A Big Data-Based Early Warning Method and System for Construction Engineering Quality and Safety

Abstract

This invention provides a method and system for early warning of construction project quality and safety based on big data. First, it acquires a multi-source operational data set of the construction project, including building structure monitoring data, construction equipment operation data, construction personnel operation data, and environmental impact data. Next, it performs cross-dimensional correlation and source tracing processing on the multi-source operational data set to establish potential risk correlations and obtain a risk correlation source tracing map of the construction project. Then, based on this map, it constructs a quality and safety risk transmission chain to identify the transmission path and impact range of risk factors. Next, it verifies the initial risk transmission chain by combining it with a historical risk event database of the construction project, generating a verified risk transmission chain. Finally, based on the verified risk transmission chain, it deduces the development trend of quality and safety risks, integrates the current operational status to generate a dynamic early warning scheme, and sends it to the monitoring terminal. This achieves real-time and dynamic early warning of construction project quality and safety, effectively improving management level and safety.

Description

Technical Field

[0001] This invention relates to the field of big data technology, and more specifically, to a method and system for early warning of construction engineering quality and safety based on big data. Background Technology

[0002] In the field of construction engineering, ensuring the quality and safety of projects is of paramount importance. As the scale and complexity of construction projects continue to expand and increase, the factors affecting the quality and safety of construction projects are becoming increasingly diverse and interconnected.

[0003] Traditional methods of quality and safety management in construction projects primarily rely on manual inspections and periodic testing. While these methods can identify quality issues to some extent, they have significant limitations. Firstly, manual inspections and periodic testing struggle to achieve real-time, comprehensive monitoring of the entire construction process, often failing to detect hidden defects or sudden quality and safety hazards in a timely manner. Secondly, traditional methods are inadequate when dealing with the complex, multi-source data in construction projects. Different types of data, such as structural monitoring data, construction equipment operation data, worker operation data, and environmental impact data, are typically isolated and lack effective correlation analysis tools, making it difficult to uncover hidden quality and safety risks within the data.

[0004] Furthermore, most existing construction engineering quality and safety early warning systems are based on single-dimensional data analysis, which cannot comprehensively consider the interaction and influence between multiple factors, resulting in insufficient accuracy and timeliness of early warnings, making it difficult to meet the stringent requirements of modern construction engineering for quality and safety management. Summary of the Invention

[0005] In view of the aforementioned problems, and in conjunction with the first aspect of the present invention, embodiments of the present invention provide a method for early warning of construction engineering quality and safety based on big data, the method comprising:

[0006] A multi-source operational data set for a construction project is obtained. This multi-source operational data set includes building structure monitoring data, construction equipment operation data, construction personnel operation data, and environmental impact data. The building structure monitoring data records stress and deformation monitoring information of key building components. The construction equipment operation data records the operating parameter monitoring information of construction machinery. The construction personnel operation data records the work process execution information of construction personnel. The environmental impact data records the monitoring information of environmental parameters such as temperature, humidity, and wind force in the construction area.

[0007] Cross-dimensional correlation and source tracing processing is performed on the multi-source operation data set of the construction project to establish potential risk correlations between the building structure monitoring data, the construction equipment operation data, the construction personnel operation data, and the environmental impact data, thereby obtaining a risk correlation source tracing map of the construction project;

[0008] Based on the risk correlation source map of the construction project, a quality and safety risk transmission chain is constructed, the transmission path and impact range of risk factors in the risk correlation source map are identified, and an initial risk transmission chain is obtained.

[0009] The initial risk transmission chain is verified by combining the historical risk event database of construction projects. Historical risk transmission records that match the initial risk transmission chain are extracted from the historical risk event database to generate a verified risk transmission chain.

[0010] Based on the verified risk transmission chain, the development trend of quality and safety risks is deduced, the risk development trend is integrated with the current operation status of the construction project, a dynamic early warning scheme for construction project quality and safety is generated, and the dynamic early warning scheme for construction project quality and safety is sent to the construction project monitoring terminal.

[0011] In another aspect, embodiments of the present invention also provide a construction engineering quality and safety early warning system based on big data, including a processor and a machine-readable storage medium connected to the processor. The machine-readable storage medium is used to store programs, instructions, or code, and the processor is used to execute the programs, instructions, or code in the machine-readable storage medium to implement the above-described method.

[0012] Based on the above, this invention acquires a multi-source operational data set for construction projects, encompassing building structure monitoring data, construction equipment operation data, construction personnel operation data, and environmental impact data. It then performs cross-dimensional correlation and tracing processing on this multi-source operational data set, establishing potential risk correlations between building structure monitoring data and other types of data. This results in a risk correlation tracing map for construction projects, enabling in-depth exploration of the intrinsic connections between different factors. Based on this risk correlation tracing map, a quality and safety risk transmission chain is constructed, identifying the transmission paths and impact range of risk factors. This clearly presents the propagation patterns of quality and safety risks in construction projects, helping to identify potential quality and safety hazards in advance. Furthermore, by combining the initial risk transmission chain with a historical risk event database for construction projects, a verified risk transmission chain is generated, further improving the accuracy and reliability of the risk transmission path. Based on the verified risk transmission chain, the development trend of quality and safety risks is deduced, and the risk development trend is integrated with the current operation status of the construction project to generate a dynamic early warning scheme. This enables real-time and dynamic early warning of the quality and safety of the construction project, and can send early warning information to the construction project monitoring terminal in a timely and accurate manner. This effectively improves the level of quality and safety management of the construction project, reduces the probability of quality and safety accidents, and ensures the smooth progress of the construction project and the safety of personnel and property. Attached Figure Description

[0013] Figure 1This is a schematic diagram of the execution flow of the construction engineering quality and safety early warning method based on big data provided in the embodiments of the present invention.

[0014] Figure 2 This is a schematic diagram of exemplary hardware and software components of the big data-based construction engineering quality and safety early warning system provided in this embodiment of the invention. Detailed Implementation

[0015] The present invention will now be described in detail with reference to the accompanying drawings. Figure 1 This is a flowchart illustrating a big data-based early warning method for construction engineering quality and safety, provided in one embodiment of the present invention. The following is a detailed description of this big data-based early warning method for construction engineering quality and safety.

[0016] Step S110: Obtain a multi-source operational data set for the construction project. The multi-source operational data set includes building structure monitoring data, construction equipment operation data, construction personnel operation data, and environmental impact data. The building structure monitoring data records stress and deformation monitoring information of key parts of the building. The construction equipment operation data records the operating parameter monitoring information of construction machinery. The construction personnel operation data records the work process execution information of construction personnel. The environmental impact data records the monitoring information of environmental parameters such as temperature, humidity, and wind force in the construction area.

[0017] During the main structure construction phase of a high-rise residential building, to achieve dynamic early warning of the building's quality and safety, it is first necessary to acquire multi-source operational data during the construction process. Building structure monitoring data is collected through a sensor network deployed at key structural components, such as fiber optic grating sensors and strain gauges installed on beams, columns, and core tubes. This data is used to monitor stress values, deformation, and other parameters in real time, forming the core of the building structure monitoring data. Construction equipment operation data comes from the control systems of various construction machinery on the construction site, such as tower cranes, construction elevators, and concrete pumps. These control systems continuously record operating parameters, including real-time load, operating speed, engine temperature, and hydraulic system pressure. Personnel operation data is acquired through smart safety helmets, RFID positioning systems, and a work process management platform deployed on the construction site. Smart safety helmets record worker work hours and voice commands; RFID positioning systems track worker movement on the site; and the work process management platform records start and end times, inspection results, and other execution information for each construction procedure. Environmental impact data is collected through environmental monitoring stations set up in different areas of the construction site. These stations are equipped with temperature and humidity sensors, wind speed sensors, noise monitors, and other equipment to record environmental parameters such as air temperature, humidity, wind speed, and wind direction in real time. All of this data is uploaded in real time to the building engineering data center via a data transmission network, where it is aggregated to form a multi-source operational data set for the building engineering project.

[0018] Step S120: Perform cross-dimensional correlation and source tracing processing on the multi-source operation data set of the building project, establish potential risk correlations between the building structure monitoring data, the construction equipment operation data, the construction personnel operation data, and the environmental impact data, and obtain a risk correlation and source tracing map of the building project.

[0019] After obtaining the aforementioned multi-source operational data set for the construction project, cross-dimensional correlation and tracing processing is required to uncover potential risk associations between different types of data. In the construction scenario of this super high-rise residential building, the safety of the building structure is comprehensively affected by the operating status of construction equipment, the operational standards of construction personnel, and environmental factors. Therefore, it is necessary to systematically analyze the intrinsic relationships between the building structure monitoring data and the other three types of data.

[0020] Step S121: Determine the cross-dimensional correlation dimensions in the multi-source operation data set of the building project. The cross-dimensional correlation dimensions include the correlation dimension between stress deformation parameters in building structure monitoring data and load parameters in construction equipment operation data, the correlation dimension between deformation rate parameters in building structure monitoring data and process execution parameters in construction personnel operation data, and the correlation dimension between stress distribution parameters in building structure monitoring data and wind parameters in environmental impact data.

[0021] In the construction scenario of this super high-rise residential building, when determining cross-dimensional correlation dimensions, the first consideration is the correlation between stress and deformation parameters in the building structure monitoring data and load parameters in the construction equipment operation data. For example, when a tower crane is hoisting building materials, changes in its load parameters directly affect the stress and deformation of the building structure below the boom; therefore, both need to be included in the correlation dimension analysis. Secondly, there is a correlation between the deformation rate parameters in the building structure monitoring data and the process execution parameters in the construction personnel operation data. For instance, in the concrete pouring process, the pouring rate and vibration frequency of the construction personnel affect the solidification process of the concrete structure, thus affecting the deformation rate of the structure; therefore, these two parameters are considered as correlation dimensions. Furthermore, there is a correlation between the stress distribution parameters in the building structure monitoring data and the wind parameters in the environmental impact data. During the construction of the super high-rise, the building structure of higher floors is affected by wind; changes in wind parameters lead to changes in the structural stress distribution; therefore, these two parameters are also identified as correlation dimensions. By determining these three cross-dimensional correlation dimensions, a clear direction is provided for subsequent correlation analysis.

[0022] Step S122: For each of the cross-dimensional correlation dimensions, extract the key correlation features from the corresponding two types of data. The key correlation features include data acquisition time synchronization features, parameter change trend correlation features, and abnormal fluctuation coordination features.

