Cultural heritage site flood resilience monitoring and early warning internet of things system

Abstract

The disclosure provides a cultural heritage site flood resilience monitoring and early warning Internet of Things system, relating to the technical field of cultural heritage site monitoring and protection. The system includes a resilience assessment module, an Internet of Things sensing module, an edge analysis module, a strategy decision module, and a response execution module. The system determines key monitoring points and resilience target values by obtaining geographic spatial data and heritage structure information and performing vulnerability analysis; multi-source sensing nodes are deployed at key points to collect multi-dimensional data such as settlement, cracks, and water accumulation; the edge analysis module processes multi-source data fusion and generates real-time flood risk state information; the strategy decision module combines a strategy library and a structure response model to perform multi-strategy combination simulation to determine the optimal response scheme; the response execution module controls flood control devices according to the target strategy and updates the model and strategy based on feedback. This technical solution can realize the intelligentization, dynamization, and precision of cultural heritage site flood protection, and improve monitoring and response efficiency.

Description

Technical Field

[0001] This disclosure relates to the field of cultural heritage site monitoring and protection technology, and more specifically, to an Internet of Things (IoT) system for monitoring and early warning of flood resilience in cultural heritage sites. Background Technology

[0002] As an important part of human civilization, cultural heritage sites have long been a focus of protection. However, many heritage sites are located in flood-prone areas and face increasingly frequent extreme weather events. How to effectively deal with flood risks has become an urgent problem to be solved.

[0003] Currently, flood monitoring solutions for cultural heritage sites largely rely on traditional hydrological and meteorological monitoring networks. These systems typically collect data from sensors at fixed locations, and the layout of monitoring points often fails to fully consider the geographical characteristics and structural attributes of the cultural heritage site, resulting in limited monitoring coverage and insufficient targeting. Furthermore, traditional methods tend to focus on single environmental parameters in data collection, lacking the integration and collaborative analysis of multi-source information, which limits the comprehensiveness and accuracy of risk assessment. Simultaneously, current monitoring and early warning systems often use predefined static thresholds for warnings, resulting in rigid response strategies that are difficult to dynamically adjust based on real-time environmental changes, and a lack of effective feedback mechanisms to optimize subsequent decisions. These limitations lead to monitoring blind spots, analytical lags, and low response efficiency in flood protection of cultural heritage sites, failing to meet the needs of long-term resilient protection.

[0004] It should be noted that the information disclosed in the background section above is only used to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0005] The purpose of this disclosure is to provide an Internet of Things (IoT) system for monitoring and early warning of flood resilience in cultural heritage sites, thereby enabling...

[0006] Other features and advantages of this disclosure will become apparent from the following detailed description, or may be learned in part from practice of this disclosure.

[0007] According to a first aspect of the present disclosure, a flood resilience monitoring and early warning Internet of Things (IoT) system for cultural heritage sites is provided, comprising: The resilience assessment module is used to acquire geospatial data and structural information of the cultural heritage site, perform vulnerability analysis on the structural information of the heritage site, generate structural vulnerability data, and combine the structural vulnerability data and the geospatial data to determine key monitoring points and corresponding resilience target values. The Internet of Things (IoT) sensing module is communicatively connected to the resilience assessment module and is used to collect environmental structural status data through multi-source sensing nodes deployed at the key monitoring points. The edge analysis module, which is communicatively connected to the IoT sensing module, is used to perform fusion analysis processing on the received environmental structure status data to determine flood risk status information; The strategy decision module is communicatively connected to the edge analysis module and the IoT sensing module. It is used to acquire simulated input data, load a preset structural response model based on a preset strategy library containing multi-level response strategies, and perform multi-strategy combination simulation in combination with the flood risk status information and the simulated input data to determine the target response strategy that meets the resilience target value. The response execution module is communicatively connected to the strategy decision module. It is used to invoke a pre-set flood control response device according to the control command generated by the target response strategy to carry out protective operations on the cultural heritage site, and to return the execution feedback result of the flood control response device to the strategy decision module to update the parameters of the strategy library and the structural response model.

[0008] In some example embodiments of this disclosure, based on the foregoing scheme, the resilience assessment module includes a vulnerability analysis unit, which is configured as follows: The structural information of the heritage site is subjected to spatial grid division to obtain multiple structural grid units; A vulnerability score matrix is ​​determined for each of the structural grid cells. The vulnerability score matrix includes a material degradation coefficient, a foundation stability coefficient, and a node connectivity coefficient. The vulnerability score matrix is ​​used as the structural vulnerability data.

[0009] In some example embodiments of this disclosure, based on the foregoing scheme, the resilience assessment module includes a key monitoring point determination unit, which is configured as follows: The vulnerability score matrix in the structural vulnerability data is matched with the digital elevation model in the geospatial data at the grid cell level to generate a spatial matching result that includes water catchment area and terrain slope parameters. Based on the spatial matching results, the monitoring priority parameters are calculated, and the monitoring priority parameters are correlated with the vulnerability score matrix to generate the monitoring weight coefficients for each structural grid unit. Each key monitoring point is determined based on the monitoring weight coefficient, and the resilience target value is calculated by combining the vulnerability score matrix of the key monitoring points.

[0010] In some example embodiments of this disclosure, based on the foregoing scheme, the IoT sensing module includes a data acquisition unit, which is configured as follows: The multi-source sensor nodes are installed at the corresponding locations according to the key monitoring points, and the structural monitoring signals are collected using the acquisition interface of each multi-source sensor node. The structural monitoring signals include at least settlement signals, crack signals, water accumulation signals and environmental signals. The structure monitoring signal is time-synchronized and noise-filtered to generate environmental structure status data.

[0011] In some example embodiments of this disclosure, based on the foregoing scheme, the edge analysis module includes a flood risk state prediction unit, which is configured as follows: Multi-source data comparison and feature parameter extraction are performed on the environmental structural state data to determine a risk feature parameter set, which includes at least the settlement change rate, crack growth rate and average water depth. Risk classification calculations are performed based on the risk feature parameter set to generate flood risk status information.

[0012] In some example embodiments of this disclosure, based on the foregoing scheme, the simulated input data includes real-time monitoring data or preset scenario data collected by the multi-source sensing nodes, and the strategy decision module includes an input data generation unit, which is configured to: The preset scenario data is generated based on historical rainfall duration-intensity-frequency data and watershed boundary conditions; and / or By combining the structured monitoring data packets collected in real time by the multi-source sensing nodes, real-time monitoring data including rainfall intensity sequence, water depth sequence and surface runoff parameters is generated.

[0013] In some example embodiments of this disclosure, based on the foregoing scheme, the strategy decision module includes a simulation unit, which is configured as follows: The multi-level response strategy of the strategy library is invoked in the loaded structural response model; The flood risk status information and the simulation input data are input into the structural response model to perform multi-strategy combination simulation, and the structural response parameter set corresponding to each strategy combination is output. Constraint calculations are performed on the set of structural response parameters to determine the target response strategy that satisfies the toughness target value.

[0014] In some example embodiments of this disclosure, based on the foregoing scheme, the simulation unit is further configured as follows: The first-level response strategy, the second-level response strategy, and the third-level response strategy are respectively called from the strategy library; wherein, the first-level response strategy includes automatic water blocking operation, the second-level response strategy includes pumping and draining operation, and the third-level response strategy includes audible and visual alarm and closure operation; and A multi-level response strategy is generated using the first-level response strategy, the second-level response strategy, and the third-level response strategy. The multi-level response strategy includes a list of candidate strategy combinations.

[0015] In some example embodiments of this disclosure, based on the foregoing scheme, the response execution module includes a flood control response control unit, which is configured to: Analyze the device identifier, execution timing, and parameter settings in the target response strategy; Based on the device identifier, the execution timing, and the parameter settings, a control instruction sequence is generated, which includes a device instruction code, a start time, and an execution cycle. The control command sequence is sent to the corresponding flood control response device, and the execution feedback results of the flood control response device are recorded.

[0016] In some example embodiments of this disclosure, based on the foregoing scheme, the response execution module includes a protection policy update unit, which is configured to: The difference between the execution feedback result and the environmental structure state data is calculated to obtain feedback deviation data; The parameter weights of the structural response model are adjusted based on the feedback deviation data, and the priority of each response strategy in the strategy library is updated.

[0017] According to a second aspect of the present disclosure, a method for monitoring and early warning of flood resilience in cultural heritage sites is provided, comprising: The geospatial data and structural information of the cultural heritage site are obtained, vulnerability analysis is performed on the structural information of the heritage site to generate structural vulnerability data, and key monitoring points and corresponding resilience target values ​​are determined by combining the structural vulnerability data and the geospatial data. Environmental structural status data are collected by deploying multi-source sensor nodes at the key monitoring points. The received environmental structure status data is subjected to fusion analysis to determine flood risk status information; Acquire simulated input data, load a preset structural response model based on a preset strategy library containing multi-level response strategies, and perform multi-strategy combination simulation by combining the flood risk status information and the simulated input data to determine the target response strategy that meets the resilience target value. The control command generated according to the target response strategy calls the pre-set flood control response device to carry out protective operations on the cultural heritage site, and returns the execution feedback result of the flood control response device to the strategy decision module to update the parameters of the strategy library and the structural response model.

[0018] The technical solutions provided in this disclosure may have the following beneficial effects: The IoT system for monitoring and early warning of flood resilience in cultural heritage sites in the example embodiments of this disclosure, on the one hand, acquires geospatial data and structural information of cultural heritage sites through a resilience assessment module, and performs vulnerability analysis to determine key monitoring points and corresponding resilience target values. This allows the deployment of monitoring points to be more closely integrated with the specific geographical features and structural attributes of the heritage site, avoiding monitoring blind spots caused by the generalization of layout in traditional methods, thereby improving the targeting and coverage integrity of the monitoring network. Furthermore, the IoT sensing module deploys multi-source sensor nodes at key monitoring points to collect environmental structural status data, which can integrate multiple physical parameters (such as water level, vibration, humidity, etc.) to achieve synchronous acquisition and coordination of multi-source environmental information, overcome the problem of information partiality under a single data acquisition method, and ensure the comprehensiveness and effectiveness of the data. On the other hand, by performing fusion analysis on multi-source environmental structural state data through the edge analysis module, flood risk characteristics can be extracted in real time and flood risk state information can be generated. This reduces data transmission and processing delays and avoids the inaccurate assessment problems caused by analysis lag or data isolation in traditional systems, thereby enhancing the timeliness and reliability of risk state identification. Simultaneously, the strategy decision module, based on a pre-set multi-level response strategy library and structural response model, combines flood risk state information with simulated input data to perform multi-strategy combination simulations. This allows for dynamic adjustment of response plans to match real-time risk changes, overcoming the rigidity and inefficiency of static threshold decisions and achieving more flexible and optimized protection decisions. Furthermore, the response execution module generates control commands based on the target response strategy to invoke flood control response devices and returns the execution feedback results to the strategy decision module to update the strategy library and model parameters. This forms a closed-loop optimization mechanism, continuously improving system response performance and avoiding the strategy rigidity problem caused by the lack of feedback adjustment in traditional methods, thereby enhancing the adaptability and accuracy of the system in long-term operation.

