A building engineering construction quality evaluation method based on big data

Abstract

The application discloses a kind of based on big data's building engineering construction quality evaluation method, it is related to building engineering quality evaluation technical field, including, from the abnormal construction behavior corresponding construction risk object of construction instability area is extracted, and according to the construction association between different construction risk objects, the risk propagation edge between different quality risk nodes is established, and quality risk dynamic propagation atlas is formed;Risk diffusion tracking is carried out to each initial quality risk node in quality risk dynamic propagation atlas, generates quality risk propagation chain, and generates risk propagation potential energy and risk aggregation index according to the risk propagation parameter in quality risk propagation chain and construction behavior entropy deviation degree;Risk propagation potential energy, risk aggregation index and construction behavior entropy deviation degree are normalized after weighting fusion, and building engineering construction quality evaluation information is formed.The application realizes the dynamic tracking of quality risk propagation process.

Description

Technical Field

[0001] This invention relates to the field of building engineering quality evaluation technology, and in particular to a building engineering construction quality evaluation method based on big data. Background Technology

[0002] As construction projects expand in scale and become more complex, quality management is increasingly moving towards digitalization. Existing quality evaluation technologies typically analyze construction quality based on construction inspection data, structural monitoring data, and BIM model data, combining IoT, big data analytics, and cloud computing technologies to achieve quality monitoring and risk warning. While some existing solutions are beginning to utilize data mining, machine learning, and big data analytics to identify abnormal construction behaviors and quality risks, most still focus on static quality indicator analysis, lacking dynamic analysis of changes in construction behavior and the propagation of quality risks.

[0003] Existing construction engineering quality evaluation technologies have the following shortcomings: First, they lack dynamic analysis of the stability of construction behavior, making it difficult to identify the impact of chaotic construction organization, frequent rework, and abnormal process switching on project quality; second, they lack the ability to analyze the transmission relationship and diffusion process of quality risks, making it difficult to identify the dynamic transmission process of quality risks between different processes and construction areas, thus resulting in insufficient quality risk early warning capabilities. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides a construction quality evaluation method based on big data to solve the problems of difficulty in identifying dynamic instability of construction behavior and difficulty in analyzing the process of quality risk propagation in existing technologies.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: This invention provides a construction quality evaluation method for building engineering based on big data, comprising: acquiring construction process data and performing time synchronization, spatial coordinate mapping, and process association processing on the construction process data to form a dynamic event set of construction behavior; classifying and statistically analyzing different construction behaviors in the dynamic event set of construction behavior, obtaining the frequency of occurrence and the probability of association of construction behaviors, calculating the dynamic construction behavior entropy, and generating a construction behavior entropy deviation degree based on the difference between the dynamic construction behavior entropy and the historical construction behavior benchmark value, forming construction instability areas and abnormal construction behaviors; extracting construction risk objects corresponding to abnormal construction behaviors from the construction instability areas, and establishing risk propagation edges between different quality risk nodes based on the construction association between different construction risk objects, forming a quality risk dynamic propagation graph; tracking the risk diffusion of each initial quality risk node in the quality risk dynamic propagation graph, generating a quality risk propagation chain, and generating risk propagation potential energy and risk aggregation index based on the risk propagation parameters in the quality risk propagation chain and the construction behavior entropy deviation degree; and normalizing and weighting the risk propagation potential energy, risk aggregation index, and construction behavior entropy deviation degree to form construction quality evaluation information for building engineering.

[0007] As a preferred embodiment of the construction quality evaluation method for building engineering based on big data described in this invention, the specific steps for forming a dynamic event set of construction behavior are as follows: Acquire construction process data and uniformly calibrate different timestamps in the construction process data to form unified time series data. At the same time, establish construction spatial units based on the spatial coordinate information in the unified time series data, and map different construction process data to the corresponding construction spatial units to form construction spatial association data. Extract the continuous execution relationship between construction procedures from the construction space association data, generate the associated procedure chain, extract the behavioral association relationship and behavior switching relationship between construction behaviors from the associated procedure chain, calculate the dynamic weight of construction behaviors based on the changes in construction behaviors, and generate a dynamic event set of construction behaviors.

[0008] As a preferred embodiment of the construction quality evaluation method for building engineering based on big data described in this invention, the specific steps for obtaining the frequency of construction behavior and the correlation probability of construction behavior are as follows: The system identifies and classifies different construction behaviors within a dynamic event set of construction activities, and then merges these behaviors according to construction area, construction procedure, and time window to form a construction behavior event flow. Statistics are collected on various construction behaviors in the construction behavior event flow. The frequency of different construction behaviors in the corresponding construction area and time window is obtained from each type of construction behavior. Based on the continuous triggering relationship between different construction behaviors in adjacent time windows, the construction behavior association probability is generated.

[0009] As a preferred embodiment of the construction quality evaluation method for building engineering based on big data described in this invention, the specific steps for calculating the dynamic construction behavior entropy and generating the construction behavior entropy deviation degree based on the difference between the dynamic construction behavior entropy and the historical construction behavior benchmark value are as follows: Based on the frequency of construction activities and the probability of association between construction activities, the distribution and association strength of different construction activities in the corresponding construction areas and time windows are statistically analyzed, and the dynamic entropy of construction activities is calculated. The deviation of construction behavior entropy is generated based on the absolute value of the difference between the dynamic construction behavior entropy and the historical construction behavior benchmark value.

[0010] As a preferred embodiment of the construction quality evaluation method for building engineering based on big data described in this invention, the specific steps for forming the construction instability zone and abnormal construction behavior are as follows: Based on the deviation of construction behavior entropy, different construction areas are continuously monitored and deviation analysis is performed, and construction areas where the deviation of construction behavior entropy exceeds the instability judgment threshold and continues to rise are identified as construction instability areas. Based on the various construction behaviors in the construction instability zone, the corresponding behavior change parameters are obtained, and the contribution of each construction behavior to the construction behavior entropy deviation is analyzed. At the same time, the magnitude of the contribution is ranked, and the construction behavior that contributes the most to the construction behavior entropy deviation is identified as abnormal construction behavior.

