BIM-driven underground pipe gallery asset dynamic management and emergency decision platform

Abstract

The application relates to the technical field of underground pipe gallery asset management, and discloses a BIM-driven underground pipe gallery asset dynamic management and emergency decision-making platform, which comprises the following: a dynamic topological mapping module, which realizes dynamic grid density adjustment of a BIM model by monitoring an environmental parameter change rate, and generates a logical topological network in real time, which is synchronous with a physical pipe gallery structure; an asset state reverse derivation unit, which reversely deduces asset deterioration based on environmental data by using a physical law library loaded dynamically in a time domain; a logic rule self-triggering engine, which constructs a multistage response rule chain containing time and space prediction, and optimizes an emergency plan; and an edge-cloud collaborative architecture, which realizes real-time decision-making with low bandwidth dependence through intelligent caching and asynchronous incremental synchronization. When the dynamic topological mapping is driven by the environmental parameter change rate, the real-time state of the underground pipe gallery can be more finely characterized, and a high-reliability data basis is provided for subsequent accurate analysis and decision-making.

Description

Technical Field

[0001] This invention relates to a BIM-driven dynamic management and emergency decision-making platform for underground utility tunnel assets, belonging to the field of underground utility tunnel asset management technology. Background Technology

[0002] Underground utility tunnels, as an important component of urban infrastructure, are widely used for the centralized layout and management of pipelines, power, and communication facilities. With the advancement of urbanization, the scale and number of underground utility tunnels are constantly increasing, and the complexity of their operation and maintenance is also rising. Traditional underground utility tunnel management mainly relies on manual inspections, fixed-time data transmission, and simple fault warnings, which are insufficient to cope with the increasingly complex operating environment and emergencies.

[0003] Existing underground utility tunnel management technologies are mostly based on BIM (Building Information Modeling) systems, using digital means to model and manage utility tunnel facilities. BIM technology can provide geometric information and structural layout of utility tunnel assets, helping managers understand the overall condition of the tunnel. However, existing BIM models often have the following problems: 1. Existing BIM models rely on manual data feedback and updates at fixed time intervals, failing to reflect the dynamic changes of the utility tunnel in real time, such as sudden leaks or equipment failures. As a result, traditional BIM systems cannot quickly respond to the actual operating conditions of the utility tunnel, resulting in poor timeliness of utility tunnel facility management and an inability to accurately predict faults or perform real-time maintenance. 2. Existing utility tunnel management systems mostly collect environmental data (such as temperature and pressure) through sensors, but this data is mainly used for simple threshold alarms, lacking further data analysis and intelligent reasoning. For example, traditional systems usually cannot establish a correlation between environmental data and the deterioration process of utility tunnel assets, nor can they accurately predict the occurrence of hidden faults. The data utilization rate is low, failing to effectively transform into in-depth analysis of asset health status. 3. In traditional utility tunnel emergency management, once a sudden event occurs (such as equipment failure or pipeline damage), managers often need to manually model and regenerate the emergency response plan. This process is not only time-consuming but also easily affected by human factors, leading to low emergency response efficiency. Industry standards stipulate that the generation time of the emergency response plan should not exceed 15 minutes, but traditional methods often fail to meet this requirement and may even result in decision-making delays.

[0004] Therefore, the technical problem to be solved by the invention is how to reduce the consumption of system computing resources while ensuring high accuracy, improve emergency response efficiency, and solve the data value gap problem in traditional solutions. Summary of the Invention

[0005] This invention provides a BIM-driven dynamic management and emergency decision-making platform for underground utility tunnel assets, the main purpose of which is to solve the problems of excessive resource consumption, insufficient response accuracy and low decision-making efficiency.

[0006] To achieve the above objectives, this invention provides a BIM-driven dynamic management and emergency decision-making platform for underground utility tunnel assets, comprising:

[0007] The dynamic topology mapping module is configured to perform real-time parsing of the BIM model using a lightweight topology parsing algorithm to generate a dynamic logical topology network synchronized with the physical utility tunnel structure. This dynamic logical topology network contains multiple grid cells, each associated with the attributes of a corresponding BIM component and receiving real-time environmental parameters. The lightweight topology parsing algorithm employs a dynamic grid density threshold adjustment mechanism, specifically including: automatically triggering subdivision of the corresponding grid cell when the rate of change of local utility tunnel environmental parameters exceeds a preset critical value; predicting the timing of grid merging based on gradient changes in physical attributes between adjacent grid cells and performing predictive merging; and establishing a dynamic balance model between grid density and edge computing resources.

[0008] The asset status reverse deduction unit is configured to use a preset library of physical laws governing the utility tunnel environment to infer the deterioration process of the utility tunnel assets using real-time environmental data of nodes in the dynamic logical topology network. The physical law library adopts a time-domain dynamic loading mechanism, specifically: dividing the utility tunnel operation cycle into three time domains: at least maintenance period, stable period, and abnormal period; dynamically activating corresponding subsets of physical laws in different time domains; and automatically matching the optimal combination of laws based on the distribution characteristics of the node status in the topology network.