[0023] For each cross-dimensional correlation dimension identified above, key correlation features need to be extracted from the corresponding two types of data. Taking the correlation dimension between stress deformation parameters in building structure monitoring data and load parameters in construction equipment operation data as an example, firstly, the data acquisition time synchronization feature is extracted, that is, analyzing whether there is synchronization between the acquisition time of building structure stress deformation data and the acquisition time of construction equipment load parameters. For example, determining whether the time point when the tower crane applies the load is consistent with the time point when the stress in the building structure begins to change. Secondly, the parameter change trend correlation feature is extracted to observe whether there is a correlation between the change trend of construction equipment load parameters and the change trend of building structure stress deformation parameters. For example, when the load parameters gradually increase, do the stress deformation parameters also show a gradual increase trend? Finally, the abnormal fluctuation coordination feature is extracted to identify whether the building structure stress deformation parameters also show abnormal fluctuations when the construction equipment load parameters show abnormal fluctuations, and whether there is coordination in the amplitude and frequency of the abnormal fluctuations of the two. For the correlation dimension between deformation rate parameters in building structure monitoring data and process execution parameters in construction personnel operation data, the data acquisition time synchronization feature, parameter change trend correlation feature, and abnormal fluctuation coordination feature are extracted in the same way as above. For example, the synchronization between the execution time of the concrete pouring process and the data acquisition time of the concrete structure deformation rate is analyzed. This includes observing whether the changing trends of the pouring rate and the deformation rate are consistent, and whether the deformation rate also shows corresponding abnormal fluctuations when the pouring rate exhibits abnormal fluctuations. Similarly, for the correlation between stress distribution parameters in building structure monitoring data and wind parameters in environmental impact data, the same three types of key correlation features are extracted. This includes analyzing the synchronization between the wind parameter acquisition time and the stress distribution parameter acquisition time, observing the correlation between the changing trends of wind magnitude and stress distribution, and whether stress distribution shows coordinated abnormal fluctuations when wind force fluctuates abnormally.

[0024] Step S123: Perform association strength analysis on the extracted key association features, and assign association importance weights to each cross-dimensional association dimension according to the requirements of building engineering quality and safety risk assessment.

[0025] After extracting the key correlation features of each cross-dimensional correlation dimension, a correlation strength analysis needs to be performed on these features. This correlation strength analysis is achieved by constructing a correlation strength evaluation model, which comprehensively considers the matching degree of data acquisition time synchronization features, the similarity of parameter change trend correlation features, and the consistency of abnormal fluctuation coordination features. For each key correlation feature, a corresponding evaluation index is set. For example, the evaluation index for time synchronization features could be the absolute value of the time difference; the evaluation index for parameter change trend correlation features could be the similarity coefficient of the trend curves; and the evaluation index for abnormal fluctuation coordination features could be the overlap of the occurrence time of abnormal fluctuations and the correlation of fluctuation amplitudes. By comprehensively calculating the above evaluation indices, the correlation strength value of each cross-dimensional correlation dimension is obtained. Based on the requirements of construction engineering quality and safety risk assessment, a correlation importance weight is assigned to each cross-dimensional correlation dimension. In this high-rise residential building construction scenario, because the load of construction equipment has a significant direct impact on the building structure, the correlation dimension between the stress deformation parameters in the building structure monitoring data and the load parameters in the construction equipment operation data is assigned a high correlation importance weight. Wind force, among environmental factors, also has a significant impact on the high-rise structure; therefore, the correlation dimension between the stress distribution parameters in the building structure monitoring data and the wind force parameters in the environmental impact data is assigned a medium correlation importance weight. The impact of construction worker operation data on the structural deformation rate is relatively indirect; therefore, the correlation dimension between the deformation rate parameters in the building structure monitoring data and the process execution parameters in the construction worker operation data is assigned a low correlation importance weight. By assigning correlation importance weights, the relative importance of different correlation dimensions can be reflected in subsequent correlation analysis.

[0026] Step S124: Based on the correlation strength analysis results and correlation importance weights, construct an initial risk correlation graph. Each node in the initial risk correlation graph corresponds to a key parameter of a type of data, and the lines between nodes represent cross-dimensional correlation relationships.

[0027] Based on the above correlation strength analysis results and correlation importance weights, an initial risk correlation map was constructed. In the initial risk correlation map, each node represents a key parameter of a type of data. For example, beam stress parameters and column deformation parameters in building structure monitoring data; tower crane load parameters and concrete pump pressure parameters in construction equipment operation data; pouring rate parameters and vibration frequency parameters in construction worker operation data; and wind force parameters and temperature parameters in environmental impact data, all of which are independent nodes in the map. The lines between nodes represent cross-dimensional correlations. The thickness of the lines is determined based on the correlation strength analysis results and correlation importance weights. The higher the correlation strength and the greater the correlation importance weight, the thicker the corresponding line. For example, there is a line between the beam stress parameter node in building structure monitoring data and the tower crane load parameter node in construction equipment operation data. The thickness of this line reflects the tightness of the correlation between the two. By constructing the initial risk correlation map, the originally scattered data parameters are organized in the form of correlations, forming a visual network structure that intuitively shows the potential risk correlations between different data parameters.

[0028] Step S125: Identify the associated fault nodes in the initial risk association map, supplement the indirect association data between the associated fault nodes, calculate the transmission strength of the indirect association, and add the indirect association relationships with the transmission strength meeting the preset association threshold to the initial risk association map.

[0029] After the initial risk correlation map is constructed, it is necessary to identify the correlation fault nodes. Correlation fault nodes refer to nodes in the map that are not directly connected to other nodes, but may have indirect connections based on the actual situation of the construction project. For example, in the initial risk correlation map, the rebar tying procedure execution parameter node in the construction worker operation data and the beam bending strength parameter node in the building structure monitoring data may not be directly connected, but the quality of the rebar tying procedure affects the structural performance of the beam, and thus its bending strength. Therefore, there is an indirect connection between these two nodes, classifying them as correlation fault nodes. After identifying the correlation fault nodes, it is necessary to supplement the indirect correlation data between these nodes. This involves analyzing the intermediate influencing factors between them to obtain data that reflects the indirect correlation. For example, for the aforementioned rebar tying procedure execution parameter node and beam bending strength parameter node, intermediate influencing factors may include rebar spacing, tying firmness, etc. By obtaining data on these intermediate factors, the indirect correlation data can be supplemented. Then, the transmission strength of the indirect correlation is calculated, taking into account the degree of influence of intermediate influencing factors on the two ends of the node and the transmission efficiency of the intermediate links. The calculated transmission strength is compared with a preset association threshold. If the transmission strength meets the preset association threshold, the indirect association is added to the initial risk association map as a dashed line, with the thickness of the dashed line determined by the magnitude of the transmission strength. By supplementing the indirect association, the initial risk association map is further improved, enabling the map to more comprehensively reflect the potential risk associations between various data parameters.

[0030] Step S126: Perform node clustering processing on the supplemented risk association map, group nodes with focused association relationships into the same risk association module, and generate a construction project risk association tracing map containing multiple risk association modules and the relationships between modules.

[0031] The risk association map, after supplementing indirect correlations, undergoes node clustering. The clustering is based on the focused correlations between nodes, meaning multiple nodes collectively focus on a specific type of construction project risk or construction phase. For example, the column stress parameter node in the building structure monitoring data, the concrete pump pressure parameter node in the construction equipment operation data, and the concrete pouring parameter node in the construction worker operation data are all related to the construction quality risk of concrete columns. Therefore, these nodes are grouped into the same risk association module, named the "Concrete Column Construction Quality Risk Module." Similarly, the core tube horizontal displacement parameter node in the building structure monitoring data, the wind force parameter node in the environmental impact data, and the tower crane hoisting position parameter node in the construction equipment operation data are all related to the horizontal stability risk of super high-rise structures. Therefore, they are grouped into the "Super High-Rise Structure Horizontal Stability Risk Module." During node clustering, a density-based clustering algorithm is used. By calculating the degree of correlation between nodes, closely related nodes are grouped together to form modules. After clustering, each risk association module contains multiple nodes with focused relationships. The modules are connected by lines in the original association graph. Through node clustering, a risk association tracing graph for construction projects is generated, which contains multiple risk association modules and their inter-module relationships. This graph can more clearly show the relationships between different risk modules in a construction project.

[0032] Step S130: Construct a quality and safety risk transmission chain based on the risk correlation source map of the construction project, identify the transmission path and impact range of risk factors in the risk correlation source map, and obtain the initial risk transmission chain.

[0033] After obtaining the risk correlation source map of the construction project, a quality and safety risk transmission chain is constructed based on this map to identify the transmission path and scope of impact of risk factors in the construction project system. Each risk correlation module and node within the module in the construction project risk correlation source map represents a different risk source and risk carrier. By analyzing the relationships between these nodes and modules, the process of risk factors from generation to transmission and then to their impact on other parts can be traced, thereby constructing a risk transmission chain. In the construction scenario of this super high-rise residential building, the construction of the risk transmission chain is of great significance for early warning of potential quality and safety accidents, helping construction managers to clarify the propagation path and scope of impact of risks, so as to take targeted prevention and control measures.

[0034] Step S131: Extract the risk starting node from the risk association tracing map of the construction project. The risk starting node is a key data node with abnormal parameter fluctuations. The abnormal parameter fluctuations are determined based on the normal operating parameter range of the construction project.

[0035] Extracting risk initiation nodes from the risk correlation tracing map of construction projects is the first step in constructing a risk transmission chain. Risk initiation nodes refer to key data nodes in the map exhibiting abnormal parameter fluctuations; these nodes are the sources of risk factors. The determination of abnormal parameter fluctuations is based on the normal operating parameter range of the construction project, which is determined according to the design specifications, construction technical standards, and historical data from similar projects. For example, in the "Concrete Column Construction Quality Risk Module," the normal operating parameter range of the column stress parameter node in the building structure monitoring data is determined based on the column bearing capacity calculation in the design drawings. If the real-time stress parameter of this node exceeds the upper or lower limit of the normal operating parameter range, then the node is determined to have abnormal parameter fluctuations and is identified as a risk initiation node. Similarly, in the "Construction Equipment Operation Risk Module," the normal operating parameter range of the tower crane load parameter node is determined based on the equipment's rated load. If the real-time load parameter exceeds the rated load, then this node is identified as a risk initiation node. When extracting risk initiation nodes, it is necessary to monitor all key data nodes in the risk correlation traceability map of construction projects in real time. Once abnormal fluctuations in node parameters are detected, they should be marked as risk initiation nodes immediately.

[0036] Step S132: Analyze the relationship type between the risk initiation node and surrounding nodes, distinguish between direct and indirect relationships, and determine the risk transmission direction corresponding to each relationship.

[0037] After identifying the risk initiation node, it is necessary to analyze the type of relationship between this node and surrounding nodes, distinguishing between direct and indirect relationships, and determining the risk transmission direction corresponding to each type of relationship. A direct relationship refers to a direct physical or logical influence between two nodes, represented by a solid line in the risk correlation source map. For example, if the risk initiation node is the tower crane load parameter node, it has a direct relationship with the beam stress parameter node in the building structure monitoring data. Because the tower crane load is directly applied to the beam, causing changes in beam stress, the risk transmission direction of this relationship is from the tower crane load parameter node to the beam stress parameter node. An indirect relationship refers to a relationship between two nodes generated through one or more intermediate nodes, represented by a dashed line in the risk correlation source map. For example, there is an indirect correlation between the load parameter node of the tower crane and the hoisting command instruction parameter node in the operation data of the construction personnel. The load change of the tower crane will be affected by the hoisting command instruction, and the hoisting command instruction will affect the operation behavior of the construction personnel. The risk transmission direction of the above correlation is from the hoisting command instruction parameter node to the tower crane load parameter node, and then from the tower crane load parameter node to other related nodes.