[0011] As a preferred embodiment of the construction quality evaluation method for building engineering based on big data described in this invention, the specific steps of extracting construction risk objects corresponding to abnormal construction behaviors from the construction instability area and determining the construction correlation between different construction risk objects are as follows: From the construction instability area, extract the construction risk objects corresponding to abnormal construction behaviors, and match the corresponding construction risk objects with the corresponding process relationship, spatial location relationship, and quality impact item relationship. Then, establish the association relationship between different construction objects to form a set of construction risk objects. Perform construction association analysis on different construction risk objects in the construction risk object set, and generate construction associations between different construction risk objects based on the construction association relationships.

[0012] As a preferred embodiment of the construction quality evaluation method for building engineering based on big data described in this invention, the specific steps for establishing risk propagation edges between different quality risk nodes to form a dynamic quality risk propagation graph are as follows: By utilizing the construction associations between different construction risk objects, corresponding quality risk nodes are established for construction risk objects with association propagation conditions, and risk propagation edges are generated based on the association direction and association strength between different quality risk nodes. The propagation direction, propagation intensity, and propagation continuity of risk propagation edges are dynamically updated, and different quality risk nodes are connected to their corresponding risk propagation edges and propagation chains are constructed to form a dynamic quality risk propagation graph.

[0013] As a preferred embodiment of the construction quality evaluation method for building engineering based on big data described in this invention, the specific steps for generating the quality risk propagation chain are as follows: The propagation status of each initial quality risk node in the dynamic propagation map of quality risk is identified, and the quality risk nodes whose construction behavior entropy deviation exceeds the instability judgment threshold are identified as risk diffusion source nodes. By using the risk propagation edges corresponding to the risk diffusion source nodes, the propagation path is searched, and based on the propagation direction, propagation intensity and propagation continuity of the risk propagation edges, the risk diffusion of downstream quality risk nodes is tracked, and a quality risk propagation chain is generated.

[0014] As a preferred embodiment of the construction quality evaluation method for building engineering based on big data described in this invention, the specific steps for generating risk propagation potential and risk clustering index are as follows: Statistical analysis was conducted on the propagation range and intensity in the quality risk propagation chain. Risk diffusion parameters were obtained based on the changes in propagation direction and intensity between different propagation nodes. Combined with the entropy deviation of construction behavior in the corresponding construction area, risk propagation potential energy was generated. Based on the overlap of propagation nodes, the overlap of propagation areas, and the continuity of propagation time among different quality risk propagation chains, cluster analysis is performed on multiple quality risk propagation chains. Based on the degree of concentration and overlap of quality risk propagation chains in the same construction area, a risk clustering index is generated.

[0015] As a preferred embodiment of the construction quality evaluation method for building engineering based on big data described in this invention, the specific steps for generating construction quality evaluation information for building engineering are as follows: The risk propagation potential, risk aggregation index, construction behavior entropy deviation and quality risk propagation chain are processed for regional correspondence and time correspondence, and the overlap of propagation nodes, crossover of propagation areas and continuity of propagation time between different quality risk propagation chains are statistically analyzed to generate quality risk propagation sequence and risk propagation coupling relationship. By utilizing the risk propagation sequence and risk propagation coupling relationship, the risk propagation potential energy, risk aggregation index and construction behavior entropy deviation are normalized and weighted and fused to generate a construction quality evaluation value for building projects. Based on the construction quality evaluation value, the quality status of different construction areas is divided to form construction quality evaluation information for building projects.

[0016] The beneficial effects of this invention are as follows: By acquiring construction process data and combining it with big data analysis to calculate dynamic construction behavior entropy and construction behavior entropy deviation, dynamic identification of unstable states and changing trends in construction behavior is achieved, enabling early detection of abnormal construction behavior and potential quality risks, and improving the dynamic monitoring capability and real-time quality evaluation of construction quality; by combining big data analysis to establish a dynamic propagation map and quality risk propagation chain for quality risks, dynamic tracking of the quality risk propagation process is achieved, thereby improving the early warning capability for quality risks and the accuracy of construction quality evaluation of building projects. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart of a construction quality evaluation method for building engineering based on big data.

[0019] Figure 2 A flowchart for creating a dynamic set of construction events.

[0020] Figure 3 A flowchart for obtaining the frequency of construction activities and the probability of their association.

[0021] Figure 4 A flowchart for identifying construction instability zones and abnormal construction behaviors. Detailed Implementation

[0022] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0023] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0024] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0025] Reference Figures 1-4 This is one embodiment of the present invention, which provides a construction quality evaluation method for building engineering based on big data, including the following steps: S1. Acquire construction process data, and perform time synchronization, spatial coordinate mapping and process association processing on the construction process data to form a dynamic event set of construction behavior.

[0026] Acquire construction process data and uniformly calibrate different timestamps in the construction process data to form unified time series data. At the same time, establish construction spatial units based on the spatial coordinate information in the unified time series data, and map different construction process data to the corresponding construction spatial units to form construction spatial association data.

[0027] The specific process includes acquiring construction process data through on-site sensing equipment, construction management platform, quality inspection terminal and construction monitoring device; reading different timestamps from the construction process data and extracting the collection time corresponding to different construction process data; converting different collection times into a unified time format; and correcting the time deviation between different collection times based on a unified time benchmark to ensure that the time information corresponding to different construction process data is consistent, thus forming a unified time series data.

[0028] Spatial coordinate information corresponding to various construction process data is read from unified time series data, and the spatial coordinate information is mapped to the floor position, axis position, component position, and work surface position of the construction area. Then, construction space units are divided according to the correspondence between construction area and construction procedure, so that each construction space unit corresponds to a defined construction scope. Based on the spatial coordinate information corresponding to the construction process data, the construction space unit to which the construction process data belongs is determined, and the construction behavior data, equipment operation data, material flow data, and quality inspection data within the same construction space unit are collected, so that construction process data from different sources form a correspondence within the same construction space unit, forming construction space association data.

[0029] It should be noted that the construction process data includes construction personnel operation data, construction equipment operation data, construction procedure execution data, construction area positioning data, and quality inspection data. Among them, construction personnel operation data is collected through personnel positioning terminals or mobile operation terminals, construction equipment operation data is collected through equipment monitoring terminals, construction procedure execution data is collected through the construction management platform, construction area positioning data is collected through positioning equipment or BIM models, and quality inspection data is collected through quality inspection records and patrol records.

[0030] Extract the continuous execution relationship between construction procedures from the construction space association data, generate the associated procedure chain, extract the behavioral association relationship and behavior switching relationship between construction behaviors from the associated procedure chain, calculate the dynamic weight of construction behaviors based on the changes in construction behaviors, and generate a dynamic event set of construction behaviors.

