structured and semantic connection with the 3D spatial model. This results in a lack of in-depth understanding and assessment of the core cultural value of heritage in conservation efforts, easily falling into the dilemma of "seeing the object but not the person," and making it difficult to effectively balance the maintenance of cultural authenticity in intervention decisions.

[0003] Structural health monitoring technology uses sensor networks to monitor physical parameters of buildings, such as displacement, cracks, vibration, temperature, and humidity, primarily to assess structural safety and material degradation. However, current assessments largely rely on step-by-step manual on-site inspections and reports, creating a static "snapshot." This approach is costly, time-consuming, and significantly lags behind actual changes, failing to achieve continuous and dynamic perception of the state of architectural heritage and its environment. Furthermore, structural safety assessments, materials science assessments, and cultural value assessments are typically conducted separately by personnel from different professional fields, resulting in isolated conclusions. There is a lack of a unified assessment framework that couples physical degradation risks with cultural value decline risks, leading to incomplete risk assessments and unscientific protection priorities.

[0004] Existing cultural heritage information management systems utilize database technology to digitize architectural drawings, archives, and repair records, enabling centralized storage and retrieval of information. However, they lack the ability to simulate and extrapolate the long-term impacts of proposed conservation measures (such as reinforcement, restoration, and functional upgrades) under various possible future scenarios (such as climate change, increased tourism pressure, and functional transformation). Decision-makers are unable to proactively assess the adaptability of different strategies to the physical survival and cultural transmission of buildings, potentially leading to unforeseen negative impacts from conservation interventions or an inability to cope with future challenges—in other words, a lack of "adaptability."

[0005] Existing digital display technologies utilize virtual reality (VR), augmented reality (AR), and other technologies to provide immersive or interactive displays of architectural heritage, primarily for public education and tourism promotion. However, these existing systems are mostly designed as professional tools, making it difficult for local residents, the public, and other heritage stakeholders to effectively participate in the value perception, status feedback, and protection oversight processes. Data flow is typically unidirectional (from professional collection to internal use), failing to form a collaborative closed loop of "data collection - public feedback - strategy optimization."

[0006] Therefore, in order to address the above issues, this paper proposes an adaptive conservation assessment system for vernacular architectural heritage based on digital narrative. Summary of the Invention

[0007] The purpose of this invention is to provide an adaptive conservation assessment system for vernacular architectural heritage based on digital narrative, aiming to build an integrated intelligent system of "perception-narrative-assessment-decision-interaction" to achieve adaptive conservation of vernacular architectural heritage from "static specimens" to "living organisms".

[0008] To achieve the above-mentioned technical effects, the present invention is implemented through the following technical solution: a digital narrative-based adaptive conservation assessment system for vernacular architectural heritage, characterized by comprising the following modules:

[0009] S1. Data Acquisition and Digital Archiving Module: Through field surveys and local participation, oral histories, documents, and ritual records related to architecture are collected, and data on architectural material space and materials are collected. After being structured by the narrative engine, the data is associated with a high-precision 3D model to construct a digital narrative knowledge graph with spatiotemporal correlation attributes.

[0010] S2, Full Life Cycle Dynamic Monitoring and Data Fusion Module: Deploy a sensor network and combine it with periodic remote sensing and survey reports to continuously acquire data on building physical status, environmental parameters and social usage, and dynamically associate and update multi-source monitoring data with the digital narrative knowledge graph;

[0011] S3, Cultural and Physical Risk Coupling Assessment Module: Based on the dynamic monitoring data and digital narrative knowledge graph, it analyzes the impact of changes in physical state on the associated narrative chain, calculates and outputs a comprehensive assessment report that includes both structural risk index and cultural value risk index through the coupling assessment model.

[0012] S4. Strategy Simulation and Decision Support Module: Based on the comprehensive evaluation report, candidate protection strategy schemes are generated from the strategy library, and the long-term implementation effects of each scheme under different future scenarios are simulated in the digital twin of architectural heritage. Visual assistance is provided to decision-makers through multi-objective optimization analysis.

[0013] S5. Feedback Iteration and Collaborative Implementation Module: Collects actual effect data and public feedback after the implementation of protection measures, feeds the data back to the risk assessment model and strategy simulation engine to optimize system parameters, and supports the implementation of protection projects and public cultural education through a hierarchical collaborative platform, forming a data closed loop of "monitoring-assessment-decision-implementation-feedback".

