Building construction safety monitoring system with automatic detection and danger early warning functions

Abstract

The invention relates to the technical field of building safety monitoring, in particular to a building construction safety monitoring system with automatic detection and danger early warning functions. Comprising a multi-dimensional data automatic acquisition unit; a dynamic risk analysis unit; a grading danger early warning unit; and an emergency linkage control unit. According to the invention, through the construction stage identification module, the weight dynamic adjustment module and the risk quantification and feature matching module, the current construction stage is determined based on the excavation depth of the foundation pit, the installation elevation of the steel structural member and the total height of the building; the weight ratio of each physical parameter is dynamically adjusted by combining the basic weight matrix and the mechanical influence factor matrix of the corresponding construction stage, then the comprehensive risk value is calculated through the single-parameter deviation degree and the deviation acceleration, the hidden danger type is matched, and the risk analysis key point can be adjusted according to the structural mechanical characteristics of different construction stages. The problems of lack of dynamic adaptation capability and limited risk assessment precision in the prior art are solved.

Description

Technical Field

[0001] This invention relates to the field of building safety monitoring technology, and more specifically, to a building construction safety monitoring system with automatic detection and hazard warning functions. Background Technology

[0002] Construction encompasses multiple stages, including foundation excavation, main structure pouring, and decoration. The safety risks at each stage differ significantly, with structural instability, environmental disturbances, and personnel violations all potentially leading to accidents. Traditional construction safety monitoring often relies on single-parameter data collection or manual inspections, resulting in incomplete data coverage, delayed risk assessment, and passive emergency response. This makes it difficult to meet the real-time, precise safety management needs of complex construction scenarios, necessitating the development of a more comprehensive monitoring and early warning system.

[0003] In the existing technology, relevant patents have been used to conduct research in the field of construction safety monitoring. For example, Chinese patent CN202210048015.X discloses a construction safety monitoring system and its monitoring method, including a control system and detectors installed vertically on the building walls. The detectors transmit data to the control system, which includes a computer and a communication module. This can solve the problem of loopholes in the perimeter management of construction sites and monitor the tilt of walls. Another Chinese patent CN202311294446.5 discloses a construction safety monitoring system and method based on the Internet of Things. For steel frame connectors and rods, different analysis methods are used to determine safety hazards according to straight and bent standard parts, thereby improving the accuracy of hazard determination.

[0004] Despite the design advantages of the aforementioned technical solutions, they also suffer from the following technical shortcomings: First, the data collection coverage is limited, failing to consider multiple key safety influencing factors: Chinese patent CN202210048015.X focuses only on site perimeter management and wall tilt monitoring, without addressing the collection of interference parameters such as harmful gases and wind speed in the construction environment; Chinese patent CN202311294446.5 only analyzes steel frame connectors and members, without incorporating personnel operation-related data. Neither can comprehensively capture the safety risks arising from structural, environmental, and personnel interactions. Second, the risk analysis does not incorporate the characteristics of the construction stage and lacks dynamic adaptability: Chinese patents CN202210048015.X and CN202311294... 446.5 failed to consider the risk differences at different construction stages (such as foundation pit excavation and main structure construction). Chinese patent CN202210048015.X consistently focused on wall monitoring, while Chinese patent CN202311294446.5 used a fixed single mode to analyze steel frame hazards, failing to adjust the analysis focus according to the structural mechanical characteristics of each construction stage, thus limiting the accuracy of risk assessment. Thirdly, they only achieved basic monitoring and hazard identification, lacking coordinated early warning and emergency response: Chinese patent CN202210048015.X did not mention adjusting early warning strategies based on risk levels, and Chinese patent CN202311294446.5 only completed hazard identification but lacked subsequent emergency control measures, neither of which could quickly intervene to prevent the escalation of danger in high-risk scenarios. Therefore, we propose a building construction safety monitoring system with automatic detection and hazard early warning functions. Summary of the Invention

[0005] The purpose of this invention is to provide a construction safety monitoring system with automatic detection and hazard warning functions, in order to solve the problems mentioned in the background art, such as limited data collection coverage, failure to take into account multiple key safety influencing factors, lack of risk analysis combined with the characteristics of the construction stage, lack of dynamic adaptation capability, and only basic monitoring and hazard identification without linkage response for early warning and emergency response.

[0006] To address the aforementioned technical problems, the present invention aims to provide a construction safety monitoring system with automatic detection and hazard warning functions, comprising: The multi-dimensional data automatic acquisition unit collects building structure stability parameters, construction environment interference parameters, and personnel work position parameters in real time through physical parameter sensors deployed in key construction areas, and transmits the collected physical parameter data to the dynamic risk analysis unit through an industrial-grade wireless encrypted transmission link. The dynamic risk analysis unit, based on the real-time physical parameters output by the multi-dimensional data automatic acquisition unit, uses an improved weighted fuzzy comprehensive evaluation model to quantitatively analyze construction safety risks. The improved weighted fuzzy comprehensive evaluation model dynamically adjusts the parameter weights according to the structural mechanical characteristics and environmental interference characteristics of the construction stage to determine whether there are any safety hazards and the risk level of the safety hazards. The graded hazard early warning unit receives the risk level results from the dynamic risk analysis unit, and constructs a multi-dimensional early warning decision model by combining the spatial distribution parameters of personnel in the construction area, the physical state parameters of equipment operation, and the physical diffusion trend parameters of hidden dangers. It dynamically adjusts the early warning reach range based on the physical impact range of the risk, adjusts the early warning intensity based on the physical hazard degree of the risk, and pushes early warning information containing the physical location of hidden dangers, physical type of risk, and priority based on physical impact to the construction management terminal. The emergency linkage control unit is linked with the graded hazard warning unit based on physical signals. When the risk level reaches the medium or high level, it sends a shutdown physical command to the control system of special equipment in the construction area through the industrial bus and simultaneously triggers the output of audio signals for on-site emergency broadcasting.

