Construction site security display method and platform based on multi-source data fusion

Through the construction site security display method and platform with multi-source data fusion, risk probability events are analyzed, multi-level fusion models are built, and monitoring sensing arrays are deployed, which solves the problem of insufficient data fusion in the construction safety monitoring system, and improves construction safety and Management efficiency.

CN119067595AInactive Publication Date: 2024-12-03XUZHOU YUNZHU CONSTR TECH CO LTD
View PDF 0 Cites 3 Cited by

Patent Information

Application Number
CN202411104766.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-13
Publication Date
2024-12-03
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The existing construction safety monitoring system lacks effective data fusion, resulting in inefficient safety management on construction sites, unable to fully utilize the complementarity and synergy between different data sources, and there are problems of time synchronization and spatial alignment.

Method used

Provide construction site security display methods and platforms for multi-source data fusion. By analyzing the risk probability events of target construction projects, determining the fusion sequence and multi-level fusion scale, building a multi-level fusion model, deploying monitoring sensing arrays for full-domain monitoring, and identifying risks The terminal visual display is carried out to achieve comprehensive monitoring and analysis of construction project risks.

Benefits of technology

By integrating multiple data sources and adopting multi-level fusion models, comprehensive monitoring and analysis of construction project risks is achieved, construction safety and efficiency are improved, and the problem of inefficient safety management caused by insufficient data fusion is solved.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119067595A_ABST
    Figure CN119067595A_ABST
Patent Text Reader

Abstract

The invention provides a multi-source data fusion construction site security display method and platform, and relates to the technical field of building safety, and the method comprises the steps: analyzing a risk probability event, determining a fusion sequence based on a collaborative analysis demand, determining a multi-level fusion scale, constructing a multi-level fusion model, deploying a monitoring sensor array, and carrying out the global monitoring of project construction. And determining a source end data group, carrying out unit matching and partition matching, carrying out construction risk decision making, taking information imbalance processing and information reconstruction as references, determining a risk target, and carrying out terminal visual display. The technical problem that the construction site safety management efficiency is low due to the fact that the safety monitoring of the construction site lacks effective data fusion in the prior art can be solved, comprehensive monitoring and analysis of the construction project risk are achieved by integrating multiple data sources and adopting a multi-level fusion model, and the construction safety and efficiency are improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of building safety technology, and in particular to a construction site security display method and platform using multi-source data fusion. Background Art

[0002] With the rapid development of the construction industry, construction safety has become an issue of increasing concern. In order to improve construction safety, various construction safety monitoring systems are widely used in construction sites. Usually rely on data sources such as video surveillance, sensor data, and geographic information systems to achieve monitoring and analysis of construction sites. However, there are some problems and limitations in the existing construction safety monitoring systems. Construction safety monitoring systems often rely on a single data source, such as video surveillance alone, and lack the integration and analysis of other types of data, resulting in incomplete and inaccurate monitoring and assessment of construction risks. There is a lack of effective data fusion mechanism, and the complementarity and synergy between different data sources cannot be fully utilized. There are problems with time synchronization and spatial alignment when processing data, which affects the accuracy and consistency of the data.

[0003] In summary, the prior art has a technical problem of low efficiency in construction site safety management due to the lack of effective data fusion in safety monitoring of construction sites. Summary of the invention

[0004] The purpose of this application is to provide a construction site security display method and platform with multi-source data fusion, so as to solve the technical problem in the prior art that the safety monitoring of the construction site lacks effective data fusion, resulting in low efficiency of construction site safety management.

[0005] In view of the above problems, the present application provides a construction site security display method and platform with multi-source data fusion.

[0006] In a first aspect, the present application provides a construction site security display method of multi-source data fusion, which is implemented through a construction site security display platform of multi-source data fusion, wherein the construction site security display method of multi-source data fusion includes: analyzing the risk probability events of the target construction project, determining a fusion sequence based on the collaborative analysis requirements, and each data fusion sequence is marked with a fusion characteristic; determining a multi-level fusion scale, taking multi-scale fusion based on fusion dimensions as the processing requirement, and constructing a multi-level fusion model based on the fusion sequence, wherein the fusion dimension includes at least a data level, a feature level, and a decision level; deploying a monitoring sensor array based on the fusion sequence; conducting full-domain monitoring of the project construction based on the monitoring sensor array to determine a source data group; transmitting back the source data group, and conducting unit matching and partition matching based on the multi-level fusion model in an event-oriented manner, making construction risk decisions, and determining risk lists; identifying the risk lists, and determining risk targets and performing terminal visualization display based on information asymmetry processing and information reconstruction.

[0007] In the second aspect, the present application also provides a construction site security display platform with multi-source data fusion, which is used to execute the construction site security display method with multi-source data fusion as described in the first aspect, wherein the construction site security display platform with multi-source data fusion includes: a data acquisition module, the data acquisition module is used to analyze the risk probability events of the target construction project, and based on the collaborative analysis requirements, determine the fusion sequence, and each data fusion sequence is marked with a fusion characteristic; a data processing module, the data processing module is used to determine the multi-level fusion scale, and take the multi-scale fusion based on the fusion dimension as the processing requirement, and construct a multi-level fusion model based on the fusion sequence, wherein the fusion dimension at least includes data level, feature level and decision level; a data fusion module, the data fusion module is used to deploy the monitoring sensor array based on the fusion sequence; a global monitoring module, the global monitoring module is used to perform global monitoring of the project construction based on the monitoring sensor array, and determine the source data group; a risk decision module, the risk decision module is used to return the source data group, event-oriented, based on the multi-level fusion model to perform unit matching and partition matching, make construction risk decisions, and determine the risk list; a visualization display module, the visualization display module is used to identify the risk list, determine the risk target and perform terminal visualization display based on information asymmetry processing and information reconstruction.