[0031] The specific process includes: extracting the process name, occurrence time, end time, and corresponding construction space unit from the construction space association data; sorting the construction processes within the same construction space unit according to their occurrence time to form a construction process sequence; extracting adjacent construction processes from the sequence, comparing the end time of the preceding process with the occurrence time of the following process, and determining whether the preceding and following processes are located in the same construction space unit and whether there is a sequential relationship between them; when the following process occurs after the preceding process has ended, and there are no other construction processes between them, the preceding and following processes are identified as having a continuous execution relationship; repeating the continuous execution relationship identification process for all construction process sequences, and connecting the construction processes with continuous execution relationships according to their execution order to generate an associated process chain.

[0032] Construction behavior information is extracted from each construction process in the associated process chain, and the construction behaviors are arranged according to the execution order of the associated process chain to form a construction behavior sequence. Adjacent construction behaviors in the sequence are compared. If two adjacent construction behaviors belong to different types and the latter occurs after the former, a behavior switching relationship is determined. If two adjacent construction behaviors belong to the same construction process chain, the same construction space unit, or correspond to the same quality impact item, a behavior association relationship is determined. The occurrence frequency, average duration, and behavior switching frequency of each type of construction behavior in the construction behavior sequence are statistically analyzed, and the dynamic weight of the corresponding construction behavior is obtained based on the occurrence frequency, average duration, and behavior switching frequency. The construction behaviors, behavior association relationships, behavior switching relationships, and dynamic weights are recorded to form a dynamic event set of construction behaviors.

[0033] The expression for calculating the dynamic weights of construction behavior is as follows: ; in, For the first Dynamic weights of construction behaviors corresponding to different types of construction behaviors; This refers to the sequence number of the construction activity type; For the first The number of times this type of construction activity occurs within the current time window; For the first The total number of times this type of construction activity occurs within the current time window; This refers to the statistical sequence number of the construction activity; This represents the total number of construction activity types. To prevent extremely small positive numbers in the denominator; For the first The average duration of this type of construction activity; The average duration of all construction activities; For the first The number of behavior switching between different types of construction behaviors.

[0034] It should be noted that, It was obtained by statistically analyzing the duration of the start and end times of construction activities in a sequence of construction activity events. It is obtained by statistically analyzing the duration of all construction activities within the current time window; It is obtained by sequentially statistically analyzing adjacent construction behaviors in the sequence of construction behavior events; the time window is a statistical interval determined according to the construction process cycle, where the time window for continuous operation processes is 1 hour, the time window for phased acceptance processes is 1 day, and the time window for special construction processes is dynamically determined according to the planned start and end times of the process.

[0035] S2. Classify and statistically analyze different construction behaviors in the dynamic event set of construction behavior, obtain the frequency of occurrence of construction behavior and the probability of association of construction behavior, calculate the dynamic construction behavior entropy, and generate the construction behavior entropy deviation degree based on the difference between the dynamic construction behavior entropy and the historical construction behavior benchmark value, forming the construction instability area and abnormal construction behavior.

[0036] Different construction behaviors in the dynamic event set of construction behavior are identified and classified, and then merged according to construction area, construction procedure and time window to form a construction behavior event flow.

[0037] The specific process includes reading the behavior name, occurrence time, construction procedure, construction area, duration and behavior status information corresponding to each construction behavior from the dynamic event set of construction behavior, matching the behavior name and behavior status information with the pre-set construction behavior type to determine the behavior type to which each construction behavior belongs, and then grouping construction behaviors belonging to the same behavior type into the same category to perform type identification and classification processing of different construction behaviors.

[0038] Different construction behaviors, after type identification and classification, are retrieved from the dynamic event set of construction behaviors, including the corresponding construction area, construction procedure, and occurrence time. The time range corresponding to the construction behavior is determined based on the time window to which the occurrence time belongs. Construction behaviors with the same construction area, the same construction procedure, and belonging to the same time window are grouped into the same event unit. Then, the construction behaviors within the same event unit are arranged in chronological order of occurrence. The arranged construction behaviors are recorded according to the time continuity relationship to form a construction behavior event flow that can reflect the change process of construction behaviors within the same construction area, the same construction procedure, and the same time window.

[0039] Statistics are collected on various construction behaviors in the construction behavior event flow. The frequency of different construction behaviors in the corresponding construction area and time window is obtained from each type of construction behavior. Based on the continuous triggering relationship between different construction behaviors in adjacent time windows, the construction behavior association probability is generated.

[0040] The specific process includes extracting different types of construction behaviors from the construction behavior event flow, classifying and summarizing them according to construction area and time window, and counting the number of occurrences of each type of construction behavior in the corresponding construction area and time window; and statistically analyzing the number of occurrences of the corresponding construction behavior in the corresponding construction area and time window with the total number of occurrences of all construction behaviors in the corresponding construction area and time window to obtain the frequency of occurrence of different construction behaviors in the corresponding construction area and time window.

[0041] The construction behaviors in the previous and subsequent time windows are extracted from the construction behavior event flow. The construction behaviors in the two time windows are matched according to the construction area and construction procedure. When a construction behavior occurs in the previous time window and another construction behavior occurs in the subsequent time window, the previous construction behavior and the subsequent construction behavior are identified as a continuous trigger relationship. The number of occurrences of the same continuous trigger relationship under the same construction area and the same construction procedure is counted, and the correspondence between the number of occurrences of continuous trigger relationships and the number of occurrences of the corresponding construction behavior in the previous time window is obtained to obtain the construction behavior association probability between different construction behaviors.

[0042] Based on the frequency of construction activities and the probability of their association, the distribution and association strength of different construction activities in corresponding construction areas and time windows are statistically analyzed, and the dynamic construction activity entropy is calculated. The expression for this entropy is: ; in, For the first Entropy of dynamic construction behavior within a time window; This is the sequence number of the time window; For the first The sum of the dynamic weights of all construction behaviors corresponding to a given construction behavior.

[0043] It should be noted that, It is obtained by summing up the dynamic weights of all construction activities within the current time window.

