A prefabricated building prefabricated component construction risk real-time evaluation system

Abstract

The present application belongs to the technical field of building construction quality evaluation, and discloses a prefabricated building prefabricated component construction risk real-time evaluation system, which comprises a construction topology module, is used for defining the prefabricated component of a construction site as a state node, constructing an edge coding relationship containing physical contact, load transmission and construction dependence between any state nodes, monotonously activating and updating the edge state corresponding to the construction dependence when the construction stage is transited, and forming a dynamic construction topology graph; an identification and conduction module is used for identifying risk sources and deducing risk conduction chains on the dynamic construction topology graph, when a plurality of risk conduction chains converge at the same node, taking the maximum influence value conduction chain as the main control conduction chain, and the rest of the conduction chains are attenuated and superimposed in descending order of influence value, so as to obtain the superimposed influence value of the node; and a risk evaluation module is used for calculating the component-level risk index according to the superimposed influence value; the accuracy of construction risk identification and the construction safety are remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of building construction quality assessment technology, and more specifically, to a real-time risk assessment system for prefabricated components in prefabricated buildings. Background Technology

[0002] In the construction of prefabricated buildings, the construction sequence and interdependencies of prefabricated components have a significant impact on overall construction safety. Existing technologies typically use a static approach to establish construction topology relationships, meaning that the construction dependencies between components are fixed during modeling and do not dynamically update as construction progresses. This leads to a problem: as the actual construction phases on-site, the topology diagram cannot reflect changes in the dependencies between components in real time, and the dynamic characteristics and stage-specific risks of construction cannot be accurately captured.

[0003] Meanwhile, existing technologies often suffer from the overwriting or loss of historically activated dependency information when updating construction dependencies, resulting in incomplete representations of construction dependencies and hindering subsequent risk transmission analysis. Furthermore, the state of construction dependencies may undergo non-monotonic changes during phase progression; that is, the system may sometimes incorrectly determine dependency rollback or inconsistency, leading to discrepancies between the dynamic construction topology diagram and the actual on-site construction progress. In addition, the lack of a real-time synchronization mechanism between the updating of construction dependencies and the risk analysis model easily causes a disconnect between the topology diagram and risk assessment data, thereby reducing the accuracy and real-time performance of risk analysis.

[0004] On the other hand, at construction sites, risks often converge at the same component node through multiple risk transmission paths. Existing technologies often handle this situation by simply adding or averaging the impact values ​​of each path. This approach fails to differentiate the structural contributions of different risk chains, potentially diluting the impact of critical risk paths and failing to accurately reflect the true sources of critical risks.

[0005] Furthermore, existing technologies lack a mechanism for prioritizing risk transmission chains, making it impossible to clearly determine which path dominates the risk at a node when multiple risk chains converge. This results in unclear risk attribution, further affecting the accuracy of construction intervention decisions. For non-dominant risk chains, existing technologies typically employ equal weighting or linear superposition, ignoring the differences in propagation intensity, structural distance, and impact priority among different risk propagation paths. This can lead to systematic overestimation or underestimation of risks in the superposition results.

[0006] In view of this, the present invention proposes a real-time risk assessment system for prefabricated components in prefabricated buildings to solve the above problems. Summary of the Invention

[0007] To overcome the aforementioned deficiencies of the prior art and to achieve the above objectives, the present invention provides the following technical solution: a real-time construction risk assessment system for prefabricated components of assembled buildings, comprising: The construction topology module is used to define prefabricated components at the construction site as state nodes, construct edge coding relationships between any state nodes including physical contact, load transfer and construction dependence, and monotonically activate and update the edge states corresponding to construction dependence during construction phase transitions to form a dynamic construction topology graph. The risk transmission identification module is used to identify risk sources and deduce risk transmission chains on a dynamic construction topology map. When several risk transmission chains converge at the same node, the transmission chain with the largest impact value is taken as the main control transmission chain, and the remaining transmission chains are superimposed according to the decreasing impact value to obtain the superimposed impact value of the node. The risk assessment module is used to calculate the component-level risk index based on the superimposed impact value and to aggregate and generate the system-level risk index based on the node betweenness centrality. The intervention decision module is used to identify state nodes whose component-level risk index exceeds a preset component-level risk index threshold as candidate intervention nodes, and to perform virtual passivation operations on each candidate intervention node. The risk output module is used to update the system-level risk index after each virtual passivation operation. When the updated system-level risk index reaches the preset shutdown conditions, a construction pause command is triggered, and the risk assessment results are output.

[0008] Preferably, the method for defining prefabricated components at the construction site as state nodes includes: The construction topology module reads the component list at the construction site and assigns a unique identifier to each prefabricated component based on the component list. The unique identifier includes a number, a QR code, or an RFID tag. Collect spatial attribute information and construction stage information for each precast component. Spatial attribute information includes spatial location information, geometric dimensions, and orientation information. The spatial attribute information is obtained through construction drawings, building information model (BIM), or on-site scanning equipment; the construction stage information represents the actual construction status of the corresponding precast component on the construction site, including the status of being transported, awaiting hoisting, being hoisted, completed in place, and completed acceptance. By aggregating unique identifiers, spatial attribute information, and construction stage information, component attribute records are generated. Based on these records, corresponding node objects are created for each precast component, and these node objects serve as the status nodes for each precast component.

[0009] Preferably, the method for constructing the edge encoding relationship includes: By reading construction drawings or building information models (BIM), the spatial adjacency, support, and construction sequence relationships between different prefabricated components are identified, thereby obtaining the relationship types between status nodes; when two prefabricated components have a direct connection, fitting, or fixing relationship after installation, a physical contact relationship is established between the corresponding status nodes. When the load of a precast component needs to be transferred through another precast component, a load transfer relationship is established between the corresponding state nodes; when the construction of a precast component requires the completion of the installation, fixing or acceptance of the preceding component as a prerequisite, a construction dependency relationship is established between the corresponding state nodes. For each relationship type, a corresponding relationship identifier is preset and written into the edge data structure between state nodes. Different relationship types are represented by different identifier codes. When several relationship types exist between two state nodes, several relationship codes are recorded in the corresponding edge data structure to form an edge code relationship that includes physical contact, load transfer and construction dependence.