[0007] As a further improvement to this technical solution, the multi-dimensional data automatic acquisition unit includes a structure sensing module, an environment sensing module, and a personnel sensing module, wherein: The structural sensing module is deployed in key construction areas and uses slope displacement sensors, scaffolding upright tilt sensors, and steel structure stress sensors to collect building structure stability parameters. The environmental sensing module is distributed within the construction area and uses harmful gas concentration sensors, environmental wind speed sensors, and precipitation sensors to collect construction environmental interference parameters. The personnel sensing module relies on the smart safety helmet and positioning base station, and uses UWB positioning sensors and contact status sensors to collect parameters of personnel's working position and safety helmet wearing status.

[0008] As a further improvement to this technical solution, the multi-dimensional data automatic acquisition unit also includes a data validity verification module. The data validity verification module is based on the physical measurement principle of sensors to judge the validity of the raw data collected by the structure sensing module, environment sensing module, and personnel sensing module. When the data is determined to be abnormal or the transmission is interrupted, a sensor fault prompt is triggered, and the abnormal data is temporarily supplemented using the physical parameters of adjacent valid periods to ensure the continuity of data transmission to the dynamic risk analysis unit.

[0009] As a further improvement to this technical solution, the dynamic risk analysis unit includes a construction stage identification module. This module identifies the current construction stage based on the physical condition parameters output by the multi-dimensional data automatic acquisition unit. The construction stage identification process includes the following steps: S210.1. Real-time acquisition of physical condition parameters of the construction area, including the excavation depth of the foundation pit, through sensor data interface. Elevation of steel structure components and total building height ; S210.2, Preset construction stage judgment threshold: when and At that time, it was determined to be the foundation pit excavation stage; when and When, it is determined to be the main structure pouring stage; when At that time, it was determined to be in the decoration and renovation stage; S210.3, Real-time data collection , Compare with the preset threshold and output the current construction stage identifier (represented by “1”, “2”, and “3” respectively, corresponding to the foundation pit excavation, main structure pouring, and decoration and finishing stages).

[0010] As a further improvement to this technical solution, the dynamic risk analysis unit also includes a weight dynamic adjustment module. This module dynamically adjusts the weight percentage of each physical parameter based on the construction stage identifier output by the construction stage identification module. The weight dynamic adjustment module performs the following steps: S220.1. Based on the construction stage identifier output by the construction stage identification module, retrieve the basic weight matrix corresponding to the construction stage. (Including basic weights for physical parameters such as slope displacement, steel structure stress, precipitation parameters, wind speed parameters, and personnel location, with the sum of the basic weights of each parameter being 1). S220.2 Retrieve the mechanical influence factor matrix corresponding to the construction stage identifier. ( (Match the construction stage identifiers "1", "2", and "3" respectively; the matrix elements are determined based on the structural mechanics test data for that stage.) S220.3, Combining the basic weight matrix With the mechanical influence factor matrix The actual weight matrix of each physical parameter is calculated. S220.4 Normalize the actual weight matrix to ensure that the sum of the actual weights of each physical parameter is 1, and output the dynamic weights for subsequent risk quantification.

[0011] As a further improvement to this technical solution, the dynamic risk analysis unit also includes a risk quantification module and a risk feature matching module, wherein: The risk quantification module calculates a comprehensive risk value based on the dynamic weights output by the dynamic weight adjustment module and the measured values ​​of physical parameters output by the multi-dimensional data automatic acquisition unit. Specifically, this includes obtaining the single-parameter deviation by comparing the measured values ​​of physical parameters with the corresponding safety thresholds. The deviation acceleration, which characterizes the rate of deterioration of parameters, is calculated based on the continuous periodic deviation variation. The single-parameter risk membership degree is obtained by combining the deviation degree and the deviation acceleration. The comprehensive risk value is then obtained by weighting the dynamic weights and the single-parameter risk membership. ; The risk feature matching module has a built-in risk physical feature library, which is used to classify risk levels and match specific safety hazard types, including: based on comprehensive risk value. The risk level is determined by comparing the risk with a preset threshold; the risk membership of each single parameter is then assigned. The combined features are compared and matched with the various types of hazard feature patterns stored in the risk physical feature library, and the specific safety hazard type that best matches the real-time parameter features is output.

[0012] As a further improvement to this technical solution, the graded hazard early warning unit includes a parameter fusion processing module. The parameter fusion processing module receives the risk level results from the dynamic risk analysis unit, combines the spatial distribution parameters of personnel in the construction area and the physical state parameters of equipment operation output by the multi-dimensional data automatic acquisition unit, and the physical diffusion trend parameters of hidden dangers output by the dynamic risk analysis unit, and performs collaborative fusion processing on the multi-source physical parameters to provide standardized parameter input for the multi-dimensional early warning decision model.

[0013] As a further improvement to this technical solution, the graded hazard warning unit also includes a warning range and intensity adjustment module. The warning range and intensity adjustment module dynamically adjusts the warning reach range and warning intensity based on the fusion parameters output by the parameter fusion processing module, combined with the physical impact range characteristics and physical hazard degree characteristics of the risk. Based on the mechanical parameters of structural hazards or the diffusion parameters of environmental hazards, the spatial range that the hazard may affect is determined as the warning reach range, and then the corresponding sound and light output and terminal prompt warning intensity form are matched according to the risk level.