[0044] The specific process includes: statistically analyzing the occurrence of different construction behaviors in corresponding construction areas and time windows based on the frequency of occurrence and the probability of association between construction behaviors; determining the distribution status of different construction behaviors based on the differences in their occurrence frequency in corresponding construction areas and time windows; determining the degree of association between different construction behaviors based on their association probability, thus forming the association strength; mapping the distribution status and association strength to different construction behaviors; and comprehensively statistically analyzing the distribution and association of different construction behaviors in corresponding construction areas and time windows to obtain the dynamic construction behavior entropy.

[0045] The deviation of construction behavior entropy is generated based on the absolute value of the difference between the dynamic construction behavior entropy and the historical construction behavior benchmark value.

[0046] The specific process includes: extracting the dynamic construction behavior entropy corresponding to the construction area and time window from the dynamic construction behavior entropy; extracting the historical construction behavior benchmark value corresponding to the construction area and time window from the construction behavior statistics information corresponding to the historical construction process data; comparing the dynamic construction behavior entropy with the historical construction behavior benchmark value to obtain the deviation of the dynamic construction behavior entropy from the historical construction behavior benchmark value; and then determining the degree of deviation of the construction behavior from the historical stable construction state based on the absolute value of the difference between the dynamic construction behavior entropy and the historical construction behavior benchmark value to generate the construction behavior entropy deviation degree.

[0047] It should be noted that the historical construction behavior benchmark value is determined based on the historical construction records corresponding to the most recent continuous construction cycle in the current construction area; the historical construction behavior entropy is obtained from multiple historical time windows corresponding to the most recent continuous construction cycle, and the average value of multiple historical construction behavior entropies is determined as the historical construction behavior benchmark value.

[0048] Based on the deviation of construction behavior entropy, different construction areas are continuously monitored and deviation analysis is performed. Construction areas where the deviation of construction behavior entropy exceeds the instability judgment threshold and continues to rise are identified as construction instability areas.

[0049] The specific process includes extracting the construction behavior entropy deviation degree corresponding to different construction areas from the construction behavior entropy deviation degree generated by the absolute value of the difference between the dynamic construction behavior entropy and the historical construction behavior benchmark value, continuously recording it according to the time window, comparing the construction behavior entropy deviation degree of the same construction area in different time windows, obtaining the deviation range between the construction behavior entropy deviation degree and the historical construction behavior benchmark value, statistically analyzing the increasing trend, decreasing trend and fluctuation of the construction behavior entropy deviation degree in the continuous time window, and conducting continuous monitoring and deviation analysis of different construction areas.

[0050] Construction areas with entropy deviations greater than the instability threshold are selected from the entropy deviations of construction behavior corresponding to different construction areas. The entropy deviations of construction behavior in the selected construction areas are compared before and after in a continuous time window. When the entropy deviation of construction behavior in the same construction area is greater than the instability threshold in a continuous time window, and the entropy deviation of construction behavior in the later time window is greater than the entropy deviation of construction behavior in the earlier time window, it is determined that the construction behavior in the corresponding construction area is in a state of continuous deviation, and the corresponding construction area is identified as a construction instability area.

[0051] It should be noted that the instability judgment threshold is determined based on the statistical information of the construction behavior entropy deviation corresponding to the normal construction state in historical construction process data. The instability judgment threshold ranges from 0.20 to 0.35. When the construction behavior entropy deviation is less than 0.20, it indicates that the construction behavior is in a stable state. When the construction behavior entropy deviation is greater than or equal to 0.20 and less than 0.35, it indicates that the construction behavior is in a fluctuating state. When the construction behavior entropy deviation is greater than or equal to 0.35, it indicates that the construction behavior is in an unstable state.

[0052] Based on the various construction behaviors in the construction instability zone, the corresponding behavior change parameters are obtained, and the contribution of each construction behavior to the construction behavior entropy deviation is analyzed. At the same time, the magnitude of the contribution is ranked, and the construction behavior that contributes the most to the construction behavior entropy deviation is identified as abnormal construction behavior.

[0053] The specific process includes extracting the frequency of occurrence, correlation probability, dynamic weight, and duration of various construction behaviors from the construction instability zone; extracting the frequency of occurrence, correlation probability, dynamic weight, and duration of different construction behaviors within a continuous time window from the dynamic event set and event flow of construction behaviors; obtaining the changes of different construction behaviors within the continuous time window; and forming behavior change parameters based on the changes in the frequency of occurrence, correlation probability, dynamic weight, and duration of construction behaviors.

[0054] The behavioral change parameters corresponding to various construction behaviors are analyzed in correspondence with the entropy deviation of construction behaviors. The influence of changes in the frequency of construction behavior occurrence, the probability of association of construction behaviors, the dynamic weight of construction behaviors, and the duration of construction behaviors on the entropy deviation of construction behaviors is determined. The influence of each factor corresponding to the same construction behavior is summarized to obtain the contribution of each type of construction behavior to the entropy deviation of construction behaviors. The various construction behaviors are ranked from high to low according to their contribution to form contribution ranking information. The construction behavior with the highest contribution is selected from the contribution ranking information and identified as an abnormal construction behavior.

[0055] S3. Extract the construction risk objects corresponding to abnormal construction behaviors from the construction instability area, and establish risk propagation edges between different quality risk nodes based on the construction correlation between different construction risk objects, forming a dynamic quality risk propagation map.

[0056] The construction risk objects corresponding to abnormal construction behaviors are extracted from the construction instability area. The corresponding construction risk objects are matched by process correspondence, spatial location correspondence, and quality impact item correspondence. Then, the relationship between different construction objects is established to form a set of construction risk objects.

[0057] The specific process includes: reading construction behavior records corresponding to abnormal construction behaviors within the identified construction instability area; extracting corresponding construction procedures, construction areas, construction components, construction materials, construction equipment, and quality inspection objects from these records; determining whether there is a procedure correspondence between different construction risk objects by utilizing the sequential relationship between the construction procedures of the abnormal construction behaviors and those of other construction risk objects; determining whether there is a spatial correspondence between different construction risk objects based on the overlapping or adjacent relationships between the construction areas of the abnormal construction behaviors and those of other construction risk objects; and determining whether there is a quality impact item correspondence between different construction risk objects by considering whether the quality inspection objects corresponding to the abnormal construction behaviors belong to the same quality impact item as the quality inspection objects corresponding to other construction risk objects.