[0009] The logical rule self-triggering engine constructs a multi-level response rule chain bound to the nodes of the dynamic logical topology network. The multi-level response rule chain includes a spatiotemporal prediction module, which is implemented as follows: after the primary rule is triggered, the response effects of similar spatiotemporal scenarios in historical handling cases are automatically retrieved; the propagation path and attenuation coefficient of the handling measures are predicted based on the current topology network connectivity; and the execution order of the contingency plan is dynamically adjusted to match the predicted spatiotemporal impact range.

[0010] The edge-cloud collaborative architecture includes a lightweight rule executor deployed at the edge and a cloud-based BIM model. The edge-side rule executor has a built-in intelligent caching strategy, which specifically includes: establishing a caching priority for handling plans based on the activity of topology nodes, wherein the activity is calculated at least in part by combining the frequency of node parameter changes and historical failure rates; dynamically adjusting the cache space allocation ratio using the rule trigger frequency; automatically reconstructing incremental synchronization data packets based on cache logs after communication is restored; and the edge-side rule executor only asynchronously and incrementally synchronizes the logical derivation results with the cloud-based BIM model to achieve real-time decision-making with low bandwidth dependence.

[0011] Preferably, the construction of the dynamic logical topology network further includes: mapping the pipe gallery components in the BIM model into weighted logical grid cells, wherein the weights are at least partially obtained by dynamically calculating the criticality level of the components and real-time monitoring data; and when a change in the physical pipe gallery structure is detected, only the affected grid cells are recalculated in terms of attributes, while the remaining unchanged areas inherit the topology weights of the adjacent grids; and establishing a dynamic balance model between grid density and edge computing resources to ensure that the peak computing power does not exceed a preset threshold.

[0012] Preferably, the working logic of the asset status reverse deduction unit includes: identifying the spatiotemporal propagation pattern of abnormal areas through the gradient distribution map of real-time environmental parameters; inferring the structural deterioration rate of hidden assets based on the stress-deformation coupling model; and comparing the inference results with the asset design life parameters recorded in the BIM model to generate a remaining life prediction curve.

[0013] Preferably, the optimization strategy of the multi-level response rule chain includes: dynamically adjusting the sensitivity coefficient of the threshold triggering condition based on the historical treatment effect database; adaptively optimizing the response level of the rule chain based on the node density of the topology network; and feeding back the execution result to the rule chain for iterative optimization after the treatment plan is generated.

[0014] Preferably, the edge rule executor has a built-in intelligent caching strategy, which includes: establishing a cache priority for handling plans based on the activity of topology nodes, wherein the activity is calculated at least in part by combining the frequency of node parameter changes and historical failure rate; dynamically adjusting the allocation ratio of cache space based on the trigger frequency of rules, wherein high-frequency rules are allocated at least 60% of the cache resources; and after communication is restored, reconstructing incremental synchronization data packets based on cache logs and prioritizing the synchronization of unexecuted cache instructions.

[0015] Preferably, the asynchronous incremental synchronization mechanism further includes: marking the timestamp and spatial identifier of the changed data field when generating the logical derivation result at the edge end; identifying data conflicts between the cloud and the edge end through a difference comparison algorithm, and prioritizing the use of the data with the latest timestamp for overwriting; and after completing data fusion at the cloud end, issuing an update instruction to the edge end and synchronously updating the model parameters of the physical law library.

[0016] Preferably, the emergency decision-making mode includes: when communication is interrupted, loading locally stored topology snapshot data, the snapshot data containing the grid attributes and environmental parameters of the most recent complete synchronization; executing a preset key handling plan according to cache priority, the key plan being generated from scenarios in historical fault data that occur more frequently than a preset threshold; and after communication is restored, performing consistency verification between local execution logs and cloud data to correct data deviations.

[0017] Preferably, the lightweight topology parsing algorithm employs the following optimization steps: mapping the geometric properties of BIM components to weighted logical grid cells, wherein the weight calculation includes at least the component's material strength, service life, and environmental corrosion coefficient; when the rate of change of local environmental parameters exceeds a preset critical value, performing octree subdivision on the target grid, with the subdivision depth dynamically determined by the gradient value of the rate of change; and merging continuous grid cells with a rate of change lower than a preset threshold based on the physical property similarity of adjacent grids.

[0018] Preferably, the update mechanism of the physical law library includes: receiving model parameter update instructions from the cloud, loading the new version of physical laws after verifying the digital signature; establishing a gray-scale release mechanism for model versions at the edge to gradually replace the old version of the model in batches; and recording the prediction error data during model operation and feeding the prediction error data back to the cloud for subsequent model training.