[0008] One or more technical solutions provided in this application have at least the following technical effects or advantages: By analyzing the risk probability events of the target construction project, based on the collaborative analysis requirements, the fusion sequence is determined, and each data fusion sequence is marked with a fusion characteristic; the multi-level fusion scale is determined, and the multi-scale fusion based on the fusion dimension is used as the processing requirement. Based on the fusion sequence, a multi-level fusion model is constructed, wherein the fusion dimension at least includes the data level, the feature level and the decision level; based on the fusion sequence, a monitoring sensor array is deployed; based on the monitoring sensor array, the whole domain monitoring of the project construction is carried out to determine the source data group; the source data group is transmitted back, and unit matching and partition matching are carried out based on the multi-level fusion model with event as the guide, and construction risk decisions are made to determine the risk list; the risk list is identified, and the risk target is determined based on information non-balanced processing and information reconstruction, and the terminal visualization display is performed. In other words, by integrating multiple data sources and adopting a multi-level fusion model, comprehensive monitoring and analysis of construction project risks are achieved, and construction safety and efficiency are improved.

[0009] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented according to the contents of the specification, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are specifically cited below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solutions in the present application or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0011] Figure 1 This is a flow chart of the construction site security display method based on multi-source data fusion in this application.

[0012] Figure 2 This is a schematic diagram of the structure of the construction site security display platform that integrates multi-source data in this application.

[0013] Explanation of the accompanying drawings: data collection module 11, data processing module 12, data fusion module 13, global monitoring module 14, risk decision module 15, visualization module 16. DETAILED DESCRIPTION

[0014] This application solves the technical problem of low efficiency of construction site safety management due to lack of effective data fusion in safety monitoring of construction sites in the prior art by providing a construction site security display method and platform with multi-source data fusion. By integrating multiple data sources and adopting a multi-level fusion model, comprehensive monitoring and analysis of construction project risks are achieved, thereby improving construction safety and efficiency.

[0015] Below, the technical solutions in the present application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments of the present application. It should be understood that the present application is not limited to the example embodiments described herein. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application. It should also be noted that, for the convenience of description, only the parts related to the present application are shown in the accompanying drawings, rather than all of them.

[0016] For example, please refer to the attached Figure 1 The present application provides a construction site security display method based on multi-source data fusion, wherein the construction site security display method based on multi-source data fusion is applied to a construction site security display platform based on multi-source data fusion, and the construction site security display method based on multi-source data fusion specifically includes the following steps: Step 1: Analyze the risk probability events of the target construction project, determine the fusion sequence based on the collaborative analysis requirements, and mark each data fusion sequence with fusion characteristics.

[0017] Specifically, identify the risk events that may occur in the target construction project, such as structural collapse, mechanical failure, material defects, casualties, environmental impact, etc. Perform a probability assessment on each identified risk event to determine its likelihood of occurrence and potential impact. Collaborative analysis refers to the comprehensive analysis of data from different data sources to improve the accuracy and efficiency of the analysis. Consider the characteristics of the data source, such as data type, data quality, data volume, etc. At the same time, consider the goal of the analysis, that is, what information needs to be extracted from the data and how to integrate this information. Determine the order and method of data fusion, as well as the data type, source, processing method and purpose of each fusion sequence to ensure that the data fusion process meets the specific needs of the project and improve the efficiency and accuracy of data fusion. Identify fusion characteristics for each data fusion sequence, including fusion method, fusion granularity, etc. Fusion method refers to how to integrate data from different data sources, such as direct fusion, fusion after feature extraction, etc. Fusion granularity refers to the level of data fusion, such as data-level fusion, feature-level fusion, etc. By analyzing the risk probability events of the target construction project and determining the fusion sequence based on the collaborative analysis requirements, possible safety risks can be identified and understood in advance, so that preventive measures can be taken.

[0018] Step 2: Determine the multi-level fusion scale, take the multi-scale fusion based on the fusion dimension as the processing requirement, and build a multi-level fusion model based on the fusion sequence, wherein the fusion dimension at least includes the data level, the feature level and the decision level.

[0019] Specifically, the multi-level fusion scale refers to the degree and granularity of data fusion at different levels. The multi-level fusion scale is determined by considering the different characteristics and requirements of the data level, feature level and decision level. The fusion dimension refers to the different aspects or levels of data fusion, such as the data level, feature level and decision level. Data-level fusion mainly integrates the original data, feature-level fusion extracts feature information from the data for fusion, and decision-level fusion makes decisions based on the results of the first two levels of fusion. Multi-scale fusion is performed based on the fusion dimension, and data is fused at different levels to meet the data fusion requirements at different levels. For example, at the data level, the original data from different data sources are integrated for subsequent analysis. At the feature level, key features in the data, such as worker activity patterns, equipment operating status, etc., are extracted and fused to facilitate a deeper understanding of the meaning behind the data. At the decision level, decisions on construction risk management are made based on the results of the first two levels of fusion, such as whether construction needs to be suspended or whether safety measures need to be added. The fusion sequence refers to the order and process of data fusion, which guides how data is processed at different fusion scales. Building a multi-level fusion model requires considering the relationship and interaction between different fusion scales to ensure that the model can effectively process and analyze multi-source data. Data-level fusion provides basic data support, feature-level fusion reveals the key information behind the data, and decision-level fusion makes effective risk decisions based on this information. Such hierarchical processing not only improves the efficiency of data fusion, but also enhances the accuracy and reliability of risk analysis.