[0058] Based on the matching information of process correspondence, spatial location correspondence, and quality impact item correspondence, the correspondence between different construction objects is summarized. When there is at least one of the following correspondences between two construction objects: process correspondence, spatial location correspondence, or quality impact item correspondence, the two construction objects are identified as having a relationship. The construction objects with a relationship are grouped according to their corresponding construction instability areas, abnormal construction behaviors, construction processes, spatial locations, and quality impact items. The grouped construction objects are then uniformly recorded as construction risk objects to form a set of construction risk objects.

[0059] Perform construction association analysis on different construction risk objects in the construction risk object set, and generate construction associations between different construction risk objects based on the construction association relationships.

[0060] The specific process includes extracting process correspondence matching information, spatial location correspondence matching information, and quality impact item correspondence matching information from the set of construction risk objects, and determining whether there are process connection relationships, spatial adjacency relationships, or quality impact transmission relationships between different construction risk objects, and determining the association status between different construction risk objects; associating construction risk objects that have process correspondence relationships, spatial location correspondence relationships, or quality impact item correspondence relationships, and determining the construction association between construction risk objects based on the association status of the correspondence relationships, and then recording the construction association between construction risk objects to generate the construction association between different construction risk objects.

[0061] By leveraging the construction associations between different construction risk objects, corresponding quality risk nodes are established for construction risk objects with associated propagation conditions, and risk propagation edges are generated based on the association direction and strength between different quality risk nodes.

[0062] The specific process includes: extracting construction risk objects with construction associations from the construction associations between different construction risk objects; extracting construction procedure information, construction area information, and quality impact item information from the construction process data corresponding to the construction risk objects; identifying construction risk objects with association propagation conditions when there is at least one relationship among the following: procedure connection relationship, spatial adjacency relationship, or quality impact transmission relationship between construction risk objects, and the corresponding quality impact item can be transmitted to other construction risk objects; determining the corresponding quality risk based on the quality impact item, quality defect record, or quality anomaly record corresponding to the construction risk object, establishing the correspondence between construction risk objects and quality risks, and identifying the corresponding construction risk object as a quality risk node.

[0063] The transmission direction and degree of association are extracted from the construction associations between different quality risk nodes. The direction that can have a quality impact from the previous quality risk node to the next quality risk node is determined as the association direction, and the closeness of the construction association between different quality risk nodes is determined as the association strength. When there are both association direction and association strength between two quality risk nodes, the previous quality risk node is taken as the risk propagation starting point, the next quality risk node is taken as the risk propagation ending point, and the association strength is taken as the propagation strength of the risk propagation edge, thus generating a risk propagation edge to represent the quality risk transmission relationship.

[0064] It should be noted that the conditions for the propagation of related risks refer to the existence of construction-related relationships between construction risk objects that can lead to the transmission of quality risks, including at least one of the following relationships: process connection relationship, spatial adjacency relationship, and quality impact transmission relationship; when there is a construction relationship between two construction risk objects, and the corresponding construction processes have a successor-successor relationship, the corresponding construction areas have a location contact relationship, or the corresponding quality impact items have an impact relationship, the conditions for the propagation of related risks are determined to be met.

[0065] The propagation direction, propagation intensity, and propagation continuity of risk propagation edges are dynamically updated, and different quality risk nodes are connected to their corresponding risk propagation edges and propagation chains are constructed to form a dynamic quality risk propagation graph.

[0066] The specific process includes: reading the corresponding propagation start and end points from the risk propagation edge; extracting the construction association information within the corresponding time period from the construction association records between different quality risk nodes; comparing the construction association information in the current time period with that in the previous time period to obtain the changes in construction association; determining whether the quality impact transmission relationship has changed based on the changes in construction association; adjusting the propagation direction when the quality impact transmission relationship changes; adjusting the propagation intensity when the degree of construction association changes; adjusting the propagation continuity when the quality impact transmission relationship continues or is interrupted; and dynamically updating the risk propagation edge by mapping the adjusted propagation direction, propagation intensity, and propagation continuity.

[0067] The process involves extracting the quality risk node identifier, construction area, construction procedure, and quality impact item from the quality risk nodes, and extracting the propagation start point, propagation end point, propagation direction, propagation intensity, and propagation continuity from the risk propagation edges. The quality risk nodes corresponding to the propagation start point and the quality risk nodes corresponding to the propagation end point are connected according to the propagation direction, and the propagation intensity and propagation continuity are recorded to the corresponding risk propagation edges. The connected quality risk nodes are then arranged continuously according to the propagation direction to form a risk propagation path. Different risk propagation paths are summarized according to construction area, construction procedure, and quality impact item, and the quality risk nodes, risk propagation edges, and risk propagation paths are recorded accordingly to form a dynamic quality risk propagation map.

[0068] It should be noted that the quality risk dynamic propagation graph is a relational structure used to characterize the risk transmission relationship between different quality risk nodes, including quality risk nodes, risk propagation edges, and risk propagation paths.

[0069] S4. Track the risk diffusion of each initial quality risk node in the dynamic propagation map of quality risk, generate a quality risk propagation chain, and generate risk propagation potential energy and risk aggregation index based on the deviation between the risk propagation parameters in the quality risk propagation chain and the entropy of construction behavior.

[0070] The propagation status of each initial quality risk node in the dynamic propagation map of quality risk is identified, and the quality risk node whose construction behavior entropy deviation exceeds the instability judgment threshold is identified as the risk diffusion source node.

[0071] The specific process includes: extracting the risk propagation edges corresponding to each initial quality risk node from the dynamic propagation graph of quality risks; reading the propagation direction, propagation intensity, and propagation continuity corresponding to the risk propagation edges from the dynamic propagation graph of quality risks; using the propagation direction to determine the risk transmission direction between quality risk nodes and other quality risk nodes; determining the degree of risk transmission between quality risk nodes and other quality risk nodes based on the propagation intensity; determining the duration of risk transmission between quality risk nodes and other quality risk nodes based on the propagation continuity; and using the risk transmission direction, risk transmission intensity, and risk transmission duration to determine whether each initial quality risk node is in a risk transmission state, thereby identifying the propagation state of each initial quality risk node.

[0072] The construction behavior entropy deviation is extracted from the construction area corresponding to each initial quality risk node, and the construction behavior entropy deviation is compared with the instability judgment threshold. When the construction behavior entropy deviation of the construction area corresponding to a certain initial quality risk node is greater than the instability judgment threshold, it indicates that the construction behavior of the construction area where the initial quality risk node is located has been in an unstable state. The quality risk node whose construction behavior has been in an unstable state is identified as the risk diffusion source node.