[0014] Furthermore, in S1, the construction of the data acquisition and digital archiving module is as follows:

[0015] S1.1 Material Space Data Acquisition: Using UAV oblique photography and ground 3D laser scanning, centimeter-accurate real-scene 3D models and point cloud data of building complexes and environment are obtained to generate real-scene 3D models or building information models (BIM).

[0016] S1.2 Material data acquisition: Infrared thermal imaging and hyperspectral scanning technology are used to supplement the acquisition of hidden information such as material defects and moisture content;

[0017] S1.3 Non-material narrative data collection and structuring: Through participatory workshops, in-depth interviews, oral history records, and digitization of folk documents, the system collects cultural information related to building construction techniques, life memories, ritual activities, and family changes. By deploying a sensor network, the system monitors physical parameters of buildings such as displacement, cracks, vibration, temperature, and humidity to assess structural safety and material deterioration.

[0018] S1.4. Using natural language processing technology, entity recognition, relation extraction, and sentiment analysis are performed on the collected text and audio transcripts to extract narrative elements such as "craftsman", "ritual", "skill", and "historical event" and label their spatiotemporal attributes, importance, and emotional value.

[0019] S1.5 Constructing a spatiotemporally related digital narrative knowledge graph: In the real-world 3D model, semantic annotation and spatial anchoring are performed on the physical components that carry the core narrative;

[0020] S1.6. Associate the narrative elements extracted in S1.5 with these spatial semantic units to establish a "narrative-space" association network of "who-where-when-what-what / what skills", forming a queryable and inferable digital narrative knowledge graph.

[0021] Furthermore, in S2, the construction of the full lifecycle dynamic monitoring and data fusion module is as follows:

[0022] S2.1 Deploy an Internet of Things (IoT) sensing network: Install displacement, tilt, and crack sensors in key structural parts; install temperature, humidity, light, and vibration sensors in typical spaces to achieve real-time monitoring of building physics and environmental conditions;

[0023] S2.2 Establish a periodic remote sensing and survey report mechanism: Conduct drone aerial photography quarterly or annually to monitor changes in the macro-pattern and surrounding environment; develop a community participation mini-program to encourage residents and managers to upload graphic reports reflecting the daily use of buildings, minor changes, or new cultural activities, forming a social perception data flow;

[0024] S2.3 Spatiotemporal Data Fusion and Correlation Update: Real-time sensor data, periodic remote sensing data, and community report data are uniformly aligned to the spatiotemporal coordinate system of the real-world 3D model and knowledge graph established in Step 1; the state attributes of relevant nodes in the knowledge graph are dynamically updated to achieve dynamic binding of physical data and cultural narrative.

[0025] Furthermore, the construction of the cultural and physical risk coupling assessment module in S3 is as follows:

[0026] S3.1, Dual-path parallel analysis:

[0027] Physical risk analysis: Based on sensor and remote sensing data, the safety risk index of each component is calculated using structural mechanics models and material aging models;

[0028] Narrative integrity analysis: Based on digital narrative knowledge graphs, analyze the degree of impact of current physical state changes on related narrative chains;

[0029] S3.2 Calculation of Coupled Risk Assessment Model: Establish a coupled assessment algorithm, which is expressed as: Comprehensive Risk Index = (Physical Risk Index, Narrative Importance Weight × Narrative Integrity Impairment);

[0030] The output results determine "where is broken" and further determine "where the break poses the greatest threat to cultural heritage"; the output is a heat map superimposed on a 3D model, which simultaneously marks the physical risk level and the cultural risk level, and generates a graded early warning report.

[0031] Furthermore, in S4, the construction of the strategy simulation and decision support module is as follows:

[0032] S4.1 Intelligent Strategy Matching and Generation: The system has a built-in strategy rule engine that integrates the principles of the "Management Measures for Cultural Relics Protection Projects" and narrative protection rules; for high-risk objects identified in S3, the engine automatically matches and combines multiple candidate protection strategy schemes from the strategy library.

[0033] S4.2 Long-term simulation based on digital twin: In the digital twin of architectural heritage constructed by integrating real-time data into a dynamic three-dimensional model, each candidate strategy is loaded; different future scenarios are set, and simulation technology is used to simulate the implementation effect of each strategy in the next 5 years;

[0034] S4.3 Multi-objective optimization and decision support: Compare the performance of different strategies in long-term simulations and conduct a comprehensive evaluation from multiple dimensions such as physical security durability, cultural narrative preservation, economic cost and implementation feasibility;

[0035] S4.4 The system presents decision-makers with long-term simulation comparison results of various strategies in the form of a visual dashboard, recommends the strategy with the best overall effectiveness and the strongest cultural adaptability, and provides detailed simulation prediction data as a basis for decision-making.