[0019] Overall, the IoT system for monitoring and early warning of flood resilience in cultural heritage sites in the example embodiments of this disclosure can realize intelligent management of the entire process from monitoring layout to response optimization, effectively addressing problems such as insufficient monitoring coverage, lack of data integration, delayed analysis, rigid response, and lack of feedback in related technologies, and can effectively improve the accuracy, efficiency, and reliability of flood protection in cultural heritage sites.

[0020] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0021] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure. It is obvious that the drawings described below are merely some embodiments of this disclosure, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort.

[0022] Figure 1 The illustration shows a schematic diagram of the composition of a cultural heritage site flood resilience monitoring and early warning Internet of Things system according to some embodiments of the present disclosure.

[0023] Figure 2 The schematic diagram illustrates the composition of a resilience assessment module according to some embodiments of the present disclosure.

[0024] Figure 3 The diagram illustrates the composition of a strategy decision module according to some embodiments of the present disclosure.

[0025] Figure 4 The illustration shows a flowchart of a method for monitoring and early warning of flood resilience in cultural heritage sites according to some embodiments of the present disclosure.

[0026] Figure 5 The schematic diagram illustrates the structural schematic of a computer system of an electronic device according to some embodiments of the present disclosure.

[0027] Figure 6 A schematic diagram of a computer-readable storage medium according to some embodiments of the present disclosure is shown.

[0028] In the accompanying drawings, the same or corresponding reference numerals indicate the same or corresponding parts. Detailed Implementation

[0029] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numerals in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this specification. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this specification as detailed in the appended claims.

[0030] The terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to be limiting of this specification. The singular forms “a,” “the,” and “the” as used in this specification and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any and all possible combinations of one or more of the associated listed items.

[0031] It should be understood that although the terms first, second, third, etc., may be used in this specification to describe various information, this information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, without departing from the scope of this specification, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the word "if" as used herein may be interpreted as "when," "when," or "in response to determination."

[0032] Furthermore, the accompanying drawings are for illustrative purposes only and are not necessarily drawn to scale. The block diagrams shown in the drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities can be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor devices and / or microcontroller devices.

[0033] In this example embodiment, an IoT system for monitoring and early warning of flood resilience in cultural heritage sites is first provided. Figure 1 This illustration schematically depicts the structure of an Internet of Things (IoT) system for monitoring and early warning of flood resilience in cultural heritage sites, according to some embodiments of this disclosure. (Reference) Figure 1 As shown, the IoT system for monitoring and early warning of flood resilience in cultural heritage sites may include a resilience assessment module 110, an IoT sensing module 120, an edge analysis module 130, a strategy decision-making module 140, and a response execution module 150, wherein: The resilience assessment module 110 is used to acquire geospatial data and structural information of the cultural heritage site, perform vulnerability analysis on the structural information of the heritage site, generate structural vulnerability data, and combine the structural vulnerability data and the geospatial data to determine key monitoring points and corresponding resilience target values.

[0034] Geospatial data refers to spatial coordinate information used to reflect the topographic features, surface cover, and drainage characteristics of cultural heritage sites. For example, geospatial data may include Digital Elevation Models (DEMs), remote sensing imagery, Geographic Information System (GIS) vector layers, and surface runoff models. DEMs can be generated through satellite altimetry or UAV laser mapping to describe topographic elevation differences and slope parameters; remote sensing imagery can be acquired using multispectral or infrared image sensors to identify surface water distribution and soil moisture; GIS vector layers can include features such as roads, building outlines, and drainage ditches. In some alternative implementations, geospatial data may also come from ground surveying data or historical structural monitoring databases, as long as their spatial resolution and geographic accuracy meet the requirements of topographic analysis. This embodiment does not impose any special limitations on this.

[0035] Structural information of heritage sites can be acquired using Building Information Modeling (BIM) or Light Detection and Ranging (LiDAR) technologies. BIM records the geometry, material properties, and node connections of cultural heritage buildings, while LiDAR provides high-precision point cloud data for extracting structural shape and surface variations. The structural information can include foundation bearing capacity parameters, wall thickness, component stress records, and maintenance records, reflecting the mechanical performance and historical condition of the heritage buildings. Optionally, the structural information can also be processed using time-series reconstruction algorithms to describe the evolution of the structural state.

[0036] Vulnerability analysis refers to the computational process of quantitatively assessing the damage sensitivity and disaster-bearing capacity of architectural structures in cultural heritage sites under flood conditions. The core of vulnerability analysis lies in establishing a response relationship model between the building structure, foundation conditions, and external environmental loads to evaluate the relative vulnerability of different components or areas under flood scenarios. It can be based on multi-parameter inputs, such as material degradation characteristics, foundation stability, node connectivity, and overall structural stiffness, and generates a vulnerability index matrix through numerical modeling or multi-index comprehensive evaluation methods.

[0037] Structural vulnerability data refers to a multidimensional dataset with spatial mapping relationships formed after vulnerability analysis. It is used to quantitatively reflect the disaster-bearing vulnerability characteristics of architectural structures in cultural heritage sites at different spatial locations. It can be represented as a structural vulnerability matrix or vulnerability score matrix. The data cells of the matrix correspond to specific structural grid cells, and each data cell contains vulnerability strength parameters and their related attributes.

[0038] Structural vulnerability data can include not only vulnerability intensity scores but also geospatial coordinates, material type identifiers, and component category information, forming a spatial data layer that can be directly visualized in a Geographic Information System (GIS). This spatial mapping method enables coupled analysis of structural safety and topographic environment. For example, vulnerability data can be overlaid on a Digital Elevation Model (DEM) to identify highly vulnerable structural units in low-lying, water-concentrated areas, providing a quantitative basis for key monitoring point deployment and resilience target value calculation. In optional implementations, structural vulnerability data can also be correlated with a time dimension to form a multi-temporal vulnerability evolution sequence, used to track the vulnerability changes of heritage structures under different seasons or rainfall conditions. Furthermore, structural vulnerability data can be linked with risk status information from the edge analysis module to achieve dynamic matching between flood risk status and structural response capabilities, giving the system predictive and adaptive capabilities.

[0039] Key monitoring points refer to representative and sensitive monitoring locations selected within the spatial scope of a cultural heritage site, based on comprehensive calculations using geospatial data and structural vulnerability data. These locations are used to install multi-source sensor nodes to achieve real-time data collection and tracking of the environmental and structural status related to flood risks. The purpose of setting up key monitoring points is to achieve efficient characterization and dynamic tracking of the overall structural safety status of the heritage site under conditions of limited monitoring resources. For example, the vulnerability score matrix in the structural vulnerability data can be spatially matched with the digital elevation model in the geospatial data to link building structural features with topographic parameters (including slope, flow direction, and catchment area) in the same coordinate system. Then, the monitoring priority parameter of each spatial unit is calculated to characterize its monitoring importance in a flood scenario. This monitoring priority parameter can be calculated by combining vulnerability intensity parameters, topographic catchment area, surface slope, and relative elevation difference, and is used to quantify the risk sensitivity of the area under flood conditions. Next, the monitoring priority parameter can be correlated with the vulnerability intensity parameter in the vulnerability score matrix to obtain the monitoring weight coefficient for each structural grid unit. The monitoring weight coefficient is a composite index formed by comprehensively considering structural vulnerability and topographic risk, used to describe the importance of different areas in the monitoring layout. Based on the distribution of the monitoring weight coefficients, representative grid units in high-weight areas can be selected as key monitoring points. Optionally, the selection of key monitoring points can also be combined with building use, cultural relic grade, or restoration history information to ensure that the monitored objects have practical protective significance.

[0040] The spatial distribution of key monitoring points is typically non-uniform, and their density can vary with terrain complexity and structural risk level. For example, in low-lying areas with poor structural connectivity, a higher density of monitoring points can be deployed to enhance data coverage; in high-altitude or structurally stable areas, the number of nodes can be appropriately reduced to lower system energy consumption. Alternatively, clustering algorithms (such as K-Means or DBSCAN) can be used to cluster and partition the monitoring weight coefficients to achieve automated point deployment optimization.

[0041] The Internet of Things (IoT) sensing module 120 is communicatively connected to the resilience assessment module 110 and is used to collect environmental structural status data through multi-source sensing nodes deployed at the key monitoring points.

[0042] Among them, multi-source sensing nodes refer to low-power intelligent sensing units installed at key monitoring points for continuous monitoring of the environment and structural status. Multi-source sensing nodes can include various types of physical sensor components, signal processing components, power supply components, and wireless communication components.

[0043] Environmental structural status data refers to the set of structural and environmental parameters generated by synchronous collection from multiple source sensing nodes and preliminary processing, used to reflect the actual operational status of cultural heritage sites at the monitoring time. Environmental structural status data includes at least settlement data, crack change data, water depth data, and environmental meteorological data, and may also include additional parameters extended according to different node types, such as foundation pore water pressure, surface reflectivity, or temperature gradient. Before data upload, the IoT sensing module can perform compression and encryption operations locally to reduce communication load and ensure data security. Compression algorithms can use differential coding or lossless compression algorithms (such as Lempel-Ziv-Welch, LZW), and encryption algorithms can use the Advanced Encryption Standard (AES).

[0044] In an optional implementation, the IoT sensing module can be configured with an adaptive sampling mechanism. When environmental parameter fluctuations exceed a set threshold (e.g., the rate of water depth rise exceeds a warning value), the sampling frequency is automatically increased and the node's operating time is extended. Conversely, when the environment is stable, it can enter a low-power sleep mode to extend its energy usage cycle. Furthermore, the IoT sensing module can form a bidirectional communication link with the resilience assessment module to dynamically adjust the node's activation status or data sampling strategy based on updated monitoring weight coefficients or risk distributions, thereby achieving hierarchical management and adaptive scheduling of monitoring resources.

[0045] The edge analysis module 130 is communicatively connected to the IoT sensing module 120 and is used to perform fusion analysis processing on the received environmental structure status data to determine flood risk status information.

[0046] Fusion analysis processing refers to the computational process of integrating different types of monitoring data from multiple source sensor nodes in multiple dimensions, removing anomalies, and extracting risk features at computing nodes close to the data acquisition end. Fusion analysis processing is supported by edge computing architecture, enabling data pre-analysis and feature extraction to be completed locally, thereby significantly reducing transmission latency and improving response speed.