[0014] As a further improvement to this technical solution, the graded hazard warning unit also includes a warning information generation and push module. The warning information generation and push module generates warning information containing the physical location of the hazard, the physical type of the risk, and the priority based on the physical impact, based on the warning reach range and warning intensity output by the warning range and intensity adjustment module, combined with the results of the physical location of the hazard and the physical type of the risk from the dynamic risk analysis unit, and pushes it accurately to the construction management terminal.

[0015] As a further improvement to this technical solution, the emergency response control unit includes a signal receiving and judging module and a device control execution module, wherein: The signal receiving and judgment module is connected to the graded hazard warning unit through a physical signal link. It receives the risk level result output by the graded hazard warning unit in real time. When the risk level is identified as medium or high, it sends a linkage trigger signal to the equipment control execution module. After receiving the linkage trigger signal, the equipment control execution module sends a standardized shutdown physical command to the control system of special equipment in the construction area through the industrial bus. At the same time, it triggers the audio signal output of the on-site emergency broadcast system through the audio control interface. The audio signal content is pre-stored in the built-in storage component of the emergency broadcast system based on the risk physical type.

[0016] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. This invention, through a multi-dimensional automatic data acquisition unit and a data validity verification module, can collect in real time building structure stability parameters, construction environment interference parameters, and personnel work position and safety helmet wearing status parameters during the foundation pit excavation, main structure pouring, and decoration stages of building construction. It can also temporarily supplement abnormal data, which can comprehensively cover the data acquisition needs of the three key safety influencing factors of structure, environment and personnel in the construction scenario. It solves the problems of limited data acquisition coverage and inability to capture the interaction risks of the three factors in the existing technology, and can also ensure the continuity of data transmission to the dynamic risk analysis unit, providing a complete and reliable data foundation for subsequent risk assessment. 2. This invention uses a construction stage identification module, a weight dynamic adjustment module, and a risk quantification and feature matching module in a dynamic risk analysis unit to determine the current construction stage based on the excavation depth of the foundation pit, the installation elevation of the steel structure components, and the total building height. It dynamically adjusts the weight ratio of each physical parameter by combining the basic weight matrix and mechanical influence factor matrix of the corresponding construction stage. Then, it calculates the comprehensive risk value and matches the type of hidden danger by using single parameter deviation degree and deviation acceleration. It can adjust the focus of risk analysis according to the structural mechanical characteristics of different construction stages, which solves the problems of lack of dynamic adaptability and limited accuracy of risk assessment in existing technologies, and improves the accuracy of safety risk assessment at each construction stage. 3. This invention utilizes the synergistic effect of a graded hazard early warning unit and an emergency response control unit. The graded hazard early warning unit integrates parameters such as risk level, personnel spatial distribution, equipment operating status, and hazard spread trend to dynamically adjust the early warning reach and intensity, and pushes early warning information containing the physical location and risk type of the hazard to the construction management terminal. When the risk level reaches medium to high, the emergency response control unit can send a shutdown command to special equipment via industrial bus and trigger on-site emergency broadcasting. This solves the problems of existing technologies lacking early warning and emergency response, and being unable to quickly intervene in high-risk scenarios. It realizes closed-loop management from hazard identification to early warning and emergency control, effectively preventing the escalation of danger. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of the system framework of the present invention; The meanings of the labels in the diagram are as follows: 100. Multi-dimensional automatic data acquisition unit; 110. Structure sensing module; 120. Environmental sensing module; 130. Personnel sensing module; 140. Data validity verification module; 200. Dynamic Risk Analysis Unit; 210. Construction Stage Identification Module; 220. Dynamic Weight Adjustment Module; 230. Risk Quantification Module; 240. Risk Feature Matching Module; 300. Graded hazard early warning unit; 310. Parameter fusion processing module; 320. Early warning range and intensity adjustment module; 330. Early warning information generation and push module; 400. Emergency linkage control unit; 410. Signal receiving and judgment module; 420. Equipment control and execution module. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0019] like Figure 1 As shown, this embodiment provides a construction safety monitoring system with automatic detection and hazard warning functions, including: The multi-dimensional data automatic acquisition unit 100 collects building structure stability parameters, construction environment interference parameters, and personnel operation position parameters in real time through physical parameter sensors deployed in key construction areas, and transmits the collected physical parameter data to the dynamic risk analysis unit 200 through an industrial-grade wireless encrypted transmission link. In this embodiment, the multi-dimensional data automatic acquisition unit 100 includes a structure sensing module 110, an environment sensing module 120, and a personnel sensing module 130, wherein: The structural sensing module 110 is deployed in key construction areas and uses slope displacement sensors, scaffolding upright tilt sensors, and steel structure stress sensors to collect building structure stability parameters. Specifically, the deployment of the structural sensing module 110 needs to accurately match the structural stress characteristics of key construction areas to ensure that the collected building structural stability parameters can truly reflect the structural safety status. This includes: Deployment Location Determination: Key monitoring areas are divided according to construction stage and structural type. Slope displacement sensors are primarily deployed at the top of the foundation pit slope (e.g., one monitoring point every 20 meters along the slope direction), at the external corners of the foundation pit, and at points where slope height changes. Scaffolding upright tilt sensors are deployed at the top, middle, and bottom key nodes of the scaffolding uprights (e.g., only one sensor needs to be installed at the middle node of each upright, with priority given to uprights at scaffolding corners and areas of concentrated load). Steel structure stress sensors are fixed at the connection points of main and secondary beams of the steel structure, column bases, and at the mid-span of steel beams exceeding 15 meters (e.g., one sensor is installed at each key connection point or location, and this installation position avoids damage to the sensor due to maximum stress at mid-span). Sensor installation methods: Slope displacement sensors are fixed with anchor bolts. The sensor base is anchored to the concrete cushion or solid soil layer at the top of the slope using expansion anchors with a diameter of 20mm, ensuring that the sensor moves synchronously with the slope structure. Scaffolding upright tilt sensors are installed with clamps. Metal clamps matching the diameter of the upright are used to tightly fix the sensor to the surface of the upright. Rubber gaskets are added between the clamp and the upright to prevent loosening from affecting the measurement. Steel structure stress sensors are fixed by welding or high-strength bolts. When welding, rust and oil stains on the surface of the steel structure components must be cleaned first, and the welding surface of the sensor must be completely in contact with the surface of the component. When fixing bolts, a torque wrench should be used to tighten them to a torque of 30N•m to avoid relative displacement between the sensor and the component. Data Acquisition and Transmission Connection: All sensors in the structural sensing module 110 are set with a unified basic acquisition cycle (1 minute / time). The acquired slope displacement value, pole inclination value, and steel structure stress value are connected to the industrial-grade wireless transmission terminal through the RS485 interface of the sensor (each wireless transmission terminal can connect 8-12 sensors). The wireless transmission terminal encrypts the data through the LoRa wireless communication protocol (a specific implementation of the industrial-grade wireless encrypted transmission link) and transmits it to the receiving gateway of the dynamic risk analysis unit 200. During the transmission process, the AES-128 encryption algorithm is used to ensure data security and prevent data from being tampered with or leaked.