[0010] Preferably, the method for forming a dynamic construction topology map includes: The construction topology module continuously monitors the construction stage information of each precast component. When the construction stage information of the corresponding precast component changes, the edge state corresponding to the construction dependency relationship is updated. When the corresponding precast component switches from the current construction stage to the next construction stage, the construction dependency relationship corresponding to the precast component is identified, and the newly activated edge state components due to the change in construction stage are determined. Perform a bitwise OR operation between the newly activated edge state component and the edge state before the update to obtain the updated edge state. The updated edge state retains the edge state components that were activated before the update and adds the newly activated edge state components, so that the edge state corresponding to the construction dependency remains monotonically increasing as the construction stage progresses. Based on the updated edge states, the activation states of the construction dependencies between state nodes are updated synchronously, thereby forming a dynamic construction topology graph for the corresponding construction stage.

[0011] Preferably, the method for identifying risk sources and deducing risk transmission chains on a dynamic construction topology map includes: The state nodes in the dynamic construction topology diagram are traversed to identify abnormal construction nodes, lagging construction stage nodes, abnormal load transfer nodes, or nodes whose construction dependencies are blocked, and the corresponding state nodes are identified as risk source nodes. Starting with the risk source node as the initial risk node, the path is traversed along the edges corresponding to the active construction dependencies between state nodes, and the risk propagation direction is determined according to the construction dependency direction, generating a risk transmission path formed by connecting several state nodes in the propagation order. The risk transmission path that meets the preset propagation conditions is continuously tracked, and the set of propagation paths that start from the risk source node, pass through a number of state nodes in sequence, and reach the target state node is determined as the risk transmission chain.

[0012] Preferably, the method for obtaining the superposition influence value of the nodes includes: When several risk transmission chains converge to the same state node, the impact value of each risk transmission chain at the corresponding state node is obtained, and the risk transmission chain with the largest impact value is determined as the main control transmission chain. The risk transmission chains other than the main control chain are sorted in descending order of impact value, and the corresponding impact value is adjusted by decreasing the attenuation according to the ranking. The impact value of the main control chain is superimposed with the impact values ​​of the other risk transmission chains after attenuation correction to obtain the superimposed impact value of the corresponding state node.

[0013] Preferably, the method for generating the system-level risk index includes: The superimposed influence value of the state node is used as the basic risk value of the corresponding precast component. Combined with the construction stage information of the corresponding precast component, the stage correction is performed to obtain the component-level risk index of the corresponding precast component. The betweenness centrality of each state node is calculated based on the dynamic construction topology graph. The betweenness centrality of each state node is used as the aggregation weight to perform weighted aggregation of the risk indices of each component level, thereby generating the system-level risk index.

[0014] Preferably, the method for obtaining the candidate intervention node includes: A preset component-level risk index threshold is set, and the component-level risk index corresponding to each state node in the dynamic construction topology is obtained. The component-level risk index corresponding to each state node is compared with the preset component-level risk index threshold one by one. When the component-level risk index of a certain state node is greater than the preset component-level risk index threshold, the state node is determined as a candidate intervention node.

[0015] Preferably, the method for performing virtual passivation operation on each candidate intervention node includes: Perform a virtual passivation operation on the candidate intervention node. The virtual passivation operation includes temporarily removing all outgoing edges of the corresponding state node of any candidate intervention node in the dynamic construction topology graph to block the risk transmission to other state nodes, while keeping other state nodes and edges in the dynamic construction topology graph unchanged. After completing the virtual passivation process for the current candidate intervention node, the edge encoding relationship of the current candidate intervention node is restored, and the virtual passivation operation is repeated for the next candidate intervention node in turn, until the virtual passivation operation is completed for all candidate intervention nodes.

[0016] Preferably, the method for outputting the risk assessment results includes: After completing the virtual passivation operation of the current candidate intervention node, the updated state node and edge encoding relationship are obtained. Based on the component-level risk index and node betweenness centrality corresponding to each state node, the system-level risk index is aggregated and calculated again to obtain the system-level risk index under the corresponding intervention state. When the updated system-level risk index reaches or exceeds the preset shutdown condition, a construction pause command is triggered to stop the current construction operation and output the corresponding risk assessment result. The preset shutdown condition is that the system-level risk index is greater than or equal to the preset system-level risk index threshold. The risk assessment result includes the corresponding candidate intervention node and the corresponding system-level risk index.

[0017] Compared with the prior art, the present invention has the following beneficial effects: This invention continuously monitors the construction stage information of precast components and updates the state of construction dependency edges, enabling the construction topology to adjust in real time as the construction stage changes, thereby accurately reflecting the dynamic dependency structure at the construction site. By performing a bitwise logical OR operation between the newly activated edge state and the edge state before the update, the historical dependency relationships are fully preserved, while the newly added dependency relationships are superimposed, improving the completeness and reliability of the expression of construction dependency information.

[0018] By ensuring that edge states monotonically increase, the irreversible accumulation of construction dependencies is guaranteed, conforming to the gradual strengthening of dependencies in the actual construction process, thus improving the logical consistency of the dynamic construction topology. By synchronously updating the activation states of construction dependencies between state nodes, the dynamic construction topology is kept consistent with the risk analysis model, providing a reliable data foundation for risk transmission analysis and component-level risk assessment, thereby enhancing the linkage and practicality of the real-time construction risk assessment system.

[0019] By introducing a master risk transmission chain identification mechanism at the same state node, the transmission chain with the largest impact value among multiple risk transmission chains converging at the same state node is identified as the master transmission chain. This highlights the dominant role of key risk propagation paths in node risk and improves the accuracy and clarity of risk source identification. Furthermore, by sorting the remaining risk transmission chains (excluding the master chain) in descending order of impact value and introducing a decreasing attenuation coefficient based on the ranking to correct the corresponding impact values, the impact contribution of different risk transmission chains exhibits a gradient attenuation distribution according to their importance, improving the granularity and robustness of the calculation of the superimposed impact value of nodes.