[0038] Step S133: For each risk initiation node, trace the associated nodes along the determined risk transmission direction, and record the parameter change magnitude and transmission delay duration of each node during the risk transmission process.

[0039] For each identified risk initiation node, the associated nodes are traced in the risk correlation tracing map of the building engineering project according to the previously determined risk transmission direction. During the tracing process, the parameter change amplitude and transmission delay duration of each associated node along the risk transmission path need to be recorded. The parameter change amplitude refers to the degree to which the parameter value of an associated node deviates from its normal operating parameter range after being affected by a risk. For example, when the load parameter node of a tower crane exhibits abnormal fluctuations, the risk transmission direction is traced to the beam stress parameter node, and the deviation of the beam stress parameter from the normal range is recorded. The transmission delay duration refers to the time it takes for the abnormal parameter fluctuations at the risk initiation node to propagate to the next associated node, causing abnormal parameter fluctuations at that associated node. For example, after the tower crane load parameter becomes abnormal, a period of time is required before the beam stress parameter becomes abnormal; this period is the transmission delay duration. When tracing associated nodes, it is necessary to proceed sequentially according to the risk transmission direction until the edge node of the map or the node where the risk impact has disappeared is reached. For each associated node, its parameter change amplitude and transmission delay duration must be accurately recorded; these data will be used to subsequently calculate the transmission strength of the risk transmission path.

[0040] Step S134: Calculate the transmission strength of each risk transmission path based on the parameter change range and transmission delay duration. The transmission strength reflects the transmission capability of the risk on the risk transmission path.

[0041] Step S1341: Extract the parameter change range of each associated node in each risk transmission path, wherein the parameter change range is the degree of deviation between the abnormal parameter value and the normal parameter value of the node.

[0042] For each risk transmission path, the parameter variation range of each associated node in the path is first extracted. The calculation of the parameter variation range is based on the normal parameter values ​​of the node, which are determined according to the normal operating parameter range of the building project. Abnormal parameter values ​​refer to the parameter values ​​actually monitored after the node is affected by the risk. The parameter variation range is the degree of deviation between the abnormal parameter value and the normal parameter value. For example, if the normal parameter value range of a beam stress parameter node is A to B, and the abnormal parameter value of the node is C after being affected by the risk, then the parameter variation range is the degree of deviation between C and the midpoint of the normal parameter value range. When extracting the parameter variation range, it is necessary to ensure that the same calculation standard is used for all associated nodes on the same risk transmission path to guarantee the comparability of the parameter variation ranges.

[0043] Step S1342: Calculate the transmission delay between adjacent nodes in each risk transmission path, where the transmission delay is the time difference between the occurrence of parameter anomalies in the previous node and the occurrence of parameter anomalies in the next node.

[0044] Calculate the transmission delay between adjacent nodes in each risk transmission path. The transmission delay directly reflects the speed of risk transmission. The start time is the time when the parameter anomaly occurs at one node, and the end time is the time when the parameter anomaly occurs at the next node. The time difference between these two times is the transmission delay. For example, in a risk transmission path, if the tower crane load parameter node experiences a parameter anomaly at time T1, and then the beam stress parameter node experiences a parameter anomaly at time T2, then the transmission delay between these two adjacent nodes is T2 minus T1. When calculating the transmission delay, it is necessary to accurately record the time point when the parameter anomaly occurs at each node to ensure the accuracy of the time difference calculation. For risk transmission paths containing multiple nodes, the transmission delay between each pair of adjacent nodes needs to be calculated sequentially to comprehensively reflect the speed of risk transmission throughout the entire path.

[0045] Step S1343: Standardize the parameter change amplitude and the transmission delay duration respectively, and convert them into dimensionless evaluation scores. The evaluation score of parameter change amplitude is positively correlated with its deviation degree, and the evaluation score of transmission delay duration is negatively correlated with its duration.

[0046] Because the magnitude of parameter changes and the duration of transmission delay have different dimensions, they cannot be directly calculated together. Therefore, they need to be standardized separately to convert them into dimensionless evaluation scores. The standardization method can use the min-max standardization approach, which converts the magnitude of each parameter change into an evaluation score between 0 and 1 based on the maximum and minimum values. The larger the magnitude of the parameter change, the higher the evaluation score, achieving a positive correlation between the evaluation score of the parameter change magnitude and its degree of deviation. For the duration of transmission delay, the same min-max standardization method is used, but with a reverse process: the duration of transmission delay is converted into an evaluation score between 0 and 1. The shorter the duration of transmission delay, the higher the evaluation score, achieving a negative correlation between the evaluation score of the duration of transmission delay and its duration. For example, for the magnitude of parameter changes, the evaluation score corresponding to the maximum deviation is set to 1, and the evaluation score corresponding to the minimum deviation is set to 0. Evaluation scores for other deviation levels are calculated through linear interpolation. For the duration of transmission delay, the evaluation score corresponding to the shortest transmission delay is set to 1, and the evaluation score corresponding to the longest transmission delay is set to 0. Evaluation scores for other transmission delay durations are also calculated through linear interpolation. Through standardization, the evaluation scores for parameter variation magnitude and transmission delay duration have the same dimensions.

[0047] Step S1344: Based on the parameter change amplitude evaluation score and the transmission delay duration evaluation score, construct the conduction strength calculation logic. In the conduction strength calculation logic, the parameter change amplitude evaluation score is positively correlated with the conduction strength, and the transmission delay duration evaluation score is positively correlated with the conduction strength.

[0048] Based on the standardized evaluation scores for parameter change magnitude and transmission delay duration, a logic for calculating transmission strength is constructed. Considering that the parameter change magnitude score reflects the degree of risk impact on nodes, and the transmission delay duration score reflects the speed of risk transmission, both jointly determine the transmission strength of the risk along the transmission path. Therefore, the transmission strength calculation logic can be set as a weighted sum of the parameter change magnitude and transmission delay duration scores, where the weights of the parameter change magnitude and transmission delay duration scores are determined according to the requirements of building engineering quality and safety risk assessment. In this high-rise residential building construction scenario, since both the degree of risk impact on nodes and the transmission speed are important, their weights can be set to be equal, i.e., the transmission strength equals the average of the parameter change magnitude and transmission delay duration scores. Through the above calculation logic, the parameter change magnitude score and the transmission strength are positively correlated; that is, the higher the parameter change magnitude score and the higher the transmission delay duration score, the greater the transmission strength and the stronger the risk transmission capability.

[0049] Step S1345: Calculate the transmission intensity value for each risk transmission path using the transmission intensity calculation logic.

[0050] Based on the constructed transmission strength calculation logic, the transmission strength value is calculated for each risk transmission path. Taking a risk transmission path as an example, this path includes a risk initiation node A, associated node B, and associated node C. First, the parameter change magnitude assessment score and transmission delay duration assessment score between node A and node B are obtained, and the local transmission strength from node A to node B is calculated according to the calculation logic. Then, the parameter change magnitude assessment score and transmission delay duration assessment score between node B and node C are obtained, and the local transmission strength from node B to node C is calculated. Finally, the local transmission strengths between all adjacent nodes on the path are combined to obtain the transmission strength value of the entire risk transmission path. The combination method can be either averaging or summing, depending on the characteristics of the risk transmission path. For example, if the local transmission strengths on the risk transmission path are relatively balanced, averaging can be used; if some local transmission strengths have a significant impact on the overall transmission strength, a weighted summation can be used.

[0051] Step S1346: Perform range normalization on the calculated transmission intensity values ​​so that the transmission intensity values ​​of different risk transmission paths are in the same characterization range.

[0052] Because different risk transmission paths vary in length and number of nodes, the calculated transmission intensity values ​​may fall within different ranges. To facilitate comparison and selection of transmission intensity values ​​across different risk transmission paths, it is necessary to normalize the transmission intensity values, transforming them to the same representation interval, such as 0 to 100. Range normalization can also employ the min-max normalization method, which transforms each transmission intensity value to the target interval based on the maximum and minimum values ​​among all risk transmission path transmission intensity values. For example, setting the maximum value to 100 and the minimum value to 0, other transmission intensity values ​​are calculated using linear interpolation to obtain normalized values ​​within the 0 to 100 interval. Range normalization makes the transmission intensity values ​​of different risk transmission paths comparable, facilitating subsequent sorting and selection based on transmission intensity.

[0053] Step S135: Screen risk transmission paths whose transmission intensity exceeds the preset transmission threshold, and determine the node range covered by each screened risk transmission path as the risk impact range.

[0054] After calculating and normalizing the transmission intensity values ​​of all risk transmission paths, it is necessary to screen out risk transmission paths whose transmission intensity exceeds a preset transmission threshold. The preset transmission threshold is determined based on the safety level of the building project, its construction stage, and historical risk event data. This preset threshold reflects the upper limit of the risk transmission capacity that the building project can withstand. For example, during the critical stage of the main structure construction of a high-rise residential building, the preset transmission threshold can be set to a higher value to strictly control the screening of risk transmission paths; while in the later stages of construction, such as the decoration stage, the preset transmission threshold can be appropriately lowered. The normalized transmission intensity value of each risk transmission path is compared with the preset transmission threshold. If the normalized transmission intensity value is greater than the preset transmission threshold, the risk transmission path is screened out. For each screened risk transmission path, the range of nodes it covers is determined as the risk impact range. The scope of risk impact refers to all related building components and construction stages involved in the risk transmission path. For example, if a selected risk transmission path covers nodes related to tower crane load parameters, beam stress parameters, and concrete strength parameters, then the scope of risk impact for this path includes the tower crane operating area, the beam structure construction area, and the concrete pouring construction stage. By determining the scope of risk impact, construction managers can clearly identify the specific locations and stages that the risk may affect.

[0055] Step S136: Sort the selected risk transmission paths from high to low according to their transmission intensity, and integrate the risk starting node, transmission node, scope of influence and transmission intensity of each risk transmission path to form an initial risk transmission chain.

[0056] The selected risk transmission paths are sorted from highest to lowest according to their normalized transmission strength values. Risk transmission paths with higher transmission strength pose a greater threat to the quality and safety of the building project and should be given priority for attention and handling. After sorting, relevant information for each risk transmission path is integrated, including the name, location, and parameter anomaly type of the risk initiation node, the order and name of the transmission nodes, the specific location and links of the risk impact range, and the normalized transmission strength value. This information is then organized into an initial risk transmission chain. The initial risk transmission chain is presented in the form of a list or directed graph, where each record represents a risk transmission path and contains the integrated information mentioned above. For example, the first record in the initial risk transmission chain might be: the risk initiation node is the tower crane load parameter node (overload), the transmission nodes are successively the beam stress parameter node and the beam deformation parameter node, the risk impact range is the beam structure from the 15th to the 20th floor, and the normalized transmission strength value is 85.