[0047] Flood risk status information refers to the structured results obtained after fusion analysis and calculation, used to characterize the flood safety status and changing trends of cultural heritage sites at the current moment. For example, flood risk status information may include risk level, trigger factor identifier, and probability weight, which can be used to determine whether to enter the warning stage or whether strategy adjustment is needed. For example, when the average water depth continues to rise and the crack growth rate exceeds a set threshold, a "moderate risk" label can be output, and this status information can be pushed to the strategy decision-making module.

[0048] In some optional implementations, the edge analysis module can be further integrated with predictive capabilities to identify potential risk trends in future time periods. The predictive capabilities can be implemented based on time series forecasting models (such as the Autoregressive Moving Average (ARIMA) model) or deep learning models (such as Long Short-Term Memory (LSTM) networks). This embodiment does not impose any specific limitations on the method of prediction. By inputting multiple time-series risk characteristic parameters, the system can output risk probability distribution curves for several future time steps, thereby triggering strategy simulation and protective preparation actions in advance.

[0049] Furthermore, the edge analysis module can maintain data interaction with the resilience assessment module. For example, when the risk status information output by the edge analysis module exceeds the medium risk threshold multiple times consecutively, the risk data can be fed back to the resilience assessment module to trigger a recalculation of the weights of key monitoring points or a dynamic correction of the resilience target value, thereby achieving real-time linkage and closed-loop optimization at the spatial distribution and decision constraint levels.

[0050] The edge analysis module integrates multi-source monitoring data, extracts risk characteristic parameters, and performs real-time risk identification, enabling the system to quickly assess the flood situation at the edge near the monitoring source. This reduces central computing latency and improves system response efficiency. Simultaneously, by leveraging feature prediction and model feedback mechanisms, it achieves dynamic integration from monitoring and analysis to risk warning, providing a highly timely and confident input foundation for the strategy decision-making module.

[0051] The strategy decision module 140 is communicatively connected to the edge analysis module 130 and the IoT sensing module 120. It is used to acquire simulated input data, load a preset structural response model based on a preset strategy library containing multi-level response strategies, and perform multi-strategy combination simulation in combination with the flood risk status information and the simulated input data to determine the target response strategy that meets the resilience target value.

[0052] The simulated input data refers to the bidirectional set of environmental and structural inputs required by the system to execute flood control strategy simulations. It includes at least one of real-time monitoring data from the IoT sensing module and preset scenario data generated based on historical environmental conditions. Real-time monitoring data reflects the current environmental and structural state of the cultural heritage site, such as rainfall intensity sequences, water depth curves, foundation pore water pressure sequences, and structural strain records. Preset scenario data is a virtual input sequence generated based on historical rainfall duration-intensity-frequency (IDF) curves, watershed boundary conditions, and typical disaster processes, used to simulate the system response under different flood scenarios. The simulated input data can be processed through data cleaning, format conversion, and time-series unification to ensure consistency in time dimension and spatial resolution between data from different sources. For example, if the monitoring data sampling frequency is 1Hz, while the historical simulation data is 0.1Hz, linear interpolation or spline interpolation can be used to resample the data to achieve time-series unification.

[0053] The strategy library refers to a pre-set database of response strategies used to store sets of protective operation instructions that can be executed under different flood scenarios. The strategy library can include multi-level response strategies, each level corresponding to different protection intensities and operational ranges. For example, a level-one response strategy may include basic control actions such as automatic raising and lowering of flood barriers and partial opening and closing of gates to cope with light water accumulation or short-term rainfall; a level-two response strategy may include starting pumping and drainage equipment, switching underground drainage pipes, and grouping control of water pumps to cope with rising water levels caused by continuous rainfall; a level-three response strategy may include triggering audible and visual alarms, closing tourist evacuation routes, and disconnecting power protection systems to cope with extreme flooding or structural safety risks. In some optional implementations, each strategy record in the strategy library may include not only the control device identifier, execution sequence, and parameter settings, but also priority weights, execution condition thresholds, and energy consumption coefficients to support multi-objective optimization calculations. The strategy library can be manually defined and corrected by expert experience, or it can be automatically generated by training a machine learning model from historical execution data to continuously optimize the effectiveness of strategy combinations; this embodiment is not limited to this.

[0054] A structural response model is a numerical calculation model used to simulate the stress, deformation, and stability response of architectural structures in cultural heritage sites under flood loads. The structural response model can be built based on the Finite Element Method (FEM). The model nodes correspond to the actual geometric positions of the building components, and the element material parameters can be derived from the structural information of the heritage site. The boundary conditions of the structural response model include external water pressure, foundation support stiffness, and structural constraints. The stress and displacement field distributions under different external loads are obtained through numerical iteration. In some alternative implementations, the structural response model can also integrate a Fluid-Structure Interaction (FSI) module to simulate the real-time response process under the interaction of flood flow and building structure, thereby improving the simulation accuracy of impact flood disasters.

[0055] Multi-strategy combination simulation refers to the process where, after loading the structural response model, the strategy decision-making module inputs flood risk status information and simulation input data into the structural response model to perform parallel simulation calculations on multi-level response strategies. For example, the simulation process may include strategy execution order resolution, parameter combination generation, and constraint evaluation. The optimal solution that simultaneously satisfies the toughness target value, safety threshold, and response time constraint can be found by searching the multi-strategy combination space using a non-dominated sorting genetic algorithm (NSGA-II) or a particle swarm optimization algorithm (PSO). The simulation output can include the structural response parameter set corresponding to each strategy combination, such as maximum displacement, stress distribution, water depth change, and execution time. The system can calculate a comprehensive response score based on these simulation results and select the combination with the highest score as the candidate strategy.

[0056] A target response strategy refers to an executable strategy scheme selected after multiple rounds of strategy combination simulation and constraint calculation. A target response strategy can include control device identification, action sequence, execution duration, and triggering conditions, and can be accompanied by a priority matrix to ensure logical consistency of the execution sequence. For example, if flood risk status information shows that the water depth is rising rapidly and the resilience target value is approaching the safety lower limit, the system can prioritize activating the automatic water blocking and drainage combination strategy, while delaying the triggering of the audible and visual alarms, in order to balance response time and resource consumption.

[0057] In some optional implementations, the strategy decision-making module can also perform multi-objective trade-off calculations based on energy consumption parameters and equipment lifespan parameters recorded in the strategy library to optimize operational energy efficiency while ensuring structural safety. The system can employ hierarchical computational logic, first performing coarse-grained strategy screening, and then performing fine-grained parameter optimization on high-priority schemes to shorten computation time and improve decision stability. Furthermore, the strategy decision-making module can form an information feedback link with the response execution module. When the effect of the target response strategy deviates from expectations, the system can return this deviation data to the strategy library, automatically adjusting the priority weights of relevant strategies, thereby achieving self-learning and strategy evolution during long-term operation.

[0058] The response execution module 150 is communicatively connected to the strategy decision module 140. It is used to call a pre-set flood control response device according to the control command generated by the target response strategy to carry out protection operations on the cultural heritage site, and return the execution feedback result of the flood control response device to the strategy decision module to update the parameters of the strategy library and the structural response model.

[0059] Control commands are logical control sequences output by the strategy decision module and encoded by the command parsing unit, which can be directly executed by the flood response devices. Control commands are typically represented in the form of digital command frames. Each command frame contains a device identifier, execution timing parameters, control variable values, and a feedback request identifier. For example, when the target response strategy requires the activation of a drainage system in a specific area and its operation for 10 minutes, the control command frame will contain the device's unique identifier, activation time, execution duration, and target flow parameters. For automatic floodgate operation, the control command will include lifting height parameters and acceleration limit values. Control commands can be sent to each flood response device via a communication interface, which can be based on the Modbus serial communication protocol, the Controller Area Network (CAN) bus protocol, or the Ethernet protocol.

[0060] Flood response devices refer to electromechanical systems deployed around cultural heritage sites for active protection or auxiliary control. For example, flood response devices may include, but are not limited to, automatic floodgates, pumping units, access control systems, audible and visual alarm devices, and power isolation control modules. This embodiment does not impose any specific limitations on these. Automatic floodgates can be hydraulic lifting, pneumatic telescopic, or gravity-driven mechanisms, achieving rapid closing or opening by receiving control signals. Pumping units can be centrifugal pumps or submersible pumps, automatically adjusting their speed according to a set flow rate. Access control systems are used to cut off visitor or equipment access when the flood risk level increases; their drive mechanism can be an electric actuator or servo motor. Audible and visual alarm devices provide on-site warnings in high-risk scenarios, and power isolation modules automatically cut off power to flooded areas in severe flooding situations to prevent secondary electrical accidents. In alternative embodiments, flood response devices may also include flexible protective dikes or mobile flood barriers for rapid deployment of temporary protective structures at specific locations within the heritage area.

[0061] The execution feedback result refers to the set of status information summarized by the response execution module after the flood control response device has completed its operation. It describes the execution result of the device and the actual system response. The execution feedback result includes data such as device execution success rate, response delay, energy consumption records, and execution deviation. This result is uploaded to the strategy decision module and compared with the prediction results from the model simulation phase to determine the deviation between the strategy execution effect and the model prediction. The feedback deviation can be calculated using a difference function. For example, the difference between the actual drainage volume of the pump unit and the target drainage volume divided by the target value can be defined as the flow deviation rate, and the difference between the actual lifting height of the sluice gate and the preset height can be defined as the execution deviation.

[0062] The strategy library and structural response model parameters are updated through joint analysis of execution feedback results and environmental structural state data. The strategy decision-making module can adjust the model parameter weights using recursive least squares (RLS) or Bayesian update algorithms to make the next decision more closely reflect the actual execution effect. If a specific response strategy is found to be highly efficient and stable in multiple executions, the system will automatically increase its priority in the strategy library; if a strategy combination with large execution deviations occurs, the system will decrease its priority and mark it as a strategy that needs to be re-evaluated, thus forming a dynamic optimization mechanism.

[0063] In some alternative implementations, the response execution module can also analyze the executed image using a visual recognition unit (e.g., a convolutional neural network (CNN)) to automatically identify whether the execution is in place or if there are physical obstacles, thereby further improving the accuracy of the feedback data.