[0036] Furthermore, in S5, the construction of the feedback iteration and collaborative implementation module is as follows:

[0037] S5.1, Hierarchical Collaboration Platform Supports Implementation: Through the professional version platform, detailed digital protection plan drawings and construction guidelines linked with the 3D model are issued to the engineering team; through the public version AR guide APP, during or after construction, "Why this needs to be restored" and "The story behind it" are shown to the public, transforming the protection process itself into a cultural and educational scenario;

[0038] S5.2 Effect Feedback and Model Iteration: After implementation, the actual effect data of the protection measures are continuously collected through sensor networks and community feedback. These follow-up data are used as new inputs and fed back to the evaluation model in S3 and the simulation rule engine in S4 to automatically calibrate and optimize the algorithm parameters, so that the system's evaluation and prediction capabilities can be continuously evolved in practice.

[0039] Furthermore, the physical components that carry the core narrative include the main hall space, the construction of specific techniques, and family events.

[0040] Furthermore, the degree of impact of the current physical state change on the associated narrative chain refers to assessing whether damage to the components carrying core memories would render the narrative incompletely perceptible.

[0041] Furthermore, the strategy library includes status quo maintenance, minimal intervention hardening, identifiable repair, and compatible function regeneration.

[0042] Furthermore, the simulation content includes: changes in structural performance, material aging process, and changes in the availability and perception of key narrative spaces.

[0043] The beneficial effects of this invention are:

[0044] This invention transforms the elusive "cultural connotation" into analyzable and displayable structured data, providing a cultural value basis for protection decisions that transcends material form; it achieves the quantification and visualization of cultural value; and it clarifies that the objects of protection are not only "bricks, tiles, wood and stones," but also the "stories and meanings" that carry them, so that subsequent protection and intervention measures can be targeted and ensure that there is "unchanging" (cultural core) within "change" (appearance changes).

[0045] This invention transforms the assessment from a "static snapshot" to a "dynamic film": overcoming the shortcomings of traditional assessment reports that are lagging and one-sided, it enables continuous tracking of the entire "change process" and can detect potential risks earlier; it enables early warning of the correlation between "physical risks" and "cultural risks": it can not only warn of structural cracks, but also assess whether the cracks endanger components that bear key narrative elements, thereby achieving risk classification management and optimizing resource allocation;

[0046] This invention elevates the decision-making model from "experience-driven" and "post-event remediation" to "data simulation-driven" and "pre-event prediction," significantly reducing the cost of trial and error in protection. By forcibly considering the impact of strategies on the narrative chain during the deduction process, it ensures that the selected solutions, while adapting to new needs, preserve and highlight core cultural identity to the greatest extent possible, avoiding "protective destruction." At the same time, it organically combines expert knowledge, management will, community memory, and public participation, enhancing the democratic nature and social recognition of protection. Through vivid narrative presentation, it greatly enhances the appeal and accessibility of heritage, making the protection system itself a powerful tool for cultural inheritance and value dissemination.

[0047] This invention is applicable to multiple scenarios, including the overall protection of traditional villages, preventive protection of cultural relics and historical buildings, community-participatory protection, and educational and research platforms. It can provide systematic archiving, dynamic monitoring, and scientific decision-making support for vernacular architectural heritage. At the same time, the system can enhance public participation and community cultural identity, realize integrated protection of "tangible and intangible" heritage, and enhance the interpretability and dissemination of heritage. Furthermore, it can improve the scientific nature and predictability of protection decisions through risk warning and simulation, reduce maintenance costs, and leverage digital content to promote cultural tourism experiences and the construction of local cultural brands, thus possessing significant social, cultural, managerial, and economic benefits. Attached Figure Description

[0048] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0049] Figure 1This is a flowchart of the adaptive conservation assessment system for vernacular architectural heritage based on digital narrative, as described in this invention.

[0050] Figure 2 This is the overall architecture diagram of the adaptive conservation assessment system for vernacular architectural heritage based on digital narrative as described in this invention;

[0051] Figure 3 This is a flowchart of the modeling process for the adaptive conservation assessment system for vernacular architectural heritage based on digital narrative, as described in this invention.