[0019] Preferably, the construction of the historical case library includes: collecting historical data on the operation and maintenance of the utility tunnel, extracting the spatiotemporal characteristics of the fault scenario and the effect parameters of the handling measures; generating typical case templates through clustering algorithms, and labeling each template with key influencing factors; and when the rule chain is triggered, matching the optimal historical case based on the similarity between the current scenario and the template.

[0020] Compared to the problems described in the background art, the beneficial effects of the present invention are:

[0021] 1. Dynamic topology mapping driven by the rate of change of environmental parameters can more accurately represent the real-time status of underground utility tunnels. Especially when emergencies occur, the model accuracy is improved by orders of magnitude, providing a highly reliable data foundation for subsequent accurate analysis and decision-making.

[0022] 2. The physical law library is dynamically loaded in a time-domain manner, which enables the system to intelligently select and load the most relevant model according to the different operating stages of the utility tunnel (maintenance period, stable period, abnormal period). While ensuring the accuracy of analysis, it significantly reduces the consumption of computing resources and improves the platform's operating efficiency and economy.

[0023] 3. The spatiotemporal prediction module integrated into the logic rule self-triggering engine can combine historical experience with the current status of the utility tunnel to predict the scope and effect of the disposal measures, thereby generating a more optimized emergency plan. This avoids the problem of uncertain disposal effects in complex pipeline networks caused by traditional methods, and improves the efficiency and reliability of emergency response.

[0024] 4. The built-in intelligent caching strategy at the edge takes into account the activity level of topology nodes and the frequency of rule triggering, and realizes priority management of the handling plan and dynamic allocation of cache space. Even in the event of network communication interruption, it can ensure the normal operation of critical handling functions, which greatly enhances the stability of the system and its adaptability in complex environments. Attached Figure Description

[0025] Figure 1 This is a flowchart illustrating the workflow of the dynamic topology mapping module of this invention.

[0026] Figure 2 This is a flowchart of the high-precision network optimization process of the present invention.

[0027] Figure 3 This is a flowchart illustrating the workflow of the present invention based on edge-cloud collaborative optimization.

[0028] Figure 4 This is a flowchart illustrating the construction process of the historical disposal case library of this invention.

[0029] The objectives, features, and advantages of this invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0030] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0031] This application provides a BIM-driven dynamic management and emergency decision-making platform for underground utility tunnel assets, including:

[0032] The dynamic topology mapping module is configured to perform real-time parsing of the BIM model using a lightweight topology parsing algorithm to generate a dynamic logical topology network synchronized with the physical utility tunnel structure. This dynamic logical topology network contains multiple grid cells, each associated with the attributes of a corresponding BIM component and receiving real-time environmental parameters. The lightweight topology parsing algorithm employs a dynamic grid density threshold adjustment mechanism, specifically including: automatically triggering subdivision of the corresponding grid cell when the rate of change of local utility tunnel environmental parameters exceeds a preset critical value; predicting the timing of grid merging based on gradient changes in physical attributes between adjacent grid cells and performing predictive merging; and establishing a dynamic balance model between grid density and edge computing resources.

[0033] The asset status reverse deduction unit is configured to use a preset library of physical laws governing the utility tunnel environment to infer the deterioration process of the utility tunnel assets using real-time environmental data of nodes in the dynamic logical topology network. The physical law library adopts a time-domain dynamic loading mechanism, specifically: dividing the utility tunnel operation cycle into three time domains: at least maintenance period, stable period, and abnormal period; dynamically activating corresponding subsets of physical laws in different time domains; and automatically matching the optimal combination of laws based on the distribution characteristics of the node status in the topology network.

[0034] The logical rule self-triggering engine constructs a multi-level response rule chain bound to the nodes of the dynamic logical topology network. The multi-level response rule chain includes a spatiotemporal prediction module, which is implemented as follows: after the primary rule is triggered, the response effects of similar spatiotemporal scenarios in historical handling cases are automatically retrieved; the propagation path and attenuation coefficient of the handling measures are predicted based on the current topology network connectivity; and the execution order of the contingency plan is dynamically adjusted to match the predicted spatiotemporal impact range.

[0035] The edge-cloud collaborative architecture includes a lightweight rule executor deployed at the edge and a cloud-based BIM model. The edge-side rule executor has a built-in intelligent caching strategy, which specifically includes: establishing a caching priority for handling plans based on the activity of topology nodes, wherein the activity is calculated at least in part by combining the frequency of node parameter changes and historical failure rates; dynamically adjusting the cache space allocation ratio using the rule trigger frequency; automatically reconstructing incremental synchronization data packets based on cache logs after communication is restored; and the edge-side rule executor only asynchronously and incrementally synchronizes the logical derivation results with the cloud-based BIM model to achieve real-time decision-making with low bandwidth dependence.