[0064] According to the IoT system for monitoring and early warning of flood resilience in cultural heritage sites in this example embodiment, on the one hand, the resilience assessment module acquires geospatial data and structural information of cultural heritage sites, and performs vulnerability analysis to determine key monitoring points and corresponding resilience target values. This allows the deployment of monitoring points to be more closely integrated with the specific geographical features and structural attributes of the heritage sites, avoiding monitoring blind spots caused by the generalization of layout in traditional methods, thereby improving the targeting and coverage integrity of the monitoring network. Furthermore, the IoT sensing module deploys multi-source sensor nodes at key monitoring points to collect environmental structural status data, which can integrate multiple physical parameters (such as water level, vibration, humidity, etc.) to achieve synchronous acquisition and coordination of multi-source environmental information, overcome the problem of information partiality under a single data collection method, and ensure the comprehensiveness and effectiveness of the data. On the other hand, by performing fusion analysis on multi-source environmental structural state data through the edge analysis module, flood risk characteristics can be extracted in real time and flood risk state information can be generated. This reduces data transmission and processing delays and avoids the inaccurate assessment problems caused by analysis lag or data isolation in traditional systems, thereby enhancing the timeliness and reliability of risk state identification. Simultaneously, the strategy decision module, based on a pre-set multi-level response strategy library and structural response model, combines flood risk state information with simulated input data to perform multi-strategy combination simulations. This allows for dynamic adjustment of response plans to match real-time risk changes, overcoming the rigidity and inefficiency of static threshold decisions and achieving more flexible and optimized protection decisions. Furthermore, the response execution module generates control commands based on the target response strategy to invoke flood control response devices and returns the execution feedback results to the strategy decision module to update the strategy library and model parameters. This forms a closed-loop optimization mechanism, continuously improving system response performance and avoiding the strategy rigidity problem caused by the lack of feedback adjustment in traditional methods, thereby enhancing the adaptability and accuracy of the system in long-term operation.

[0065] The following will provide a further explanation of the IoT system for monitoring and early warning of flood resilience in cultural heritage sites in this example embodiment.

[0066] In one example embodiment of this disclosure, the resilience assessment module includes a vulnerability analysis unit, which is configured to: The structural information of the heritage site is spatially divided into multiple structural grid units. The vulnerability score matrix corresponding to each structural grid unit is determined. The vulnerability score matrix includes the material degradation coefficient, the foundation stability coefficient, and the node connectivity coefficient. The vulnerability score matrix is ​​used as the structural vulnerability data.

[0067] Spatial mesh generation refers to the discretization of cultural heritage buildings or their complex structures according to geometric features and material properties, dividing the entire building into multiple structural mesh units to facilitate subsequent local vulnerability calculations and spatial analysis. The size and shape of the structural mesh units can be determined based on the building's complexity and monitoring accuracy requirements. For example, rectangular or triangular meshes can be used for linear or planar structures such as walls and foundations; non-uniform three-dimensional meshes can be used for curved structures such as domes and arches. This embodiment does not impose any special limitations on this. In one example implementation, the system can perform automated mesh generation based on a finite element mesh generation algorithm, generating a set of structural units with node coordinates and topological connections by inputting a three-dimensional geometric model of the building (from Building Information Modeling (BIM) or Light Detection and Ranging (LiDAR) data). Optionally, the mesh density can be increased in high-vulnerability areas (such as the junction of arches, load-bearing columns, or foundations) and appropriately decreased in low-risk areas (such as decorative components or non-load-bearing walls) to balance calculation accuracy and data volume.

[0068] The material degradation coefficient is a quantitative indicator used to characterize the degree of mechanical property degradation of building materials under long-term environmental exposure. It can be obtained through physical measurements or inversion of historical data. For example, for stone heritage, the degree of weathering can be assessed by combining infrared spectral reflectance with surface hardness testing; for wood or brick heritage, the degree of wood corrosion or brick erosion can be obtained through moisture content sensors and micro-drill resistance testing. The system can convert the measurement results into a coefficient value between 0 and 1 using a normalization function, where a value close to 1 indicates that the material properties remain good, and a value close to 0 indicates significant degradation. Alternatively, the material degradation coefficient can also be estimated using empirical formulas based on environmental exposure parameters (such as humidity, acid rain frequency, and salt spray concentration) to quickly establish assessment models for heritage sites with insufficient data.

[0069] The foundation stability coefficient is a comprehensive parameter used to describe the bearing capacity of a building foundation and the stability characteristics of the soil strata. Foundation stability can be determined through various monitoring methods, such as obtaining dynamic monitoring data through pore water pressure sensors, inclinometers, or settlement sensors installed in the foundation; or obtaining soil shear strength, density, and compression modulus through geotechnical drilling experiments. The system can synthesize the above parameters into a single stability coefficient using a multivariate regression model or the Analytic Hierarchy Process (AHP). The foundation stability coefficient is also expressed using normalization. When the foundation stability is high, the coefficient is close to 1; when there are weak interlayers, high water content, or uneven settlement risks, the coefficient decreases accordingly.

[0070] The node connectivity coefficient is an index used to assess the continuity of force transmission between structural components, reflecting the integrity of a building structure under external loads (such as flood impact and foundation deformation). Node connectivity can be obtained through a structural analysis model. The system can input node stress and strain data into finite element analysis software to calculate the differences in eigenvalues ​​of stress gradients or stiffness matrices between nodes. A connectivity coefficient closer to 1 indicates uniform stress distribution and good structural integrity at the nodes; a significant decrease in the coefficient suggests potential problems such as loose connections, crack propagation, or interrupted force transmission. In an optional implementation, the node connectivity coefficient can also be calculated using the phase difference of vibration signals measured by edge sensing nodes, enabling non-contact dynamic connectivity analysis of the structure.

[0071] A vulnerability score matrix is ​​a two-dimensional or three-dimensional data structure formed by weighting and summing the three coefficients mentioned above according to preset weights. It is used to describe the comprehensive vulnerability intensity of different locations of a building under the influence of multiple parameters. The weights can be set based on expert knowledge and experience or obtained through machine learning model training. For example, using historical disaster damage samples, a multi-layer perceptron (MLP) can be used to calculate the contribution of each coefficient to the probability of structural damage and automatically adjust the weights. Each cell of the vulnerability score matrix corresponds one-to-one with a structural grid cell, and its values, when visualized, form a "vulnerability heatmap," reflecting the risk distribution of different locations. Optionally, the system can perform spatial interpolation on the matrix to generate a continuous vulnerability field for subsequent overlay and matching calculations with geospatial data.

[0072] In some alternative implementations, the system can also incorporate a time dimension, storing vulnerability score matrices from different periods as sequential data to construct a structural vulnerability evolution model. This model can predict future vulnerability trends through time series analysis or Long Short-Term Memory (LSTM) networks, enabling early identification of building aging rates and potential risk points.

[0073] By dividing the space into grids and calculating multi-dimensional parameters of cultural heritage buildings through vulnerability analysis units, the structural safety problem is transformed into a quantifiable and comparable matrix form. Through the collaborative modeling of material degradation coefficient, foundation stability coefficient and node connectivity coefficient, the accurate characterization from the performance of local components to the overall structural safety can be achieved, improving the accuracy and rationality of the optimized layout of key monitoring points and the quantitative calculation of toughness target values.

[0074] In one example embodiment of this disclosure, the resilience assessment module includes a key monitoring point determination unit, which is configured to: The vulnerability score matrix in the structural vulnerability data is matched with the digital elevation model in the geospatial data at the grid cell level to generate spatial matching results that include water catchment area and terrain slope parameters. Based on the spatial matching results, monitoring priority parameters are calculated, and the monitoring priority parameters are correlated with the vulnerability score matrix to generate monitoring weight coefficients for each structural grid cell. Based on the monitoring weight coefficients, key monitoring points are determined, and resilience target values ​​are calculated by combining the vulnerability score matrix of the key monitoring points.

[0075] Grid cell-level matching refers to the process of calculating a one-to-one correspondence between the spatial distribution of structural vulnerability data and the topographic features of geospatial data under a unified coordinate system. Grid cells for structural vulnerability data can be obtained through building structure partitioning, and their coordinate accuracy can be corrected using LiDAR point cloud data or Building Information Modeling (BIM). The digital elevation model (DEM) in the geospatial data provides information on the land surface elevation, slope, and runoff direction. In one embodiment, the system maps both to the same projected coordinate system using a coordinate transformation algorithm (such as a seven-parameter affine transformation or Helmert transformation), and then performs a spatial grid overlay operation. Each structural grid cell finds its corresponding topographic cell in geospatial space, generating a spatial matching result containing topographic slope, runoff area, flow direction vector, and elevation difference. Optionally, to eliminate the impact of resolution differences on matching accuracy, the system can use bilinear interpolation or Kriging interpolation algorithms to resample the DEM, ensuring that the grid size is consistent with the structural vulnerability data.

[0076] The monitoring priority parameter is an indicator used to quantify the necessity and urgency of monitoring different areas of a heritage site under flood risk scenarios. This parameter can be calculated by comprehensively considering factors such as the degree of depression, catchment area, surface slope, and structural vulnerability level. For example, areas with larger catchment areas and gentler slopes are more prone to water accumulation, and their monitoring priority parameter should be higher than that of higher ground or areas with good drainage. Subsequently, the system correlates the monitoring priority parameter with the vulnerability score matrix to generate monitoring weight coefficients. The correlation calculation can be implemented using methods such as linear weighting, fuzzy inference, or Bayesian fusion. Optionally, to avoid weight bias, the system can also use normalization to map all monitoring weight coefficients to the [0, 1] interval for subsequent threshold screening. The spatial distribution of the monitoring weight coefficients is visualized to form a "monitoring importance heatmap," reflecting the sensitivity of different areas in the cultural heritage site to flood risk.

[0077] Key monitoring points can be determined based on the distribution of monitoring weight coefficients. Key monitoring points refer to monitoring locations within the spatial scope of the heritage site that represent high risk and high importance. Their selection process can be completed through a combination of threshold screening and clustering algorithms. For example, a monitoring weight threshold can be set, and monitoring points with a weight coefficient greater than 0.7 are automatically marked as candidate monitoring points. Subsequently, K-Means clustering or density-based spatial clustering of applications with noise (DBSCAN) algorithms are used to spatially aggregate candidate points to avoid overly dense or duplicated monitoring points. The final location of key monitoring points can be adjusted using optimal placement algorithms (such as Voronoi diagram partitioning or greedy optimization algorithms) to achieve maximum spatial coverage and minimum information redundancy with a limited number of nodes. In some alternative implementations, the selection of key monitoring points can also refer to the cultural heritage protection level and architectural functional area information, for example, prioritizing their placement near key protected structures (such as ancient bridge arches, foundation nodes, mural halls, etc.) to achieve differentiated allocation of monitoring resources.

[0078] After identifying key monitoring points, resilience target values ​​can be calculated by combining the vulnerability score matrix corresponding to these monitoring points. The resilience target value is a quantitative indicator reflecting the maximum safe response limit that a building structure can withstand under flood conditions. Its calculation is based on the joint analysis results of the vulnerability intensity and topographic risk level of each monitoring point. When the topographic risk of the monitoring area is high (i.e., the monitoring priority parameter is large), the resilience target value is correspondingly reduced, indicating that the structure has a lower safety margin in that area. The system can adjust the function form according to different building materials or structural forms to reflect the impact of material strength differences on resilience.

