Engineering construction safety monitoring method
By collecting historical data to train LSTM models and construct BIM three-dimensional models, combined with multi-source detection and wavelet noise reduction technology, accurate risk warnings for construction sites are achieved, solving the problem of large errors in existing monitoring systems and ensuring construction safety.
Patent Information
- Application Number
- CN202510928878.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-07
- Publication Date
- 2025-10-17
AI Technical Summary
The existing engineering construction monitoring system has large detection errors and is unable to make accurate judgments based on the differences in construction sites and building types, resulting in frequent accidents at construction sites and endangering worker safety.
By collecting historical monitoring data of similar projects, using LSTM model training to generate safety thresholds, and combining the detection module and BIM engine to construct a three-dimensional model, accurate graded hazard warnings can be achieved, including multi-source detection of displacement deformation, stress data and environmental data. Wavelet noise reduction and edge computing are used for data processing to generate risk characteristic values, and warning information is displayed on the AR terminal.
It achieves accurate risk warning for construction sites, reduces false alarms, and can timely prevent construction hazards, ensure worker safety, and reduce economic losses.
Smart Images
Figure CN120806929A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the technical field of civil engineering, more particularly, it relates to an engineering construction safety monitoring method. BACKGROUND
[0002] Civil engineering refers to the science and technology of building various land engineering facilities; it refers to the materials, equipment and technical activities such as surveying, designing, constructing, maintaining and repairing, and also refers to the object of engineering construction; that is, various engineering facilities built on land or underground, on land, directly or indirectly serving human life, production, military and scientific research.
[0003] The topographical environment of a large civil engineering construction site is generally very complex, and because of its large construction scale, fast speed, long construction period, complex cross-operation, and a large number of people from the owner, construction and supervision parties, the people, machines and environment are complex and changeable, so that various accidents occur on the construction site from time to time, causing serious social impact and significant economic losses, endangering the safety of workers' lives, so it is necessary to monitor the construction site by using an engineering construction safety monitoring system, but the existing monitoring system has a large detection error, and the existing monitoring system uses a general monitoring method, which cannot make accurate judgments according to the actual construction site and the different types of buildings.
[0004] Therefore, a new scheme needs to be proposed to solve this problem. SUMMARY
[0005] The purpose of the embodiment of the present application is to provide an engineering construction safety monitoring method to solve the above problems.
[0006] The above technical purpose of the embodiment of the present application is realized by the following technical scheme:
[0007] In a first aspect, historical monitoring data of the same type of engineering is collected, including structural displacement value, stress value, environmental parameter and accident record;
[0008] A plurality of detection modules are arranged at key nodes of the construction site, and displacement deformation data, stress data and environmental data of the building are collected by the detection modules;
[0009] The displacement deformation data, stress data and environmental data measured by the detection modules are obtained based on a risk feature analysis module to generate a risk feature value;
[0010] An LSTM prediction model is trained based on the historical monitoring data, and a safety threshold is generated by inputting the displacement deformation data, stress data and environmental data, and when the risk feature value exceeds the safety threshold, a graded danger warning information is triggered;
[0011] The three-dimensional model of the construction site is constructed through a BIM engine, and the danger early warning information is displayed on an AR terminal.
[0012] The application further provides that the generation of the risk characteristic value comprises:
[0013] The displacement deformation data is wavelet denoised based on a wavelet denoising model.
[0014] The stress data and the environmental data are analyzed based on an edge computing gateway, and the risk characteristic value is output in combination with wavelet denoising.
[0015] The application further provides that the risk characteristic value calculation formula is:
[0016]
[0017] wherein G represents the risk characteristic value, Delta F represents the stress change amount per unit time, Delta t represents the detection time interval, sigma represents the standard dry strength of the building material, beta represents the humidity sensitivity coefficient of the building material, RH represents the relative humidity, and RH0 represents the optimal humidity of the building material.
[0018] When the difference between the theoretical stress data and the measured actual stress data is less than 10%, the Delta F adopts the stress change amount of the actual stress data; when the difference between the theoretical stress data and the measured actual stress data is greater than 10%, the Delta F adopts the stress change amount of the theoretical stress data.