[0073] It should be noted that the construction behavior entropy deviation is generated corresponding to the construction area. When establishing a quality risk node, the information of its construction area is recorded, and the corresponding construction behavior entropy deviation is obtained based on the construction area. When the construction behavior entropy deviation of the construction area to which the quality risk node belongs exceeds the instability judgment threshold, the quality risk node is identified as a risk diffusion source node.

[0074] By using the risk propagation edges corresponding to the risk diffusion source nodes, the propagation path is searched, and based on the propagation direction, propagation intensity and propagation continuity of the risk propagation edges, the risk diffusion of downstream quality risk nodes is tracked, and a quality risk propagation chain is generated.

[0075] The specific process includes: reading the propagation start point, propagation end point, and propagation direction from the risk propagation edge corresponding to the risk diffusion source node; identifying the quality risk node corresponding to the propagation end point as the downstream quality risk node with a risk transmission relationship to the risk diffusion source node; continuing to read the propagation end point and propagation direction from the risk propagation edge corresponding to the downstream quality risk node, and finding the next quality risk node with a connection relationship to the downstream quality risk node; searching for quality risk nodes with a connection relationship to the current quality risk node level by level according to the propagation direction corresponding to the risk propagation edge, and recording the risk propagation edges and risk transmission order between different quality risk nodes to obtain the propagation path corresponding to the risk diffusion source node.

[0076] The process involves extracting downstream quality risk nodes corresponding to the risk diffusion source nodes from the propagation path, and simultaneously reading the risk propagation edges connecting the risk diffusion source nodes and downstream quality risk nodes. The propagation direction of the risk propagation edges is used to determine whether the quality risk originates from the risk diffusion source node and points to the downstream quality risk node. The propagation strength of the risk propagation edges is used to determine the degree of impact of the quality risk transmission to the downstream quality risk node. The propagation continuity of the risk propagation edges is used to determine whether the quality risk continues to propagate between adjacent quality risk nodes. When the propagation direction, propagation strength, and propagation continuity all meet the risk diffusion conditions, the corresponding downstream quality risk node is included in the risk diffusion tracking information, and the next downstream quality risk node is tracked along the corresponding risk propagation edge. The risk diffusion source node, the downstream quality risk nodes included in the risk diffusion tracking information, and the corresponding risk propagation edges are connected according to the risk transmission order to generate a quality risk propagation chain.

[0077] Statistical analysis is performed on the propagation range and intensity in the quality risk propagation chain. Risk diffusion parameters are obtained based on the changes in propagation direction and intensity between different propagation nodes. Combined with the entropy deviation of construction behavior in the corresponding construction area, risk propagation potential energy is generated.

[0078] The specific process includes: extracting sequentially connected quality risk nodes and risk propagation edges from the quality risk propagation chain; extracting corresponding construction area information from the quality risk nodes; and extracting corresponding propagation intensity information from the risk propagation edges; summarizing the different construction areas traversed by the quality risk propagation chain to obtain the propagation range of the quality risk propagation chain; summarizing the propagation intensity corresponding to each risk propagation edge in the quality risk propagation chain to obtain the propagation intensity of the quality risk propagation chain; and characterizing the scale of diffusion of quality risks among different construction areas and the degree of impact during the propagation process based on the propagation range and propagation intensity.

[0079] The propagation direction and intensity of the corresponding risk propagation edges between adjacent propagation nodes are extracted sequentially from the quality risk propagation chain. The propagation direction between adjacent risk propagation edges is compared to obtain the change in the direction of quality risk during propagation, and the propagation intensity between adjacent risk propagation edges is compared to obtain the change in the intensity of quality risk during propagation. Based on the changes in propagation direction and intensity, the degree of stability and enhancement of the diffusion of quality risk in the quality risk propagation chain are determined, and risk diffusion parameters are obtained. The risk diffusion parameters are correlated and compared with the construction behavior entropy deviation degree in the construction area corresponding to the quality risk propagation chain. The diffusion changes are obtained from the change records of the risk diffusion parameters in a continuous time window, and the instability changes are obtained from the change records of the construction behavior entropy deviation degree in a continuous time window. Based on the diffusion state reflected by the risk diffusion parameters and the construction instability state reflected by the construction behavior entropy deviation degree, the probability of continued propagation of quality risk is determined, and risk propagation potential energy is generated.

[0080] It should be noted that risk diffusion parameters are a set of parameters used to characterize the diffusion state of quality risk in the quality risk propagation chain, including parameters for changes in propagation direction and parameters for changes in propagation intensity; risk propagation potential energy is an evaluation parameter used to characterize the ability of quality risk to continue to spread in the quality risk propagation chain, including the degree of diffusion stability, the degree of diffusion enhancement, and the deviation of construction behavior entropy.

[0081] Based on the overlap of propagation nodes, the overlap of propagation areas, and the continuity of propagation time among different quality risk propagation chains, cluster analysis is performed on multiple quality risk propagation chains. Based on the degree of concentration and overlap of quality risk propagation chains in the same construction area, a risk clustering index is generated.

[0082] The specific process includes: extracting corresponding quality risk nodes, construction areas, and risk propagation time records from multiple quality risk propagation chains; comparing the quality risk nodes in different quality risk propagation chains to determine the overlap of propagation nodes; comparing the construction areas corresponding to different quality risk propagation chains to determine the overlap of propagation areas; comparing the risk propagation time records corresponding to different quality risk propagation chains to determine the continuity of propagation time; and based on the overlap of propagation nodes, the overlap of propagation areas, and the continuity of propagation time, determining whether there is clustering propagation among multiple quality risk propagation chains, determining the clustering among multiple quality risk propagation chains, and performing clustering analysis of multiple quality risk propagation chains.

[0083] The number of quality risk propagation chains appearing in the same construction area is counted from the construction areas corresponding to multiple quality risk propagation chains. The concentration of quality risk propagation chains in the same construction area is determined based on the number of quality risk propagation chains. The recurrence of the same quality risk nodes and the same risk propagation edges is compared from multiple quality risk propagation chains, and the degree of propagation overlap between quality risk propagation chains is determined based on the recurrence of the same quality risk nodes and the same risk propagation edges. The concentration degree and the degree of propagation overlap are comprehensively analyzed to obtain a risk clustering index to characterize the clustering state of multiple quality risk propagation chains in the same construction area.