[0079] By identifying key monitoring points as units, structural vulnerability and topographic risk can be spatially integrated and calculated, thus achieving a precise mapping from the data layer to the spatial layer. By establishing a quantitative relationship between monitoring priority parameters and weighting coefficients, it is ensured that the monitoring point layout conforms to the laws of topographic hydrodynamics while also taking into account structural vulnerability characteristics. Combined with the calculation of resilience target values, the subsequent system can perform risk identification and strategy decision-making under the constraint of quantitative targets, thereby making the monitoring network targeted, efficient, and adaptive, providing a scientific and systematic deployment basis for the flood resilience protection of cultural heritage sites.

[0080] In one example embodiment of this disclosure, the IoT sensing module includes a data acquisition unit, which is configured to: Multi-source sensor nodes are installed at the corresponding locations based on key monitoring points, and structural monitoring signals are collected using the acquisition interfaces of each multi-source sensor node. The structural monitoring signals include at least settlement signals, crack signals, water accumulation signals, and environmental signals. Time synchronization and noise filtering operations are performed on the structural monitoring signals to generate environmental structural status data.

[0081] The multi-source sensing node can include various types of physical sensor components, signal processing components, power supply components, and wireless communication components. For example, physical sensor components can include settlement sensors for detecting foundation displacement, crack sensors for monitoring changes in crack opening, water level sensors for monitoring surface or groundwater accumulation, and environmental sensors for collecting meteorological factors such as temperature, humidity, and air pressure. Settlement sensors can employ inductive displacement gauges or fiber Bragg grating (FBG) sensors; crack sensors can employ linear variable differential transformer (LVDT) sensors or resistance strain gauges; and water level sensors can employ ultrasonic ranging sensors, pressure level gauges, or radar altimeters. The signal processing component can integrate an analog-to-digital converter (ADC) and a microcontroller unit (MCU) for digitizing and initially filtering the acquired signals. The microcontroller unit can execute multi-level filtering algorithms, such as moving average, median filtering, or wavelet denoising, to eliminate environmental noise and random errors. The processed data is framed with timestamps, node numbers, parameter types, and values, and is then uploaded to the edge analysis module. The wireless communication component enables data transmission between multi-source sensor nodes and the host module, supporting Narrow Band Internet of Things (NB-IoT), Long Range (LoRa), or Wi-Fi protocols. For heritage site scenarios with complex terrain and severe signal obstruction, a relay network mechanism can be optionally employed, achieving wide-area coverage through multi-hop transmission between sub-nodes and the master node. In some alternative implementations, the communication component can also integrate a satellite communication module to ensure continuous transmission of monitoring data in remote areas. The energy supply component can be a standalone energy module to continuously provide operating power to the sensor nodes. For example, the energy module can use a combination of solar panels and lithium batteries, optimizing photovoltaic conversion efficiency through Maximum Power Point Tracking (MPPT) algorithms; or it can use micro-wind or geothermal power units to enhance the system's continuous operation capability in extreme climates. To extend the lifespan of the node, an energy recovery circuit can optionally be included in the energy supply components to store excess electrical energy during sensor dormancy.

[0082] Multi-source sensing nodes trigger synchronous sampling of each sensor at a preset sampling frequency within the acquisition period. To ensure the temporal consistency of multi-source signals, a unified time synchronization protocol (such as Network Time Protocol, NTP, or Precision Time Protocol, PTP) can be used to calibrate the clocks of all nodes, keeping the structural monitoring signals aligned within millisecond-level time errors. The acquired raw structural monitoring signals are processed locally at each node and encapsulated in the form of structured data packets. These packets typically include the acquisition timestamp, sensor number, measured value, confidence interval, and node geographic coordinates. For example, a typical structural monitoring data packet can be represented as: {Node_ID: K1, Time: 2025-10-19 14:00:01, Type: CrackWidth, Value: 0.45mm, Confidence: 98%, Position: (120.36, 31.12)}. This format ensures that the data can be quickly indexed and parsed in the edge analysis module.

[0083] Noise filtering is used to remove random interference and invalid fluctuations from the acquired signals, improving the signal-to-noise ratio and analytical stability. Filtering can be performed at the node or edge level; for example, noise filtering operations may include moving average filtering, Kalman filtering, and median filtering. In one implementation, the system can use a sliding window algorithm at the node level to smooth continuously sampled data in real time, reducing the impact of short-term interference. The edge analysis module can further perform multi-source comparison, using cross-validation of similar signals from different sensor nodes to identify abnormal readings or drift values. Furthermore, for environments with periodic interference (such as electromagnetic waves or mechanical vibration), wavelet transform can be used to decompose the signal in the frequency domain, retaining only low-frequency components related to structural changes, thereby ensuring the accuracy of the monitoring data.

[0084] After time synchronization and noise filtering are completed, multi-source structural monitoring signals can be fused to generate environmental structural state data. Environmental structural state data refers to a comprehensive dataset containing multi-dimensional parameters such as building structural deformation, foundation stability, hydrological conditions, and meteorological environment. This data records the comprehensive physical state of the cultural heritage site at a specific moment in time-series format and can be directly used as input for the edge analysis module. The data structure can be in Time-Series Database (TSDB) format, supporting high-frequency writing and historical backtracking queries.

[0085] By deploying multi-source sensor nodes with multi-parameter acquisition and self-organizing communication capabilities at key monitoring points through data acquisition units, a dynamic monitoring network covering the structure and environmental elements of cultural heritage sites is formed. Through a unified time synchronization, noise filtering and data fusion mechanism, the system can achieve high-precision and low-latency acquisition of environmental structure status information in complex environments, thereby improving the stability, accuracy and effectiveness of environmental structure status data.

[0086] In one example embodiment of this disclosure, the edge analysis module includes a flood risk state prediction unit, which is configured to: Multi-source data comparison and feature parameter extraction are performed on environmental structural status data to determine the risk feature parameter set, which includes at least the settlement change rate, crack growth rate, and average water depth. Risk classification calculation is performed based on the risk feature parameter set to generate flood risk status information.

[0087] Multi-source data comparison refers to the process of aligning and correlating data from different types of sensor nodes in terms of time and space to ensure the comparability and consistency of various signals during analysis. The system first performs spatial registration and temporal interpolation of various signals in the environmental structural status data based on the geographic coordinates and sampling timestamps of the sensor nodes. For example, the system can use a timestamp-based nearest neighbor matching algorithm or a Kalman time synchronization model to dynamically correct the monitoring data from different nodes, enabling different monitoring types (such as settlement and water accumulation) to form a unified data window within the same time period. Through multi-source data comparison, the correlation characteristics between various monitoring signals within the same area can be identified, such as the spatial proximity of foundation settlement points and water accumulation points, and the synchronous changes in crack propagation and rainfall intensity, thus providing data support for subsequent risk feature extraction.

[0088] Feature parameter extraction refers to the quantitative calculation and statistical analysis of dynamic indicators related to flood risk in multi-source data to form a set of risk feature parameters that can be used as input to the model. The system can use the sliding window method or time series analysis to calculate indicators such as settlement change rate, crack growth rate, and average water depth from continuously monitored data. The settlement change rate can be defined as the ratio of the difference in vertical displacement of the foundation or structure between adjacent sampling periods to the time interval. The crack growth rate can be calculated from the change in crack width between adjacent periods, and the average water depth is calculated from the average output of the level sensor over a period of time, reflecting the persistence of flooding. In alternative implementations, extended features such as the rate of change in ambient humidity, surface runoff velocity, or water conductivity can also be extracted to further improve the model's sensitivity to risk evolution.

[0089] Risk classification refers to the process of calculating and judging the flood risk status of cultural heritage sites based on the numerical combination of risk feature parameters. The system can establish a mapping relationship between multidimensional features and risk levels based on fuzzy logic reasoning or Support Vector Machine (SVM) classification models. For example, in fuzzy logic reasoning, the system defines three fuzzy membership functions—"low," "medium," and "high"—for the settlement change rate, crack growth rate, and mean water depth, respectively, and performs comprehensive reasoning through a fuzzy rule base to output the corresponding risk level (e.g., safe, warning, dangerous). In the SVM classification method, the system establishes a risk decision boundary in a high-dimensional feature space using training samples, enabling rapid determination of the risk status. The risk classification output can include risk level labels (e.g., L1-L3 levels), risk confidence scores, and corresponding spatial coordinates.

[0090] Flood risk status information refers to the multidimensional risk representation output by risk classification calculations, used to describe the comprehensive safety status of cultural heritage sites at the current moment. This information includes not only risk level and geographical location, but also weighted results and trend indicators of risk characteristic parameters. For example, the system can mark "the crack growth rate continues to rise and the water depth has not decreased" in the output to indicate that the risk in the area is escalating. In an alternative implementation, the system can also generate a risk heat map, dynamically visualizing the flood risk status information on the geographic information system interface, so that operation and maintenance personnel can grasp the spatial distribution of risks in the heritage site in real time.

[0091] By incorporating a flood risk prediction unit within the edge analysis module, the system extracts features and performs multi-source comparisons on environmental structural status data collected by multi-source sensor nodes. This allows the system to identify real-time trends in key parameters such as settlement rate, crack growth rate, and water depth at the edge. Based on a multi-dimensional feature classification model, it generates flood risk status information, enabling simultaneous identification of structural response and environmental changes. This module transforms risk assessment from single-parameter judgment to multi-source fusion calculation, significantly improving the accuracy and speed of risk identification. Furthermore, by performing data screening and risk classification at the edge, it reduces the computational load on the central processing unit, providing more timely and spatially targeted risk inputs for subsequent strategy decision-making modules. This enables rapid early warning and dynamic assessment of flood conditions in cultural heritage sites.

[0092] In one example embodiment of this disclosure, the simulated input data includes real-time monitoring data collected by multi-source sensing nodes or preset scenario data. The strategy decision module includes an input data generation unit, which is configured to: Preset scenario data is generated based on historical rainfall duration-intensity-frequency data and watershed boundary conditions; and / or combined with structured monitoring data packets collected in real time by multi-source sensor nodes to generate real-time monitoring data including rainfall intensity sequences, water depth sequences and surface runoff parameters.

[0093] Historical rainfall duration-intensity-frequency data refers to statistical model data used to describe the relationship between the intensity and return period of rainfall events of different durations. The system can obtain long-term rainfall records from national meteorological databases or regional meteorological monitoring stations and generate rainfall intensity curves for different return periods (e.g., 2 years, 5 years, 50 years) by fitting probability distributions (such as Pearson Type III distribution or log-normal distribution). In one implementation, the IDF curve can be used as an input template, and different intensities of flood scenarios can be constructed by controlling rainfall duration and return period parameters, such as typical rainstorm patterns like "short-duration high-intensity" or "long-duration medium-intensity".