[0020] By superimposing the impact value of the main risk transmission chain with the impact values ​​of the remaining risk transmission chains after attenuation correction, a comprehensive impact value expression method that includes both the contribution of the dominant risk and the contribution of the secondary risk is formed. This makes the node risk assessment results highlight the key risk propagation path and take into account the comprehensive effect of multiple risk sources, thereby improving the interpretability and engineering applicability of the risk assessment results. Attached Figure Description

[0021] Figure 1 This is a schematic diagram of the structure of a real-time construction risk assessment system for prefabricated building components according to the present invention. Figure 2 This is a schematic diagram of a real-time risk assessment method for prefabricated components in prefabricated buildings according to the present invention. Detailed Implementation

[0022] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0023] Example 1

[0024] Please see Figure 1 As shown, this embodiment provides a real-time risk assessment system for prefabricated components in prefabricated buildings. The system includes a processor, a memory, a communication interface, and a system bus. The construction topology module, identification and transmission module, risk assessment module, intervention decision module, and risk output module are all stored in the memory in the form of computer-executable program instructions and are called and executed by the processor.

[0025] The construction topology module is used to define prefabricated components at the construction site as state nodes, construct edge coding relationships between any state nodes including physical contact, load transfer and construction dependence, and monotonically activate and update the edge states corresponding to construction dependence during construction phase transitions to form a dynamic construction topology graph. The risk transmission identification module is used to identify risk sources and deduce risk transmission chains on a dynamic construction topology map. When several risk transmission chains converge at the same node, the transmission chain with the largest impact value is taken as the main control transmission chain, and the remaining transmission chains are superimposed according to the decreasing impact value to obtain the superimposed impact value of the node. The risk assessment module is used to calculate the component-level risk index based on the superimposed impact value and to aggregate and generate the system-level risk index based on the node betweenness centrality. The intervention decision module is used to identify state nodes whose component-level risk index exceeds a preset component-level risk index threshold as candidate intervention nodes, and to perform virtual passivation operations on each candidate intervention node. The risk output module is used to update the system-level risk index after each virtual passivation operation. When the updated system-level risk index reaches the preset shutdown conditions, a construction pause command is triggered, and the risk assessment results are output.

[0026] The memory establishes a node table, an edge table, an outgoing adjacency table, and an incoming adjacency table. The node table is used to store the unique identifier, spatial attributes, construction stage, and component-level risk index of the state node. The edge table is used to store the edge coding relationship between two state nodes. The outgoing and incoming adjacency tables are used to support forward traversal and reverse lookup of the risk propagation path.

[0027] The edge encoding relationship is stored using a three-bit binary mask. The first bit represents the physical contact relationship, the second bit represents the load transfer relationship, and the third bit represents the construction dependency relationship. The processor reads the specified relationship bits through a bitwise AND operation, writes the newly added active relationship bits through a bitwise OR operation, and determines whether only one of the construction dependency relationship and physical contact relationship exists through an XOR operation. When the third bit is active and the first bit is inactive, the system marks the corresponding edge as a cross-location pure logical dependency edge.

[0028] Construction phase transitions are written to the event queue in the form of event data packets. Each event data packet includes a state node identifier, the construction phase before the transition, the construction phase after the transition, and a timestamp. The processor sequentially reads the event data packets from the event queue, locates the corresponding state node in the node table, and queries the edge table for the construction dependency edges related to that state node. If the construction phase after the transition meets the edge activation conditions, a new activation bit is generated, and a bitwise OR operation is performed between the new activation bit and the original edge code to obtain the updated edge code. To avoid the topology graph being rewritten simultaneously during real-time calculations, the system generates a version number for each updated dynamic construction topology graph and uses the topology graph snapshot as the calculation object for subsequent risk identification and risk assessment.

[0029] During the risk propagation chain simulation, the processor starts from the risk source node, writes the starting point into the traversal queue, and reads the adjacent edges that are in an active state according to the outgoing adjacency list. For each propagable edge, the corresponding adjacent state nodes, the sequence of nodes already visited, the edge sequence, and the current impact value are written into the traversal queue until the terminal node is reached or the preset traversal depth is reached. During the traversal, an access flag is configured for each risk propagation chain, and state nodes that have already appeared in the current chain are not written into the chain again to avoid forming a circular propagation path.

[0030] When several risk transmission chains converge at the same state node, the processor writes the impact values ​​of each transmission chain into a temporary array, sorts the temporary array in descending order of impact value, uses the first impact value as the main control transmission chain impact value, and configures a decreasing attenuation coefficient for the remaining impact values ​​according to their sorting order, thus calculating the superimposed impact value of that state node. This superimposed impact value is written back to the node table as input for the risk assessment module to calculate the component-level risk index.

[0031] In the intervention decision module, the processor writes state nodes whose component-level risk index exceeds a preset component-level risk index threshold into the candidate intervention node set. When performing a virtual passivation operation on any candidate intervention node, the processor does not rewrite the original dynamic construction topology diagram. Instead, it copies a snapshot of the current topology diagram to form a temporary computation graph and temporarily masks all outgoing edges of the candidate intervention node in the temporary computation graph, so that the candidate intervention node no longer outputs risk impact to subsequent state nodes. Subsequently, the risk transmission chain deduction and system-level risk index calculation are re-executed based on the temporary computation graph.

[0032] The processor calculates the risk reduction for each candidate intervention node and generates a priority queue based on the risk reduction amount, from largest to smallest. The risk reduction amount is the difference between the system-level risk index corresponding to the original topology snapshot and the system-level risk index corresponding to the temporary calculation graph. After the calculation is completed, the temporary calculation graph is released, while the original dynamic construction topology remains unchanged.

[0033] The risk output module first determines whether the system-level risk index corresponding to the original topology snapshot has reached the preset system-level risk index threshold. If it has, a construction pause command is triggered directly. If it has not, candidate intervention nodes, risk reduction amount, priority queue, and corresponding system-level risk index are output. If the system-level risk index obtained after temporary calculation of all candidate intervention nodes still reaches the preset system-level risk index threshold, a construction pause command is triggered, and the triggering reason, topology map version number, candidate intervention node identifier, and calculation result are written to the risk assessment log.