[0020] The environmental sensing module 120 is distributed in the construction area and uses harmful gas concentration sensors, environmental wind speed sensors, and precipitation sensors to collect construction environmental interference parameters. Specifically, the environmental sensing module 120 needs to cover the entire construction area, from underground to ground level to high altitude, to ensure that there are no blind spots in the collection of environmental interference parameters. This includes: Sensor distribution rules: The distribution of hazardous gas concentration sensors (monitoring common construction hazardous gases such as carbon monoxide, hydrogen sulfide, and methane) needs to be combined with the type of work space. In underground construction areas (such as inside foundation pits or underground garage work surfaces), one sensor should be placed every 50 square meters, and it should be close to locations where workers frequently stay (such as operating platforms inside foundation pits or entrances and exits of underground work surfaces). In ground construction areas (such as around the main structure or material storage areas), one sensor should be placed every 100 square meters, with a focus on covering areas prone to generating hazardous gases, such as painting and welding work areas. The environmental wind speed sensor should be installed at an unobstructed high location within the construction area, preferably at the top of the construction tower crane (1.5 meters from the top of the tower crane), the top of the scaffolding (1 meter from the top of the scaffolding), or the top of a temporary monitoring pole (the pole height should not be less than 8 meters) to ensure that the collected wind speed data reflects the true wind environment of the construction area. The precipitation sensor should be placed in low-lying areas of the construction site (such as the lowest point of the site's drainage slope) and near the drainage outlets around the foundation pit (within 2 meters of the drainage outlet), while avoiding areas crushed by construction machinery to prevent damage to the sensor. Data acquisition triggering mechanism: The environmental sensor module 120 collects data at a default cycle of 2 minutes / time. Dynamic acquisition triggering conditions are also set. For example, when the hazardous gas concentration sensor detects that the concentration of a certain gas reaches 80% of its safety threshold, the sensor automatically shortens the acquisition cycle to 30 seconds / time and marks "early warning ready state" in the data. When the environmental wind speed sensor detects a wind speed exceeding 5 m / s, it simultaneously records wind direction data (divided into 8 directions, such as east, south, west, north, northeast, southeast, northwest, and southwest), and updates the wind speed and direction data every 30 seconds. When the precipitation sensor detects precipitation exceeding 0.5 mm / 10 minutes, the acquisition cycle automatically adjusts to 10 seconds / time, and the precipitation intensity is recorded (divided into light rain, moderate rain, and heavy rain, based on 10-minute precipitation). Data transmission connectivity: All sensors in the environmental sensing module 120 are connected to an industrial-grade wireless encrypted transmission link wirelessly. The hazardous gas concentration sensor and precipitation sensor are connected to the nearest wireless transmission terminal via Bluetooth Low Energy (BLE) protocol. The environmental wind speed sensor, due to its high installation position, communicates directly with the receiving gateway of the dynamic risk analysis unit 200 via LoRa protocol. All environmental parameter data are transmitted after being bound with the collection timestamp, sensor number, and installation location information, which facilitates subsequent data validity verification and risk analysis.

[0021] The personnel sensing module 130 relies on the smart safety helmet and positioning base station, and uses UWB positioning sensors and contact status sensors to collect parameters of personnel's working position and safety helmet wearing status.