[0020] Step 3: Based on the fusion sequence, deploy a monitoring sensor array.

[0021] Specifically, we need to deeply understand the data needs and processing requirements of each step in the fusion sequence, select appropriate monitoring equipment, such as video surveillance cameras, temperature sensors, sound sensors, vibration sensors, etc., and collect data from different angles and aspects of the construction site. Based on the situation on the construction site, we design the layout of the monitoring sensor array, and according to the layout plan, we arrange the monitoring equipment on the construction site to ensure coverage of key areas and risk points. When deploying the monitoring sensor array, we also need to consider the order and process of data fusion. For example, if some data sources need to be fused before other data sources, the corresponding monitoring sensors should be deployed first. At the same time, we also need to ensure that the coverage of the monitoring sensors is wide enough to collect data from the entire construction site. By deploying the monitoring sensor array, we can ensure that high-quality and highly relevant data can be collected.

[0022] Step 4: Based on the monitoring sensor array, conduct full-area monitoring of the project construction and determine the source data group.

[0023] Specifically, the deployed monitoring sensor array is used to conduct global monitoring of the construction project, including video monitoring, environmental monitoring, equipment status monitoring, etc., to timely discover and deal with potential risks. The raw data collected from the monitoring sensor array is sorted and classified according to certain rules and standards to form a group of data sets for subsequent processing and analysis. These data groups are divided according to the data source, data type or monitoring purpose. For example, data from the same area are divided into a data group, or data of the same type (such as temperature data) are divided into a data group. Preliminary quality control is performed on the collected data, such as removing noise and correcting errors, to ensure the quality and reliability of the data set. At the same time, it is also necessary to ensure the integrity of the data set, that is, to include all required data types and monitoring points, so as to facilitate subsequent data fusion and analysis. Through global monitoring and the determination of source data groups, a comprehensive understanding and monitoring of the construction site can be achieved.

[0024] Step 5: The source data group is transmitted back, and unit matching and partition matching are performed based on the multi-level fusion model in an event-oriented manner to make construction risk decisions and determine risk orders.

[0025] Specifically, the collected construction site data is transmitted from the monitoring sensor array to the data processing center, and risk-related events such as personnel violations and equipment abnormalities are identified from the source data group. Unit matching refers to matching the collected data with the corresponding units in the multi-level fusion model so that the model can correctly process and analyze the data. Partition matching refers to matching the data with different fusion partitions in the model so that the model can perform corresponding fusion processing according to the characteristics and attributes of the data. In the process of unit matching and partition matching, the multi-level fusion model will process the data according to preset rules and algorithms to identify potential construction risks. Based on the analysis results of the multi-level fusion model, construction risk decisions are made and risk lists are determined. Risk lists refer to specific risk events identified, including risk type, risk level, possible impact and recommended countermeasures. After determining the risk list, corresponding preventive or countermeasures can be taken in time to ensure the safety of the construction site. Based on the source data group collected by the monitoring sensor array, a multi-level fusion model is used to perform unit matching and partition matching, make construction risk decisions, and finally determine the risk list, which helps the project team to promptly identify and respond to risks in the construction process and ensure the safety and smooth progress of the project.

[0026] Step 6: Identify the risk list, determine the risk target based on information imbalance processing and information reconstruction, and perform terminal visualization.

[0027] Specifically, risk orders are identified based on the results of risk decision-making. Specific construction risk events are determined from the analysis results of the multi-level fusion model, including different types of risks such as worker injuries, equipment failures, and material defects. Each risk order contains information such as risk type, risk level, possible impact, and recommended countermeasures. Information asymmetry processing refers to taking appropriate processing measures when facing incomplete, inaccurate, or inconsistent information to ensure the accuracy of risk identification and prevention. For example, if the data of a monitoring sensor is lost or abnormal, it needs to be supplemented or corrected by data from other sensors. Information reconstruction refers to reorganizing and reconstructing the collected data on the basis of information asymmetry processing to facilitate more accurate identification and understanding of risks. The most valuable information is extracted from the data to provide support for risk identification and prevention. Based on the results of information asymmetry processing and information reconstruction, risk targets are determined, that is, specific risks that need to be paid attention to and managed. Terminal visualization is performed to display information such as the type, level, possible impact, and recommended countermeasures of equipment failures through charts or images on terminal devices, so that managers can understand the risk situation in a timely manner and take corresponding measures. By defining targets and presenting them visually, we ensure that risks can be accurately identified and understood, and provide managers with an intuitive and easy-to-understand display of risk information.

[0028] Furthermore, step one of this application includes: Identify the data fusion sequence and determine the data phase rule, wherein the data phase rule includes a monitoring trigger phase and a synchronous return phase; based on the monitoring trigger phase, perform construction monitoring management on the monitoring sensor array; based on the synchronous return phase, perform data transmission management on the communication network in the construction site area.