[0036] Preferably, the construction of the dynamic logical topology network further includes: mapping the pipe gallery components in the BIM model into weighted logical grid cells, wherein the weights are at least partially obtained by dynamically calculating the criticality level of the components and real-time monitoring data; and when a change in the physical pipe gallery structure is detected, only the affected grid cells are recalculated in terms of attributes, while the remaining unchanged areas inherit the topology weights of the adjacent grids; and establishing a dynamic balance model between grid density and edge computing resources to ensure that the peak computing power does not exceed a preset threshold.

[0037] Preferably, the working logic of the asset status reverse deduction unit includes: identifying the spatiotemporal propagation pattern of abnormal areas through the gradient distribution map of real-time environmental parameters; inferring the structural deterioration rate of hidden assets based on the stress-deformation coupling model; and comparing the inference results with the asset design life parameters recorded in the BIM model to generate a remaining life prediction curve.

[0038] Preferably, the optimization strategy of the multi-level response rule chain includes: dynamically adjusting the sensitivity coefficient of the threshold triggering condition based on the historical treatment effect database; adaptively optimizing the response level of the rule chain based on the node density of the topology network; and feeding back the execution result to the rule chain for iterative optimization after the treatment plan is generated.

[0039] Preferably, the edge rule executor has a built-in intelligent caching strategy, which includes: establishing a cache priority for handling plans based on the activity of topology nodes, wherein the activity is calculated at least in part by combining the frequency of node parameter changes and historical failure rate; dynamically adjusting the allocation ratio of cache space based on the trigger frequency of rules, wherein high-frequency rules are allocated at least 60% of the cache resources; and after communication is restored, reconstructing incremental synchronization data packets based on cache logs and prioritizing the synchronization of unexecuted cache instructions.

[0040] Preferably, the asynchronous incremental synchronization mechanism further includes: marking the timestamp and spatial identifier of the changed data field when generating the logical derivation result at the edge end; identifying data conflicts between the cloud and the edge end through a difference comparison algorithm, and prioritizing the use of the data with the latest timestamp for overwriting; and after completing data fusion at the cloud end, issuing an update instruction to the edge end and synchronously updating the model parameters of the physical law library.

[0041] Preferably, the emergency decision-making mode includes: when communication is interrupted, loading locally stored topology snapshot data, the snapshot data containing the grid attributes and environmental parameters of the most recent complete synchronization; executing a preset key handling plan according to cache priority, the key plan being generated from scenarios in historical fault data that occur more frequently than a preset threshold; and after communication is restored, performing consistency verification between local execution logs and cloud data to correct data deviations.

[0042] Preferably, the lightweight topology parsing algorithm employs the following optimization steps: mapping the geometric properties of BIM components to weighted logical grid cells, wherein the weight calculation includes at least the component's material strength, service life, and environmental corrosion coefficient; when the rate of change of local environmental parameters exceeds a preset critical value, performing octree subdivision on the target grid, with the subdivision depth dynamically determined by the gradient value of the rate of change; and merging continuous grid cells with a rate of change lower than a preset threshold based on the physical property similarity of adjacent grids.

[0043] Preferably, the update mechanism of the physical law library includes: receiving model parameter update instructions from the cloud, loading the new version of physical laws after verifying the digital signature; establishing a gray-scale release mechanism for model versions at the edge to gradually replace the old version of the model in batches; and recording the prediction error data during model operation and feeding the prediction error data back to the cloud for subsequent model training.

[0044] Preferably, the construction of the historical case library includes: collecting historical data on the operation and maintenance of the utility tunnel, extracting the spatiotemporal characteristics of the fault scenario and the effect parameters of the handling measures; generating typical case templates through clustering algorithms, and labeling each template with key influencing factors; and when the rule chain is triggered, matching the optimal historical case based on the similarity between the current scenario and the template.

[0045] Example 1: This example describes the specific application of a BIM-driven dynamic management and emergency decision-making platform for underground utility tunnel assets in urban underground utility tunnel management. This technical solution, by introducing dynamic topology mapping, intelligent inference algorithms, and an edge-cloud collaborative architecture, solves the problems of insufficient accuracy, response delay, and excessive resource consumption in traditional utility tunnel management systems when facing emergencies. For example, an underground utility tunnel is undergoing long-term pipeline monitoring and facility maintenance. This tunnel includes multiple important systems such as water supply, power supply, and communication. The operational status of these systems directly affects the normal operation of the city; therefore, the monitoring and maintenance of underground utility tunnels are crucial. Traditional management models rely on periodic manual inspections and real-time data feedback from sensors. While this can provide fault alarms, the response speed and accuracy for emergencies are often insufficient.