[0022] Specifically, the personnel sensing module 130 needs to achieve the dual functions of "personnel positioning and status monitoring," forming a complete personnel monitoring network based on the smart safety helmet and positioning base station, specifically including: Configuration of UWB positioning base stations and smart safety helmets: The deployment of UWB positioning base stations must meet the positioning coverage requirements of the construction area. They should be evenly distributed within the construction area at 20m x 20m intervals. Base stations must be installed on fixed structures at a height of no less than 2.5 meters (such as the top of construction fences, the exterior walls of temporary office buildings, or scaffold crossbars), and obstructions must be avoided (the direction of base station signal propagation should not be obstructed by reinforced concrete walls, large machinery, or other objects). Each positioning base station is connected to a network switch within the construction area via a network cable. The switch then connects to an industrial-grade wireless encrypted transmission link via fiber optic cable to ensure the connection between the base station and the dynamic risk analysis unit. 200 real-time data interaction; each worker needs to be equipped with one smart safety helmet. The safety helmet has a built-in UWB positioning sensor (integrated at the back of the helmet) and a contact status sensor (installed at the helmet buckle). The positioning accuracy of the UWB positioning sensor needs to match the personnel management requirements of the construction area (no specific accuracy value needs to be marked, only area positioning and trajectory tracking are required). The contact status sensor is linked to the closed state of the buckle through metal contacts - when the buckle is fully closed, the sensor outputs a "wearing" signal; when the buckle is not closed or half closed, the sensor outputs a "not wearing" signal.

[0023] Personnel Data Collection and Association: UWB positioning sensors collect personnel location data (data format is X, Y, Z three-dimensional coordinates in the construction area coordinate system) at a frequency of 1 second / time, and contact-type status sensors collect safety helmet wearing status data at a frequency of 5 seconds / time. The built-in wireless module of the smart safety helmet (supporting LoRa protocol) packages the location data, wearing status data, unique safety helmet ID (each safety helmet is assigned a unique ID, which is pre-bound in the system with the worker's name and job information), and collection timestamp, and transmits them to the dynamic risk analysis unit 200 through an industrial-grade wireless encrypted transmission link. After receiving the data, the dynamic risk analysis unit 200 automatically maps the location data to the electronic map of the construction area to realize the real-time visualization of personnel location, and at the same time records the trajectory of each person's safety helmet wearing status change.

[0024] In this embodiment, the multi-dimensional data automatic acquisition unit 100 also includes a data validity verification module 140. The data validity verification module 140 judges the validity of the raw data collected by the structure sensing module 110, the environment sensing module 120, and the personnel sensing module 130 based on the physical measurement principle of the sensors. When the data is determined to be abnormal or the transmission is interrupted, a sensor fault prompt is triggered, and the abnormal data is temporarily supplemented using the physical parameters of the adjacent valid period to ensure the continuity of data transmission to the dynamic risk analysis unit 200.

[0025] Specifically, the criteria for judging data validity include: For the raw data from the structural sensing module 110: the judgment is based on the measurement range of the sensor and the normal working range of the structure. For example, the range of the slope displacement sensor is 0-500mm. If the collected value is less than 0 or greater than 500mm, it is directly judged as data abnormality. The normal measurement range of the scaffolding upright tilt angle sensor is 0-15° (within the allowable deviation range of the upright verticality). If the collected value exceeds 15° or the collected values ​​are completely consistent for 3 consecutive times (excluding the static situation of the structure after the concrete has solidified during the main structure pouring stage, it is necessary to combine the stage identification module 210 to assist in the judgment. For example, the upright tilt angle should remain stable during the decoration and finishing stage, and 3 consecutive consistent values ​​are normal. If the upright tilt angle is consistent for 3 consecutive times during the foundation pit excavation stage, it may be data abnormality), it is judged as data abnormality. For the raw data from the environmental sensing module 120: based on the physical characteristics of the sensors and the logical judgment of environmental parameters, if the value collected by the harmful gas concentration sensor is negative or exceeds the maximum measurement value of the sensor (e.g., the maximum measurement value of the carbon monoxide sensor is 1000ppm), it is judged as data abnormality; if the wind speed value collected by the wind speed sensor is negative, or the precipitation value collected by the precipitation sensor is negative, it is judged as data abnormality; if a sensor of the environmental sensing module 120 has no data transmitted to the wireless transmission terminal for 10 consecutive seconds, it is judged as transmission interruption. For the raw data from the personnel sensing module 130: based on positioning logic and status logic, if the location data collected by the UWB positioning sensor exceeds the boundary of the electronic map of the construction area, or if there is no location data feedback for 5 consecutive seconds, it is determined to be data abnormal or transmission interrupted; if the contact status sensor outputs two contradictory signals, "worn" and "not worn", it is determined to be data abnormal.

[0026] Furthermore, the fault indication and data completion mechanism specifically includes: Fault indication: When data abnormality or transmission interruption is detected, the data validity verification module 140 immediately generates fault information. The fault information includes the abnormal sensor number, installation location, abnormality type and abnormality occurrence timestamp. It is pushed to the construction management terminal through the industrial-grade wireless encrypted transmission link. After receiving the information, the construction management terminal will pop up a pop-up window and mark the location of the abnormal sensor with a red icon on the display screen to facilitate timely troubleshooting by management personnel. Data Completion: For data determined to be abnormal or interrupted, the data validity verification module 140 uses the "average of adjacent valid period data completion" method. If the abnormal data lasts for only one acquisition period (e.g., the acquisition period of the structural sensing module is 1 minute / time, and the abnormality lasts for 1 minute), the average of the acquisition data of the previous valid period and the next valid period is calculated as the temporary completion data for the abnormal period. If the abnormal data lasts for multiple acquisition periods (e.g., 5 minutes), during the period of abnormality, each acquisition period uses the "data of the previous valid period" for temporary completion (e.g., if the data is abnormal in the 2nd minute, the valid data of the 1st minute is used to complete the 2nd minute; if it is still abnormal in the 3rd minute, the data of the 1st minute is used to complete the 2nd minute until the sensor returns to normal). When the abnormal sensor resumes normal acquisition, the data validity verification module 140 automatically replaces the previous completion data with the recovered real acquisition data and marks the completion time period and replacement status in the data record to ensure that the data transmitted to the dynamic risk analysis unit 200 is continuous and as close to the real value as possible, avoiding the impact of data loss on the risk analysis results.