[0084] It should be noted that the risk clustering index is an evaluation parameter used to characterize the clustering level of multiple quality risk propagation chains in the same construction area. The larger the risk clustering index, the higher the degree of clustering of the quality risk propagation chains; the smaller the risk clustering index, the lower the degree of clustering of the quality risk propagation chains.

[0085] S5. After normalizing the risk propagation potential, risk aggregation index and construction behavior entropy deviation, the data are weighted and integrated to form construction quality evaluation information for building projects.

[0086] The risk propagation potential, risk aggregation index, construction behavior entropy deviation and quality risk propagation chain are processed for regional correspondence and time correspondence. The overlap of propagation nodes, crossover of propagation areas and continuity of propagation time between different quality risk propagation chains are statistically analyzed to generate quality risk propagation sequences and risk propagation coupling relationships.

[0087] The specific process includes: extracting the corresponding construction area and time records from the risk propagation potential energy, risk clustering index, and construction behavior entropy deviation; extracting the construction area and risk propagation time records corresponding to the quality risk nodes from the quality risk propagation chain; comparing the construction areas corresponding to the risk propagation potential energy, risk clustering index, and construction behavior entropy deviation with the construction areas corresponding to the quality risk propagation chain, and mapping the risk propagation potential energy, risk clustering index, and construction behavior entropy deviation with the construction areas to the quality risk propagation chain for area mapping; and comparing the time records corresponding to the risk propagation potential energy, risk clustering index, and construction behavior entropy deviation with the risk propagation time records corresponding to the quality risk propagation chain, and mapping the risk propagation potential energy, risk clustering index, and construction behavior entropy deviation corresponding to the time records to the quality risk propagation chain for time mapping.

[0088] The quality risk node identifiers in different quality risk propagation chains are matched one by one according to node name, node type, and construction area to which the node belongs. The occurrence of the same quality risk node is counted to obtain the propagation node overlap. The construction area information corresponding to different quality risk propagation chains is matched to obtain the occurrence of the same or adjacent construction areas to obtain the propagation area intersection. The risk propagation time records corresponding to different quality risk propagation chains are matched one by one according to the risk propagation start time and risk propagation end time, and the consecutive risk propagation time records are counted to obtain the propagation time continuity. The quality risk propagation chains are arranged according to the quality risk node propagation order and risk propagation time records to generate a quality risk propagation sequence. Based on the propagation node overlap, propagation area intersection, and propagation time continuity, the risk propagation coupling relationship is generated.

[0089] It should be noted that risk propagation coupling relationship is a relational information used to characterize the degree of association between different quality risk propagation chains, including the overlap of propagation nodes, the crossover of propagation areas, and the continuity of propagation time.

[0090] By utilizing the risk propagation sequence and risk propagation coupling relationship, the risk propagation potential energy, risk aggregation index and construction behavior entropy deviation are normalized and weighted and fused to generate a construction quality evaluation value for building projects. Based on the construction quality evaluation value, the quality status of different construction areas is divided to form construction quality evaluation information for building projects.

[0091] The specific process includes: reading the time sequence records corresponding to risk propagation potential energy, risk clustering index, and construction behavior entropy deviation from the risk propagation sequence; reading the degree of correlation between different quality risk propagation chains from the risk propagation coupling relationship; determining the weights corresponding to risk propagation potential energy, risk clustering index, and construction behavior entropy deviation based on the degree of correlation; weighting the normalized risk propagation potential energy, normalized risk clustering index, and normalized construction behavior entropy deviation using the corresponding weights; and summarizing the weighted information to obtain the construction quality evaluation value of the building project.

[0092] The construction quality evaluation values ​​for different construction areas are extracted from the construction quality evaluation values ​​for those areas. These values ​​are then summarized. A comparative analysis of these values ​​is conducted to determine the magnitude and differences between the construction quality evaluation values ​​for different areas. Based on the magnitude of the construction quality evaluation values, the differences in quality status between different construction areas are determined, and the quality status of each construction area is classified according to these differences. Finally, the construction quality evaluation values ​​and quality status classification results for each construction area are linked and recorded to form construction quality evaluation information.

[0093] It should be noted that the risk propagation coupling relationship is obtained through comprehensive analysis of the overlap of propagation nodes, the crossover of propagation areas, and the continuity of propagation time between different quality risk propagation chains; the construction quality evaluation value of the building project is a comprehensive evaluation parameter used to characterize the overall construction quality status of the construction area. It is normalized by risk propagation potential energy, risk aggregation index, and construction behavior entropy deviation, and the corresponding weights are determined according to the risk propagation coupling relationship. The normalized risk propagation potential energy, risk aggregation index, and construction behavior entropy deviation are then weighted and fused; the construction quality evaluation information of the building project is the evaluation information used to characterize the construction quality status of different construction areas, including the construction quality evaluation value of the building project, quality status classification information, and corresponding construction area information.

[0094] In summary, this invention achieves dynamic identification of unstable construction behavior states and trends by acquiring construction process data and combining it with big data analysis to calculate dynamic construction behavior entropy and construction behavior entropy deviation. This enables the early detection of abnormal construction behaviors and potential quality risks, improving the dynamic monitoring capability of construction quality and the real-time nature of quality evaluation. Furthermore, by combining big data analysis to establish a dynamic propagation map and quality risk propagation chain, it achieves dynamic tracking of the quality risk propagation process, thereby improving the early warning capability of quality risks and the accuracy of construction quality evaluation.

[0095] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present 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 present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A construction quality evaluation method for building engineering based on big data, characterized in that, include: Acquire construction process data, and perform time synchronization, spatial coordinate mapping and process association processing on the construction process data to form a dynamic event set of construction behavior; Different construction behaviors in the dynamic event set of construction behavior are classified and statistically analyzed to obtain the frequency of occurrence and the probability of association of construction behaviors. The dynamic construction behavior entropy is calculated, and the construction behavior entropy deviation is generated based on the difference between the dynamic construction behavior entropy and the historical construction behavior benchmark value, forming the construction instability area and abnormal construction behavior. The construction risk objects corresponding to abnormal construction behaviors are extracted from the construction instability area, and risk propagation edges between different quality risk nodes are established based on the construction correlation between different construction risk objects, forming a dynamic quality risk propagation map. Risk diffusion is tracked for each initial quality risk node in the dynamic quality risk propagation map, a quality risk propagation chain is generated, and risk propagation potential and risk aggregation index are generated based on the deviation between the risk propagation parameters in the quality risk propagation chain and the entropy of construction behavior. After normalizing the risk propagation potential, risk aggregation index and construction behavior entropy deviation, a weighted fusion is performed to form construction quality evaluation information for building projects.