[0052] Figure 4 This is a flowchart of the risk assessment and decision-making process for the adaptive conservation assessment system for vernacular architectural heritage based on digital narrative, as described in this invention.

[0053] Figure 5 This is a diagram illustrating the narrative content design scheme of the adaptive conservation assessment system for vernacular architectural heritage based on digital narrative, as described in this invention.

[0054] Figure 6 This is a flowchart of the "Bai Shou Tang" application example of Embodiment 2 of the digital narrative-based adaptive protection assessment system for vernacular architectural heritage described in this invention. Detailed Implementation

[0055] 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.

[0056] Example 1

[0057] A digital narrative-based assessment system for the adaptive conservation of vernacular architectural heritage, characterized by comprising the following modules:

[0058] S1. Data Acquisition and Digital Archiving Module: Through field surveys and local participation, oral histories, documents, and ritual records related to architecture are collected, and data on architectural material space and materials are collected. After being structured by the narrative engine, the data is associated with a high-precision 3D model to construct a digital narrative knowledge graph with spatiotemporal correlation attributes.

[0059] S2, Full Life Cycle Dynamic Monitoring and Data Fusion Module: Deploy a sensor network and combine it with periodic remote sensing and survey reports to continuously acquire data on building physical status, environmental parameters and social usage, and dynamically associate and update multi-source monitoring data with the digital narrative knowledge graph;

[0060] S3, Cultural and Physical Risk Coupling Assessment Module: Based on the dynamic monitoring data and digital narrative knowledge graph, it analyzes the impact of changes in physical state on the associated narrative chain, calculates and outputs a comprehensive assessment report that includes both structural risk index and cultural value risk index through the coupling assessment model.

[0061] S4. Strategy Simulation and Decision Support Module: Based on the comprehensive evaluation report, candidate protection strategy schemes are generated from the strategy library, and the long-term implementation effects of each scheme under different future scenarios are simulated in the digital twin of architectural heritage. Visual assistance is provided to decision-makers through multi-objective optimization analysis.

[0062] S5. Feedback Iteration and Collaborative Implementation Module: Collects actual effect data and public feedback after the implementation of protection measures, feeds the data back to the risk assessment model and strategy simulation engine to optimize system parameters, and supports the implementation of protection projects and public cultural education through a hierarchical collaborative platform, forming a data closed loop of "monitoring-assessment-decision-implementation-feedback".

[0063] In S1, a digital narrative knowledge graph with spatiotemporal correlation attributes is constructed, as follows:

[0064] S1.1 Material Space Data Acquisition: Using UAV oblique photography and ground 3D laser scanning, centimeter-accurate real-scene 3D models and point cloud data of building complexes and environment are obtained to generate real-scene 3D models or building information models (BIM).

[0065] S1.2 Material data acquisition: Infrared thermal imaging and hyperspectral scanning technology are used to supplement the acquisition of hidden information such as material defects and moisture content;

[0066] S1.3 Non-material narrative data collection and structuring: Through participatory workshops, in-depth interviews, oral history records, and digitization of folk documents, the system collects cultural information related to building construction techniques, life memories, ritual activities, and family changes. By deploying a sensor network, the system monitors physical parameters of buildings such as displacement, cracks, vibration, temperature, and humidity to assess structural safety and material deterioration.

[0067] S1.4. Using natural language processing technology, entity recognition, relation extraction, and sentiment analysis are performed on the collected text and audio transcripts to extract narrative elements such as "craftsman", "ritual", "skill", and "historical event" and label their spatiotemporal attributes, importance, and emotional value.

[0068] S1.5 Constructing a spatiotemporally related digital narrative knowledge graph: In the real-world 3D model, semantic annotation and spatial anchoring are performed on the physical components that carry the core narrative;

[0069] S1.6. Associate the narrative elements extracted in S1.5 with these spatial semantic units to establish a "narrative-space" association network of "who-where-when-what-what / what skills", forming a queryable and inferable digital narrative knowledge graph.

[0070] In S2, the construction of the full lifecycle dynamic monitoring and data fusion module is as follows:

[0071] S2.1 Deploy an Internet of Things (IoT) sensing network: Install displacement, tilt, and crack sensors in key structural parts; install temperature, humidity, light, and vibration sensors in typical spaces to achieve real-time monitoring of building physics and environmental conditions;

[0072] S2.2 Establish a periodic remote sensing and survey report mechanism: Conduct drone aerial photography quarterly or annually to monitor changes in the macro-pattern and surrounding environment; develop a community participation mini-program to encourage residents and managers to upload graphic reports reflecting the daily use of buildings, minor changes, or new cultural activities, forming a social perception data flow;

[0073] S2.3 Spatiotemporal Data Fusion and Correlation Update: Real-time sensor data, periodic remote sensing data, and community report data are uniformly aligned to the spatiotemporal coordinate system of the real-world 3D model and knowledge graph established in Step 1; the state attributes of relevant nodes in the knowledge graph are dynamically updated to achieve dynamic binding of physical data and cultural narrative.