[0034] Methods for defining prefabricated components at the construction site as state nodes include: The construction topology module reads the component list at the construction site and assigns a unique identifier to each prefabricated component based on the component list. The unique identifier includes a number, a QR code, or a radio frequency identification (RFID) tag. Collect spatial attribute information and construction stage information for each precast component. Spatial attribute information includes spatial location information, geometric dimensions, and orientation information. The spatial attribute information is obtained through construction drawings, building information model (BIM), or on-site scanning equipment (such as laser scanning or drone aerial photography); the construction stage information represents the actual construction status of the corresponding precast components on the construction site, including the status of transportation, waiting to be hoisted, hoisting, completed placement, and completed acceptance; the actual construction status can be obtained by direct observation on the construction site. By aggregating unique identifiers, spatial attribute information, and construction stage information, component attribute records are generated. Based on these records, corresponding node objects are created for each precast component, and these node objects serve as the status nodes for each precast component.

[0035] In this embodiment, it should be noted that the construction topology module reads the component attribute record of the corresponding precast component and extracts its unique identifier, spatial attribute information, and construction stage information. Subsequently, the construction topology module requests corresponding data storage space in system memory or a database and generates a data entity framework corresponding to the unique index value. This data entity framework includes an index area, a spatial attribute area, and a status area. The construction topology module writes the unique identifier from the component attribute record into the index area, enabling the system to quickly locate the corresponding data entity using the unique identifier; it writes the spatial location, geometric dimensions, and orientation information into the spatial attribute area to characterize the spatial characteristics of the corresponding precast component; and it writes the construction stage information into the status area to characterize the current construction status of the corresponding precast component. After writing is complete, the data entity forms the corresponding node object.

[0036] For example, if there are three precast components A, B, and C at a construction site, the construction topology module first reads the component list from the construction site and assigns a unique identifier to each component: component A is numbered "PC-001", component B is numbered "PC-002", and component C is numbered "PC-003", and each is attached with a QR code or radio frequency identification (RFID) tag. The system then collects the spatial attribute information and construction stage information for each component. Component A's spatial location is (x1, y1, z1), its geometric dimensions are 5 meters long, 2 meters wide, and 0.5 meters high, and its orientation is along the X-axis; the spatial attribute information for components B and C is also obtained accordingly. The construction stage information shows that component A is being hoisted, component B is being transported, and component C has not yet arrived on site.

[0037] Subsequently, a node object is created for each precast component based on the component attribute record. For example, a node object Node-A is generated for component A, and the number "PC-001" is written into the index field, the spatial location, size and orientation information is written into the spatial attribute field, and the construction status "lifting in progress" is written into the status field, thereby forming the status node of the corresponding component A.

[0038] Methods for constructing edge encoding relationships include: By reading construction drawings or building information models (BIM), the spatial adjacency, support, and construction sequence relationships between different prefabricated components are identified, thereby obtaining the relationship types between status nodes; when two prefabricated components have a direct connection, fitting, or fixing relationship after installation, a physical contact relationship is established between the corresponding status nodes. When the load of a precast component needs to be transferred through another precast component, a load transfer relationship is established between the corresponding state nodes; when the construction of a precast component requires the completion of the installation, fixing or acceptance of the preceding component as a prerequisite, a construction dependency relationship is established between the corresponding state nodes. For each relationship type, a corresponding relationship identifier is preset and written into the edge data structure between state nodes. Different relationship types are represented by different identifier codes. When several relationship types exist between two state nodes, several relationship codes are recorded in the corresponding edge data structure to form an edge code relationship that includes physical contact, load transfer and construction dependence.

[0039] For example, in a construction site, precast wall panel A is installed on precast composite slab B. Wall panel A can only be hoisted and installed after composite slab B is fixed in place. The construction topology module determines that there are physical contact, load transfer, and construction dependency relationships between state nodes Node-A and Node-B. Specifically, wall panel A and composite slab B are in direct contact after installation, corresponding to the physical contact relationship; the self-weight of wall panel A is borne by composite slab B, corresponding to the load transfer relationship; and the installation of wall panel A is contingent upon the prior fixing of composite slab B, corresponding to the construction dependency relationship. Subsequently, the construction topology module generates corresponding edge codes between Node-A and Node-B and writes the above three types of relationships into the corresponding edge data structure.

[0040] Methods for generating dynamic construction topology diagrams include: The construction topology module continuously monitors the construction stage information of each precast component. When the construction stage information of the corresponding precast component changes, the edge state corresponding to the construction dependency relationship is updated. When the corresponding precast component switches from the current construction stage to the next construction stage, the construction dependency relationship corresponding to the precast component is identified, and the newly activated edge state components due to the change in construction stage are determined. Perform a bitwise OR operation between the newly activated edge state component and the edge state before the update to obtain the updated edge state. Updated edge state: ;in, Indicates the first stage after the construction phase jump. The state node and the first Updated edge states between state nodes; Indicates the first The state node and the first The edge state vector between each state node is used to represent the construction dependency relationship state between the corresponding prefabricated components. Indicates the index identifier of the state node; Indicates the relationship with the first The state node with a relationship The index identifier of each state node; Indicates the first stage before the jump during the construction phase. The state node and the first The edge states between state nodes; This represents the newly activated edge state components due to changes in the construction phase. Indicates the current construction phase; This indicates the moment after the construction phase transition is completed; This indicates the point in time before the construction phase jump is completed; Indicates the next construction phase; This indicates the stage index identifier corresponding to the current construction stage; This indicates the stage index identifier corresponding to the next construction stage; This indicates that the corresponding precast component has transitioned from the current construction stage to the next construction stage; Indicates that during the construction phase, by Leap to The newly activated edge state components are added at that time; This represents a bitwise logical OR operation, used to perform a logical OR calculation on corresponding bits in the edge state vector; when at least one of the corresponding bits involved in the operation is in an active state, the corresponding bit of the operation result is in an active state; when all the corresponding bits involved in the operation are inactive states, the corresponding bit of the operation result is inactive. The updated edge state retains the edge state components that were activated before the update and adds the newly activated edge state components, so that the edge state corresponding to the construction dependency remains monotonically increasing as the construction stage progresses. Based on the updated edge states, the activation states of the construction dependencies between state nodes are updated synchronously, thereby forming a dynamic construction topology graph for the corresponding construction stage.