[0094] Watershed boundary conditions refer to the set of surface hydrological characteristics and topographic parameters that influence rainfall-runoff processes. These conditions can include factors such as catchment area, surface roughness coefficient, slope distribution, river cross-sectional geometry, and drainage system capacity. Surface slope and flow direction can be extracted based on a digital elevation model, and watershed boundaries can be automatically identified using watershed delineation algorithms (such as D8 or D-Infinity algorithms). Simultaneously, land use type and drainage facility distribution information can be extracted from geographic information system (GIS) layers to determine surface runoff coefficients. Combining these factors, the input data generation unit can use hydrological simulation models (such as the unit hydrograph method or the SCS curve number method) to calculate watershed runoff characteristics, ultimately outputting a time-series rainfall-runoff process for constructing preset scenario data.

[0095] Preset scenario data refers to a set of virtual environment inputs generated based on historical statistical information and terrain conditions. These sets are used in the strategy decision-making module to simulate system responses under different types of flooding events. For example, the system can set a scenario of "50-year return period heavy rainfall superimposed with rising tides" to simulate the response performance of heritage site drainage systems under extreme conditions. Each preset scenario dataset may include rainfall intensity sequences, runoff depth sequences, water accumulation evolution curves, and external water level boundary conditions. The time step can be synchronized with the calculation step of the structural response model (e.g., 10 seconds or 1 minute) to ensure the temporal consistency of the simulation input. Alternatively, the system can also introduce a probabilistic scenario generation algorithm to generate multiple possible rainfall combinations through Monte Carlo random sampling, thereby forming a multi-scenario test set.

[0096] Real-time monitoring data refers to environmental structural state data collected by multi-source sensing nodes of the IoT sensing module at key monitoring points, which is then preprocessed and transformed into a set of input parameters that can be directly used by the structural response model. The input data generation unit formats and structures the real-time data, transforming the raw signal into time-series features. For example, rainfall intensity sequences can be obtained by calculating the difference in cumulative rainfall output from rain gauges; water depth sequences can be formed from the time changes in water level height measured by level sensors; and surface runoff parameters can be obtained by inversion from flow velocity sensors or remote sensing image data.

[0097] In some optional implementations, the input data generation unit can also perform spatial interpolation and data resampling operations when generating real-time monitoring data to address the spatial resolution differences caused by uneven node distribution. For example, when data is missing for some monitoring areas, the system can estimate the parameters of the missing points using Kriging interpolation or Radial Basis Function (RBF) interpolation methods; for nodes with inconsistent sampling frequencies, the system can achieve data synchronization through time resampling (such as linear interpolation or spline fitting). Furthermore, the input data generation unit can establish a fusion mechanism between real-time monitoring data and preset scenario data. When the system detects that the actual rainfall intensity is close to a historical extreme scenario, it can automatically fuse the real-time data with the corresponding preset scenario data to form a semi-realistic input dataset, balancing real-time performance and predictability. This fused input can improve the accuracy of simulation predictions, enabling the strategy decision-making module to enter a preventative mode in advance for upcoming high-risk scenarios.

[0098] By combining historical rainfall duration-intensity-frequency data with watershed boundary conditions through the input data generation unit, preset scenario data with realistic hydrodynamic characteristics are generated. At the same time, structured monitoring data collected in real time by multi-source sensor nodes are formatted and fused to construct a simulation input set with consistent time and spatial dimensions. This two-way fusion of historical patterns and real-time monitoring enables the system to have both predictive and real-time capabilities during the strategy simulation phase, ensuring that the computational input of the structural response model under different flood scenarios is more continuous and comparable, thereby improving the accuracy and stability of flood evolution simulation and strategy decision-making.

[0099] In one example embodiment of this disclosure, the strategy decision module includes a simulation unit, which is configured to: The multi-level response strategies in the strategy library are called in the loaded structural response model; the flood risk status information and simulation input data are input into the structural response model to perform multi-strategy combination simulation, and the structural response parameter set corresponding to each strategy combination is output; the constraint calculation is performed on the structural response parameter set to determine the target response strategy that meets the toughness target value.

[0100] The import of simulated input data and flood risk status information can be completed through the input interface of the structural response model. The simulated input data can be a set of environmental variables in time-series format, including rainfall intensity, water depth, and surface runoff velocity; the flood risk status information represents structural state parameters and risk level indicators. After receiving both, the simulation unit introduces them as boundary conditions and load inputs into the structural response model. In a specific embodiment, the system can perform dynamic integration on the structural response model using a matrix solver (such as direct sparse matrix solving or iterative conjugate gradient method) to obtain structural response results under different strategy combinations. The output data corresponding to each strategy combination can include stress distribution matrix, maximum displacement vector, strain energy curve, and nodal reaction force sequence, etc.

[0101] Constraint calculation refers to the process by which, after generating the structural response parameter set, the system filters and optimizes various strategy combinations based on preset toughness target values ​​and operating conditions. Constraints may include structural safety constraints (e.g., stress not exceeding the material yield strength), response time constraints (e.g., the start-up delay of the drainage system not exceeding 5 minutes), and energy consumption constraints (e.g., the maximum power consumption of the flood control device not exceeding a limit). In one embodiment, the system can employ a Non-dominated Sorting Genetic Algorithm II (NSGA-II) for multi-objective optimization calculations. The algorithm uses the comprehensive response score as the fitness function to iteratively filter strategy combinations, generating a Pareto non-dominated solution set, and selecting the optimal strategy that simultaneously satisfies both structural safety and toughness target values. In alternative implementations, the system can also introduce a hierarchical constraint mechanism to enhance the decision-making controllability of the optimization process. For example, priority constraints can ensure that first-level response strategies (e.g., automatic water blocking) are activated in any combination, while third-level strategies (e.g., closure alarms) can only be triggered under high-risk conditions. Through this hierarchical constraint, the system can maintain the rationality of the response logic and the feasibility of execution while optimizing the strategy.

[0102] The target response strategy is the optimal solution set after constraint calculation and filtering, and its corresponding control action has the best balance between structural safety, response timeliness, and energy efficiency. The system can generate a corresponding execution parameter table at output, including the strategy number, control device ID, execution order, and duration, and store it in the strategy library for subsequent protection execution calls. If the system detects that the strategy execution result deviates from expectations during actual operation, it can adjust the strategy parameters through a feedback update mechanism to maintain the self-learning characteristics of the strategy library.

[0103] In some alternative implementations, the simulation unit can also employ a distributed computing architecture, splitting complex simulation tasks into multiple edge nodes for parallel execution to improve computational efficiency. For example, when the cultural heritage site is large or contains multiple independent building units, the simulation tasks of each unit can be computed in parallel, and the results can be aggregated at the central node to achieve real-time strategy generation.

[0104] By loading multi-level response strategies into the structural response model through simulation units, and combining flood risk status information with simulated input data to perform multi-strategy combination simulations, the system can quantitatively compare the structural response characteristics of different strategy combinations in a virtual environment. After constraint calculation, the system outputs target response strategies that meet the resilience target value, realizing a logical closed loop from multi-source risk identification to strategy optimization selection. This process not only enables quantitative screening of strategy effects, but also ensures comprehensive coordination of response actions with structural safety, response timeliness, and energy consumption constraints at the model level, thereby transforming the generation of protection strategies from an empirical process to a data-driven decision optimization process.

[0105] In one example embodiment of this disclosure, the simulation unit is further configured to: The system calls first-level, second-level, and third-level response strategies from the strategy library. The first-level response strategy includes automatic water blocking, the second-level response strategy includes pumping and draining, and the third-level response strategy includes audible and visual alarms and closure. The system also generates a multi-level response strategy using the first-level, second-level, and third-level response strategies, which includes a list of candidate strategy combinations.

[0106] The primary response strategy serves as the system's foundational protection layer, addressing initial or mild flooding risks. Its operation is typically automated, localized, and rapid-response. For example, automatic water-blocking can be performed by a hydraulic lifting gate or a gravity-type flood barrier. When an environmental water level sensor detects a rise in water level exceeding a threshold, the system automatically issues a gate-raising command, elevating the water-blocking structure to a preset height. In an alternative implementation, the automatic water-blocking device can be automatically zoned and controlled based on the distribution of sensor nodes. This means that edge nodes are independently triggered based on local water level changes, achieving a distributed response and reducing overall response delay. The execution parameters of the automatic water-blocking action include lifting height, opening / closing speed, and duration, which can be automatically adjusted based on the mechanical boundary conditions fed back from the structural response model.

[0107] The secondary response strategy is a proactive intervention designed to address moderate flood risks. Its core objective is to restore or maintain the drainage balance of heritage sites through external equipment intervention. Pumping operations can be performed by underground or mobile pump units. The system activates the corresponding pump units based on control commands generated by the strategy decision module, draining accumulated water into designated drainage channels or storage tanks. To ensure efficient water delivery, the system monitors pump speed, current load, and water output parameters in real time and automatically adjusts operating power using closed-loop control algorithms (such as proportional-integral-derivative control, PID). In some embodiments, the secondary response strategy can also be linked with the primary strategy. After automatic water blocking operations are executed, the system automatically detects the water depth in the stagnant area. If the depth remains above a safety threshold, pumping is triggered, thus forming a combined "water blocking + drainage" protection mechanism.

[0108] The Level 3 response strategy serves as the emergency response layer, initiating comprehensive response measures in situations of severe flooding or structural damage risk. Audible and visual alarms can be activated via fixed warning lights, loudspeakers, and mobile broadcasting systems for on-site personnel evacuation and safety warnings. Closure refers to the physical isolation of specific areas (such as visitor passageways, basement entrances, or equipment rooms), which can be implemented using electric access control, retractable partitions, or hydraulic gates. In one embodiment, the Level 3 response strategy can communicate in real-time with the edge analytics module. When a sharp increase in crack growth rate or a foundation settlement rate exceeding a safety threshold is detected, the system immediately triggers audible and visual alarms and closure actions to prevent further structural damage. Alternatively, the Level 3 strategy can also be linked to the power system to perform power isolation or backup power switching in hazardous conditions to ensure the continuous operation of core sensing nodes and communication equipment.

[0109] The generation of multi-level response strategies refers to the process by which the simulation unit, after invoking strategies at different levels, forms multiple combination schemes based on the logical dependencies and execution conditions in the strategy library. The system first reads the strategies at each level corresponding to the risk level from the strategy library through the strategy invocation module, establishing a strategy dependency tree. This tree structure defines the temporal relationships, conditional triggering relationships, and parallel constraints between strategies. For example, the automatic water blocking strategy must be executed before the drainage strategy, while the audible and visual alarm strategy can be executed in parallel with drainage but requires a delayed trigger. The system can encode the strategy dependency tree using a Directed Acyclic Graph (DAG) modeling method to ensure the conflict-free execution logic during simulation. In an optional embodiment, the system uses a combinatorial generation algorithm (such as full permutation screening or a Genetic Algorithm (GA)) to traverse the combination possibilities between strategies at each level, forming a candidate strategy combination list. Each record in the candidate list includes parameters such as strategy combination identifier, execution order, action duration, and expected resilience improvement value. For example, one combination might be "automatic water blocking + drainage + audible and visual alarm", while another combination might be "automatic water blocking + sealing + audible and visual alarm". The system can quantitatively evaluate the response effect of each combination in subsequent simulations.