[0027] The dynamic risk analysis unit 200 uses the real-time physical parameters output by the multi-dimensional data automatic acquisition unit 100 to perform quantitative analysis of construction safety risks using an improved weighted fuzzy comprehensive evaluation model. The improved weighted fuzzy comprehensive evaluation model dynamically adjusts the parameter weights according to the structural mechanical characteristics and environmental interference characteristics of the construction stage to determine whether there are any safety hazards and the risk level of the safety hazards. In this embodiment, the dynamic risk analysis unit 200 includes a construction stage identification module 210. The construction stage identification module 210 identifies the current construction stage based on the physical condition parameters output by the multi-dimensional data automatic acquisition unit 100. The construction stage identification module 210 performs the following steps: S210.1. Real-time acquisition of physical condition parameters of the construction area, including the excavation depth of the foundation pit, through sensor data interface. Elevation of steel structure components and total building height ; S210.2, Preset construction stage judgment threshold: when and At that time, it was determined to be the foundation pit excavation stage; when and When, it is determined to be the main structure pouring stage; when At that time, it was determined to be in the decoration and renovation stage; S210.3, Real-time data collection , Compare with the preset threshold and output the current construction stage identifier (represented by “1”, “2”, and “3” respectively, corresponding to the foundation pit excavation, main structure pouring, and decoration and finishing stages).

[0028] Specifically, the methods for obtaining the parameters of each physical condition are as follows: Excavation depth of foundation pit The depth sensor deployed at the bottom of the foundation pit in the structural sensing module 110 (working in conjunction with the slope displacement sensor, with depth benchmarks set in advance around the foundation pit) collects data in real time. The sensor converts the relative height difference between the bottom of the foundation pit and the benchmark into excavation depth data, which is then synchronized to the parameter receiving interface of the construction phase identification module 210 via an industrial-grade wireless encrypted transmission link. Steel structure component installation elevation Based on the installation location information of the steel structure stress sensor in the structural sensing module 110 (the design elevation of the corresponding component was recorded in the system during sensor installation), and combined with the "Steel Structure Component Installation Completion Record" in the construction log (the system is linked with the construction management platform; after installation, on-site personnel mark it on the platform, and the construction phase identification module 210 automatically reads the marking information), the highest elevation of the currently installed steel structure component is obtained as... The value of ; Total building height During the system initialization phase, administrators input the total building height data from the project design drawings (e.g., if the total design height is 100 meters, then...). (The preset value is 100 meters). This data is stored in the local database of the construction stage identification module 210 and does not need to be collected in real time. It is only used as a fixed threshold for stage determination.

[0029] Specifically, the execution logic and cycle of the phase determination include: The construction phase identification module 210 is set to a phase judgment cycle of "5 minutes / time" (matching the longest data acquisition cycle of the multi-dimensional data automatic acquisition unit 100 to avoid misjudgment due to outdated parameters), and is verified before each judgment. , Data validity (if the data timestamp is more than 10 minutes from the current time, it is considered invalid, and the latest data is retrieved again).

[0030] Perform the judgment based on the preset threshold: like (The foundation pit has not been excavated or excavation work has been suspended and the depth has been reduced to zero) and If there are no records of steel structure component installation, then output the construction stage identifier "1" (corresponding to the foundation pit excavation stage). like (The foundation pit has undergone at least one excavation operation) and (If the highest elevation of the installed steel structure does not exceed 50% of the total building height), then output the construction stage identifier "2" (corresponding to the main structure pouring stage); like (The highest elevation of the installed steel structure exceeds 50% of the total building height), regardless of Regardless of the value, the output will always be the construction stage identifier "3" (corresponding to the decoration and renovation stage).

[0031] At the same time, if the judgment process occurs or If data is missing (e.g., sensor malfunction and lack of supplementary data), the construction phase identification module 210 temporarily stores the phase identifier of the previous cycle and pushes a "parameter missing, phase determination delayed" prompt to the construction management terminal until the data is recovered and re-determined.

[0032] Specifically, after the determination is completed, the construction stage identification module 210 synchronizes the stage identifiers ("1", "2", "3") to the weight dynamic adjustment module 220 through the internal data bus. At the same time, the current stage information is displayed on the "Risk Analysis Panel" of the construction management terminal (e.g., "Current construction stage: Main structure pouring stage (identifier 2), determination basis: foundation pit excavation depth"). Steel structure installation elevation Total building height This allows managers to easily trace the basis for their judgments.

[0033] In this embodiment, the dynamic risk analysis unit 200 further includes a weight dynamic adjustment module 220. The weight dynamic adjustment module 220 dynamically adjusts the weight percentage of each physical parameter based on the construction stage identifier output by the construction stage identification module 210. The weight dynamic adjustment module 220 performs the following steps: S220.1 Based on the construction stage identifier output by the construction stage identification module 210, retrieve the basic weight matrix corresponding to the construction stage. (Including basic weights for physical parameters such as slope displacement, steel structure stress, precipitation parameters, wind speed parameters, and personnel location, with the sum of the basic weights of each parameter being 1). S220.2 Retrieve the mechanical influence factor matrix corresponding to the construction stage identifier. ( (Match the construction stage identifiers "1", "2", and "3" respectively; the matrix elements are determined based on the structural mechanics test data for that stage.) S220.3, Combining the basic weight matrix With the mechanical influence factor matrix The actual weight matrix of each physical parameter is calculated. S220.4 Normalize the actual weight matrix to ensure that the sum of the actual weights of each physical parameter is 1, and output the dynamic weights for subsequent risk quantification.