[0041] Methods for identifying risk sources and deduce risk transmission chains on dynamic construction topology maps include: The state nodes in the dynamic construction topology diagram are traversed to identify abnormal construction nodes, lagging construction stage nodes, abnormal load transfer nodes, or nodes whose construction dependencies are blocked, and the corresponding state nodes are identified as risk source nodes. In this embodiment, it should be noted that the identification and transmission module reads the construction stage information, edge state information and association information corresponding to each state node in turn according to the connection relationship of the state nodes in the dynamic construction topology graph, and compares the current state with the preset construction state to identify different types of abnormal nodes.

[0042] Among them, construction anomaly nodes are used to characterize the state nodes where the corresponding precast component has an abnormal construction state, specifically including installation offset, incomplete fixing, abnormal connection, or inconsistency between the construction state and the actual construction progress; construction stage lag nodes are used to characterize the state nodes where the current construction stage of the corresponding precast component lags behind the preset planned construction stage; load transfer anomaly nodes are used to characterize the state nodes where there is an abnormal state in the load transfer path of the corresponding precast component, specifically including the absence of support, broken load-bearing relationship, or inactive load transfer edge; construction dependency obstruction nodes are used to characterize the state nodes where the construction dependency conditions of the corresponding precast component are not met, specifically including the situation where the subsequent component cannot continue construction because the previous component has not been installed, fixed, or accepted.

[0043] During the traversal, the identification and transmission module first reads the construction stage information corresponding to the target state node and determines whether the construction stage is consistent with the preset construction plan. If inconsistent, the corresponding state node is marked as a lagging construction stage node. Then, it reads the construction dependencies and load transfer relationships corresponding to the state node. For construction dependencies, it determines whether the corresponding construction dependency edge state meets the activation conditions. If the corresponding construction dependency edge state is not activated, the corresponding state node is marked as a construction dependency obstructed node. For load transfer relationships, it determines whether there are missing supports, load interruptions, or abnormal load-bearing conditions in the corresponding load transfer path. If an abnormal load transfer state exists, the corresponding state node is marked as a load transfer abnormal node. If construction abnormalities such as installation offset, connection abnormalities, or incomplete fixing are detected in the corresponding state node, the corresponding state node is marked as a construction abnormal node.

[0044] For example, in a construction site, precast wall panel A corresponds to state node Node-A, whose preset construction stage is "completed in place," but the current construction stage is still "lifting in progress." The identification and transmission module then marks Node-A as a lagging construction stage node. Simultaneously, the preceding composite slab B corresponding to Node-A has not yet been fixed, causing the edge states of the construction dependency relationship corresponding to Node-A to be inactive. Therefore, the identification and transmission module further marks Node-A as a construction dependency obstruction node. Furthermore, if the system detects that wall panel A has excessive installation tilt, it marks Node-A as a construction anomaly node. Through these methods, the identification and transmission module can identify different types of risk source nodes in the dynamic construction topology map.

[0045] Starting with the risk source node as the initial risk node, the path is traversed along the edges corresponding to the active construction dependencies between state nodes, and the risk propagation direction is determined according to the construction dependency direction, generating a risk transmission path formed by connecting several state nodes in the propagation order. The risk propagation path that meets the preset propagation conditions is continuously tracked, and the set of propagation paths that start from the risk source node, pass through several state nodes in sequence, and reach the target state node is determined as the risk propagation chain. The risk propagation chain includes the initial risk node, intermediate propagation nodes, and terminal impact nodes, which are used to characterize the propagation process of risk in the dynamic construction topology map.

[0046] Methods for obtaining the superposition effect value of nodes include: When several risk transmission chains converge to the same state node, the impact value of each risk transmission chain at the corresponding state node is obtained, and the risk transmission chain with the largest impact value is determined as the main control transmission chain. The influence value of the master control transmission chain is: ;in, This indicates the impact value corresponding to the main control transmission chain; This indicates the operation of finding the maximum value. Indicates the risk transmission chain index; Indicates the first The risk transmission chain is in the first The influence value at each state node; Indicates convergence to the first The total number of risk transmission chains for each state node; The risk transmission chains other than the main control chain are sorted in descending order of impact value, and the corresponding impact value is adjusted by decreasing the attenuation according to the ranking. The impact value of the main control chain is superimposed with the impact values ​​of the other risk transmission chains after attenuation correction to obtain the superimposed impact value of the corresponding state node.

[0047] The cumulative effect of state nodes: ;in, Indicates the first The superimposed impact value of each state node is used to characterize the comprehensive risk impact after multiple risk transmission chains converge at the same state node; This indicates that the main risk transmission chain is excluded, and only the remaining risk transmission chains are calculated; This represents a decreasing attenuation coefficient, used to attenuate and correct the impact value of non-controlling risk transmission chains, and satisfies the following conditions: Indicates the first The risk transmission chains are ranked from largest to smallest according to their impact value, with the main control chain corresponding to... The remaining chains increase sequentially; This represents an exponential decay factor based on the ranking, used to achieve the effect that the influence value gradually decreases as the ranking decreases.

[0048] In this embodiment, it should be noted that the formula for calculating the node superposition impact value is a weighted attenuation summation of the impact values ​​of each risk transmission chain. Each impact value has the same risk dimension, the attenuation coefficient is a dimensionless ratio, and the exponent is a dimensionless integer. Therefore, the dimension of the product remains consistent with the original impact value. The dimensions of all items in the summation operation are consistent, and the dimension of the final superposition impact value is the same as that of a single impact value. Simultaneously, the edge state update formula performs a bitwise OR operation on the logical activation state of the construction-dependent edges. The operation object is a binary logic quantity, whose value is "activated" or "inactive." Therefore, the dimensions of all items within the formula are consistent, and there are no dimensional conflicts. In summary, the dimensions within and between formulas are matched, and there are no issues of dimensional inconsistency or incompatibility.