[0110] The candidate strategy combination list is a key output of the simulation unit, used for subsequent strategy optimization and target response determination. Each candidate combination can be considered a set of simulation test scenarios. The system will input each scenario into the structural response model to perform multi-strategy combination simulations, and calculate the resilience target achievement rate, response delay, and energy consumption indicators based on the output results. In this way, the system can complete virtual drills before strategy execution, enabling predictive evaluation and optimization selection of strategy effectiveness.

[0111] By introducing a multi-level response strategy system (including automatic water blocking, pumping and drainage, audible and visual alarms, and closure operations) into the simulation unit, and using a hierarchical strategy invocation and combination generation mechanism to form a candidate strategy list, the system can dynamically generate multi-level protection combinations under a single risk input. This hierarchical invocation and combination modeling approach enables the system to achieve progressive control logic from local protection to full-domain response based on the risk level. At the same time, virtual drills and feasibility assessments of candidate strategies are conducted during the simulation phase to improve the adaptability and synergy of the strategies.

[0112] In one example embodiment of this disclosure, the response execution module includes a flood response control unit, which is configured to: The system analyzes the device identifier, execution sequence, and parameter settings in the target response strategy; it generates a control command sequence based on the device identifier, execution sequence, and parameter settings, which includes the device instruction code, start time, and execution cycle; it sends the control command sequence to the corresponding flood control response device and records the execution feedback results of the flood control response device.

[0113] Specifically, the system can parse and read the control target list from the target response strategy, extracting key elements such as device ID, action type, parameter settings, and execution sequence. The device ID uniquely identifies the flood control response device installed on-site, such as a sluice gate, drainage pump, audible and visual alarm system, or access control equipment. Parameter settings can represent the action range or operating conditions of each device during execution, such as pump speed (rpm), gate lifting height (mm), or alarm duration (seconds). Execution sequence indicates the order and time delay relationship between device actions; for example, drainage action must be initiated 30 seconds after the automatic sluice gate action is completed.

[0114] The generation of control command sequences can be based on the structured instruction set output by the strategy parsing module, combined with the device communication protocol and execution environment to generate digital control commands. Control command sequences can be organized in a hierarchical structure; for example, the upper layer of the control command sequence consists of strategy-level commands (such as starting the pumping module), the middle layer consists of device group commands (such as starting pump group 3), and the lower layer consists of device-level command frames (such as device command codes, parameters, and timestamps). For example, a control command sequence can be represented as: {Device_ID: “Pump_03”, Command: “Start”, FlowRate: 80 L / s, Start_Time: 14:00:35, Duration: 600 s}. The system can perform syntax and logic checks while generating commands to ensure there are no time or physical resource conflicts between commands. When multiple devices are detected sharing the same power supply or hydraulic pipeline, the system can automatically adjust the execution timing or allocate power priorities to avoid overload caused by simultaneous startup.

[0115] After executing commands, the flood control response device will return execution feedback signals in real time. These signals may include execution status (success, delay, fault), execution parameters (such as actual flow rate, gate displacement), and exception codes. The system receives and stores these signals through the feedback acquisition module and aggregates them to generate execution feedback results. The execution feedback results record may include a timestamp, device ID, the difference between the target parameter and the actual parameter, the action duration, and the execution status flag.

[0116] In an alternative implementation, the flood response control unit can also incorporate real-time monitoring and closed-loop regulation mechanisms. When an execution deviation is detected, the system can immediately correct the control parameters and reissue adjustment commands. For example, when the pump output is lower than the target value, the system can increase the operating frequency or extend the operating time; when the floodgate is not fully closed, the system can resend a "fine-tuning" command until it is fully closed. To prevent mechanical fatigue caused by over-adjustment, the maximum number of adjustments and a safety margin can be set according to the equipment life model.

[0117] By performing device-level analysis, control command generation, and dispatch scheduling of the target response strategy through the flood control response control unit, high-precision conversion from virtual decision results to the physical execution layer can be achieved. The introduction of command verification, communication redundancy, and feedback monitoring mechanisms during the control process ensures high reliability and verifiability of control command transmission and execution. Combined with real-time feedback and automatic correction algorithms, the system forms a closed-loop control link during the action execution phase, effectively reducing the lag in the action of the protection device, execution deviation, and energy consumption redundancy, and ensuring that the strategy execution effect is consistent with the model prediction response.

[0118] In one example embodiment of this disclosure, the response execution module includes a protection policy update unit, which is configured to: The difference between the execution feedback results and the environmental structural state data is calculated to obtain feedback deviation data; the parameter weights of the structural response model are adjusted based on the feedback deviation data, and the priority of each response strategy in the strategy library is updated.

[0119] The difference calculation refers to the process by which the protection strategy update unit, after receiving execution feedback results from the response execution module and environmental structural state data from the IoT sensing module, quantitatively analyzes the numerical differences between the two within the same time window. The execution feedback results reflect the actual state parameters of the flood control response device after executing control commands, such as actual pump flow, gate displacement, or alarm duration; the environmental structural state data describes the environmental changes in the heritage site after the execution of actions, such as changes in water depth, foundation settlement rate, and crack propagation rate. Through difference calculation, the system can identify the degree of deviation between "command execution and environmental response."

[0120] Feedback deviation data is the output of the difference calculation, used to describe the direction and amount of deviation of the actual performance from the model prediction. Feedback deviation data includes not only numerical differences but also trend indicators, such as a lack of significant decrease in water depth (indicating insufficient drainage efficiency) or an increasing crack propagation rate (indicating the water-retaining structure is not fully effective). When recording feedback deviation data, the system can include geographic coordinates, equipment identification, and execution time period to support spatial and temporal analysis. For example, the system can identify specific locational problems such as "lagging pump response in the south drainage ditch area" and "incomplete gate lifting on the north side," providing targeted basis for parameter correction.

[0121] Parameter adjustment refers to the process by which the system dynamically corrects the weights of key parameters in the structural response model based on feedback deviation data. The parameter weights of the structural response model determine the model's sensitivity to different input variables (such as flood pressure, foundation deformation, and flow velocity impact). If the model prediction deviates significantly from the actual measurement, the system will use recursive least squares (RLS) or a Bayesian update algorithm to correct the parameters of the structural response model. In one embodiment, when the actual drainage volume of the pumping operation is lower than the model's expected value, the system automatically increases the weight of the flow resistance coefficient in the model; when the water-blocking device fails to reach the preset gate height, the system adjusts the structural stiffness parameters and execution time constant accordingly.

[0122] Strategy optimization refers to the process by which the system dynamically updates the priority of each response strategy in the strategy library based on the output results of the model after parameter adjustment. The system first recalculates the response performance score of each strategy combination based on the updated structural response model. The scoring dimensions may include resilience target achievement rate, energy efficiency, and response timeliness. Subsequently, the system reorders the strategy priorities in the strategy library based on the scoring results. For example, an automatic water-blocking strategy that performs stably and has a low deviation rate in multiple executions will have its priority increased; while a pumping and drainage strategy that repeatedly exhibits response delays or excessive energy consumption will have its priority reduced or be marked for re-evaluation.

[0123] In optional implementations, the protection strategy update unit can also perform structural optimizations on the strategy library, such as removing strategies with consistently poor performance or introducing new strategy combinations. It can also establish a strategy lifecycle management mechanism based on a historical strategy performance database, statistically analyzing strategy activation frequency, average deviation, and energy consumption indicators to automatically determine whether a strategy has entered a "degradation phase." When a strategy's performance is detected to be declining over a long period, the system will remove it from the high-priority list or replace it with an equivalent strategy regenerated based on the latest parameters.

[0124] The protection strategy update unit calculates the difference between the execution feedback results and the environmental structural status data, generating feedback deviation data. Based on the deviation results, it automatically corrects the structural response model parameters and strategy priorities. This not only gradually improves the model's prediction accuracy but also dynamically optimizes the calling order and weight of each strategy in the strategy library, achieving closed-loop optimization from passive response to continuous learning. The system can automatically adapt to environmental changes and equipment aging during long-term operation, continuously improving the stability and foresight of flood resilience monitoring and protection decisions.

[0125] It should be noted that although several modules or units of the Internet of Things system for monitoring and early warning of flood resilience in cultural heritage sites have been mentioned in the detailed description above, this division is not mandatory. In fact, according to the embodiments of this disclosure, the features and functions of two or more modules or units described above can be embodied in one module or unit. Conversely, the features and functions of one module or unit described above can be further divided and embodied by multiple modules or units.

[0126] Furthermore, this example embodiment also provides a method for monitoring and early warning of flood resilience in cultural heritage sites. (Refer to...) Figure 4 As shown, the flood resilience monitoring and early warning method for cultural heritage sites may specifically include the following steps: Step S410: Obtain geospatial data and heritage site structural information of the cultural heritage site, perform vulnerability analysis on the heritage site structural information to generate structural vulnerability data, and combine the structural vulnerability data and the geospatial data to determine key monitoring points and corresponding resilience target values. Step S420: Collect environmental structural status data by deploying multi-source sensor nodes at the key monitoring points; Step S430: Perform fusion analysis processing on the received environmental structure status data to determine flood risk status information; Step S440: Obtain simulated input data, load a preset structural response model based on a preset strategy library containing multi-level response strategies, and perform multi-strategy combination simulation by combining the flood risk status information and the simulated input data to determine the target response strategy that meets the resilience target value. Step S450: The control command generated according to the target response strategy is used to invoke the preset flood control response device to carry out protection operations on the cultural heritage site, and the execution feedback result of the flood control response device is returned to the strategy decision module to update the parameters of the strategy library and the structural response model.

[0127] The specific details of each step in the above-mentioned method for monitoring and early warning of flood resilience in cultural heritage sites have been described in detail in the corresponding Internet of Things system for monitoring and early warning of flood resilience in cultural heritage sites, so they will not be repeated here.

[0128] It should be noted that although the steps of the method in this disclosure are described in a specific order in the accompanying drawings, this does not require or imply that the steps must be performed in that specific order, or that all the steps shown must be performed to achieve the desired result. Additional or alternative steps may be omitted, multiple steps may be combined into one step, and / or a step may be broken down into multiple steps.

[0129] Furthermore, in an exemplary embodiment of this disclosure, an electronic device is also provided that can implement the above-described Internet of Things method for monitoring and early warning of flood resilience in cultural heritage sites.

[0130] Those skilled in the art will understand that various aspects of this disclosure can be implemented as a system, method, or program product. Therefore, various aspects of this disclosure can be embodied in the following forms: a completely hardware embodiment, a completely software embodiment (including firmware, microcode, etc.), or an embodiment combining hardware and software aspects, collectively referred to herein as a "circuit," "module," or "system."