[0046] In this embodiment, a BIM-driven dynamic management and emergency decision-making platform for underground utility tunnel assets is deployed. This platform first generates a dynamic logical topology network synchronized with the tunnel structure in real time through a dynamic topology mapping module. Unlike the static characteristics of traditional BIM models, the platform can adjust the density of the topology mesh according to real-time changes in the tunnel environment. For example, assuming a sudden pressure change occurs in a section of water supply pipeline within the tunnel, the platform automatically triggers a subdivision operation of the relevant mesh by detecting a sharp increase in the rate of pressure change, thus improving the accuracy of that area. At this time, the mesh subdivision mechanism can accurately locate the problem area based on the rate of change of environmental parameters and predict the timing of mesh merging through gradient changes in the physical properties of adjacent meshes, thereby achieving accurate and efficient model updates.

[0047] As the dynamic topology mapping unfolds, the asset status reverse deduction unit begins operation. By integrating with a physical law library, this unit uses real-time environmental data from the utility tunnel to deduce the potential asset degradation process. In this scenario, the platform automatically calculates the degradation rate of the water supply pipeline section based on a pre-set physical law library, such as a stress-deformation coupling model. This calculation does not rely on direct monitoring of the asset status by external sensors; instead, it is derived from environmental data of the utility tunnel (such as temperature, humidity, and pressure). This innovation avoids the problem of insufficient sensor deployment density in traditional technologies and significantly reduces reliance on physical sensors, thereby reducing equipment installation and maintenance costs.

[0048] When the system detects signs of deterioration in an area, the logic-based self-triggering engine automatically generates an emergency response plan. Through a spatiotemporal prediction module, combined with similar scenarios from a historical case library, it predicts the effectiveness of the response measures. For example, in the event of a similar fault, the platform can quickly predict the path the problem will take in the utility tunnel, thereby deriving the optimal maintenance sequence and steps. This intelligent prediction avoids the cumbersome steps of manually reconstructing emergency plans required by traditional methods, improving the efficiency and accuracy of emergency response.

[0049] This system not only provides real-time monitoring of utility tunnels under normal circumstances but also continues to perform emergency tasks even when network communication is interrupted. During implementation, the edge-end rule executor dynamically adjusts cache space allocation based on the activity of topology nodes, historical fault data, and rule triggering frequency. For example, when an emergency occurs within the utility tunnel, the system automatically prioritizes caching frequently occurring fault scenarios and their handling solutions, regardless of network bandwidth limitations. Even if communication is interrupted, the system can still execute the most critical emergency response tasks, significantly improving the system's stability and emergency response capabilities in extreme environments.

[0050] Through the above implementations, the platform effectively addresses several pain points in traditional underground utility tunnel management systems. First, the dynamic topology mapping mechanism ensures the tunnel model maintains high accuracy in the face of emergencies, avoiding the problem of lagging model updates common in traditional BIM systems. Second, the asset status reverse derivation unit deduces the asset deterioration process through environmental data, not only avoiding reliance on numerous sensors but also improving the system's accuracy in predicting the health status of utility tunnel assets. Finally, the intelligent emergency response mechanism improves the efficiency of emergency plan generation and reduces the need for manual intervention through spatiotemporal predictions based on historical cases.

[0051] Example 2: In this example, the dynamic topology mapping module uses a lightweight topology parsing algorithm to parse the BIM model in real time, generating a dynamic logical topology network synchronized with the physical utility tunnel structure. Specifically, the module's dynamic grid density threshold adjustment mechanism includes the following steps: Grid subdivision and merging: When the rate of change of environmental parameters exceeds a preset threshold, the corresponding grid cell subdivision operation is automatically triggered. This subdivision operation is based on the rate of change of environmental parameters (such as temperature, humidity, pressure, etc.), using an octree algorithm as the core, to dynamically adjust the grid subdivision depth. The subdivision depth is related to the gradient value of the rate of change, ensuring more refined state description and prediction; Grid merging logic: The gradient change of physical attributes of adjacent grid cells predicts the timing of grid merging. The merging operation is based on the similarity of adjacent grids. If the rate of change is lower than a set threshold, predictive merging is performed to reduce unnecessary computational resource consumption; Dynamic balancing model: During the network topology density adjustment process, the dynamic balance of computational resources at the edge is considered. This balancing model ensures that the allocation of computational resources matches the demand under different grid densities, avoiding problems of excessive or insufficient resource consumption.

[0052] The asset status reverse derivation unit, combined with a physical law database, uses environmental data from nodes in the topology network to infer the degradation process of utility tunnel assets. The unit's workflow includes: Environmental parameter gradient map analysis: The system identifies the spatiotemporal propagation patterns of abnormal areas through real-time environmental parameter gradient distribution maps. This pattern, combined with the physical laws of the utility tunnel (such as stress-deformation coupling models), can accurately derive the degradation rate of hidden assets, avoiding over-reliance on expensive sensor deployments; Temporal separation and subset loading: The physical law database employs a time-domain dynamic loading mechanism, dividing the utility tunnel's operating cycle into maintenance, stable, and abnormal periods. The system loads the most applicable subset of physical laws in different time domains, ensuring efficient use of computing resources while avoiding unnecessary data computation; Lifespan prediction and comparison: By comparing the inferred asset degradation rate with the asset design life parameters recorded in the BIM model, a remaining lifespan prediction curve is generated. This process further provides accurate decision support for the maintenance of utility tunnel assets.