[0049] Methods for generating system-level risk indices include: The superimposed influence value of the state node is used as the basic risk value of the corresponding precast component. Combined with the construction stage information of the corresponding precast component, the stage correction is performed to obtain the component-level risk index of the corresponding precast component. In this embodiment, it should be noted that in the risk assessment module, the system first uses the superimposed impact value of each state node as the basic risk value of the corresponding precast component. The basic risk value characterizes the comprehensive risk level borne by a single precast component at the current construction stage. Subsequently, the system obtains the construction stage information of the corresponding precast component, such as "in transport", "arrived on site for hoisting", "in hoisting", "positioning completed" or "acceptance completed", and performs stage correction on the basic risk value according to the risk characteristics of each construction stage.

[0050] Stage corrections can be achieved by multiplying by a stage coefficient, where different values ​​can be preset using historical construction data to reflect the sensitivity of construction risks at different stages, thus obtaining the component-level risk index for the corresponding precast component: ;in, Indicates the first The component-level risk index of the prefabricated component corresponding to each state node, that is, the actual risk value after stage correction. This represents the stage correction factor related to the construction stage. The risk sensitivity of precast components differs at different construction stages, therefore... Adjustments were made to the underlying risks; This represents the construction stage index, indicating the current construction stage of the component. For example: Corresponding to "in transit" This corresponds to "entering the site awaiting hoisting".

[0051] The betweenness centrality of each state node is calculated based on the dynamic construction topology graph. The betweenness centrality characterizes the degree to which the corresponding state node is a transit point in the risk propagation path. The shortest paths between all node pairs in the construction topology graph are statistically analyzed, counting those passing through the state nodes. The betweenness centrality of a node can be obtained by measuring the proportion of the number of shortest paths to the total number of shortest paths. The higher the betweenness centrality of a node, the more significant the central role of the corresponding prefabricated component in the risk transmission chain, and the greater the impact of its risk on the system.

[0052] Using the node betweenness centrality corresponding to each state node as the aggregation weight, the risk indices of each component level are weighted and aggregated to generate a system-level risk index.

[0053] System-level risk index: ;in, This represents the system-level risk index, which is obtained by weighted aggregation of the risk indices at the component level. Indicates the total number of state nodes; Indicates the first The betweenness centrality of each state node.

[0054] For example, there are three precast components A, B, and C on a construction site, and their cumulative influence values ​​are respectively... The corresponding coefficients for the construction stage are as follows: (In the process of hoisting) (Arriving at the site for hoisting) (In transit). The component-level risk indices are respectively: Assume the node betweenness centrality is respectively The system-level risk index is: This method enables the system to effectively combine the risk of an individual component with its propagation impact in the construction topology map, forming a comprehensive system risk assessment.

[0055] Methods for obtaining candidate intervention nodes include: A preset component-level risk index threshold is set, and the component-level risk index corresponding to each state node in the dynamic construction topology is obtained. The component-level risk index corresponding to each state node is compared with the preset component-level risk index threshold one by one. When the component-level risk index of a certain state node is greater than the preset component-level risk index threshold, the state node is determined as a candidate intervention node.

[0056] Methods for performing virtual passivation operations on candidate intervention nodes one by one include: Perform a virtual passivation operation on the candidate intervention node. The virtual passivation operation includes temporarily removing all outgoing edges of the corresponding state node of any candidate intervention node in the dynamic construction topology graph to block the risk transmission to other state nodes, while keeping other state nodes and edges in the dynamic construction topology graph unchanged. After completing the virtual passivation process for the current candidate intervention node, the edge encoding relationship of the current candidate intervention node is restored, and the virtual passivation operation is repeated for the next candidate intervention node in turn, until the virtual passivation operation is completed for all candidate intervention nodes.

[0057] For example, suppose there are five prefabricated components A, B, C, D, and E on a construction site, corresponding to state nodes mA, mB, mC, mD, and mE in the dynamic construction topology diagram. The edge coding relationships between components represent construction dependencies and load transfer relationships, for example: mA→mB (B can only be installed after A is installed, and B bears part of the load of A). mB→mC; mC→mD; mD→mE; During the risk assessment process, the system calculates the component-level risk index and finds that the risk indices of components B and D exceed the preset component-level risk index threshold. Therefore, state nodes mB and mD are identified as candidate intervention nodes.

[0058] When performing virtual passivation: Perform virtual passivation on mB: temporarily remove the outgoing edge of mB (i.e., mB→mC) to block the risk transmission from B to C; keep other edges such as mA→mB and mC→mD unchanged; recalculate the system-level risk index under this state to obtain the system-level risk result under the corresponding intervention scenario, and restore the edge encoding relationship of mB (mB→mC).

[0059] Virtual passivation is performed on mD: the outgoing edge of mD is temporarily removed (i.e., mD→mE) to block the risk propagation from D to E; other edges remain unchanged; in this state, the system-level risk index is recalculated, and the edge encoding relationship of mD is restored (mD→mE). Through this method of virtual passivation, the system can independently evaluate the intervention effect of each candidate intervention node, ensuring that other nodes and topological relationships are not disturbed.

[0060] Methods for outputting risk assessment results include: After completing the virtual passivation operation of the current candidate intervention node, the updated state node and edge encoding relationship are obtained. Based on the component-level risk index and node betweenness centrality corresponding to each state node, the system-level risk index is aggregated and calculated again to obtain the system-level risk index under the corresponding intervention state. When the updated system-level risk index reaches or exceeds the preset shutdown condition, a construction pause command is triggered to stop the current construction operation and output the corresponding risk assessment result. The preset shutdown condition is that the system-level risk index is greater than or equal to the preset system-level risk index threshold. The risk assessment result includes the corresponding candidate intervention node and the corresponding system-level risk index.

[0061] The preset component-level risk index threshold is set by staff based on historical data analysis results. This historical analysis process includes the system collecting multiple component-level risk indices and calculating their average value as a reference to obtain the preset component-level risk index threshold. Similarly, the preset system-level risk index threshold is also set and adjusted by staff based on the system's historical operating data and the specific application scenario requirements.