[0019] The application further provides that the calculation formula of the theoretical stress data is:
[0020]
[0021] wherein E represents the material elastic modulus, Delta L represents the change amount of the wavelet denoised displacement deformation data, and L0 represents the initial length of the building.
[0022] The application further provides that the graded danger early warning information comprises:
[0023] When the risk characteristic value is less than or equal to 0.8 times of the safety threshold value, the construction is in a safe state, and normal construction is carried out; when the risk characteristic value is greater than 0.8 times of the safety threshold value and less than or equal to the safety threshold value, a first-level early warning is started, and a warning alarm is issued; when the risk characteristic value is greater than the safety threshold value and less than or equal to 1.5 times of the safety threshold value, the person in charge is notified and timely maintenance is carried out; when the risk characteristic value is greater than or equal to 1.5 times of the safety threshold value, the power is cut off and the workers are urgently evacuated.
[0024] The application further provides that the detection module comprises a displacement sensor, a stress sensor and an environmental temperature and humidity sensor.
[0025] In a second aspect, a computer readable storage medium is provided, the computer readable storage medium storing a computer program, the computer program, when executed by a computer or a processor, implementing the method of any one of the preceding claims.
[0026] In a third aspect, a computer program product is provided, the computer program product comprising a computer program, the computer program, when executed by a computer or a processor, causing the computer or the processor to perform the method of any one of the preceding claims.
[0027] In summary, the present application has the following beneficial effects:
[0028] By collecting historical monitoring data of similar projects and using the LSTM model to train and predict the historical data, the early warning threshold of such projects can be generated, thereby helping to effectively warn and prevent safety accidents.
[0029] By detecting the construction site through the detection module, various risks of the construction site can be coupled from multiple aspects, false positives caused by single data can be eliminated, early warning monitoring can be more accurate, different early warning alarms can be sent for different degrees of building deformation through the setting of the hierarchical dangerous early warning information, more accurate judgments can be made on the deformation, the building can be maintained and repaired under the condition of reducing losses, and through the BIM engine to construct the three-dimensional model of the construction site, any place of the construction site can be quickly and accurately surveyed, the place of the danger early warning can be deployed in time, and effective and rapid protection can be performed. BRIEF DESCRIPTION OF DRAWINGS
[0030] Figure 1 FIG. 1 is a flowchart of the engineering construction safety monitoring method of the present application;
[0031] Figure 2 FIG. 1 is a flowchart of the engineering construction safety monitoring method of the present application; DETAILED DESCRIPTION
[0032] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative labor fall within the scope of the present application.
[0033] In a feasible embodiment, referring to FIGS. 1, 2 and 3, Figure 1 and Figure 2 An engineering construction safety monitoring method, the engineering construction safety monitoring method comprising:
[0034] Step 101: Collect historical monitoring data of similar projects, including structural displacement values, stress values, environmental parameters, and accident records;
[0035] Step 102: Arrange multiple detection modules at key nodes of the construction site, and collect displacement deformation data, stress data, and environmental data of the building through the detection modules;
[0036] Step 103: Obtain displacement deformation data, stress data, and environmental data based on the risk feature analysis module to generate risk features;
[0037] Step 104: Train an LSTM prediction model based on historical monitoring data, input displacement deformation data, stress data, and environmental data to generate a safety threshold. When the risk characteristic exceeds the safety threshold, trigger a graded hazard warning message.
[0038] Step 105: Construct a three-dimensional model of the construction site through the BIM engine and display the danger warning information on the AR terminal.
[0039] Specifically, by collecting historical monitoring data from similar projects and using the LSTM model to train and predict the historical data, and inputting displacement deformation data, stress data, and environmental data, it is possible to generate warning thresholds for similar projects, thereby facilitating effective early warning and prevention of safety accidents.
[0040] Furthermore, through the detection module, multi-source detection is carried out on the construction site, so that various risks on the construction site can be coupled from multiple aspects, false alarms caused by single data can be eliminated, and early warning monitoring can be made more accurate. At the same time, by setting graded danger warning information, different early warning alarms are issued for different degrees of building deformation, so that more accurate judgments can be made on the deformation sites, and the buildings can be maintained and repaired while reducing losses. At the same time, a three-dimensional model of the construction site is constructed through the BIM engine, so that any part of the construction site can be quickly and accurately surveyed, so that the danger warning locations can be deployed in time for effective and rapid protection.