[0053] The logical rule-based self-triggering engine automatically triggers emergency plans by constructing a multi-level response rule chain, and optimizes decisions in conjunction with the spatiotemporal prediction module. The specific implementation path is as follows: Application of the historical case library: The engine automatically retrieves the response effects of similar scenarios by searching the historical case library and combining it with the current status of the utility tunnel and spatiotemporal characteristics. This effect is used to predict the propagation path and attenuation coefficient of the current response measures, and further dynamically adjusts the execution order of the plan to optimize the emergency response strategy; Intelligent adjustment of response levels: Based on the node density of the topology network, the system adaptively adjusts the response levels of the rule chain to ensure that emergency responses at key nodes have higher priority and avoid delays.

[0054] The edge-end rule executor collaborates with the cloud-based BIM model through an intelligent caching strategy to ensure real-time decision-making under low bandwidth conditions. The implementation steps include: dynamic cache priority management: based on the activity level of topology nodes, combined with the historical failure rate and change frequency of the nodes, the system dynamically adjusts the caching priority of contingency plans. Frequently occurring fault scenarios and handling solutions will be cached first, allowing the system to continue executing critical emergency tasks even when the network is disconnected; asynchronous incremental synchronization: after communication is restored, the edge end reconstructs incremental synchronization data packets through cached logs. This process prioritizes synchronizing unexecuted cached instructions to ensure the continuous consistency of the system state and avoid data deviations. These are all extended implementation methods known to those skilled in the art.

[0055] Example 3: This example, in conjunction with the accompanying drawings, further details the workflow and implementation mechanism of the BIM-driven underground utility tunnel asset dynamic management and emergency decision-making platform. Figures 1 to 4 The core technical solutions of the platform and the collaborative working mechanism between its various modules were demonstrated.

[0056] Figure 1 The workflow of the dynamic topology mapping module is demonstrated. First, the system acquires environmental data of the utility tunnel by monitoring environmental parameters in real time and judges the rate of change of these parameters. When the monitored rate exceeds a preset threshold, the system triggers a subdivision operation of the grid cells, thereby refining the accuracy of the BIM model. If the rate does not exceed the threshold, the system continues to monitor the changes in the physical properties of the target grid cells. Next, the system predicts the timing of grid merging based on the gradient changes in physical properties between adjacent grids and executes the merging operation, establishing a dynamic equilibrium model to ensure a balance between edge computing resources and grid density. This generates a dynamic logical topology network synchronized with the physical utility tunnel structure in real time, providing reliable data support for utility tunnel asset management.

[0057] Figure 2 A flowchart of high-precision network optimization is presented. In this process, the system first monitors the rate of change of environmental parameters in real time to determine if it exceeds a preset threshold. If the threshold is exceeded, the system initiates a subdivision trigger operation to perform subdivision calculations of the grid cells, further improving network accuracy. Next, the system performs deep tuning to optimize network accuracy and improve its stability, ultimately generating a high-precision network topology. Through these optimization steps, the system ensures more accurate emergency response and decision support during the dynamic management of utility tunnels.

[0058] Figure 3The diagram illustrates a workflow based on edge-cloud collaborative optimization. When an anomaly is detected at the edge, it reports the event to the cloud. The cloud then returns a corresponding case template based on the received anomaly information. The system enters a local decision-making phase, loading cached contingency plans and combining them with the actual situation, utilizing path influence rules for prediction. During this process, the system continuously loops to optimize decisions until an optimized emergency response command is generated. Finally, the optimized command is issued through the executor, ensuring that the system can still perform critical tasks under low bandwidth or communication difficulties, improving emergency response efficiency and system stability.

[0059] Figure 4 The system demonstrates the flowchart for constructing the historical case study database. First, it collects historical operational data of the utility tunnel and extracts the spatiotemporal characteristics of fault scenarios to provide data support for subsequent fault handling and emergency response. The system also acquires the effectiveness parameters of the handling measures, conducts effectiveness evaluations, and generates case templates. Next, the system generates typical case templates using clustering algorithms and labels each template with key influencing factors. These templates will provide efficient reference for the utility tunnel's emergency response, ensuring that emergency responses can handle emergencies quickly and accurately.

[0060] Example 4: To improve the model's adaptability to dynamic environments, the system's dynamic topology mapping module introduces an adaptive mesh density adjustment mechanism to analyze environmental changes in the BIM model in real time. This mechanism not only triggers mesh subdivision operations when local environmental parameters (such as temperature, humidity, and pressure) change, but also dynamically adjusts the mesh subdivision depth according to the rate of change. The combination of subdivision depth and the gradient of environmental parameter changes ensures a balance between mesh update accuracy and computational resource consumption, avoiding unnecessary waste of computational resources.