[0062] In this embodiment, by continuously monitoring the construction stage information of prefabricated components and updating the state of construction dependency edges, the construction topology can be adjusted in real time as the construction stage changes, thereby accurately reflecting the dynamic dependency structure at the construction site. By performing a bitwise logical OR operation between the newly added active edge state and the edge state before the update, the historical dependency relationship is fully preserved, while the newly added dependency relationship is superimposed, improving the completeness and reliability of the expression of construction dependency information.

[0063] By ensuring that edge states monotonically increase, the irreversible accumulation of construction dependencies is guaranteed, conforming to the gradual strengthening of dependencies in the actual construction process, thus improving the logical consistency of the dynamic construction topology. By synchronously updating the activation states of construction dependencies between state nodes, the dynamic construction topology is kept consistent with the risk analysis model, providing a reliable data foundation for risk transmission analysis and component-level risk assessment, thereby enhancing the linkage and practicality of the real-time construction risk assessment system.

[0064] By introducing a master risk transmission chain identification mechanism at the same state node, the transmission chain with the largest impact value among multiple risk transmission chains converging at the same state node is identified as the master transmission chain. This highlights the dominant role of key risk propagation paths in node risk and improves the accuracy and clarity of risk source identification. Furthermore, by sorting the remaining risk transmission chains (excluding the master chain) in descending order of impact value and introducing a decreasing attenuation coefficient based on the ranking to correct the corresponding impact values, the impact contribution of different risk transmission chains exhibits a gradient attenuation distribution according to their importance, improving the granularity and robustness of the calculation of the superimposed impact value of nodes.

[0065] By superimposing the impact value of the main risk transmission chain with the impact values ​​of the remaining risk transmission chains after attenuation correction, a comprehensive impact value expression method that includes both the contribution of the dominant risk and the contribution of the secondary risk is formed. This makes the node risk assessment results highlight the key risk propagation path and take into account the comprehensive effect of multiple risk sources, thereby improving the interpretability and engineering applicability of the risk assessment results.

[0066] Example 2 Please see Figure 2 As shown, for parts not described in detail in this embodiment, please refer to the description in Embodiment 1. A method for real-time assessment of construction risks of prefabricated components for prefabricated buildings is provided, including: S1. Define the prefabricated components at the construction site as state nodes. Construct edge coding relationships between any state nodes, including physical contact, load transfer and construction dependence. When transitioning during the construction phase, monotonically activate and update the edge states corresponding to the construction dependence to form a dynamic construction topology graph. S2. Identify risk sources and deduce risk transmission chains on the dynamic construction topology map. When several risk transmission chains converge at the same node, take the transmission chain with the largest impact value as the main control transmission chain, and add the remaining transmission chains according to the decreasing and attenuating impact values ​​to obtain the superimposed impact value of the node. S3. Calculate the component-level risk index based on the superimposed impact value, and aggregate the system-level risk index based on the node betweenness centrality. S4. Select state nodes whose component-level risk index exceeds the preset component-level risk index threshold as candidate intervention nodes, and perform virtual passivation operation on each candidate intervention node. S5. Update the system-level risk index after each virtual passivation operation. When the updated system-level risk index reaches the preset shutdown conditions, trigger the construction pause command and output the risk assessment results.

[0067] Example 3 This embodiment discloses an electronic device, including a memory, a processor, and a computer program stored in the memory and capable of running on the processor. When the processor executes the computer program, it implements the operation mode of the real-time construction risk assessment system for prefabricated building components provided above.

[0068] Since the electronic device described in this embodiment is the one used in implementing the real-time risk assessment system for prefabricated building construction based on the embodiments of this application, those skilled in the art can understand the specific implementation methods and various variations of the electronic device in this embodiment based on the real-time risk assessment system for prefabricated building construction based on the embodiments of this application. Therefore, how the electronic device implements the method in the embodiments of this application will not be described in detail here. Any electronic device used by those skilled in the art in implementing the real-time risk assessment system for prefabricated building construction based on the embodiments of this application falls within the scope of protection of this application.

[0069] It should be noted that all formulas in this manual are calculated by removing dimensions and taking their numerical values. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters and thresholds in the formulas are set by those skilled in the art according to the actual situation.

[0070] The above description is merely a preferred embodiment of the present invention, and the scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for users of ordinary technical skills, any improvements and modifications made without departing from the principles of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. A real-time risk assessment system for prefabricated components in prefabricated buildings, characterized in that, include: The construction topology module is used to define prefabricated components at the construction site as state nodes, construct edge coding relationships between any state nodes including physical contact, load transfer and construction dependence, and monotonically activate and update the edge states corresponding to construction dependence during construction phase transitions to form a dynamic construction topology graph. The risk transmission identification module is used to identify risk sources and deduce risk transmission chains on a dynamic construction topology map. When several risk transmission chains converge at the same node, the transmission chain with the largest impact value is taken as the main control transmission chain. The remaining risk transmission chains other than the main control chain are arranged in descending order of their impact values, and the corresponding impact values ​​are adjusted by decreasing attenuation according to the ranking. The impact value corresponding to the main control chain is superimposed with the impact values ​​of the remaining risk transmission chains after attenuation correction to obtain the superimposed impact value of the corresponding state node. The risk assessment module is used to calculate the component-level risk index based on the superimposed impact value and to aggregate and generate the system-level risk index based on the node betweenness centrality. The intervention decision module is used to identify state nodes whose component-level risk index exceeds a preset component-level risk index threshold as candidate intervention nodes, and to perform virtual passivation operations on each candidate intervention node. The method for performing virtual passivation operation on each candidate intervention node includes: Perform a virtual passivation operation on the candidate intervention node. The virtual passivation operation includes temporarily removing all outgoing edges of the corresponding state node of any candidate intervention node in the dynamic construction topology graph to block the risk transmission to other state nodes, while keeping other state nodes and edges in the dynamic construction topology graph unchanged. After completing the virtual passivation process for the current candidate intervention node, restore the edge encoding relationship of the current candidate intervention node, and repeat the virtual passivation operation for the next candidate intervention node in turn, until the virtual passivation operation is completed for all candidate intervention nodes. The risk output module is used to update the system-level risk index after each virtual passivation operation. When the updated system-level risk index reaches the preset shutdown conditions, a construction pause command is triggered, and the risk assessment results are output.