[0029] Specifically, the characteristics of each data fusion sequence are identified, including sequence identification, data source, fusion characteristics (fusion mode, fusion granularity), etc., and the data phase rules are determined, including the monitoring trigger phase and the synchronous return phase. The monitoring trigger phase refers to the moment when each sensor in the monitoring sensor array starts to collect data. It is necessary to consider the synchronization and real-time of data collection to ensure that all sensors start collecting data at the same time point or according to the predetermined time sequence. Synchronicity ensures the consistency of data, while real-time ensures that the data can reflect the current situation of the construction site. According to the needs of data fusion, set specific trigger conditions for each sensor, such as timed trigger, event trigger, etc. For example, an environmental sensor can trigger monitoring when the wind speed exceeds a certain threshold. The monitoring sensor array is a series of sensors deployed at the construction site, responsible for collecting various types of data, such as video surveillance, temperature, humidity, vibration, sound, etc. Through a precise time synchronization mechanism, ensure that all sensors start collecting data at a predetermined time point or according to a specific time sequence, such as using the Network Time Protocol (NTP) or other time synchronization technologies. The purpose of synchronization is to ensure data consistency and avoid data analysis errors caused by time deviation. The Network Time Protocol (NTP) is a protocol used to synchronize computer clocks so that the computer time is consistent with UTC (Coordinated Universal Time). According to the monitoring trigger phase, the sampling frequency and monitoring strategy of the sensor are adjusted. Monitoring management also includes monitoring and maintaining the status of the sensor to ensure normal operation during the monitoring period, such as regularly checking the working status of the sensor, calibrating the sensor to maintain its accuracy, and replacing or repairing the sensor in a timely manner when it fails.

[0030] According to the specific conditions of the construction site and the data transmission requirements, the communication network in the construction site area is determined. It is the bridge connecting the sensors and the central processing system, responsible for transmitting the collected data from the sensors to the central processing system. The synchronous return phase refers to the moment when the collected data is transmitted from the sensor to the central processing system. The synchronization and real-time nature of the data also need to be considered to ensure that the data can reach the central processing system in time for analysis and processing. Synchronicity ensures the consistency of the data, while real-time ensures that the data can reflect the changes on the construction site in a timely manner, providing support for risk identification and decision-making. Set a time window for data return according to the data transmission requirements and network conditions. Ensure that the communication network in the construction site area supports the data transmission requirements, including bandwidth, latency, etc. Arrange the priority and timing of data transmission according to the synchronous return phase to avoid data congestion. Provide accurate time synchronization for the entire network through the NTP server to ensure the accuracy and timeliness of data return. By identifying the data fusion sequence and determining the data phase law, the consistency and real-time nature of data collection and transmission can be ensured, thereby improving the quality and availability of data.

[0031] Further, step 2 of this application includes: Based on the fusion granularity, a first fusion constraint is determined; based on the information loss degree, a second fusion constraint is determined; based on the multi-source fusion subject and the subordinate fusion item, a third fusion constraint is determined, wherein the multi-source fusion subject is any at least one item in the multi-source data, which is determined based on the fusion ratio; the first fusion constraint, the second fusion constraint and the third fusion constraint are randomly combined, and the parallel fusion requirement is determined in combination with the fusion sequence; based on the parallel fusion requirement, the multi-level fusion model is supervised for training.

[0032] Specifically, the fusion granularity refers to the level of detail of data fusion, which determines the fineness of data processing during the fusion process. Depending on specific needs, you can choose coarse data fusion or fine data fusion. Coarse data fusion refers to the rough processing and analysis of data during the data fusion process, which is suitable for scenarios with low requirements for data accuracy, such as trend analysis, overview analysis and other broad task analysis. Fine data fusion refers to the very fine processing and analysis of data during the data fusion process, which is suitable for scenarios with high requirements for fusion fineness, such as accurate event detection, detailed data analysis, etc. This fusion is based on the smallest data unit, such as the fusion of a single sensor data point.

[0033] Determine the first fusion constraint based on the specific data fusion tasks and requirements. The first fusion constraint is a rule or restriction applied in the initial stage of data fusion. It is a constraint on the level of detail or temporal resolution of data fusion to ensure the correctness and consistency of data fusion. For example, in the safety monitoring of a construction site, if the analysis task is to monitor the overall safety status of the site, a looser data cleaning and format unification rule, that is, coarse data fusion, can be adopted to quickly obtain the overall safety trend of the site. If the analysis task is to detect safety incidents in a specific area, a stricter data cleaning and format unification rule, that is, fine data fusion, may be adopted to ensure that safety incidents can be accurately detected and located. By determining the first fusion constraint based on the fusion granularity, the correctness and consistency of the data fusion process can be ensured, thereby improving the accuracy of data fusion.

[0034] Data loss refers to the degree of information loss caused by data processing and analysis during the data fusion process. According to the data loss, the second fusion constraint is determined, that is, the acceptable degree of loss during the fusion process, so as to balance the efficiency of fusion and the accuracy of the results. Under the premise of ensuring that the core information is not lost, a certain degree of secondary information loss is allowed to optimize the speed of data processing and resource consumption. The multi-source fusion subject refers to the data item that plays a leading role in the multi-source data, while the subordinate fusion item is other data item that complements the dominant data item. Determine the weight or proportion of the main data source and the subordinate data source in the fusion process, clarify which data sources are primary and which are auxiliary, and adjust the fusion strategy accordingly, determine the third fusion constraint, and clarify the constraints on the master-slave relationship between different data sources in multi-source data fusion.