[0061] During mesh merging, the system relies on the gradient changes in the physical properties of adjacent meshes for intelligent prediction, thereby merging mesh cells with low change rates at appropriate times. This optimization strategy effectively reduces computational burden and improves real-time response capabilities while ensuring model accuracy. In the derivation and prediction of asset degradation, the system introduces a more precise temporal dynamic loading mechanism, enabling the automatic loading of appropriate subsets of physical laws based on the specific environment and operating status of the utility tunnel during different operational cycles. This mechanism divides the utility tunnel's lifecycle into three time domains: maintenance, stable, and abnormal. The physical law library within each time domain is dynamically activated based on real-time data, ensuring efficient utilization of system computing resources.

[0062] By combining the design life data from the BIM model, the system can reverse-engineer the actual deterioration rate of the utility tunnel assets and generate a remaining life prediction curve. This derivation process relies on accurate analysis of environmental data, such as temperature, humidity, and stress, ensuring high accuracy and low cost in predicting the condition of the utility tunnel assets, while avoiding excessive reliance on large-scale sensor equipment.

[0063] To improve the efficiency and accuracy of emergency response, the system's logical rule-based self-triggering engine incorporates a more refined spatiotemporal prediction mechanism. When a fault or anomaly occurs, the system can automatically generate an optimized emergency plan based on the current status of the utility tunnel and similar scenarios in the historical case database. By combining the spatiotemporal characteristics of historical response cases, the system can predict the scope of impact and propagation path of response measures, thereby dynamically adjusting the execution order and priority to optimize emergency response.

[0064] Furthermore, after the emergency plan is generated, the system automatically feeds back the execution results to the rule chain and performs self-optimization. This mechanism can continuously revise the emergency response strategy based on feedback from real-time environmental data and historical data, ensuring optimal response speed and accuracy.

[0065] Edge computing and cloud-based collaborative architectures play a crucial role in emergency response. In the event of poor network conditions or communication interruptions, the system can continue to perform critical tasks relying on intelligent caching strategies at the edge. The edge rule executor dynamically adjusts cache priorities based on the activity level and historical failure rate of topology nodes, ensuring that high-frequency failure scenarios are handled first.

[0066] Through an asynchronous incremental synchronization mechanism, the edge device performs consistency checks between local execution logs and cloud data after communication is restored, preventing data discrepancies. This process not only optimizes bandwidth utilization but also ensures high availability and stability of the system.

[0067] To ensure the platform maintains excellent performance even under high loads, the system establishes a dynamic balance model between grid density and edge computing resources, adjusting the allocation of computing resources in real time. This model monitors the system's real-time computing demands and resource consumption, automatically optimizing resource allocation during grid subdivision and merging, thus preventing system crashes or response delays caused by computing resource overload. This model allows the system to flexibly adjust computing resource usage according to the actual needs of the utility tunnel, ensuring stable operation under different workloads. These are all extended implementation methods known to those skilled in the art.

[0068] This invention is not limited to the details of the exemplary embodiments described above, and can be implemented in other specific forms without departing from the spirit or essential characteristics of the invention. Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the invention and not to limit it. Although the invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the invention without departing from the spirit and scope of the invention.

Claims

1. A BIM-driven dynamic management and emergency decision-making platform for underground utility tunnel assets, characterized in that, include: The dynamic topology mapping module is configured to perform real-time parsing of the BIM model using a lightweight topology parsing algorithm to generate a dynamic logical topology network synchronized with the physical utility tunnel structure. This dynamic logical topology network contains multiple grid cells, each associated with the attributes of a corresponding BIM component and receiving real-time environmental parameters. The lightweight topology parsing algorithm employs a dynamic grid density threshold adjustment mechanism, specifically including: automatically triggering subdivision of the corresponding grid cell when the rate of change of local utility tunnel environmental parameters exceeds a preset critical value; predicting the timing of grid merging based on gradient changes in physical attributes between adjacent grid cells and performing predictive merging; and establishing a dynamic balance model between grid density and edge computing resources. The asset status reverse deduction unit is configured to use a preset library of physical laws governing the utility tunnel environment to infer the deterioration process of the utility tunnel assets using real-time environmental data of nodes in the dynamic logical topology network. The physical law library adopts a time-domain dynamic loading mechanism, specifically: dividing the utility tunnel operation cycle into three time domains: at least maintenance period, stable period, and abnormal period; dynamically activating corresponding subsets of physical laws in different time domains; and automatically matching the optimal combination of laws based on the distribution characteristics of the node status in the topology network. The logical rule self-triggering engine constructs a multi-level response rule chain bound to the nodes of the dynamic logical topology network. The multi-level response rule chain includes a spatiotemporal prediction module, which is implemented as follows: after the primary rule is triggered, the response effects of similar spatiotemporal scenarios in historical handling cases are automatically retrieved; the propagation path and attenuation coefficient of the handling measures are predicted based on the current topology network connectivity; and the execution order of the contingency plan is dynamically adjusted to match the predicted spatiotemporal impact range. The edge-cloud collaborative architecture includes a rule executor deployed at the edge and a cloud-based BIM model. The edge rule executor has a built-in intelligent caching strategy, which specifically includes: establishing a caching priority for handling plans based on the activity of topology nodes, wherein the activity is calculated at least in part by combining the frequency of node parameter changes and historical failure rates; dynamically adjusting the cache space allocation ratio using the rule trigger frequency; automatically reconstructing incremental synchronization data packets based on cache logs after communication is restored; and the edge rule executor only asynchronously and incrementally synchronizes the logical derivation results with the cloud-based BIM model to achieve real-time decision-making with low bandwidth dependence.