2. The construction quality evaluation method for building engineering based on big data as described in claim 1, characterized in that, The specific steps for forming a dynamic event set of construction behavior are as follows: Acquire construction process data and uniformly calibrate different timestamps in the construction process data to form unified time series data. At the same time, establish construction spatial units based on the spatial coordinate information in the unified time series data, and map different construction process data to the corresponding construction spatial units to form construction spatial association data. Extract the continuous execution relationship between construction procedures from the construction space association data, generate the associated procedure chain, extract the behavioral association relationship and behavior switching relationship between construction behaviors from the associated procedure chain, calculate the dynamic weight of construction behaviors based on the changes in construction behaviors, and generate a dynamic event set of construction behaviors.

3. The construction quality evaluation method for building engineering based on big data as described in claim 2, characterized in that, The specific steps for obtaining the frequency of construction activities and the probability of association between construction activities are as follows: The system identifies and classifies different construction behaviors within a dynamic event set of construction activities, and then merges these behaviors according to construction area, construction procedure, and time window to form a construction behavior event flow. Statistics are collected on various construction behaviors in the construction behavior event flow. The frequency of different construction behaviors in the corresponding construction area and time window is obtained from each type of construction behavior. Based on the continuous triggering relationship between different construction behaviors in adjacent time windows, the construction behavior association probability is generated.

4. The construction quality evaluation method for building engineering based on big data as described in claim 1 or 3, characterized in that, The specific steps for calculating the dynamic construction behavior entropy and generating the construction behavior entropy deviation based on the difference between the dynamic construction behavior entropy and the historical construction behavior benchmark value are as follows: Based on the frequency of construction activities and the probability of association between construction activities, the distribution and association strength of different construction activities in the corresponding construction areas and time windows are statistically analyzed, and the dynamic entropy of construction activities is calculated. The deviation of construction behavior entropy is generated based on the absolute value of the difference between the dynamic construction behavior entropy and the historical construction behavior benchmark value.

5. The construction quality evaluation method for building engineering based on big data as described in claim 4, characterized in that, The specific steps for forming construction instability zones and abnormal construction behaviors are as follows: Based on the deviation of construction behavior entropy, different construction areas are continuously monitored and deviation analysis is performed, and construction areas where the deviation of construction behavior entropy exceeds the instability judgment threshold and continues to rise are identified as construction instability areas. Based on the various construction behaviors in the construction instability zone, the corresponding behavior change parameters are obtained, and the contribution of each construction behavior to the construction behavior entropy deviation is analyzed. At the same time, the magnitude of the contribution is ranked, and the construction behavior that contributes the most to the construction behavior entropy deviation is identified as abnormal construction behavior.

6. The construction quality evaluation method for building engineering based on big data as described in claim 1 or 5, characterized in that, The specific steps for extracting construction risk objects corresponding to abnormal construction behaviors from the construction instability area and based on the construction correlation between different construction risk objects are as follows: From the construction instability area, extract the construction risk objects corresponding to abnormal construction behaviors, and match the corresponding construction risk objects with the corresponding process relationship, spatial location relationship, and quality impact item relationship. Then, establish the association relationship between different construction objects to form a set of construction risk objects. A construction association analysis is performed on different construction risk objects in the set of construction risk objects, and construction associations between different construction risk objects are generated based on the construction association relationships.

7. The construction quality evaluation method for building engineering based on big data as described in claim 6, characterized in that, The specific steps for establishing risk propagation edges between different quality risk nodes to form a dynamic quality risk propagation graph are as follows: By utilizing the construction associations between different construction risk objects, corresponding quality risk nodes are established for construction risk objects with association propagation conditions, and risk propagation edges are generated based on the association direction and association strength between different quality risk nodes. The propagation direction, propagation intensity, and propagation continuity of risk propagation edges are dynamically updated, and different quality risk nodes are connected to their corresponding risk propagation edges and propagation chains are constructed to form a dynamic quality risk propagation graph.

8. The construction quality evaluation method for building engineering based on big data as described in claim 1 or 7, characterized in that, The specific steps for generating the quality risk propagation chain are as follows: The propagation status of each initial quality risk node in the dynamic propagation map of quality risk is identified, and the quality risk nodes whose construction behavior entropy deviation exceeds the instability judgment threshold are identified as risk diffusion source nodes. By using the risk propagation edges corresponding to the risk diffusion source nodes, the propagation path is searched, and based on the propagation direction, propagation intensity and propagation continuity of the risk propagation edges, the risk diffusion of downstream quality risk nodes is tracked, and a quality risk propagation chain is generated.

9. The construction quality evaluation method for building engineering based on big data as described in claim 8, characterized in that, The specific steps for generating the risk propagation potential and risk clustering index are as follows: Statistical analysis was conducted on the propagation range and intensity in the quality risk propagation chain. Risk diffusion parameters were obtained based on the changes in propagation direction and intensity between different propagation nodes. Combined with the entropy deviation of construction behavior in the corresponding construction area, risk propagation potential energy was generated. Based on the overlap of propagation nodes, the overlap of propagation areas, and the continuity of propagation time among different quality risk propagation chains, cluster analysis is performed on multiple quality risk propagation chains. Based on the degree of concentration and overlap of quality risk propagation chains in the same construction area, a risk clustering index is generated.

10. The construction quality evaluation method for building engineering based on big data as described in claim 1 or 9, characterized in that, The specific steps for generating construction quality evaluation information for building projects are as follows: The risk propagation potential, risk aggregation index, construction behavior entropy deviation and quality risk propagation chain are processed for regional correspondence and time correspondence, and the overlap of propagation nodes, crossover of propagation areas and continuity of propagation time between different quality risk propagation chains are statistically analyzed to generate quality risk propagation sequence and risk propagation coupling relationship. By utilizing the risk propagation sequence and risk propagation coupling relationship, the risk propagation potential energy, risk aggregation index and construction behavior entropy deviation are normalized and weighted and fused to generate a construction quality evaluation value for building projects. Based on the construction quality evaluation value, the quality status of different construction areas is divided to form construction quality evaluation information for building projects.

Images