[0035] The first fusion constraint, the second fusion constraint, and the third fusion constraint focus on the granularity of data fusion, the degree of information loss, and the relationship between the main and subordinate data, respectively. During the fusion process, these three constraints are combined and applied in different ways to generate different fusion strategies to explore the best fusion effect. The fusion sequence refers to the steps and order of data fusion, which determines the processing order of different data sources and data levels during the fusion process. The parallel fusion requirement means that during the data fusion process, different data sources and data levels can be processed in parallel to improve the efficiency and real-time performance of data fusion. Determine which fusion strategies can be executed in parallel at each stage of data fusion to ensure that the data can be fused quickly and accurately while meeting the requirements of various fusion constraints. According to the parallel fusion requirement, supervised training Furthermore, the present application also includes the following steps: Based on the parallel fusion requirement, an initialization fusion unit is constructed, and the initialization fusion unit includes multiple fusion partitions; the initialization fusion unit is subjected to unit twin processing to determine a parallel unit structure, wherein there is a corresponding relationship between the parallel unit structure and the fusion dimension, and the parallel unit structure is expandable; based on the parallel fusion requirement, the risk probability events are clustered and segmented, training samples are determined, the parallel unit structure is matched, and data-driven training is performed until convergence to generate the multi-level fusion model.

[0036] Specifically, an initialization fusion unit is constructed based on the parallel fusion requirements, which includes multiple fusion partitions that are responsible for processing different types of data and ensuring the real-time and efficiency of data fusion. Each fusion partition can be designed based on different data sources or fusion characteristics. For example, one partition may focus on the fusion of video surveillance data, while another partition may focus on the fusion of sensor data. The initialization fusion unit is twinned, that is, a virtual copy corresponding to the initialization fusion unit is created. The parallel unit structure refers to a structure that processes multiple data sources and data levels in parallel, and there is a corresponding relationship with the fusion dimension. The fusion dimensions include data level, feature level, and decision level, and each dimension corresponds to one or more parallel units. For example, data-level fusion corresponds to one parallel unit, feature-level fusion corresponds to another parallel unit, and decision-level fusion corresponds to a third parallel unit. The parallel unit structure can be expanded according to data processing requirements, which means that it can adapt to possible changes and expansions in the future.

[0037] Based on the parallel fusion requirements, the risk probability events are clustered and segmented, and similar risk events are classified into one category. Cluster segmentation refers to clustering similar risk probability events together to form different clusters or classes. Features are extracted from the clustered risk events to form a training sample set. The clustered training samples are matched with the parallel unit structure to ensure that each fusion partition can handle the corresponding risk event type. Data-driven training is performed on each fusion partition using the training samples, and the model is trained using machine learning algorithms such as deep learning, support vector machines, and random forests. Model parameters are adjusted to optimize model performance until the model converges. By matching the parallel unit structure and performing data-driven training, a multi-level fusion model that can accurately identify and predict safety risks at construction sites can be generated, and a model that handles data fusion of different levels and types can be generated. Through a data-driven approach, the model can learn how to identify and predict safety risks from a large number of samples, thereby improving the accuracy of data fusion.

[0038] Furthermore, the present application also includes the following steps: The risk probability events are traversed, and the phases are integrated as the project progresses to determine the risk event chain of the entire project cycle; the risk event chain is traversed to determine the pre-risk characteristics of the project construction, wherein the pre-risk characteristics are determined based on the preset time zone of the event link node; based on the pre-risk characteristics, the risk event chain is identified, and a fusion partition mapping between the risk event chain and the multi-level fusion model is established.

[0039] Specifically, all possible risk events in the project are systematically enumerated and evaluated to fully understand the possible safety risks of the project, and to classify and evaluate them. As the project progresses, the risk probability events are integrated according to the project timeline to track the evolution of risk probability events as the project progresses. The risk event chain refers to connecting risk probability events in chronological order and logical relationships to form a complete sequence of risk events. By determining the risk event chain of the entire project cycle, the evolution process and impact of risk events are fully understood, so as to timely discover and respond to potential safety risks. Comprehensively identify and record various risk events that may be encountered at the construction site. By traversing the risk event chain, the interrelationships and impacts between risk events are identified. Pre-risk characteristics refer to risk characteristics that affect construction safety before the project is constructed. Before the risk event occurs, potential risk signs or indicators are obtained by observing and analyzing historical data and similar projects, as well as the specific conditions of the construction site. Set a time window. If specific risk signs or indicators are observed within this time window, it can be considered that a risk event will occur. Pre-risk characteristics are determined by pushing the event link node forward to the preset time zone. That is to say, by analyzing historical data, we can find out some common characteristics before risk events occur, and use these characteristics as the basis for risk prediction. Once the pre-risk characteristics are observed, we need to take corresponding measures to prevent the corresponding risk events in time and avoid or reduce the impact of risk events.