[0074] In S3, the construction of the cultural and physical risk coupling assessment module is as follows:

[0075] S3.1, Dual-path parallel analysis:

[0076] Physical risk analysis: Based on sensor and remote sensing data, the safety risk index of each component is calculated using structural mechanics models and material aging models;

[0077] Narrative integrity analysis: Based on digital narrative knowledge graphs, analyze the degree of impact of current physical state changes on related narrative chains;

[0078] S3.2 Calculation of Coupled Risk Assessment Model: Establish a coupled assessment algorithm, which is expressed as: Comprehensive Risk Index = (Physical Risk Index, Narrative Importance Weight × Narrative Integrity Impairment);

[0079] The output results determine "where is broken" and further determine "where the break poses the greatest threat to cultural heritage"; the output is a heat map superimposed on a 3D model, which simultaneously marks the physical risk level and the cultural risk level, and generates a graded early warning report.

[0080] In S4, the construction of the strategy simulation and decision support module is as follows:

[0081] S4.1 Intelligent Strategy Matching and Generation: The system has a built-in strategy rule engine that integrates the principles of the "Management Measures for Cultural Relics Protection Projects" and narrative protection rules; for high-risk objects identified in S3, the engine automatically matches and combines multiple candidate protection strategy schemes from the strategy library.

[0082] S4.2 Long-term simulation based on digital twin: In the digital twin of architectural heritage constructed by integrating real-time data into a dynamic three-dimensional model, each candidate strategy is loaded; different future scenarios are set, and simulation technology is used to simulate the implementation effect of each strategy in the next 5 years;

[0083] S4.3 Multi-objective optimization and decision support: Compare the performance of different strategies in long-term simulations and conduct a comprehensive evaluation from multiple dimensions such as physical security durability, cultural narrative preservation, economic cost and implementation feasibility;

[0084] S4.4 The system presents decision-makers with long-term simulation comparison results of various strategies in the form of a visual dashboard, recommends the strategy with the best overall effectiveness and the strongest cultural adaptability, and provides detailed simulation prediction data as a basis for decision-making.

[0085] In S5, the construction of the feedback iteration and collaborative implementation module is as follows:

[0086] S5.1, Hierarchical Collaboration Platform Supports Implementation: Through the professional version platform, detailed digital protection plan drawings and construction guidelines linked with the 3D model are issued to the engineering team; through the public version AR guide APP, during or after construction, "Why this needs to be restored" and "The story behind it" are shown to the public, transforming the protection process itself into a cultural and educational scenario;

[0087] S5.2 Effect Feedback and Model Iteration: After implementation, the actual effect data of the protection measures are continuously collected through sensor networks and community feedback. These follow-up data are used as new inputs and fed back to the evaluation model in S3 and the simulation rule engine in S4 to automatically calibrate and optimize the algorithm parameters, so that the system's evaluation and prediction capabilities can be continuously evolved in practice.

[0088] The physical components that carry the core narrative include the main hall space, the construction of specific techniques, and family events.

[0089] The extent to which the change in the current physical state affects the associated narrative chain refers to assessing whether damage to the components carrying core memories would render the narrative incompletely perceptible.

[0090] The strategy library includes status quo maintenance, minimal intervention hardening, identifiable repair, and compatible function regeneration.

[0091] The simulation content of the simulation and deduction includes: changes in structural performance, the process of material aging, and changes in the availability and perception of key narrative spaces.

[0092] Embodiment 2

[0093] Apply the adaptive protection evaluation system for vernacular architectural heritage based on digital narrative described in Embodiment 1 to the "Hand-Waving Hall" in a certain village. The specific steps are as follows:

[0094] Step 1: Digital narrative archiving and baseline model construction

[0095] (1) Data collection: Use drones to conduct oblique photography of the settlement to generate a real-scene 3D model; conduct fine 3D laser scanning of the "Hand-Waving Hall"; at the same time, the research group stays in the village and records the ceremony process of the "Hand-Waving Dance", the formula for building the ridgepole, the oral history of the successive additions, reconstructions of the building through interviews with old craftsmen and elders in the village, and collect relevant old photos.