[0057] Step S140: Verify the transmission path of the initial risk transmission chain by combining the historical risk event database of construction projects, extract historical risk transmission records in the historical risk event database that match the initial risk transmission chain, and generate the verified risk transmission chain.

[0058] After the initial risk transmission chain is constructed, it needs to be verified against a historical risk event database for construction projects to ensure its accuracy and reliability. This database stores detailed information on numerous past quality and safety risk events, including risk transmission paths, scope of impact, and handling outcomes. By comparing and matching the initial risk transmission chain with historical risk transmission records, the rationality of the risk transmission path in the initial chain can be verified, potential deviations or omissions can be identified, and the initial risk transmission chain can be corrected and improved, ultimately generating a verified risk transmission chain.

[0059] Step S141: Obtain the historical risk event database for construction projects. The historical risk event database for construction projects contains the historical risk transmission path, historical risk impact scope, and historical risk handling results of multiple historical quality and safety risk events.

[0060] A historical risk event database for construction projects is obtained through a construction project management platform or database system. This database is established by collecting and organizing quality and safety risk events that occurred during the construction of similar high-rise residential buildings both domestically and internationally, and contains detailed information on multiple historical quality and safety risk events. For each historical risk event, the recorded content includes the time, location, and construction stage of the event; the historical risk transmission path (i.e., the path from the starting point to other points); the historical risk impact scope (i.e., which parts and aspects of the construction project were affected by the risk event); and the historical risk handling results (including the handling measures taken and their effects). For example, a historical risk event is recorded as follows: When a high-rise residential building was under construction up to the 25th floor, the tower crane was overloaded, causing excessive stress in the beams and resulting in cracks. The historical risk transmission path was: tower crane load parameter node — beam stress parameter node — beam crack parameter node; the historical risk impact scope was the beam structure of the 25th floor; the historical risk handling result was: the hoisting operation was immediately stopped, the beams were reinforced, and the structural performance returned to normal after reinforcement.

[0061] Step S142: Compare the features of each risk transmission path in the initial risk transmission chain with the historical risk transmission paths in the historical risk event database, and extract historical risk transmission paths with similar path structures and consistent node types as candidate matching paths.

[0062] Step S1421: Analyze the structural features of each risk transmission path in the initial risk transmission chain. The structural features include the number of nodes in the risk transmission path, the association hierarchy between nodes, and the association relationship type.

[0063] The structural characteristics of each risk transmission path in the initial risk transmission chain are analyzed. First, the number of nodes in each risk transmission path is counted. For example, a path may contain a risk initiation node and two transmission nodes, for a total of three nodes. Second, the association hierarchy between nodes is analyzed. The association hierarchy reflects the order and hierarchical relationship of nodes in the risk transmission process. For example, the risk initiation node is the first layer, the node directly transmitted to is the second layer, and the node transmitted from the second layer node is the third layer, and so on. Finally, the type of association between nodes is determined, i.e., whether the nodes are directly or indirectly associated. Direct associations are represented by solid lines, and indirect associations by dashed lines. By analyzing the structural characteristics, each initial risk transmission path is described using the number of nodes, association hierarchy, and association type, forming structured path characteristic information. For example, the structural characteristics of an initial risk transmission path can be described as: 3 nodes, 3 association levels (initial node is the first layer, transmission node 1 is the second layer, and transmission node 2 is the third layer), and the association type is that the initial node and transmission node 1 are directly associated, and transmission node 1 and transmission node 2 are directly associated.

[0064] Step S1422: Analyze the structural features of each historical risk transmission path in the historical risk event database so that the structural feature analysis dimensions of the historical risk transmission path are consistent with those of the initial risk transmission path.

[0065] Following the same dimensions used in analyzing the structural features of the initial risk transmission path, the structural features of each historical risk transmission path in the historical risk event database are analyzed. Specifically, the analysis considers the number of nodes, the hierarchy of relationships between nodes, and the types of relationships, ensuring consistency between the analytical dimensions of the historical risk transmission path and the initial risk transmission path. For example, the structural features of a certain historical risk transmission path might be analyzed as follows: 3 nodes, 3 hierarchy levels, and a direct relationship between the starting node and transmission node 1, and an indirect relationship between transmission node 1 and transmission node 2. By using consistent analytical dimensions, the structural features of the historical risk transmission path and the initial risk transmission path are comparable, facilitating subsequent calculations of structural feature similarity.

[0066] Step S1423: Calculate the similarity of the initial risk transmission path and the historical risk transmission path in terms of structural features. The similarity is calculated based on the matching degree of the number of nodes, the overlap of the association level, and the consistency of the association relationship type.

[0067] The similarity of the initial risk transmission path and the historical risk transmission path in terms of structural features is calculated. First, the node number matching degree is calculated, i.e., how close the number of nodes in the initial risk transmission path is to the number of nodes in the historical risk transmission path; the closer the number of nodes, the higher the matching degree. Second, the association hierarchy overlap degree is calculated, comparing whether the association hierarchy structures of the two are overlapping, such as whether the number of nodes in the first layer, the number of nodes in the second layer, etc., are the same; the higher the overlap degree, the higher the similarity. Finally, the association relationship type consistency is calculated, judging whether the association relationship types between corresponding nodes are consistent; direct associations are consistent with direct associations, and indirect associations are consistent with indirect associations; the higher the consistency, the higher the similarity. Node number matching degree, association hierarchy overlap degree, and association relationship type consistency are each assigned a certain weight, and then the structural feature similarity is calculated by weighted summation. For example, the weight of node number matching degree is 0.3, the weight of association hierarchy overlap degree is 0.4, and the weight of association relationship type consistency is 0.3. Each score is multiplied by its corresponding weight and then summed to obtain the comprehensive score of structural feature similarity.

[0068] Step S1424: Extract historical risk transmission paths whose structural feature similarity exceeds a preset structural threshold, and further compare the node types of the initial risk transmission path with those of the historical risk transmission paths; wherein, the node type comparison includes determining whether the building structure nodes, equipment nodes, personnel nodes, and environmental nodes in the initial risk transmission path are consistent with the corresponding node types in the historical risk transmission paths.

[0069] A preset structural threshold is set, determined based on the calculation range of structural feature similarity, for example, 0.7 (assuming a similarity score range of 0 to 1). Historical risk transmission paths with a comprehensive structural feature similarity score exceeding the preset threshold are extracted; these paths are structurally similar to the initial risk transmission path. Then, node type comparisons are performed between these historical and initial risk transmission paths. Node type comparisons are performed sequentially according to the order of nodes in the path, determining whether each node type (building structure node, equipment node, personnel node, environment node) in the initial risk transmission path matches the corresponding node type in the historical risk transmission path. For example, if the first node in the initial risk transmission path is an equipment node (tower crane load parameter node), then the first node in the historical risk transmission path must also be an equipment node; otherwise, the node types are inconsistent. Node type comparisons ensure consistency in node attribute types between the historical and initial risk transmission paths.

[0070] Step S1425: Retain historical risk transmission paths with completely identical node types and structural feature similarity, and use these historical risk transmission paths as candidate matching paths.

[0071] After comparing node types, historical risk transmission paths with completely identical node types and structural feature similarity exceeding a preset structural threshold are retained and identified as candidate matching paths. For example, if a historical risk transmission path has a structural feature similarity score of 0.85, exceeding the preset structural threshold of 0.7, and its node type is completely identical to that of the initial risk transmission path (both are equipment node—structure node—structure node), then this historical risk transmission path is retained as a candidate matching path. Through this step, historical risk transmission paths that are highly similar to the initial risk transmission path in both structural features and node type are selected.

[0072] Step S143: Analyze the historical risk events, environmental conditions, and construction conditions corresponding to the candidate matching paths, and determine the similarity between the historical scenarios and the current construction project scenarios.

[0073] For each candidate matching path, the background, environmental conditions, and construction status of its corresponding historical risk events are analyzed to determine the similarity between the historical scenario and the current construction project scenario. The background includes the building type, construction stage, and project progress at the time of the historical risk event; environmental conditions include the weather conditions (temperature, humidity, wind speed, etc.) and geographical location; and construction status includes the technical solutions used, the type and condition of the construction equipment, and the skill level of the construction personnel. The above historical scenario information is compared with the corresponding information in the current high-rise residential building construction scenario to analyze the degree of similarity. For example, the current construction project scenario involves the main structure of a high-rise residential building being constructed to the 30th floor, using climbing formwork construction technology, with a certain type of tower crane as the construction equipment, and recent weather being mainly sunny with low wind speeds; the historical risk event background corresponding to a certain candidate matching path involves the main structure of a high-rise residential building being constructed to the 28th floor, also using climbing formwork construction technology, with the same type of tower crane as the construction equipment, and the weather also being sunny with low wind speeds. A comparison reveals that the two scenarios share similar backgrounds, environmental conditions, and construction processes, indicating a high degree of similarity between the historical and current construction projects.

[0074] Step S144: Filter candidate matching paths whose scene similarity exceeds a preset scene threshold, and extract historical risk transmission records corresponding to the filtered candidate matching paths whose scene similarity exceeds the preset scene threshold. The historical risk transmission records include the parameter change patterns of historical risk transmission nodes and the trend of historical risk transmission intensity changes.

[0075] A preset scenario threshold is set, determined based on the similarity evaluation criteria between historical and current scenarios, for example, set to 0.6 (assuming the scenario similarity score ranges from 0 to 1). The similarity score between the historical scenario and the current construction project scenario corresponding to each candidate matching path is compared with the preset scenario threshold, and candidate matching paths with scenario similarity exceeding the threshold are selected. For these selected candidate matching paths, their corresponding historical risk transmission records are extracted. Historical risk transmission records are detailed risk transmission information about the historical risk event in the historical risk event database, including the parameter change patterns of historical risk transmission nodes, i.e., how the parameters of each transmission node in the historical risk event changed over time, such as whether the parameters gradually increased, suddenly increased, or changed periodically; and the trend of historical risk transmission intensity changes, i.e., how the transmission intensity of the historical risk transmission path changed with the risk transmission process, whether it gradually increased, gradually decreased, or remained stable. For example, in the historical risk transmission records corresponding to a certain selected candidate matching path, the parameter change pattern of the historical risk transmission node is as follows: the load parameter of the tower crane suddenly increases, followed by the gradual increase of the beam stress parameter and the slow increase of the beam deformation parameter; the trend of the historical risk transmission intensity is as follows: the transmission intensity gradually increases from the starting node to the first transmission node, and the transmission intensity remains stable from the first transmission node to the second transmission node.