[0040] Identify these pre-risk features and identify the risk event chain, that is, a series of risk events caused by these features. Use a unified identification system to ensure the traceability and consistency of the risk event chain. Develop a set of rules to map the events in the risk event chain to the corresponding fusion partitions in the multi-level fusion model according to their characteristics and requirements. Assign one or more fusion partitions to each risk event chain node to ensure that risk information can be properly processed and analyzed. As the project progresses and risk events change, dynamically adjust the fusion partition mapping to adapt to the new risk situation. Through this mapping relationship, the multi-level fusion model can more accurately identify and predict potential risk events. When pre-risk features are detected, the model can automatically trigger corresponding risk management measures, such as alarms, warnings, or preventive measures. By determining the risk event chain of the entire project cycle, the risks of the entire project cycle can be effectively managed, potential risks can be identified in advance, and monitored and analyzed through the multi-level fusion model, thereby improving the safety and efficiency of project construction.

[0041] Furthermore, step six of this application includes: Identify the risk list and determine the risk factors of the first risk event; based on the risk factors, determine the visualization standard, wherein the visualization standard includes the display method and information value; based on the visualization standard, reconstruct the information of the first risk event and perform terminal visualization display.

[0042] Specifically, according to the results of risk decision-making, the risk list is identified, the projects or areas with potential risks are identified, and the first risk event, that is, the primary risk listed in the risk list, is determined. Risk factors refer to the key factors or conditions that lead to the occurrence of risk events. The first risk event is analyzed in depth to determine the key factors that lead to its occurrence, such as the source of risk, possible impact, probability of occurrence, etc. For example, the risk factors that lead to worker injuries may include the worker's own safety awareness, the safety equipment used, the safety status of the working environment, etc. The display method refers to the method of displaying risk information in visual elements such as charts, images, and colors. Different risk factors may require different display methods. When determining the display method, it is necessary to consider the characteristics of the risk factor and the needs of risk management. For example, if the risk factor is related to the geographical location, then a map may be a suitable display method; if the risk factor is time-related, then a time series chart may be a suitable display method. Information value refers to the importance or priority of risk information display. When determining information value, it is necessary to consider the importance and impact of risk factors. For example, the information value of high-risk factors is higher and should be displayed in a more conspicuous way; the information value of low-risk factors is lower and can be displayed in a less conspicuous way.

[0043] According to the visualization standards, the information in the risk list is reconstructed to make it more in line with the display requirements. Information reconstruction refers to the reorganization and construction of the collected data based on the information non-balanced processing, so as to more accurately identify and understand the risks. The reconstructed information is visualized on the terminal device (such as a computer, mobile phone, etc.) to ensure that the user can intuitively understand the risk situation. For example, visual elements such as charts, images, and colors are used to display risk information in an intuitive and easy-to-understand way. By identifying the risk list, determining the risk factors, and reconstructing and terminal displaying the information according to the visualization standards, decision makers can better understand and respond to project risks.

[0044] Furthermore, the present application also includes the following steps: Based on the information value, a first-layer weight distribution is determined for risk factors and non-risk factors, wherein the weight distribution of the risk factors is higher than that of the non-risk factors; based on the information value, a second-layer weight distribution is determined for factor strength and factor correlation; based on the first-layer weight distribution and the second-layer weight distribution, the information structure is determined based on the display method and an unbalanced reconstruction is performed to generate the risk target.

[0045] Specifically, risk factors refer to factors that directly cause or affect construction risks, such as workers' safety awareness and equipment maintenance status. Non-risk factors refer to factors that are not directly related to construction risks but have a certain impact on the construction process, such as weather conditions and construction progress. Different weights are assigned to risk factors and non-risk factors according to their impact on construction safety to ensure that high-risk factors receive sufficient attention. In this process, the weight distribution of risk factors is usually higher than that of non-risk factors because they are directly related to the risk level of construction safety. Factor intensity refers to the impact of risk factors or non-risk factors on construction safety. Factor correlation refers to the relationship and impact between risk factors or non-risk factors. Different weights are assigned to risk factors and non-risk factors according to factor intensity and factor correlation. In this process, risk factors and non-risk factors with higher factor intensity and factor correlation should be given higher weights to ensure that they receive sufficient attention and processing in the data analysis and decision-making process.

[0046] Reasonable deployment is carried out according to the value of information to ensure that key risk information can be highlighted. At the same time, redundant information is removed to improve the clarity and comprehensibility of information. Information structure refers to the organization and display form of risk information, which is determined based on the display method. For example, if the complex relationship between risk factors needs to be displayed, information mapping is a suitable display method; if the risk distribution in three-dimensional space needs to be displayed, three-dimensional picture reconstruction is a suitable display method. Unbalanced reconstruction refers to reorganizing and adjusting information according to the importance and priority of risk factors to highlight key risk factors. In this way, risk targets can be generated, that is, specific goals that need to be achieved in the risk management process, such as reducing risks and improving safety. According to the reconstructed risk information, specific and quantifiable risk management goals are set. Determining the information structure based on the first-layer weight distribution and the second-layer weight distribution, as well as the display method and performing unbalanced reconstruction can help improve the efficiency and effectiveness of construction safety management, thereby ensuring the safety and smooth progress of the construction site.