[0034] Specifically, the basic weight matrix The specific preset and retrieval rules are as follows: During system initialization, the local database of the dynamic weight adjustment module 220 presets three sets of basic weight matrices (corresponding to stage identifiers "1", "2", and "3" respectively). The matrices contain the basic weights of five core physical parameters: slope displacement, steel structure stress, precipitation parameters, wind speed parameters, and personnel location. The sum of the weights of all parameters in each matrix is ​​1. The specific presets are determined based on industry risk statistics. Corresponding to the foundation pit excavation stage (marked "1"): (Slope displacement has the highest weight because the risk of slope instability is the greatest during the foundation pit stage; steel structure stress has the lowest weight because there is no steel structure installation work during this stage.) Corresponding to the main structure pouring stage (marked "2"): (The stress weight of the steel structure is the highest because the installation and stress deformation risks of the steel structure are prominent during the main construction stage; the displacement weight of the slope is reduced because the foundation pit has been excavated and the support is stable.) Corresponding to the decoration and renovation stage (marked "3"): (Personnel location has the highest weight because the work area is wide and the risk of unauthorized entry into dangerous areas is high during the decoration phase; structural parameters have a lower weight because the main structure has been accepted and approved.) Meanwhile, after receiving the construction stage identifier, the weight dynamic adjustment module 220 automatically retrieves the corresponding matrix. For example, if the received identifier is "3", it retrieves the matrix. If the identification transmission is interrupted, the basic weight matrix retrieved in the previous cycle will be used temporarily until the identification is restored.

[0035] Specifically, The matrix elements are determined based on the structural mechanics test reports of similar building construction projects (e.g., collecting mechanical test data from the foundation pit stage, main structure stage, and decoration stage of 10 residential projects of similar scale, and statistically analyzing the actual impact coefficients of each parameter on structural safety). After being preset, they are stored in the weight dynamic adjustment module 220, as follows: Corresponding to the foundation pit excavation stage : (The slope displacement factor is the highest because the slope displacement rate has the most significant impact on the collapse risk during the foundation pit stage; the steel structure stress factor is the lowest because there is no requirement for the steel structure to bear the load). Corresponding to the main structure pouring stage : (The stress influence factor is the highest for steel structures because the steel structure bears the construction load during the main construction stage, and the stress change has the most significant impact on the deformation risk.) Corresponding decoration and renovation stage : (Personnel location has the highest impact factor, as personnel activities during the decoration phase have the strongest direct correlation with safety accidents). Meanwhile, the weight dynamic adjustment module 220 matches the construction stage identifier. If the identifier is "2", then it matches This ensures that the influencing factors are compatible with the mechanical properties at the current stage.

[0036] Furthermore, this embodiment employs an element-wise multiplication method to multiply the basic weight matrix. With the mechanical influence factor matrix Multiplying the corresponding elements, we obtain the initial actual weights of each physical parameter, as shown in the formula: ( These correspond to slope displacement, steel structure stress, precipitation parameters, wind speed parameters, and personnel location, respectively.

[0037] This embodiment provides the following example (main structure pouring stage, marked "2"): Initial actual weights of slope displacement: ; Initial actual weights of stress in steel structures: ; Initial actual weights of precipitation parameters: ; Initial actual weights for wind speed parameters: ; Initial actual weight of personnel position: .

[0038] Normalization: First, calculate the sum of the initial actual weights (in the example, the sum is...). Then, divide each initial actual weight by the sum to obtain the final actual weight (in the example, the final weight of the steel structure stress is ). This ensures that the final sum of weights is 1.

[0039] Weight output: The weight dynamic adjustment module 220 pushes the final actual weight (dynamic weight) to the risk quantification module 230 through the internal interface, and generates a "weight adjustment log" (including adjustment time, stage identifier, initial weight and final weight of each parameter) in the system background to facilitate subsequent risk analysis and traceability.

[0040] Understandably, the core improvements of the improved weighted fuzzy comprehensive evaluation model are: first, dynamic weighting, which retrieves the corresponding foundation weight matrix and mechanical influence factor matrix based on the construction stage (excavation, main structure pouring, and decoration), and calculates and normalizes them to obtain the actual weights; second, upgraded quantification dimensions, adding a deviation acceleration to characterize the rate of deterioration on the basis of single-parameter static deviation, thus correcting the single-parameter risk membership degree. This solves the problems of traditional models where fixed weights are not suitable for risks at different stages (e.g., the excavation stage does not focus on slope displacement, and the main structure stage does not emphasize steel structure stress), where static deviation warnings are lagging, and where there is no specific matching of hazard types. For example, the excavation stage model assigns a slope displacement weight of 0.4, accurately capturing the risk of slope instability; when the slope displacement of a certain excavation exceeds the threshold and deteriorates rapidly, the model provides an early warning 5-10 minutes in advance, and can also match hazard types, improving the accuracy of assessment and the targeted nature of response.