[0076] Step S145: Verify the consistency between the parameter change patterns and transmission intensity change trends in the historical risk transmission records and the parameter changes and transmission intensity of the corresponding risk transmission paths in the initial risk transmission chain. Retain the risk transmission paths that pass the consistency verification, correct the transmission intensity and impact range of the risk transmission paths that fail the consistency verification, and integrate the corrected risk transmission paths to generate the verified risk transmission chain.

[0077] The consistency verification process involves comparing the parameter change patterns in historical risk transmission records with the parameter changes along the corresponding risk transmission path in the initial risk transmission chain. This involves comparing whether the parameter change trends are consistent (e.g., both gradually increasing or both increasing suddenly). Simultaneously, the consistency verification is performed between the transmission intensity change trends in historical risk transmission records and the transmission intensity of the corresponding risk transmission path in the initial risk transmission chain. This involves comparing whether the direction of transmission intensity change is consistent (e.g., both gradually increasing or both remaining stable). If the consistency of both parameter change patterns and transmission intensity change trends is high, the consistency verification passes, and the initial risk transmission path is retained. If the consistency verification fails (e.g., the parameter change trend in historical risk transmission records is gradually increasing, while the parameter change in the initial risk transmission path is a sudden increase), the transmission intensity and impact range of the initial risk transmission path need to be corrected. During correction, the relationship between parameter change magnitude and transmission intensity in historical risk transmission records is referenced to adjust the calculated transmission intensity of the initial risk transmission path; simultaneously, the risk impact range of the initial risk transmission path is corrected based on the relationship between the historical risk impact range and transmission intensity. For example, based on historical data, when a parameter changes suddenly, the transmission strength should be proportionally higher than when it gradually increases; the transmission strength of the initial risk transmission path should be adjusted accordingly. Historically, the impact range of risks expands when the transmission strength is high; the impact range of the initial risk transmission path should be adjusted accordingly. All retained or modified risk transmission paths are integrated and arranged in descending order of transmission strength to generate a validated risk transmission chain.

[0078] Step S150: Based on the verified risk transmission chain, deduce the development trend of quality and safety risks, integrate the risk development trend with the current operation status of the construction project, generate a dynamic early warning scheme for the quality and safety of the construction project, and send the dynamic early warning scheme for the quality and safety of the construction project to the construction project monitoring terminal.

[0079] After verifying the risk transmission chain, the development trend of construction project quality and safety risks is deduced based on this chain. The risk development trend includes changes in the intensity of risk transmission and the expansion of its impact range over a future period. Simultaneously, considering the current operational status of the construction project, such as construction progress, structural load, and equipment status, a comprehensive analysis of the risk development trend and the current operational status is conducted to formulate a construction project quality and safety early warning scheme that can adapt to dynamic changes. Finally, the generated dynamic early warning scheme is sent to the construction project monitoring terminal so that construction management personnel can promptly understand the risk situation and take corresponding measures. In this high-rise residential building construction scenario, the generation and transmission of the dynamic early warning scheme enables real-time monitoring and rapid response to quality and safety risks.

[0080] Step S151: Extract the current transmission strength, current influence range, and current node parameter change rate of each risk transmission path in the verified risk transmission chain.

[0081] Key parameters for each risk transmission path are extracted from the verified risk transmission chain, including the current transmission intensity, current impact range, and current node parameter change rate. The current transmission intensity refers to the normalized value of the transmission intensity of the risk transmission path at the current moment after verification and correction. The current impact range refers to the building components and stages currently affected by the risk transmission path. The current node parameter change rate refers to the degree of change of parameters at each node on the risk transmission path per unit time, such as the change in stress value per hour or the change in deformation per minute. For example, if the current transmission intensity of a certain risk transmission path is 75, the current impact range is the beam structure from the 28th to the 30th floor, and the current node parameter change rate is a certain increase in beam stress parameter per hour and a certain increase in beam deformation parameter per hour.

[0082] Step S152: Based on the current transmission intensity and the current node parameter change rate, predict the transmission intensity change trend and node parameter change trend of each risk transmission path within a preset time period in the future.

[0083] Step S152: Based on the current transmission intensity and the current node parameter change rate, predict the transmission intensity change trend and node parameter change trend of each risk transmission path within a preset time period in the future.

[0084] In this high-rise residential building construction scenario, for each risk transmission path in the verified risk transmission chain, it is necessary to predict the trend of transmission intensity change and node parameter change within a preset time period (e.g., the next 48 hours) based on its current transmission intensity and the rate of change of current node parameters. The current transmission intensity has been normalized and falls within the range of 0 to 100. The rate of change of current node parameters reflects the degree of change of parameters of each related node per unit time, such as the hourly change of beam stress parameters and the hourly change of column lateral displacement parameters. The prediction process needs to combine the transmission patterns of similar paths in historical risk events with the construction characteristics of the current building project to ensure the accuracy and applicability of the prediction results.

[0085] Step S1521: For each risk transmission path, establish a correlation model between the current transmission intensity and the rate of change of the current node parameters, and determine the mutual influence relationship between the two.

[0086] For each risk transmission path, samples of the current transmission intensity and the rate of change of the current node parameters are collected from historical data. These samples are derived from risk monitoring records of similar construction projects at similar construction stages. Correlation analysis is performed on these sample data to determine the mutual influence between the current transmission intensity and the rate of change of the current node parameters. For example, when the current transmission intensity is high (e.g., between 80 and 100), the rate of change of the node parameters may show an accelerating trend, as high-intensity risk transmission further exacerbates abnormal fluctuations in the node parameters. When the current transmission intensity is medium (e.g., between 50 and 80), the rate of change of the node parameters may maintain a stable growth. When the current transmission intensity is low (e.g., between 0 and 50), the rate of change of the node parameters may gradually slow down or even stabilize. Through this correlation analysis, a correlation model is established between the two, which can describe the changing patterns of the rate of change of the node parameters under different transmission intensities.

[0087] Step S1522: Collect data on the transmission intensity change and node parameter change of the risk transmission path in similar historical scenarios as a reference for prediction.

[0088] Historical risk events with similar structural features, node types, and scenario conditions to the current risk transmission path are selected from the historical risk event database for building engineering. Data on the transmission intensity changes and node parameter changes along this path are extracted from these historical events. The selection criteria for similar historical scenarios include: the same building structure type (e.g., super high-rise residential buildings), the same construction stage (e.g., the main structure is constructed to approximately 30 stories), similar risk initiation node types (e.g., equipment load nodes), and similar environmental conditions (e.g., wind force and temperature range). The collected historical data should include the time-varying sequence of transmission intensity from the occurrence to the end of the risk event, as well as the time-varying sequence of node parameters (e.g., stress, deformation, displacement, etc.). For example, in a certain similar historical scenario, the transmission intensity of the risk transmission path is 60 in the initial stage, then gradually increases to 90 within 24 hours, and the rate of change of node parameters increases from 2 units per hour to 5 units per hour. This data will serve as an important reference for current predictions.

[0089] Step S1523: Based on the correlation model and historical reference data, construct a trend prediction logic. In the trend prediction logic, if the current transmission strength increases, the rate of change of node parameters will accelerate, and if the current transmission strength decreases, the rate of change of node parameters will slow down.

[0090] Combining the correlation model established in step S1521 and the historical reference data collected in step S1522, a trend prediction logic is constructed. This trend prediction logic needs to clarify how the direction of change in the current transmission intensity affects the rate of change of node parameters, and how the change in the rate of change of node parameters, in turn, affects the transmission intensity. For example, the trend prediction logic is set as follows: when the current transmission intensity shows an increasing trend, according to the correlation model, the rate of change of node parameters will accelerate accordingly; and the acceleration of the rate of change of node parameters will further enhance the transmission ability of risk in the path, causing the transmission intensity to continue to increase, forming a positive feedback loop. Conversely, when the current transmission intensity shows a decreasing trend, the rate of change of node parameters will slow down, thereby causing the growth momentum of transmission intensity to weaken or even begin to decline. At the same time, referring to the lag effect of transmission intensity and node parameter changes in historical data, for example, after the transmission intensity increases, the acceleration of the rate of change of node parameters may have a delay of 2 to 4 hours, the above lag effect is incorporated into the trend prediction logic to improve the time accuracy of prediction.

[0091] Step S1524: Set multiple future time nodes, and calculate the predicted value of the conduction intensity and the predicted value of the node parameter change rate for each future time node through the trend prediction logic.

[0092] A predetermined future time period (e.g., 48 hours) is divided into multiple equally spaced future time nodes, for example, one time node every 6 hours, for a total of 8 time nodes (t1=6h, t2=12h, ..., t8=48h). For each future time node, based on the current conduction strength, the current node parameter change rate, the correlation model, and the trend prediction logic, the predicted values ​​of the conduction strength and the node parameter change rate for that time node are calculated. During the calculation process, the interaction between the conduction strength and the node parameter change rate within each time interval needs to be deduced step by step. For example, starting from the current time t0, based on the conduction strength and node parameter change rate at t0, combined with the trend prediction logic, the predicted value of the conduction strength at time t1 is predicted; then, based on the predicted value of the conduction strength at time t1, the predicted value of the node parameter change rate at time t1 is predicted through the correlation model; then, based on the conduction strength and node parameter change rate at time t1, the relevant value at time t2 is predicted, and so on, until the predicted values ​​for all future time nodes are calculated.

[0093] Step S1525: Connect the predicted conduction intensity values ​​for each future time node in chronological order to form a conduction intensity change trend curve.

[0094] The predicted transmission intensity values ​​for each future time node calculated in step S1524 are plotted in a two-dimensional coordinate system according to time sequence (t0—t1—t2—...—t8), where the horizontal axis represents time and the vertical axis represents the predicted transmission intensity value. Then, using linear interpolation or smooth curve fitting, the plotted points are connected to form a continuous curve, i.e., the transmission intensity change trend curve. This transmission intensity change trend curve can intuitively reflect whether the transmission intensity of the risk transmission path gradually increases, remains stable, or gradually weakens within a preset future time period. For example, if the predicted transmission intensity values ​​for each time node are 65, 72, 78, 83, 87, 89, 90, and 90 respectively, the resulting transmission intensity change trend curve shows a pattern of rapid growth in the early stage and gradual flattening in the later stage, indicating that the risk transmission capacity will continuously increase in the next 48 hours, but the growth rate will gradually slow down and eventually reach saturation.

[0095] Step S1526: Connect the predicted values ​​of the node parameter change rate of each future time node in chronological order to form a node parameter change trend curve. The transmission intensity change trend curve and the node parameter change trend curve together constitute the trend prediction result of the risk transmission path.