[0047] In summary, the construction site security display method based on multi-source data fusion provided by this application has the following technical effects: By analyzing the risk probability events of the target construction project, based on the collaborative analysis requirements, the fusion sequence is determined, and each data fusion sequence is marked with a fusion characteristic; the multi-level fusion scale is determined, and the multi-scale fusion based on the fusion dimension is used as the processing requirement. Based on the fusion sequence, a multi-level fusion model is constructed, wherein the fusion dimension at least includes the data level, the feature level and the decision level; based on the fusion sequence, a monitoring sensor array is deployed; based on the monitoring sensor array, the whole domain monitoring of the project construction is carried out to determine the source data group; the source data group is transmitted back, and unit matching and partition matching are carried out based on the multi-level fusion model with event as the guide, and construction risk decisions are made to determine the risk list; the risk list is identified, and the risk target is determined based on information non-balanced processing and information reconstruction, and the terminal visualization display is performed. In other words, by integrating multiple data sources and adopting a multi-level fusion model, comprehensive monitoring and analysis of construction project risks are achieved, and construction safety and efficiency are improved.

[0048] Embodiment 2, based on the same inventive concept as the construction site security display method of multi-source data fusion in the previous embodiment, this application also provides a construction site security display platform of multi-source data fusion, please refer to the attached Figure 2 The multi-source data fusion construction site security display platform includes: The data acquisition module 11 is used to analyze the risk probability events of the target construction project, determine the fusion sequence based on the collaborative analysis requirements, and each data fusion sequence is marked with a fusion characteristic.

[0049] The data processing module 12 is used to determine the multi-level fusion scale, take the multi-scale fusion based on the fusion dimension as the processing requirement, and build a multi-level fusion model based on the fusion sequence, wherein the fusion dimension at least includes the data level, the feature level and the decision level.

[0050] The data fusion module 13 is used to deploy a monitoring sensor array based on the fusion sequence.

[0051] The global monitoring module 14 is used to perform global monitoring of the project construction based on the monitoring sensor array and determine the source end data group.

[0052] The risk decision module 15 is used to transmit the source data group back, perform unit matching and partition matching based on the multi-level fusion model in an event-oriented manner, make construction risk decisions, and determine risk orders.

[0053] The visualization display module 16 is used to identify the risk list, determine the risk target and perform terminal visualization based on information imbalance processing and information reconstruction.

[0054] Furthermore, the data acquisition module 11 in the multi-source data fusion construction site security display platform is also used for: Identify the data fusion sequence and determine the data phase rule, wherein the data phase rule includes a monitoring trigger phase and a synchronous return phase; based on the monitoring trigger phase, perform construction monitoring management on the monitoring sensor array; based on the synchronous return phase, perform data transmission management on the communication network in the construction site area.

[0055] Furthermore, the data processing module 12 in the multi-source data fusion construction site security display platform is also used for: Based on the fusion granularity, a first fusion constraint is determined; based on the information loss degree, a second fusion constraint is determined; based on the multi-source fusion subject and the subordinate fusion item, a third fusion constraint is determined, wherein the multi-source fusion subject is any at least one item in the multi-source data, which is determined based on the fusion ratio; the first fusion constraint, the second fusion constraint and the third fusion constraint are randomly combined, and the parallel fusion requirement is determined in combination with the fusion sequence; based on the parallel fusion requirement, the multi-level fusion model is supervised for training.

[0056] Furthermore, the multi-source data fusion construction site security display platform also includes a fusion model building module for: Based on the parallel fusion requirement, an initialization fusion unit is constructed, and the initialization fusion unit includes multiple fusion partitions; the initialization fusion unit is subjected to unit twin processing to determine a parallel unit structure, wherein there is a corresponding relationship between the parallel unit structure and the fusion dimension, and the parallel unit structure is expandable; based on the parallel fusion requirement, the risk probability events are clustered and segmented, training samples are determined, the parallel unit structure is matched, and data-driven training is performed until convergence to generate the multi-level fusion model.

[0057] Furthermore, the data processing module 12 in the multi-source data fusion construction site security display platform is also used for: The risk probability events are traversed, and the phases are integrated as the project progresses to determine the risk event chain of the entire project cycle; the risk event chain is traversed to determine the pre-risk characteristics of the project construction, wherein the pre-risk characteristics are determined based on the preset time zone of the event link node; based on the pre-risk characteristics, the risk event chain is identified, and a fusion partition mapping between the risk event chain and the multi-level fusion model is established.

[0058] Furthermore, the visualization display module 16 in the multi-source data fusion construction site security display platform is also used for: Identify the risk list and determine the risk factors of the first risk event; based on the risk factors, determine the visualization standard, wherein the visualization standard includes the display method and information value; based on the visualization standard, reconstruct the information of the first risk event and perform terminal visualization display.

[0059] Furthermore, the multi-source data fusion construction site security display platform also includes an information reconstruction module for: Based on the information value, a first-layer weight distribution is determined for risk factors and non-risk factors, wherein the weight distribution of the risk factors is higher than that of the non-risk factors; based on the information value, a second-layer weight distribution is determined for factor strength and factor correlation; based on the first-layer weight distribution and the second-layer weight distribution, the information structure is determined based on the display method and an unbalanced reconstruction is performed to generate the risk target.

[0060] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. Figure 1 The construction site security display method and specific examples of multi-source data fusion in Example 1 are also applicable to the construction site security display platform of multi-source data fusion in this embodiment. Through the detailed description of the construction site security display method of multi-source data fusion, those skilled in the art can clearly know the construction site security display platform of multi-source data fusion in this embodiment, so for the sake of brevity of the specification, it will not be described in detail here. For the platform disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description.

[0061] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

[0062] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application belong to the scope of the present application and its equivalent technology, the present application is also intended to include these modifications and variations.