2. The BIM-driven dynamic management and emergency decision-making platform for underground utility tunnel assets according to claim 1, characterized in that, The construction of the dynamic logical topology network further includes: mapping the pipe gallery components in the BIM model into weighted logical grid cells, wherein the weights are at least partially obtained by dynamically calculating the criticality level of the components and real-time monitoring data; and when a change in the physical pipe gallery structure is detected, only the affected grid cells are recalculated in terms of attributes, while the remaining unchanged areas inherit the topology weights of the adjacent grids; and establishing a dynamic balance model between grid density and edge computing resources.

3. The BIM-driven dynamic management and emergency decision-making platform for underground utility tunnel assets according to claim 1, characterized in that, The working logic of the asset status reverse derivation unit includes: identifying the spatiotemporal propagation pattern of abnormal areas through the gradient distribution map of real-time environmental parameters; inferring the structural deterioration rate of hidden assets based on the stress-deformation coupling model; and comparing the inference results with the asset design life parameters recorded in the BIM model to generate a remaining life prediction curve.

4. The BIM-driven dynamic management and emergency decision-making platform for underground utility tunnel assets according to claim 1, characterized in that, The optimization strategy of the multi-level response rule chain includes: dynamically adjusting the sensitivity coefficient of the threshold triggering condition based on the historical treatment effect database; adaptively optimizing the response level of the rule chain based on the node density of the topology network; and feeding back the execution result to the rule chain for iterative optimization after the treatment plan is generated.

5. The BIM-driven dynamic management and emergency decision-making platform for underground utility tunnel assets according to claim 1, characterized in that, The edge rule executor has a built-in intelligent caching strategy, which includes: establishing a cache priority for handling plans based on the activity of topology nodes, wherein the activity is calculated at least in part by combining the frequency of node parameter changes and historical failure rate; dynamically adjusting the allocation ratio of cache space based on the trigger frequency of rules, wherein high-frequency rules are allocated at least 60% of the cache resources; and after communication is restored, reconstructing incremental synchronization data packets based on cache logs and prioritizing the synchronization of unexecuted cache instructions.

6. The BIM-driven dynamic management and emergency decision-making platform for underground utility tunnel assets according to claim 1, characterized in that, The asynchronous incremental synchronization mechanism further includes: when generating logical derivation results at the edge, marking the timestamp and spatial identifier of the changed data fields; identifying data conflicts between the cloud and the edge through a difference comparison algorithm, and prioritizing the use of the data with the latest timestamp for overwriting; and after completing data fusion at the cloud, issuing an update instruction to the edge and synchronously updating the model parameters of the physical law library.

7. The BIM-driven dynamic management and emergency decision-making platform for underground utility tunnel assets according to claim 1, characterized in that, It also includes an emergency decision-making mode, which includes: loading locally stored topology snapshot data when communication is interrupted, the snapshot data containing grid attributes and environmental parameters from the most recent complete synchronization; executing a preset critical handling plan according to cache priority, the critical handling plan being generated from scenarios in historical fault data that occur more frequently than a preset threshold; and after communication is restored, performing consistency verification between local execution logs and cloud data to correct data discrepancies.

8. The BIM-driven dynamic management and emergency decision-making platform for underground utility tunnel assets according to claim 1, characterized in that, The update mechanism of the physical law library includes: receiving model parameter update instructions from the cloud, loading the new version of physical laws after verifying the digital signature; establishing a gray-scale release mechanism for model versions at the edge to gradually replace the old version of the model in batches; and recording the prediction error data during model operation and feeding the prediction error data back to the cloud for subsequent model training.

9. The BIM-driven dynamic management and emergency decision-making platform for underground utility tunnel assets according to claim 1, characterized in that, The construction of the historical case database includes: collecting historical data on the operation and maintenance of utility tunnels, extracting the spatiotemporal characteristics of fault scenarios and the effect parameters of disposal measures; generating typical case templates through clustering algorithms and labeling key influencing factors for each template; and matching the optimal historical case based on the similarity between the current scenario and the template when the rule chain is triggered.

Images