[0096] Similarly, the predicted rate of change of node parameters for each future time node calculated in step S1524 is plotted in chronological order on a two-dimensional coordinate system (horizontal axis represents time, vertical axis represents the predicted rate of change of node parameters), and connected to form a continuous trend curve of node parameter change. This trend curve reflects the change in the rate of change of node parameters within a predetermined future time period, such as whether it is accelerating, growing at a constant rate, or decelerating. The trend curve of transmission intensity change and the trend curve of node parameter change together constitute the trend prediction result of the risk transmission path. They corroborate each other and can more comprehensively reflect the development trend of the risk. For example, if the trend curve of transmission intensity change shows that the transmission intensity is continuously increasing, while the trend curve of node parameter change shows that the rate of change of node parameters is also increasing synchronously, this indicates that the risk is in an escalating phase; if the transmission intensity tends to stabilize, and the rate of change of node parameters also stabilizes, this indicates that the risk development has entered a plateau period.

[0097] Step S153: Based on the current impact range and the predicted transmission intensity change trend, deduce the expansion trend of the impact range of each risk transmission path within the future preset time period.

[0098] By combining the current impact range and predicted transmission intensity trends of each risk transmission path, the expansion trend of the impact range within a predetermined time period is derived. The expansion of the impact range is closely related to changes in transmission intensity; higher transmission intensity indicates stronger risk transmission capacity and faster expansion of the impact range, while weaker transmission intensity slows down the expansion. Furthermore, the expansion of the impact range is also limited by the structural layout and construction procedures of the building project. For example, the risk impact range may expand along the connection direction of structural components or extend to new construction areas as construction progresses. Based on the relationship between the expansion of the impact range and changes in transmission intensity in historical risk events, an impact range expansion model is established. By inputting the current impact range and predicted transmission intensity trends into this model, the expansion path and speed of the impact range within a predetermined time period are obtained. For example, if the current impact range is the beam structure from the 28th to the 30th floor, and the predicted transmission intensity continues to increase within the next 12 hours, the impact range expansion model predicts that the impact range will expand upwards to the 31st floor beam structure and downwards to the 27th floor beam structure within the next 12 hours. By deriving the impact range expansion trend, construction managers can make advance regional planning for risk prevention and control.

[0099] Step S154: Integrate the trends of transmission intensity, node parameter changes, and impact range expansion of each risk transmission path to form an overall quality and safety risk development trend.

[0100] The trends in transmission intensity, key node parameters, and the expansion of the impact range for each risk transmission path are integrated to form the overall quality and safety risk development trend of the building project. During integration, the mutual influence and superposition effects between different risk transmission paths need to be considered. For example, two different risk transmission paths may affect the same building section; in this case, a comprehensive assessment of the total risk impact on that section is required. A weighted integration method can be used, assigning a corresponding weight based on the current transmission intensity of each risk transmission path. The higher the transmission intensity of a risk transmission path, the greater its weight in the overall risk development trend. Through integration, the overall risk transmission intensity, key node parameter, and overall risk impact range expansion trend of the building project over a predetermined future time period are obtained. For example, the overall quality and safety risk development trend might manifest as follows: within the next 24 hours, the overall risk transmission intensity of the building project gradually increases, key node parameters such as beam stress and deformation continue to increase, and the risk impact range gradually expands from the current 28th to 30th floors to the 25th to 32nd floors.

[0101] Step S155: Obtain the current construction project operation status, which includes the current construction progress stage, the current critical structural load status, and the current construction equipment health status.

[0102] The current operational status of a building project is obtained through a building project management system and on-site monitoring equipment. The current construction progress stage refers to the current construction process and completed work volume, such as the main structure reaching the 30th floor and masonry work reaching the 20th floor. The current critical structural load status refers to the current load on key structural components (such as beams, columns, and core tubes), including dead load, live load, and temporary construction loads. The current construction equipment health status refers to the operational status of major construction equipment (such as tower cranes, construction elevators, and concrete pumps), including equipment failure rates, whether performance parameters are normal, and maintenance records. For example, the current construction project status might be: the main structure has reached the 30th floor, and construction of the 31st floor is expected to begin in 3 days; the current critical structural load status is that the temporary construction load on the 30th-floor beam structure is relatively large, approaching the design limit; the current construction equipment health status is that the tower crane is operating normally, the construction elevator had a minor malfunction that has been repaired, and the concrete pump is performing well.

[0103] Step S156: Conduct a correlation analysis between the overall quality and safety risk development trend and the current construction project operation status to identify the key constraints and accelerating factors of risk development under the current operation status.

[0104] This study analyzes the correlation between the overall quality and safety risk development trend and the current operational status of the construction project, exploring how the current operational status affects the risk development trend and the potential repercussions of the risk development trend on the current operational status. The correlation analysis includes: whether the current construction phase will introduce new risk factors or alter the impact of existing risks (e.g., will the upcoming construction of the 31st floor increase the load on the 28th to 30th floors, thus accelerating the expansion of the risk's impact); whether the current critical structural load can withstand the additional loads brought about by the risk development (e.g., the current load on the 30th-floor beam structure is nearing its limit, and whether the increased stress caused by the risk will lead to structural failure); and whether the current health status of construction equipment will affect the transmission and handling of risks (e.g., if a tower crane malfunctions, will it affect the transportation of materials and emergency reinforcement work in the risk area). Through this correlation analysis, key constraints and accelerating factors of risk development under the current operational status are identified. Key constraints are factors that can mitigate risk development, such as the current slow construction progress providing sufficient time for risk handling; key accelerating factors are factors that will accelerate risk development, such as the current critical structural load approaching its limit, which will accelerate the expansion of the risk's impact.

[0105] Step S157: Based on the correlation analysis results, formulate early warning level classification standards and corresponding risk response measures for different risk transmission paths, and integrate early warning levels, response measures and risk development trends to generate a dynamic early warning scheme for the quality and safety of building engineering.

[0106] Based on the correlation analysis results, early warning level classification standards and corresponding risk response measures were formulated for each risk transmission path. The early warning level classification standards comprehensively consider the current transmission intensity, future transmission intensity trend, current impact range, future impact range expansion trend, and the influence of key limiting factors and accelerating factors of the risk transmission path. Early warning levels can be divided into Level 1, Level 2, and Level 3, with higher levels indicating more severe risks. For example, if the current transmission intensity of a risk transmission path exceeds 80, and the transmission intensity continues to increase within the next 24 hours, extending the impact range to critical load-bearing structures, and multiple accelerating factors exist, its early warning level is classified as Level 1. Corresponding risk response measures are formulated for different early warning levels, including monitoring measures (such as increasing monitoring frequency), control measures (such as limiting construction loads and adjusting construction procedures), and emergency measures (such as stopping work, organizing personnel evacuation, and reinforcing structures). For example, response measures for a Level 1 warning include immediately halting construction work within the risk impact area, organizing experts to conduct a risk assessment, developing and implementing an emergency reinforcement plan, and monitoring key node parameters hourly. Response measures for a Level 2 warning include limiting construction loads, increasing monitoring frequency (monitoring every two hours), and preparing emergency reinforcement materials. The warning level, corresponding risk response measures, and risk development trends (transmission intensity change trends, node parameter change trends, and impact range expansion trends) for each risk transmission path are integrated to form a dynamic early warning scheme for building construction quality and safety. This dynamic early warning scheme for building construction quality and safety is presented in the form of a report, containing detailed warning information and response measures for each risk transmission path.

[0107] Step S158: Send the dynamic early warning scheme for the quality and safety of the building project to the building project monitoring terminal.

[0108] The generated dynamic early warning plan for construction quality and safety is sent to the construction project monitoring terminal through the construction project management information system. The monitoring terminal includes a monitoring center display screen at the construction site and mobile terminals (such as mobile phones and tablets) for construction management personnel. Encrypted transmission is used during transmission to ensure the information security of the early warning plan. On the monitoring terminal, the early warning plan is displayed intuitively, such as a risk map (marking the risk impact range and warning level), a trend curve (showing the transmission intensity and node parameter change trends), and a list of response measures. Construction management personnel can view the early warning plan in a timely manner through the monitoring terminal, understand the current risk status and future development trends, and organize and implement risk prevention and control work according to the response measures. For example, the monitoring center display screen shows the risk warning level of each floor in real time: red indicates a level one warning area, yellow indicates a level two warning area, and green indicates a normal area; construction management personnel receive early warning information push notifications on their mobile terminals and can view detailed early warning plans and response measures by clicking on them. By sending the early warning plan to the monitoring terminal, rapid transmission and sharing of risk information is achieved.

[0109] Furthermore, the building structure monitoring data in the multi-source operational data set of the building project includes building beam monitoring data, building column monitoring data, and building foundation monitoring data. When performing cross-dimensional correlation and tracing processing on the multi-source operational data set of the building project, the sub-types of the building structure monitoring data include:

[0110] Step S201: Differentiate between building beam monitoring data, building column monitoring data, and building foundation monitoring data in the building structure monitoring data, and determine the monitoring location and parameter type corresponding to each subdivision of data.

[0111] Building structure monitoring data is categorized, clearly distinguishing between beam monitoring data, column monitoring data, and foundation monitoring data. Beam monitoring data refers to data obtained from monitoring the beam structure, including main beams, secondary beams, and cantilever beams. Parameter types primarily include beam stress, deflection, and rotation values. Column monitoring data monitors the column structure, including frame columns, irregularly shaped columns, and structural columns. Parameter types include column axial force, bending moment, and lateral displacement values. Foundation monitoring data monitors the foundation structure, including raft foundations, pile foundations, and strip foundations. Parameter types include foundation settlement, inclination, and earth pressure values. For example, stress and deflection values ​​of the main beams on the 28th to 30th floors are selected from the building structure monitoring data and classified as building beam monitoring data; axial force and lateral displacement values ​​of the frame columns on the 28th to 30th floors are selected and classified as building column monitoring data; and settlement and tilt data of the building raft foundation are selected and classified as building foundation monitoring data.

[0112] Step S202: Establish the correlation between the monitoring data of the building beam and the lifting equipment load data in the construction equipment operation data, analyze the influence of the change of lifting equipment load on the stress deformation of the beam, and obtain the first subdivision correlation.

[0113] For example, step S2021: extract the beam stress value, beam deflection value and data acquisition time from the building beam monitoring data, and extract the real-time load value of the lifting equipment, load application location and load application time from the construction equipment operation data.

[0114] Key parameters are extracted from the building beam monitoring data, including beam stress values ​​(such as tensile and compressive stresses on the upper and lower surfaces of the beam), beam deflection values ​​(vertical displacement of the beam under load), and the data acquisition time for these parameters (accurate to minutes or seconds). Simultaneously, the real-time load values ​​of the lifting equipment (the weight of the load currently being lifted by the lifting equipment), the load application location (determined by the position sensors of the lifting equipment on the building plane), and the load application time (the time when the load begins to be applied to the beam) are extracted from the construction equipment operation data. For example, a building beam monitoring data record might be extracted as follows: beam stress value is a certain amount, beam deflection value is a certain amount, and data acquisition time is a certain moment; the corresponding lifting equipment operation data record might be as follows: real-time load value is a certain amount, load application location is the mid-span of the beam, and load application time is 5 minutes prior to that moment.