[0096] (2) Narrative modeling: In the software, integrate the scanned high-precision model of the "Hand-Waving Hall" with the real-scene model of the settlement; use natural language processing tools to analyze the interview text and extract key narrative elements (such as: "a certain craftsman", "major repair in a certain year", "beam sacrifice ceremony", "hand-waving activities on specific festivals"); subsequently, in the 3D model, anchor the "beam sacrifice ceremony" narrative to the middle column, anchor the core activity area of the "hand-waving dance" to a specific area in the center of the hall, and associate the historical scenes in the old photos to the corresponding spaces through superposition and comparison; all association relationships are stored in a graph database (such as Neo4j) to form an initial digital narrative knowledge graph.

[0097] (3) Output: A digital twin baseline model of the "Hand-Waving Hall" and its environment containing geometric information and cultural semantics.

[0098] Step 2: Dynamic perception and data fusion

[0099] (1) Deploy perception devices: Install wireless inclination sensors and temperature and humidity sensors at the key wooden frame nodes of the "hall".

[0100] (2) Establish participation channels: Open "Settlement Guardian" mini-program accounts for the village committee and enthusiastic villagers, and encourage them to regularly upload photos of the current building situation, report problems such as insect damage and water leakage, or share videos of current cultural activities.

[0101] (3) Data platform fusion: All sensor data is uploaded to the cloud platform in real time through the Internet of Things gateway; the content uploaded by the mini-program is also imported into the platform after being reviewed and lightly annotated by the background; the platform automatically associates and updates these dynamic data with the corresponding components (such as the beams and columns where the sensors are located) or spaces (such as the hall corresponding to the activity video) in the baseline model.

[0102] (4) Output: A continuously pulsating, "living" digital twin that reflects the current situation.

[0103] Step 3: Coupling Risk Assessment and Early Warning

[0104] (1) Model calculation: The system continuously analyzes sensor data and finds that the tilt angle of a load-bearing column in the southwest corner has a continuous and slow increasing trend during the humid season, triggering a physical risk warning;

[0105] (2) Narrative association analysis: The system automatically queries the knowledge graph and finds that the structure is associated with the oral narrative of "installed by a famous craftsman", and the narrative importance weight is high;

[0106] (3) Coupled Assessment Output: After comprehensive calculation by the coupled assessment model, it is determined that this problem is not only a medium-level structural safety hazard, but also a high-level cultural value risk—once the column fails, the physical evidence of the specific craftsmanship memory associated with it will be severely damaged; the system highlights the structure in the 3D model and pushes an alert to the administrator's mobile phone: "Continuous deformation of the core cultural carrier component (southwest corner column, associated with the XX craftsman narrative) has been detected. It is recommended to prioritize intervention and investigation;"

[0107] (4) Output: Priority warning reports with precise positioning and accompanying cultural value interpretation.

[0108] Step 4: Adaptive Strategy Simulation and Decision Making

[0109] (1) Strategy Generation: For the local construction of the early warning, the system strategy engine combines the principles of "minimum intervention" and "reversibility" with the rule of "protection of core narrative carrier" to generate three candidate strategies:

[0110] A) Add reversible steel structure auxiliary support;

[0111] B) Localized patching and reinforcement;

[0112] C) Replace the column during the overall disassembly and overhaul;

[0113] (2) Digital twin simulation: Three strategies are simulated in the digital twin; the next 30 years are simulated with two scenarios: "increased tourism load" and "increased rainfall"; the simulation shows that strategy A has the least impact on the current activities, the lowest cost, and can stabilize the situation in the long term, but has a slight impact on the building's appearance; strategy C is the most thorough, but it is costly, has a long construction period, and will interrupt the holding of the ceremony for a long time.

[0114] (3) Decision support: Decision-makers can view the simulation results comparison chart through the system dashboard, and comprehensively consider the urgency of protection, the continuity of community activities and the funding situation. Finally, they select strategy A as the short-term adaptive intervention measure and strategy C as the long-term vision plan alternative.

[0115] (4) Output: Optimized scientific protection decision-making schemes that have been verified by long-term impact simulation.