Claims

1. A construction site security display method based on multi-source data fusion, characterized in that: include: Analyze the risk probability events of the target construction project, determine the fusion sequence based on the collaborative analysis requirements, and mark each data fusion sequence with fusion characteristics; Determine a multi-level fusion scale, take multi-scale fusion based on fusion dimensions as a processing requirement, and construct a multi-level fusion model based on the fusion sequence, wherein the fusion dimensions at least include a data level, a feature level, and a decision level; deploying a monitoring sensor array based on the fusion sequence; Based on the monitoring sensor array, the whole-area monitoring of the project construction is carried out to determine the source end data group; The source data group is transmitted back, and unit matching and partition matching are performed based on the multi-level fusion model in an event-oriented manner, so as to make construction risk decisions and determine risk orders; Identify the risk list, determine the risk target and perform terminal visualization based on information imbalance processing and information reconstruction.

2. The construction site security display method of multi-source data fusion according to claim 1, characterized in that: After determining the data fusion sequence, it includes: Identify the data fusion sequence and determine the data phase rule, wherein the data phase rule includes monitoring trigger phase and synchronous return phase; Based on the monitoring trigger phase, performing construction monitoring management on the monitoring sensor array; Based on the synchronous return phase, data transmission management is performed on the communication network in the construction site area.

3. The construction site security display method of multi-source data fusion according to claim 1, characterized in that: The multi-level fusion model is constructed, comprising: Based on the fusion granularity, determine a first fusion constraint; Determining a second fusion constraint based on the information loss degree; Determining a third fusion constraint based on the multi-source fusion subject and the subordinate fusion item, wherein the multi-source fusion subject is any at least one item in the multi-source data, determined based on the fusion ratio; Randomly combining the first fusion constraint, the second fusion constraint and the third fusion constraint, and combining the fusion sequence to determine a parallel fusion requirement; Based on the parallel fusion requirement, supervise the training of the multi-level fusion model.

4. The construction site security display method of multi-source data fusion as claimed in claim 3, characterized in that: Based on the parallel fusion requirement, supervised training of the multi-level fusion model includes: Based on the parallel fusion requirement, construct an initialization fusion unit, wherein the initialization fusion unit includes multiple fusion partitions; Performing unit twinning processing on the initialized fusion unit to determine a parallel unit structure, wherein the parallel unit structure corresponds to the fusion dimension and the parallel unit structure is scalable; Based on the parallel fusion requirements, the risk probability events are clustered and segmented, training samples are determined, the parallel unit structure is matched, and data-driven training is performed until convergence to generate the multi-level fusion model.

5. The construction site security display method of multi-source data fusion as claimed in claim 4, characterized in that: Also includes: Traverse the risk probability events, integrate them as the project progresses, and determine the risk event chain for the entire project cycle; Traversing the risk event chain to determine the pre-risk characteristics of the project construction, wherein the pre-risk characteristics are determined based on the event link node forward by a preset time zone; Based on the pre-risk feature, the risk event chain is identified, and a fusion partition mapping between the risk event chain and the multi-level fusion model is established.

6. The construction site security display method of multi-source data fusion as claimed in claim 1, characterized in that: The determination of risk targets and terminal visualization display include: Identify the risk list and determine the risk factors of the first risk event; Based on the risk factors, determining visualization standards, wherein the visualization standards include display methods and information value; Based on the visualization standard, information of the first risk event is reconstructed and displayed visually on a terminal.

7. The construction site security display method of multi-source data fusion as claimed in claim 6, characterized in that: Based on the visualization standard, information reconstruction of the first risk event includes: Based on the information value, a weight distribution layer is determined for risk factors and non-risk factors, wherein the weight distribution of the risk factors is higher than that of the non-risk factors; Based on the information value, a two-layer weight distribution is determined for the element intensity and the element relevance; Based on the first-layer weight distribution and the second-layer weight distribution, the information structure is determined based on the display method and an unbalanced reconstruction is performed to generate the risk target.

8. The construction site security display platform with multi-source data fusion is characterized by: The steps for implementing the construction site security display method of multi-source data fusion as described in any one of claims 1 to 7, wherein the construction site security display platform of multi-source data fusion comprises: A data acquisition module, wherein the data acquisition module is used to analyze the risk probability events of the target construction project, and determine the fusion sequence based on the collaborative analysis requirements, and each data fusion sequence is marked with a fusion characteristic; A data processing module, the data processing module is used to determine the multi-level fusion scale, take the multi-scale fusion based on the fusion dimension as the processing requirement, and build a multi-level fusion model based on the fusion sequence, wherein the fusion dimension at least includes the data level, the feature level and the decision level; A data fusion module, the data fusion module is used to deploy a monitoring sensor array based on the fusion sequence; A global monitoring module, the global monitoring module is used to perform global monitoring of project construction based on the monitoring sensor array and determine the source end data group; A risk decision module, which is used to transmit the source data group back, perform unit matching and partition matching based on the multi-level fusion model in an event-oriented manner, make construction risk decisions, and determine risk orders; A visualization display module is used to identify the risk list, determine the risk target and perform terminal visualization based on information imbalance processing and information reconstruction.

Citation Information

Cited By

  • Titanium rod surface wear morphology detection method and system based on multispectral image fusion

    CN120125573A

  • Titanium rod surface wear morphology detection method and system based on multispectral image fusion

    CN120125573B

  • Intelligent data fusion system and method for construction project

    CN121935863A