[0115] Step S2022: Synchronize the beam data acquisition time with the lifting equipment load application time. For each set of beam parameters and load parameters after synchronization, analyze the variation law of beam stress value and beam deflection value when the lifting equipment load value changes.

[0116] Synchronizing the beam data acquisition time with the crane load application time involves identifying the corresponding beam and load parameters in time. Since the stress and deflection of the beam require time to change after the load is applied, a time delay needs to be determined based on mechanical principles and historical data to match the load application time with the start time of beam parameter changes. For example, historical data analysis reveals that beam stress and deflection begin to change significantly approximately 5 minutes after the crane load is applied. Therefore, adding 5 minutes to the load application time and matching it with the beam data acquisition time achieves synchronization. For each set of synchronized beam parameters (stress value, deflection value) and load parameters (load value), the analysis examines the direction (increase or decrease), magnitude (proportional relationship with the load change magnitude), and rate of change (parameter change rate per unit load change) of beam stress and deflection values ​​when the crane load value increases or decreases. For example, the analysis found that for every certain increase in load value, the stress value of the beam increases by a certain amount, the deflection value increases by a certain amount, and the rate of change remains stable.

[0117] Step S2023: Identify the synchronicity between the change in load value and the changes in beam stress value and beam deflection value, and determine whether the beam stress and deflection change in the same direction when the load increases.

[0118] Based on the analysis of the changing patterns, further identification is needed of the synchronicity between load changes and changes in beam stress and deflection. Synchronicity includes the synchronicity of the start time of the change and the synchronicity of the change process. Synchronicity of the start time refers to whether the time when the load begins to change coincides with the time when the beam stress and deflection begin to change (considering the aforementioned time delay); synchronicity of the change process refers to whether the beam stress and deflection continuously change along with the load as it continues to change. Simultaneously, it is determined whether the beam stress and deflection change in the same direction when the load increases, i.e., whether the stress and deflection increase when the load increases, and whether they decrease when the load decreases. If the load changes are synchronous with the changes in beam stress and deflection in time and in the same direction, it indicates a strong correlation between the two. For example, if the load value gradually increases from a certain value, after a 5-minute delay, the stress and deflection values ​​of the beam also gradually increase. Furthermore, as the load increases, the stress and deflection continue to increase, indicating that the three values ​​change synchronously and in the same direction.

[0119] Step S2024: Calculate the proportional relationship between the load change range, the beam stress change range, and the beam deflection change range to determine the degree of influence of load change on beam stress deformation.

[0120] This calculation establishes the proportional relationship between the load variation amplitude of the lifting equipment and the stress and deflection amplitudes of the beam. The load variation amplitude refers to the maximum change in load value (e.g., from one value to another, the amplitude is the difference between the two values); the stress variation amplitude refers to the maximum change in beam stress value within a corresponding time period; and the deflection variation amplitude refers to the maximum change in beam deflection value within a corresponding time period. The ratios of the load variation amplitude to the stress variation amplitude, and the ratio of the load variation amplitude to the deflection variation amplitude, are calculated. These ratios reflect the degree of stress and deflection change caused by a unit load change, i.e., the degree of influence of load change on beam stress deformation. For example, when the load variation amplitude is a certain value, the stress variation amplitude is also a certain value, and their ratio is constant; when the load variation amplitude is a certain value, the deflection variation amplitude is also a certain value, and their ratio is constant. The larger the ratio, the greater the influence of load change on beam stress deformation. By calculating these proportional relationships, the influence of load change on beam stress deformation is quantified.

[0121] Step S2025: Based on the degree of influence and the pattern of change, establish a description of the correlation between the building beam monitoring data and the lifting equipment load data, determine the influence threshold and response delay between the two, and obtain the first subdivision correlation.

[0122] Based on the degree and pattern of influence of load changes on beam stress and deformation, a correlation description is established between building beam monitoring data and lifting equipment load data. This correlation description includes the direction of influence (same direction or opposite direction), the degree of influence (proportional relationship), and the pattern of change (linear or nonlinear). Simultaneously, the influence threshold and response delay between the two are determined. The influence threshold refers to the critical value at which a significant change in beam stress or deflection begins after a certain load change; the response delay refers to the time required for beam stress or deflection to begin changing after a load change. For example, the correlation description could be: lifting equipment load data and building beam monitoring data are linearly correlated in the same direction; the ratio of load change amplitude to stress change amplitude is a constant, and the ratio to deflection change amplitude is also a constant; the influence threshold is when the load change amplitude reaches a certain value, at which point beam stress and deflection begin to change significantly; the response delay is 5 minutes. By establishing the above correlation description, the first subdivided correlation is obtained.

[0123] Step S203: Establish the correlation between the monitoring data of the building column and the concrete pouring parameter data in the operation data of the construction personnel, analyze the influence of concrete pouring rate and vibration frequency on the structural strength of the column, and obtain the second subdivision correlation.

[0124] Establishing a correlation between monitoring data of building columns and concrete pouring parameters from construction personnel's operational data, i.e., the second subdivided correlation, is crucial. Concrete pouring is a key process in column construction, and operational parameters such as the pouring rate and vibration frequency directly affect the density and strength of the concrete, thus influencing the load-bearing capacity and deformation performance of the column structure. Therefore, analyzing the impact of these pouring parameters on the strength of the column structure is essential for establishing the correlation between the two. In this high-rise residential building construction scenario, the strength of the column structure is key to ensuring the overall structural stability; therefore, establishing the second subdivided correlation has significant engineering implications.

[0125] Step S204: Establish the correlation between the building foundation monitoring data and the groundwater level data and soil moisture content data in the environmental impact data, analyze the impact of groundwater level changes and soil moisture content changes on foundation settlement, and obtain the third subdivision correlation.

[0126] Establishing a correlation between building foundation monitoring data and groundwater level and soil moisture content data from environmental impact data, i.e., the third subdivision correlation, is crucial. Foundation settlement is closely related to groundwater level and soil moisture content. Fluctuations in groundwater level alter the effective stress of the foundation soil, while changes in soil moisture content affect soil compressibility and shear strength, both leading to variations in foundation settlement. Analyzing the impact of groundwater level and soil moisture content changes on foundation settlement is essential for accurately predicting foundation deformation and assessing overall building stability. In this high-rise residential building construction scenario, foundation stability is particularly critical due to the building's height; therefore, establishing the third subdivision correlation is necessary.

[0127] Step S205: Calculate the association strength of the first subdivision association, the second subdivision association, and the third subdivision association respectively, and add the subdivision associations that meet the association strength criteria to the cross-dimensional association dimension system.

[0128] Using a correlation strength analysis method similar to that in step S123, the correlation strength of the first, second, and third sub-correlation relationships is calculated respectively. For the first sub-correlation relationship (building beam monitoring data and lifting equipment load data), the correlation strength calculation considers factors such as the degree of influence of load changes on beam stress deformation and the consistency of change patterns. For the second sub-correlation relationship (building column monitoring data and concrete pouring parameter data), factors such as the significance of the influence of pouring parameters on column structural strength and data synchronization are considered. For the third sub-correlation relationship (building foundation monitoring data and groundwater level and soil moisture content data), factors such as the sensitivity of groundwater level and soil moisture content changes on foundation settlement and the correlation of change trends are considered. A correlation strength threshold is set, and the calculated correlation strength of each sub-correlation relationship is compared with this threshold. If the correlation strength meets the standard (i.e., greater than or equal to the threshold), the sub-correlation relationship is added to the cross-dimensional correlation dimension system. For example, if the correlation strength of the first sub-relationship is 0.8, the correlation strength of the second sub-relationship is 0.75, and the correlation strength of the third sub-relationship is 0.65, and the correlation strength threshold is set to 0.6, then the correlation strength of all three sub-relationships meets the standard and they are all added to the cross-dimensional correlation dimension system.

[0129] Step S206: Based on the detailed correlation relationships and corresponding correlation strengths, supplement and improve the correlation connections between structural monitoring nodes and other types of nodes in the risk correlation tracing map of building engineering.

[0130] Based on the first, second, and third sub-categories of correlation relationships incorporated into the cross-dimensional correlation system, and their corresponding correlation strengths, the risk correlation source tracing map for building engineering is supplemented and improved. Building upon the original risk correlation source tracing map, correlation connections are added between building beam monitoring data nodes and lifting equipment load data nodes, with the line thickness determined by the correlation strength of the first sub-categories; correlation connections are added between building column monitoring data nodes and concrete pouring parameter data nodes, with the line thickness determined by the correlation strength of the second sub-categories; and correlation connections are added between building foundation monitoring data nodes and groundwater level and soil moisture content data nodes, with the line thickness determined by the correlation strength of the third sub-categories. For example, if the correlation strength of the first sub-categories is high, the connection between the building beam monitoring data node and the lifting equipment load data node is thicker; if the correlation strength of the third sub-categories is relatively low, the connection between the building foundation monitoring data node and the groundwater level data node is thinner. By supplementing these connections, the relationships between structural monitoring nodes and other types of nodes in the risk correlation tracing map of building engineering are enriched and made more accurate, thereby improving the map's ability to represent the risk correlation of building engineering.

[0131] Figure 2 The illustration shows exemplary hardware and software components of a big data-based construction project quality and safety early warning system 100, which can implement the ideas of this application, according to some embodiments of this application. For example, a processor 120 can be used in the big data-based construction project quality and safety early warning system 100 and to perform the functions in this application.

[0132] For example, a big data-based construction project quality and safety early warning system 100 may include a network port 110 connected to a network, one or more processors 120 for executing program instructions, a communication bus 130, and various forms of storage media 140, such as a disk, ROM, or RAM, or any combination thereof. Exemplarily, the big data-based construction project quality and safety early warning system 100 may also include program instructions stored in ROM, RAM, or other types of non-transitory storage media, or any combination thereof. The methods of this application can be implemented according to the program instructions. The big data-based construction project quality and safety early warning system 100 also includes an I / O interface 150 between the computer and other input / output devices.

[0133] Furthermore, this embodiment of the invention also provides a readable storage medium, wherein computer-executable instructions are preset in the readable storage medium, and when the processor executes the computer-executable instructions, the above-mentioned big data-based construction engineering quality and safety early warning method is implemented.

[0134] It should be noted that, in order to simplify the description of the present invention and thus help to understand one or more embodiments of the invention, multiple features may sometimes be grouped into one embodiment, drawing or description thereof in the foregoing description of the embodiments of the present invention.