2. The real-time construction risk assessment system for prefabricated components of assembled buildings according to claim 1, characterized in that, The method for defining prefabricated components at the construction site as state nodes includes: The construction topology module reads the component list at the construction site and assigns a unique identifier to each prefabricated component based on the component list. The unique identifier includes a number, a QR code, or an RFID tag. Collect spatial attribute information and construction stage information for each precast component. Spatial attribute information includes spatial location information, geometric dimensions, and orientation information. The spatial attribute information is obtained through construction drawings, building information model (BIM), or on-site scanning equipment; the construction stage information represents the actual construction status of the corresponding precast component on the construction site, including the status of being transported, awaiting hoisting, being hoisted, completed in place, and completed acceptance. By aggregating unique identifiers, spatial attribute information, and construction stage information, component attribute records are generated. Based on these records, corresponding node objects are created for each precast component, and these node objects serve as the status nodes for each precast component.

3. The real-time construction risk assessment system for prefabricated components of assembled buildings according to claim 2, characterized in that, The method for constructing the edge encoding relationship includes: By reading construction drawings or building information models (BIM), the spatial adjacency, support, and construction sequence relationships between different prefabricated components are identified, thereby obtaining the relationship types between status nodes; when two prefabricated components have a direct connection, fitting, or fixing relationship after installation, a physical contact relationship is established between the corresponding status nodes. When the load of a precast component needs to be transferred through another precast component, a load transfer relationship is established between the corresponding state nodes; when the construction of a precast component requires the completion of the installation, fixing or acceptance of the preceding component as a prerequisite, a construction dependency relationship is established between the corresponding state nodes. For each relationship type, a corresponding relationship identifier is preset and written into the edge data structure between state nodes. Different relationship types are represented by different identifier codes. When several relationship types exist between two state nodes, several relationship codes are recorded in the corresponding edge data structure to form an edge code relationship that includes physical contact, load transfer and construction dependence.

4. The real-time construction risk assessment system for prefabricated components of assembled buildings according to claim 3, characterized in that, The method for generating a dynamic construction topology map includes: The construction topology module continuously monitors the construction stage information of each precast component. When the construction stage information of the corresponding precast component changes, the edge state corresponding to the construction dependency relationship is updated. When the corresponding precast component switches from the current construction stage to the next construction stage, the construction dependency relationship corresponding to the precast component is identified, and the newly activated edge state components due to the change in construction stage are determined. Perform a bitwise OR operation between the newly activated edge state component and the edge state before the update to obtain the updated edge state. The updated edge state retains the edge state components that were activated before the update and adds the newly activated edge state components, so that the edge state corresponding to the construction dependency remains monotonically increasing as the construction stage progresses. Based on the updated edge states, the activation states of the construction dependencies between state nodes are updated synchronously, thereby forming a dynamic construction topology graph for the corresponding construction stage.

5. The real-time construction risk assessment system for prefabricated components of assembled buildings according to claim 4, characterized in that, The method for identifying risk sources and deducing risk transmission chains on a dynamic construction topology map includes: The state nodes in the dynamic construction topology diagram are traversed to identify abnormal construction nodes, lagging construction stage nodes, abnormal load transfer nodes, or nodes whose construction dependencies are blocked, and the corresponding state nodes are identified as risk source nodes. Starting with the risk source node as the initial risk node, the path is traversed along the edges corresponding to the active construction dependencies between state nodes, and the risk propagation direction is determined according to the construction dependency direction, generating a risk transmission path formed by connecting several state nodes in the propagation order. The risk transmission path that meets the preset propagation conditions is continuously tracked, and the set of propagation paths that start from the risk source node, pass through a number of state nodes in sequence, and reach the target state node is determined as the risk transmission chain.

6. The real-time construction risk assessment system for prefabricated components of assembled buildings according to claim 5, characterized in that, The method for obtaining the superimposed influence value of the corresponding state node includes: When several risk transmission chains converge to the same state node, the impact value of each risk transmission chain at the corresponding state node is obtained, and the risk transmission chain with the largest impact value is determined as the main control transmission chain. The risk transmission chains other than the main control chain are sorted in descending order of impact value, and the corresponding impact value is adjusted by decreasing the attenuation according to the ranking. The impact value of the main control chain is superimposed with the impact values ​​of the other risk transmission chains after attenuation correction to obtain the superimposed impact value of the corresponding state node.

7. A real-time construction risk assessment system for prefabricated components of assembled buildings according to claim 6, characterized in that, The method for generating the system-level risk index includes: The superimposed influence value of the state node is used as the basic risk value of the corresponding precast component. Combined with the construction stage information of the corresponding precast component, the stage correction is performed to obtain the component-level risk index of the corresponding precast component. The betweenness centrality of each state node is calculated based on the dynamic construction topology graph. The betweenness centrality of each state node is used as the aggregation weight to perform weighted aggregation of the risk indices of each component level, thereby generating the system-level risk index.

8. A real-time construction risk assessment system for prefabricated components of assembled buildings according to claim 7, characterized in that, The method for obtaining the candidate intervention nodes includes: A preset component-level risk index threshold is set, and the component-level risk index corresponding to each state node in the dynamic construction topology is obtained. The component-level risk index corresponding to each state node is compared with the preset component-level risk index threshold one by one. When the component-level risk index of a certain state node is greater than the preset component-level risk index threshold, the state node is determined as a candidate intervention node.

9. A real-time construction risk assessment system for prefabricated components of assembled buildings according to claim 8, characterized in that, The methods for outputting risk assessment results include: After completing the virtual passivation operation of the current candidate intervention node, the updated state node and edge encoding relationship are obtained. Based on the component-level risk index and node betweenness centrality corresponding to each state node, the system-level risk index is aggregated and calculated again to obtain the system-level risk index under the corresponding intervention state. When the updated system-level risk index reaches or exceeds the preset shutdown condition, a construction pause command is triggered to stop the current construction operation and output the corresponding risk assessment result. The preset shutdown condition is that the system-level risk index is greater than or equal to the preset system-level risk index threshold. The risk assessment result includes the corresponding candidate intervention node and the corresponding system-level risk index.

Images