[0041] Specifically, in step 102, the detection module includes a displacement sensor, a stress sensor, an environmental temperature and humidity sensor, and an image acquisition module. The displacement sensor is used to monitor the displacement deformation data of the building, the stress sensor is used to monitor the stress data of the building, and the environmental data on site is monitored by the environmental humidity sensor, mainly including the temperature and humidity of the environment. The image acquisition module is used to identify the human portrait.
[0042] Specifically, in step 103, the risk feature analysis module obtains various data collected by the detection module in real time through the wireless transmission module, and generates risk features including:
[0043] Based on the YOLOV5 model, the image data is analyzed in real time to identify safety violations, that is, through image data monitoring, whether workers wear safety helmets and other behaviors, to protect the safety of workers;
[0044] Based on the wavelet denoising model, the displacement deformation data is wavelet denoised. Wavelet denoising can extract the true structure deformation signal from strong noise, provide pure data base for accurate early warning, reduce external dryness, and make the obtained displacement deformation data more accurate.
[0045] Based on the edge computing gateway, the stress data and environmental data are analyzed, combined with wavelet denoising, and the risk characteristic value is output. By analyzing the stress data and environmental data, the change value of stress and the thermal expansion and contraction caused by temperature change are measured, so that more accurate risk characteristic value is obtained, and more accurate data is provided for subsequent graded danger early warning information.
[0046] Specifically, the risk characteristic value calculation formula is:
[0047]
[0048] Where G represents the risk characteristic value, ΔF represents the change of stress per unit time, Δt represents the detection time interval, σ represents the standard dry strength of building materials, β represents the humidity sensitivity coefficient of building materials, RH represents the relative humidity, and RH0 represents the best humidity of building materials.
[0049] When the difference between the theoretical stress data and the measured actual stress data is less than 10%, then ΔF adopts the stress change of the actual stress data; when the difference between the theoretical stress data and the measured actual stress data is greater than 10%, then ΔF adopts the stress change of the theoretical stress data. Since the actual measured stress data will be disturbed by many factors, there is a certain difference between the actual stress, so the theoretical stress data is introduced. When calculating the risk characteristic value, dynamically select the theoretical stress data or the actual measured stress data, which can obtain more accurate risk characteristic value.
[0050] Further, the calculation formula of the theoretical stress data is:
[0051]
[0052] Where E represents the elastic modulus of the material, ΔL represents the change of the wavelet denoised displacement deformation data, and L0 represents the initial length of the building.
[0053] Specifically, when the monitoring object is steel, the E of the steel is 210 Gpa, the support length of the steel is 10 m, and the AL after the wavelet denoising processing is 1.2 mm, and the theoretical stress data is 25.2 Mpa, and when the measured stress data is between 22.6 Mpa and 26.4 Mpa, the measured stress data is brought into the risk characteristic value for calculation; when the measured stress data is less than 22.6 Mpa or greater than 26.4 Mpa, the theoretical stress data is brought into the risk characteristic value for calculation.
[0054] Specifically, in step 104, the safety threshold generated by training the LSTM prediction model based on the historical monitoring data can be more suitable for such engineering, thereby helping to effectively warn and prevent safety accidents. When different types of buildings are constructed, the historical monitoring data of different buildings needs to be collected to generate different safety thresholds.
[0055] Specifically, please refer to Figure 2 As shown in the figure, the hierarchical risk warning information includes:
[0056] When the risk characteristic value is less than or equal to 0.8 times the safety threshold, the construction is in a safe state at this time, and the construction is normal;
[0057] When the risk characteristic value is greater than 0.8 times the safety threshold and less than or equal to the safety threshold, start a first-level warning and issue a warning alarm;
[0058] When the risk characteristic value is greater than the safety threshold and less than or equal to 1.5 times the safety threshold, notify the person in charge and stop work in time for maintenance;
[0059] When the risk characteristic value is greater than or equal to 1.5 times the safety threshold, cut off the power and evacuate the workers in an emergency;
[0060] By setting the hierarchical risk warning information, accurate judgments can be made for different dangerous situations, and effective and rapid protection of the building can be achieved.