[0116] Step 5: Collaborative Implementation and Closed-Loop Optimization

[0117] (1) Collaborative implementation: The selected strategy A will generate construction drawings with three-dimensional annotations to guide the construction team to carry out precise construction; at the same time, through the AR guide APP, the "past and present" story of the structure will be shown to tourists, and the significance of the current protection measures will be explained, turning "interference" into "education".

[0118] (2) Feedback iteration: After the project is completed, the data from the new sensor on the column remains stable; local villagers provide feedback on “protection views and opinions” through the mini-program; positive aftereffect data is recorded by the system and used to strengthen the recommendation weight of strategy A in similar situations and optimize the strategy rule engine.

[0119] (3) Output: A complete protection practice loop enhances the system’s future assessment and recommendations.

Claims

1. A digital narrative-based assessment system for the adaptive conservation of vernacular architectural heritage, characterized in that, Includes the following modules: S1. Data Acquisition and Digital Archiving Module: Through field surveys and local participation, oral histories, documents, and ritual records related to architecture are collected, and data on architectural material space and materials are collected. After being structured by the narrative engine, the data is associated with a high-precision 3D model to construct a digital narrative knowledge graph with spatiotemporal correlation attributes. S2, Full Life Cycle Dynamic Monitoring and Data Fusion Module: Deploy a sensor network and combine it with periodic remote sensing and survey reports to continuously acquire data on building physical status, environmental parameters and social usage, and dynamically associate and update multi-source monitoring data with the digital narrative knowledge graph; S3, Cultural and Physical Risk Coupling Assessment Module: Based on the dynamic monitoring data and digital narrative knowledge graph, it analyzes the impact of changes in physical state on the associated narrative chain, calculates and outputs a comprehensive assessment report that includes both structural risk index and cultural value risk index through the coupling assessment model. S4. Strategy Simulation and Decision Support Module: Based on the comprehensive evaluation report, candidate protection strategy schemes are generated from the strategy library, and the long-term implementation effects of each scheme under different future scenarios are simulated in the digital twin of architectural heritage. Visual assistance is provided to decision-makers through multi-objective optimization analysis. S5. Feedback Iteration and Collaborative Implementation Module: Collects actual effect data and public feedback after the implementation of protection measures, feeds the data back to the risk assessment model and strategy simulation engine to optimize system parameters, and supports the implementation of protection projects and public cultural education through a hierarchical collaborative platform, forming a data closed loop of "monitoring-assessment-decision-implementation-feedback".

2. The adaptive conservation assessment system for vernacular architectural heritage based on digital narrative as described in claim 1, characterized in that, In S1, the data acquisition and digital archiving module is constructed as follows: S1.1 Material Space Data Acquisition: Using UAV oblique photography and ground 3D laser scanning, centimeter-accurate real-scene 3D models and point cloud data of building complexes and environment are obtained to generate real-scene 3D models or building information models (BIM). S1.2 Material data acquisition: Infrared thermal imaging and hyperspectral scanning technology are used to supplement the acquisition of hidden information such as material defects and moisture content; S1.3 Non-material narrative data collection and structuring: Through participatory workshops, in-depth interviews, oral history records, and digitization of folk documents, the system collects cultural information related to building construction techniques, life memories, ritual activities, and family changes. By deploying a sensor network, the system monitors physical parameters of buildings such as displacement, cracks, vibration, temperature, and humidity to assess structural safety and material deterioration. S1.

4. Using natural language processing technology, entity recognition, relation extraction, and sentiment analysis are performed on the collected text and audio transcripts to extract narrative elements such as "craftsman," "ritual," "skill," and "historical event," and to label their spatiotemporal attributes, importance, and emotional value. S1.5 Constructing a spatiotemporally related digital narrative knowledge graph: In the real-world 3D model, semantic annotation and spatial anchoring are performed on the physical components that carry the core narrative; S1.

6. Associate the narrative elements extracted in S1.5 with these spatial semantic units to establish a "narrative-space" association network of "who-where-when-what-what / what skills", forming a queryable and inferable digital narrative knowledge graph.