Claims

1. A big data-based construction engineering quality safety early warning method, characterized in that, The method comprises: Step S110: acquiring a building engineering multi-source operation data set; Step S120: performing cross-dimension association traceability processing on the building engineering multi-source operation data set, establishing potential risk association between building structure monitoring data and construction equipment operation data, construction personnel operation data, and environmental influence data, and obtaining a building engineering risk association traceability graph; Step S130: constructing a quality and safety risk transmission chain based on the building engineering risk association traceability graph, identifying a transmission path and an influence range of a risk factor in the risk association traceability graph, and obtaining an initial risk transmission chain; Step S140: verifying the transmission path of the initial risk transmission chain in combination with a building engineering historical risk event library, extracting a historical risk transmission record in the historical risk event library that matches the initial risk transmission chain, and generating a verified risk transmission chain; Step S150: deducing a quality and safety risk development trend based on the verified risk transmission chain, integrating the risk development trend and a current building engineering operation state, generating a building engineering quality and safety dynamic early warning scheme, and sending the building engineering quality and safety dynamic early warning scheme to a building engineering monitoring terminal; The step 120 comprises determining cross-dimension association dimensions in the building engineering multi-source operation data set, and the cross-dimension association dimensions comprise an association dimension between a stress deformation parameter in the building structure monitoring data and a load parameter in the construction equipment operation data, an association dimension between a deformation rate parameter in the building structure monitoring data and a work procedure execution parameter in the construction personnel operation data, and an association dimension between a stress distribution parameter in the building structure monitoring data and a wind force parameter in the environmental influence data; For each of the cross-dimension association dimensions, key association features in the corresponding two types of data are extracted, and the key association features comprise data collection time synchronization features, parameter change trend association features, and abnormal fluctuation coordination features; The extracted key association features are subjected to association strength analysis, and each of the cross-dimension association dimensions is assigned an association importance weight according to building engineering quality and safety risk assessment requirements; Based on the association strength analysis result and the association importance weight, an initial risk association graph is constructed, each node in the initial risk association graph corresponds to a key parameter of a type of data, and a connection between nodes represents a cross-dimension association relationship; Association fault nodes in the initial risk association graph are identified, indirect association data between the association fault nodes are supplemented, the transmission strength of the indirect association is calculated, and indirect association relationships with transmission strength meeting a preset association threshold are added to the initial risk association graph; The supplemented risk association graph is subjected to node clustering processing, nodes with focused association relationships are classified into the same risk association module, and a building engineering risk association traceability graph comprising multiple risk association modules and inter-module association relationships is generated; The step S130 comprises: Extracting a risk starting node from the building engineering risk association traceability graph, the risk starting node being a key data node with parameter abnormal fluctuation, and the parameter abnormal fluctuation being determined based on a normal operation parameter range of the building engineering; Analyze the association type between the risk starting node and the surrounding nodes, distinguish the direct association and the indirect association, and determine the risk transmission direction corresponding to each association type; For each risk starting node, track the associated nodes along the determined risk transmission direction, and record the parameter variation amplitude and the transmission delay duration of each node in the risk transmission process; According to the parameter variation amplitude and the transmission delay duration, calculate the conduction strength of each risk transmission path, which reflects the transmission ability of the risk on the risk transmission path; Screen the risk transmission paths with the conduction strength exceeding the preset conduction threshold, and determine the node range covered by each screened risk transmission path as the risk influence range; Sort the screened risk transmission paths according to the conduction strength from high to low, integrate the risk starting node, the transmission node, the influence range and the conduction strength of each risk transmission path, and form an initial risk conduction chain.

2. The big data-based construction engineering quality safety early warning method according to claim 1, characterized in that, The calculation of the conduction strength of each risk transmission path according to the parameter variation amplitude and the transmission delay duration includes: Extract the parameter variation amplitude of each associated node in each risk transmission path, which is the deviation degree of the abnormal parameter value and the normal parameter value of the node; Calculate the transmission delay duration between adjacent nodes in each risk transmission path, which is the time difference between the occurrence of parameter abnormality of the previous node and the occurrence of parameter abnormality of the next node; Standardize the parameter variation amplitude and the transmission delay duration respectively, and convert them into dimensionless evaluation scores, wherein the parameter variation amplitude evaluation score is positively correlated with the deviation degree, and the transmission delay duration evaluation score is negatively correlated with the time length; Based on the parameter variation amplitude evaluation score and the transmission delay duration evaluation score, construct a conduction strength calculation logic, wherein the parameter variation amplitude evaluation score is positively correlated with the conduction strength, and the transmission delay duration evaluation score is negatively correlated with the conduction strength; Calculate the conduction strength value of each risk transmission path through the conduction strength calculation logic; Range-normalize the calculated conduction strength value to make the conduction strength values of different risk transmission paths in the same representation interval. 3.The big data-based construction engineering quality safety early warning method according to claim 1, characterized in that, The conduction path verification of the initial risk conduction chain in combination with the building engineering historical risk event library includes: Obtain a building engineering historical risk event library, which contains the historical risk conduction path, the historical risk influence range and the historical risk disposal result of multiple historical quality and safety risk events; Compare the features of each risk transmission path in the initial risk conduction chain with the historical risk conduction path in the historical risk event library, and extract the historical risk conduction path with similar path structure and consistent node type as the candidate matching path; Analyze the occurrence background, environmental conditions and construction conditions of the historical risk event corresponding to the candidate matching path, and judge the similarity between the historical scene and the current building engineering scene; The candidate matching path with a scene similarity exceeding a preset scene threshold is screened, and a historical risk conduction record corresponding to the candidate matching path with the scene similarity exceeding the preset scene threshold is extracted, the historical risk conduction record including a parameter variation law of a historical risk transmission node and a historical risk conduction intensity variation trend; The parameter variation law and the conduction intensity variation trend in the historical risk conduction record are subjected to consistency verification with a parameter variation condition and a conduction intensity of a corresponding risk transmission path in the initial risk conduction chain, a risk transmission path passing the consistency verification is retained, a conduction intensity and an influence range of a risk transmission path failing the consistency verification are corrected, and the corrected risk transmission path is integrated to generate a verified risk conduction chain.

4. The big data-based construction engineering quality safety early warning method according to claim 3, characterized in that, The feature comparison between each risk transmission path in the initial risk conduction chain and a historical risk conduction path in the historical risk event library includes: The structural features of each risk transmission path in the initial risk conduction chain are analyzed, the structural features including a number of nodes included in the risk transmission path, an association level between the nodes, and a type of association relationship between the nodes; The structural features of each historical risk conduction path in the historical risk event library are analyzed to make the structural feature analysis dimensions of the historical risk conduction path and the initial risk transmission path consistent; The similarity between the initial risk transmission path and the historical risk conduction path in the structural features is calculated, the similarity being calculated based on a node number matching degree, an association level coincidence degree, and a type of association relationship consistency; The historical risk conduction path with a structural feature similarity exceeding a preset structural threshold is extracted, and the node types of the initial risk transmission path and the historical risk conduction path are further compared; the node type comparison includes judging whether a building structure node, a device node, a personnel node, and an environment node in the initial risk transmission path are consistent with corresponding node types in the historical risk conduction path; The historical risk conduction path with a completely consistent node type and a satisfactory structural feature similarity is retained, and the historical risk conduction path is taken as a candidate matching path.

5. The big data-based construction engineering quality safety early warning method according to claim 1, characterized in that, The quality and safety risk development trend is derived based on the verified risk conduction chain, the risk development trend is integrated with a current construction project operation state, and a construction project quality and safety dynamic early warning scheme is generated, including: The current conduction intensity, the current influence range, and the current node parameter variation rate of each risk transmission path in the verified risk conduction chain are extracted; Based on the current conduction intensity and the current node parameter variation rate, the conduction intensity variation trend and the node parameter variation trend of each risk transmission path in a future preset time period are predicted; Based on the current influence range and the predicted conduction intensity variation trend, the influence range expansion trend of each risk transmission path in the future preset time period is derived; The conduction intensity variation trend, the node parameter variation trend, and the influence range expansion trend of each risk transmission path are integrated to form an overall quality and safety risk development trend. Obtain a current construction project operation state, the current construction project operation state including a current construction progress stage, a current key structure load state and a current construction equipment health state; Correlation analysis is performed on the overall quality safety risk development trend and the current construction project operation state to identify key restriction factors and acceleration factors of risk development under the current operation state; Based on the correlation analysis result, warning level division standards and corresponding risk response measures are formulated for different risk transmission paths, and a construction project quality safety dynamic warning scheme is generated by integrating the warning levels, response measures and risk development trends.

6. The big data-based construction engineering quality safety early warning method according to claim 5, characterized in that, The prediction of the conduction intensity change trend and the node parameter change trend of each risk transmission path in the future preset time period based on the current conduction intensity and the current node parameter change rate includes: For each risk transmission path, an association model of the current conduction intensity and the current node parameter change rate is established to determine the mutual influence relationship therebetween; The conduction intensity change data and the node parameter change data of the risk transmission path under historical similar scenarios are collected as prediction reference; Based on the association model and the historical reference data, a trend prediction logic is constructed, in which the current conduction intensity increases and the node parameter change rate accelerates, and the current conduction intensity decreases and the node parameter change rate slows down; A plurality of future time nodes are set, and the conduction intensity prediction value and the node parameter change rate prediction value corresponding to each future time node are calculated through the trend prediction logic; The conduction intensity prediction values of each future time node are connected in time sequence to form a conduction intensity change trend curve; The node parameter change rate prediction values of each future time node are connected in time sequence to form a node parameter change trend curve, and the conduction intensity change trend curve and the node parameter change trend curve together constitute the trend prediction result of the risk transmission path. 7.The big data-based construction engineering quality safety early warning method according to claim 1, characterized in that, The building structure monitoring data in the building engineering multi-source operation data set includes building beam monitoring data, building column monitoring data and building foundation monitoring data, and when the building engineering multi-source operation data set is processed by cross-dimension correlation tracing, for the subdivided types of building structure monitoring data, including: The building beam monitoring data, the building column monitoring data and the building foundation monitoring data in the building structure monitoring data are distinguished to determine the monitoring parts and parameter types corresponding to each subdivided data; An association relationship between the building beam monitoring data and the hoisting equipment load data in the construction equipment operation data is established to analyze the influence of hoisting equipment load change on beam stress deformation to obtain a first subdivided association relationship; An association relationship between the building column monitoring data and the concrete pouring parameter data in the construction personnel operation data is established to analyze the influence of concrete pouring rate and vibration frequency on column structure strength to obtain a second subdivided association relationship; An association relationship between the building foundation monitoring data and the underground water level data and the soil moisture content data in the environmental influence data is established to analyze the influence of underground water level change and soil moisture content change on foundation settlement to obtain a third subdivided association relationship; The association strength of the first, second and third sub-association relationships is calculated respectively, and the sub-association relationship with the association strength reaching a standard is added to the cross-dimension association dimension system; Based on the sub-association relationship and the corresponding association strength, the association connection between the structure monitoring node and other types of nodes in the building engineering risk association traceability graph is supplemented and improved.

8. A big data-based construction engineering quality safety early warning system, characterized in that, The building engineering quality and safety early warning system based on big data comprises a processor and a memory, the memory is connected with the processor, the memory is used for storing programs, instructions or codes, and the processor is used for executing the programs, instructions or codes in the memory to realize the building engineering quality and safety early warning method based on big data in any one of claims 1-7.

Images