[0061] The embodiment of the application also provides a computer readable storage medium, which can be any available medium that can be stored by a computing device or a data storage device such as a data center containing one or more available media, which can be a magnetic medium (for example, a floppy disk, a hard disk, a magnetic tape), an optical medium (for example, a DVD), or a semiconductor medium (for example, a solid state disk), etc. The computer readable storage medium contains instructions for instructing the computing device to execute the aforementioned time synchronization method.
[0062] The embodiment of the present application further provides a computer program product containing instructions, which can be software or program product containing instructions, capable of running on a computing device or being stored in any available medium, and when the computer program product runs on the computing device, the computing device is caused to execute the foregoing time synchronization method
[0063] The preferred embodiments of the application disclosed above are only used to help explain the application. The preferred embodiments do not describe all the details and limit the application to the specific embodiments described. Obviously, many modifications and variations can be made in light of the contents of the specification. The specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the application, so that those skilled in the art can well understand and utilize the application.
Claims
1. A method for monitoring engineering construction safety, characterized in that: The engineering construction safety monitoring method comprises: Collect historical monitoring data of similar projects, including structural displacement values, stress values, environmental parameters and accident records; Multiple detection modules are deployed at key nodes on the construction site to collect displacement deformation data, stress data, and environmental data of the building; The risk characteristic analysis module obtains the displacement deformation data, stress data and environmental data measured by the detection module to generate a risk characteristic value; An LSTM prediction model is trained based on the historical monitoring data, and displacement deformation data, stress data, and environmental data are input to generate a safety threshold. When the risk characteristic value exceeds the safety threshold, a graded danger warning message is triggered; A three-dimensional model of the construction site is constructed through the BIM engine, and the danger warning information is displayed on the AR terminal.
2. A construction safety monitoring method according to claim 1, characterized in that: Generating the risk characteristic value includes: Perform wavelet denoising on displacement deformation data based on wavelet denoising model; Based on the edge computing gateway, stress data and environmental data are analyzed, combined with wavelet noise reduction, and risk characteristic values are output.
3. A construction safety monitoring method according to claim 2, characterized in that: The risk characteristic value calculation formula is: Where G represents the risk characteristic value, ΔF represents the change in stress per unit time, Δt represents the detection time interval, σ represents the standard dry strength of building materials, β represents the humidity sensitivity coefficient of building materials, RH represents relative humidity, and RH0 represents the optimal humidity of the material; When the difference between the theoretical stress data and the measured actual stress data is less than 10%, ΔF adopts the stress change of the actual stress data; When the difference between the theoretical stress data and the measured actual stress data is greater than 10%, ΔF adopts the stress change of the theoretical stress data.
4. A construction safety monitoring method according to claim 3, characterized in that: The calculation formula of the theoretical stress data is: Where E represents the elastic modulus of the building material, ΔL represents the change in displacement deformation data after wavelet denoising, and L0 represents the initial length of the building.
5. The method for monitoring engineering construction safety according to claim 3, wherein: The graded danger warning information includes: When the risk characteristic value is less than or equal to 0.8 times the safety threshold, the construction is in a safe state and can proceed normally. When the risk characteristic value is greater than 0.8 times the safety threshold and less than or equal to the safety threshold, the first-level warning is activated and a warning alarm is issued; When the risk characteristic value is greater than the safety threshold and less than or equal to 1.5 times the safety threshold, notify the person in charge and stop work for maintenance in a timely manner; When the risk characteristic value is greater than or equal to 1.5 times the safety threshold, cut off the power supply and evacuate the workers urgently.
6. A construction safety monitoring method according to claim 1, characterized in that: The detection module includes a displacement sensor, a stress sensor and an ambient temperature and humidity sensor.
7. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a computer or a processor, the method according to any one of claims 1 to 6 is implemented.
8. A computer program product, characterized in that: The computer program product comprises a computer program, and when the computer program is executed by a computer or a processor, the computer or the processor is caused to perform the method according to any one of claims 1 to 6.