[0131] The following reference Figure 5 To describe an electronic device 500 according to such an embodiment of the present disclosure. Figure 5 The electronic device 500 shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments disclosed herein.

[0132] like Figure 5 As shown, the electronic device 500 is presented in the form of a general-purpose computing device. The components of the electronic device 500 may include, but are not limited to: at least one processing unit 510, at least one storage unit 520, a bus 530 connecting different system components (including storage unit 520 and processing unit 510), and a display unit 540.

[0133] The storage unit stores program code that can be executed by the processing unit 510, causing the processing unit 510 to perform the steps described in the "Exemplary Methods" section of this specification according to various exemplary embodiments of this disclosure. For example, the processing unit 510 can perform actions such as... Figure 4Step S410 involves acquiring geospatial data and structural information of the cultural heritage site, performing vulnerability analysis on the structural information to generate structural vulnerability data, and combining the structural vulnerability data and the geospatial data to determine key monitoring points and corresponding resilience target values. Step S420 involves collecting environmental structural state data through multi-source sensor nodes deployed at the key monitoring points. Step S430 involves performing fusion analysis on the received environmental structural state data to determine flood risk status information. Step S440 involves acquiring simulated input data, loading a preset structural response model based on a preset strategy library containing multi-level response strategies, and performing multi-strategy combination simulations based on the flood risk status information and the simulated input data to determine a target response strategy that meets the resilience target value. Step S450 involves calling a preset flood control response device according to the control command generated by the target response strategy to perform protective operations on the cultural heritage site, and returning the execution feedback result of the flood control response device to the strategy decision module to update the parameters of the strategy library and the structural response model.

[0134] Storage unit 520 may include readable media in the form of volatile storage units, such as random access memory (RAM) 521 and / or cache memory 522, and may further include read-only memory (ROM) 523.

[0135] Storage unit 520 may also include a program / utility 524 having a set (at least one) program module 525, such program module 525 including but not limited to: operating system, one or more application programs, other program modules and program data, each or some combination of these examples may include an implementation of a network environment.

[0136] Bus 530 can represent one or more of several types of bus structures, including a memory cell bus or memory cell controller, a peripheral bus, a graphics acceleration port, a processing unit, or a local bus using any of the various bus structures.

[0137] Electronic device 500 can also communicate with one or more external devices 570 (e.g., keyboard, pointing device, Bluetooth device, etc.), and with one or more devices that enable a user to interact with electronic device 500, and / or with any device that enables electronic device 500 to communicate with one or more other computing devices (e.g., router, modem, etc.). This communication can be performed via input / output (I / O) interface 550. Furthermore, electronic device 500 can also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN), and / or public networks, such as the Internet) via network adapter 560. As shown, network adapter 560 communicates with other modules of electronic device 500 via bus 530. It should be understood that, although not shown in the figures, other hardware and / or software modules can be used in conjunction with electronic device 500, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0138] From the above description of the embodiments, those skilled in the art will readily understand that the exemplary embodiments described herein can be implemented by software or by combining software with necessary hardware. Therefore, the technical solutions according to the embodiments of this disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, external hard drive, etc.) or on a network, including several instructions to cause a computing device (such as a personal computer, server, terminal device, or network device, etc.) to execute the methods according to the embodiments of this disclosure.

[0139] In exemplary embodiments of this disclosure, a computer-readable storage medium is also provided, on which a program product capable of implementing the methods described above is stored. In some possible embodiments, various aspects of this disclosure may also be implemented as a program product including program code that, when the program product is run on a terminal device, causes the terminal device to perform the steps described in the "Exemplary Methods" section of this specification according to various exemplary embodiments of this disclosure.

[0140] refer to Figure 6 As shown, a program product 600 for implementing the above-described Internet of Things (IoT) system for monitoring and early warning of flood resilience in cultural heritage sites, according to embodiments of the present disclosure, is described. It may employ a portable compact disk read-only memory (CD-ROM) and include program code, and may run on a terminal device, such as a personal computer. However, the program product of this disclosure is not limited thereto. In this document, the readable storage medium may be any tangible medium containing or storing a program that may be used by or in conjunction with an instruction execution system, apparatus, or device.

[0141] The program product may employ any combination of one or more readable media. A readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires, a portable disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0142] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, carrying readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A readable signal medium may also be any readable medium other than a readable storage medium, capable of sending, propagating, or transmitting programs for use by or in conjunction with an instruction execution system, apparatus, or device.

[0143] The program code contained on the readable medium may be transmitted using any suitable medium, including but not limited to wireless, wired, optical fiber, RF, etc., or any suitable combination thereof.

[0144] Program code for performing the operations of this disclosure can be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java and C++, and conventional procedural programming languages ​​such as C or similar languages. The program code can execute entirely on the user's computing device, partially on the user's computing device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0145] Furthermore, the above figures are merely illustrative of the processes included in the method according to exemplary embodiments of this disclosure and are not intended to be limiting. It is readily understood that the processes shown in the above figures do not indicate or limit the temporal order of these processes. Additionally, it is readily understood that these processes may be executed synchronously or asynchronously, for example, in multiple modules.

[0146] From the above description of the embodiments, those skilled in the art will readily understand that the exemplary embodiments described herein can be implemented by software or by combining software with necessary hardware. Therefore, the technical solutions according to the embodiments of this disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, external hard drive, etc.) or on a network, including several instructions to cause a computing device (such as a personal computer, server, touch terminal, or network device, etc.) to execute the methods according to the embodiments of this disclosure.

[0147] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and embodiments are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the claims.

[0148] It should be understood that this disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims.

Claims

1. A flood resilience monitoring and early warning Internet of Things system for cultural heritage sites, characterized in that, include: The resilience assessment module is used to acquire geospatial data and structural information of the cultural heritage site, perform vulnerability analysis on the structural information of the heritage site, generate structural vulnerability data, and combine the structural vulnerability data and the geospatial data to determine key monitoring points and corresponding resilience target values. The Internet of Things (IoT) sensing module is communicatively connected to the resilience assessment module and is used to collect environmental structural status data through multi-source sensing nodes deployed at the key monitoring points. The edge analysis module, which is communicatively connected to the IoT sensing module, is used to perform fusion analysis processing on the received environmental structure status data to determine flood risk status information; The strategy decision module is communicatively connected to the edge analysis module and the IoT sensing module. It is used to acquire simulated input data, load a preset structural response model based on a preset strategy library containing multi-level response strategies, and perform multi-strategy combination simulation in combination with the flood risk status information and the simulated input data to determine the target response strategy that meets the resilience target value. The response execution module is communicatively connected to the strategy decision module. It is used to invoke a pre-set flood control response device according to the control command generated by the target response strategy to carry out protective operations on the cultural heritage site, and to return the execution feedback result of the flood control response device to the strategy decision module to update the parameters of the strategy library and the structural response model.

2. The system according to claim 1, characterized in that, The resilience assessment module includes a vulnerability analysis unit, which is configured as follows: The structural information of the heritage site is subjected to spatial grid division to obtain multiple structural grid units; A vulnerability score matrix is ​​determined for each of the structural grid cells. The vulnerability score matrix includes a material degradation coefficient, a foundation stability coefficient, and a node connectivity coefficient. The vulnerability score matrix is ​​used as the structural vulnerability data.

3. The system according to claim 2, characterized in that, The resilience assessment module includes a key monitoring point determination unit, which is configured to: The vulnerability score matrix in the structural vulnerability data is matched with the digital elevation model in the geospatial data at the grid cell level to generate a spatial matching result that includes water catchment area and terrain slope parameters. Based on the spatial matching results, the monitoring priority parameters are calculated, and the monitoring priority parameters are correlated with the vulnerability score matrix to generate the monitoring weight coefficients for each structural grid unit. Each key monitoring point is determined based on the monitoring weight coefficient, and the resilience target value is calculated by combining the vulnerability score matrix of the key monitoring points.

4. The system according to claim 1, characterized in that, The IoT sensing module includes a data acquisition unit, which is configured to: The multi-source sensor nodes are installed at the corresponding locations according to the key monitoring points, and the structural monitoring signals are collected using the acquisition interface of each multi-source sensor node. The structural monitoring signals include at least settlement signals, crack signals, water accumulation signals and environmental signals. The structure monitoring signal is time-synchronized and noise-filtered to generate environmental structure status data.

5. The system according to claim 1, characterized in that, The edge analysis module includes a flood risk state prediction unit, which is configured as follows: Multi-source data comparison and feature parameter extraction are performed on the environmental structural state data to determine a risk feature parameter set, which includes at least the settlement change rate, crack growth rate and average water depth. Risk classification calculations are performed based on the risk feature parameter set to generate flood risk status information.

6. The system according to claim 1, characterized in that, The simulated input data includes real-time monitoring data or preset scenario data collected by the multi-source sensing nodes. The strategy decision module includes an input data generation unit, which is configured to: The preset scenario data is generated based on historical rainfall duration-intensity-frequency data and watershed boundary conditions; and / or By combining the structured monitoring data packets collected in real time by the multi-source sensing nodes, real-time monitoring data including rainfall intensity sequence, water depth sequence and surface runoff parameters is generated.

7. The system according to claim 1, characterized in that, The strategy decision-making module includes a simulation unit, which is configured as follows: The multi-level response strategy of the strategy library is invoked in the loaded structural response model; The flood risk status information and the simulation input data are input into the structural response model to perform multi-strategy combination simulation, and the structural response parameter set corresponding to each strategy combination is output. Constraint calculations are performed on the set of structural response parameters to determine the target response strategy that satisfies the toughness target value.

8. The system according to claim 7, characterized in that, The simulation unit is also configured to: The first-level response strategy, the second-level response strategy, and the third-level response strategy are respectively called from the strategy library; wherein, the first-level response strategy includes automatic water blocking operation, the second-level response strategy includes pumping and draining operation, and the third-level response strategy includes audible and visual alarm and closure operation; and A multi-level response strategy is generated using the first-level response strategy, the second-level response strategy, and the third-level response strategy. The multi-level response strategy includes a list of candidate strategy combinations.

9. The system according to claim 1, characterized in that, The response execution module includes a flood control response control unit, which is configured to: Analyze the device identifier, execution timing, and parameter settings in the target response strategy; Based on the device identifier, the execution timing, and the parameter settings, a control instruction sequence is generated, which includes a device instruction code, a start time, and an execution cycle. The control command sequence is sent to the corresponding flood control response device, and the execution feedback results of the flood control response device are recorded.

10. The system according to claim 1, characterized in that, The response execution module includes a protection policy update unit, which is configured as follows: The difference between the execution feedback result and the environmental structure state data is calculated to obtain feedback deviation data; The parameter weights of the structural response model are adjusted based on the feedback deviation data, and the priority of each response strategy in the strategy library is updated.

Images