3. The adaptive conservation assessment system for vernacular architectural heritage based on digital narrative as described in claim 1, characterized in that, In S2, the construction of the full lifecycle dynamic monitoring and data fusion module is as follows: S2.1 Deploy an Internet of Things (IoT) sensing network: Install displacement, tilt, and crack sensors in key structural parts; install temperature, humidity, light, and vibration sensors in typical spaces to achieve real-time monitoring of building physics and environmental conditions; S2.2 Establish a periodic remote sensing and survey report mechanism: Conduct drone aerial photography quarterly or annually to monitor changes in the macro-pattern and surrounding environment; Develop a community participation mini-program to encourage residents and managers to upload graphic and text reports reflecting the daily use of the building, minor changes, or new cultural activities, forming a social perception data stream; S2.3 Spatiotemporal Data Fusion and Correlation Update: Real-time sensor data, periodic remote sensing data, and community report data are uniformly aligned to the spatiotemporal coordinate system of the real-world 3D model and knowledge graph established in Step 1; the state attributes of relevant nodes in the knowledge graph are dynamically updated to achieve dynamic binding of physical data and cultural narrative.

4. The adaptive conservation assessment system for vernacular architectural heritage based on digital narrative as described in claim 1, characterized in that, In S3, the construction of the cultural and physical risk coupling assessment module is as follows: S3.1, Dual-path parallel analysis: Physical risk analysis: Based on sensor and remote sensing data, the safety risk index of each component is calculated using structural mechanics models and material aging models; Narrative integrity analysis: Based on digital narrative knowledge graphs, analyze the degree of impact of current physical state changes on related narrative chains; S3.2 Calculation of Coupled Risk Assessment Model: Establish a coupled assessment algorithm, which is expressed as: Comprehensive Risk Index = (Physical Risk Index, Narrative Importance Weight × Narrative Integrity Impairment); The output results determine "where is broken" and further determine "where the break poses the greatest threat to cultural heritage"; the output is a heat map superimposed on a 3D model, which simultaneously marks the physical risk level and the cultural risk level, and generates a graded early warning report.

5. The adaptive conservation assessment system for vernacular architectural heritage based on digital narrative as described in claim 1, characterized in that, In S4, the construction of the strategy simulation and decision support module is as follows: S4.1 Intelligent Strategy Matching and Generation: The system has a built-in strategy rule engine that integrates the principles of the "Management Measures for Cultural Relics Protection Projects" and narrative protection rules; for high-risk objects identified in S3, the engine automatically matches and combines multiple candidate protection strategy schemes from the strategy library. S4.2 Long-term simulation based on digital twin: In the digital twin of architectural heritage constructed by integrating real-time data into a dynamic three-dimensional model, each candidate strategy is loaded; different future scenarios are set, and simulation technology is used to simulate the implementation effect of each strategy in the next 5 years; S4.3 Multi-objective optimization and decision support: Compare the performance of different strategies in long-term simulations and conduct a comprehensive evaluation from multiple dimensions such as physical security durability, cultural narrative preservation, economic cost and implementation feasibility; S4.4 The system presents decision-makers with long-term simulation comparison results of various strategies in the form of a visual dashboard, recommends the strategy with the best overall effectiveness and the strongest cultural adaptability, and provides detailed simulation prediction data as a basis for decision-making.

6. The adaptive conservation assessment system for vernacular architectural heritage based on digital narrative as described in claim 1, characterized in that, In S5, the construction of the feedback iteration and collaborative implementation module is as follows: S5.1, Hierarchical Collaborative Platform Supports Implementation: Through the professional version platform, detailed digital protection plan drawings and construction guidelines linked with the 3D model are issued to the engineering team; through the public version AR guide APP, during or after construction, "Why this needs to be restored" and "The story behind it" are shown to the public, transforming the protection process itself into a cultural and educational scenario; S5.2 Effect Feedback and Model Iteration: After implementation, the actual effect data of the protection measures are continuously collected through sensor networks and community feedback. These follow-up data are used as new inputs and fed back to the evaluation model in S3 and the simulation rule engine in S4 to automatically calibrate and optimize the algorithm parameters, so that the system's evaluation and prediction capabilities can be continuously evolved in practice.

7. A digital narrative-based assessment system for the adaptive conservation of vernacular architectural heritage according to claim 2, characterized in that, The physical components that carry the core narrative include the main hall space, the construction of specific techniques, and family events.

8. A digital narrative-based assessment system for the adaptive conservation of vernacular architectural heritage according to claim 4, characterized in that, The degree of impact of the current physical state change on the associated narrative chain refers to assessing whether damage to the component carrying the core memory results in the narrative being unable to be fully perceived; the strategy library includes status quo maintenance, minimal intervention reinforcement, identifiable repair, and compatibility function regeneration.

9. A digital narrative-based assessment system for the adaptive conservation of vernacular architectural heritage according to claim 5, characterized in that, The simulation includes changes in structural performance, material aging processes, and changes in the availability and perception of key narrative spaces.

Images