[0041] In this embodiment, the dynamic risk analysis unit 200 further includes a risk quantification module 230 and a risk feature matching module 240, wherein: The risk quantification module 230 calculates a comprehensive risk value based on the dynamic weights output by the weight dynamic adjustment module 220 and the measured values ​​of physical parameters output by the multi-dimensional data automatic acquisition unit 100. Specifically, this includes obtaining the single-parameter deviation by comparing the measured values ​​of physical parameters with the corresponding safety thresholds. The deviation acceleration, which characterizes the rate of deterioration of parameters, is calculated based on the continuous periodic deviation variation. The single-parameter risk membership degree is obtained by combining the deviation degree and the deviation acceleration. The comprehensive risk value is then obtained by weighting the dynamic weights and the single-parameter risk membership. ; Specifically, in the threshold database of the risk quantification module 230, safety thresholds for five categories of physical parameters are preset according to the "Construction Safety Inspection Standard" (JGJ59) and the project's specific construction plan, including: The slope displacement safety threshold is 50mm (determined based on the allowable displacement of the foundation pit support design), the steel structure stress safety threshold is 235MPa (determined based on the yield strength of Q235 steel), the precipitation parameter safety threshold is 10mm / hour (determined based on the foundation pit drainage capacity), the wind speed parameter safety threshold is 10m / s (determined based on the safe wind speed limit for high-altitude operations), and the personnel location safety threshold is the "dangerous area boundary line" (preset through the electronic map of the construction area; if the personnel coordinates exceed the boundary, it is judged as exceeding the threshold).

[0042] Specifically, single-parameter deviation The calculation logic is as follows: retrieve the measured values ​​of physical parameters transmitted by the multi-dimensional data automatic acquisition unit 100. ,like ( (for the corresponding safety threshold), then (No parameter deviation, no risk contribution); if ,but (The deviation is positively correlated with the proportion of the measured value exceeding the threshold; the larger the proportion, the higher the deviation.) For example, when the measured slope displacement is 65mm, .

[0043] Specifically, deviation acceleration The calculation requires single-parameter deviation data from three consecutive acquisition cycles (the acquisition cycle is consistent with the multi-dimensional data automatic acquisition unit 100, such as structural parameters 1 minute / time, environmental parameters 2 minutes / time), denoted as... (Cycle 1) (Cycle 2) (3rd cycle), the cycle interval is First, calculate the rate of change of deviation between adjacent periods. , then calculate ,in This indicates that the rate of deterioration of parameter deviation is accelerating (risk is increasing). This indicates that the rate of deterioration has slowed or stabilized (risk is manageable), for example... hour, .

[0044] Specifically, single-parameter risk membership A composite algorithm of "deviation degree dominance + deviation acceleration correction" is adopted, and the formula is as follows: , A preset correction coefficient (valued at 1000, used to amplify the effect of acceleration and prevent the correction from being ineffective due to excessively small acceleration values) is set. The range of values ​​is When the calculation result is greater than 1, it is counted as 1 (representing that the risk of this parameter has reached the limit value), and when it is less than 0, it is counted as 0 (representing no risk). For example... hour, .

[0045] Specifically, comprehensive risk value The calculation is performed using a "dynamic weighted summation" method, and the formula is as follows: ( , (The dynamic weights output by the dynamic weight adjustment module 220). For example, during the main structure pouring stage, dynamic weights , Membership of each parameter hour, .

[0046] Furthermore, the risk quantification module 230 will integrate the risk value. and the risk membership degree of each individual parameter The data is synchronized to the risk feature matching module 240 via an internal data interface, and relevant data (such as "comprehensive risk value") is displayed in real time on the "risk quantification panel" of the construction management terminal. Slope displacement membership Stress membership degree of steel structures (”).

[0047] The risk feature matching module 240 has a built-in risk physical feature library, which is used to classify risk levels and match specific safety hazard types, including: based on comprehensive risk values. The risk level is determined by comparing the risk with a preset threshold; the risk membership of each single parameter is then assigned. The combined features are compared and matched with the various types of hazard feature patterns stored in the risk physical feature library, and the specific safety hazard type that best matches the real-time parameter features is output.

[0048] Specifically, the risk feature matching module 240 has a built-in risk physical feature library, which is divided into 5 categories according to the type of construction safety hazards. Each category of hazard corresponds to a set of "single-parameter membership degree combination feature patterns". The feature patterns are determined based on the "Safety Standard for Prevention and Control of Accidents in Construction" (JGJ / T429) and a summary of historical hazard cases, as follows: Slope collapse risk response and (Large slope displacement deviations accompanied by precipitation can easily lead to soil instability and collapse); Corresponding risks of steel structure deformation. and (Steel structure stress exceeding the threshold and accompanied by strong winds can easily cause component bending deformation), corresponding to the risk of flooding during rainfall. and (If rainfall exceeds drainage capacity and people are in low-lying areas, flooding accidents are likely to occur.) Strong wind impact risk response and (Wind speed exceeding safety limits and steel structure under stress can easily cause components to sway or fall.) Risks of personnel violating regulations And others (Personnel entering the pre-designated danger zone); This feature library allows managers to add new hazard types (such as adding "scaffold tilt risk" and the membership feature of the corresponding scaffold upright tilt angle parameter) through the "feature library maintenance interface". (This will be automatically included in the matching range after the update).

[0049] Specifically, the risk level is determined based on the "Standard for Safety Inspection of Building Construction" (JGJ59-2011) and preset thresholds in conjunction with engineering practice: Low risk (Corresponding to the "excellent" category of safety inspection, with controllable risks and no need to initiate early warning); The medium risk is (Corresponding to "qualified" in safety inspection but with local hidden dangers, and it is necessary to initiate a secondary early warning); The high risk is (Corresponding to "unqualified" in safety inspection, with urgent risks, and it is necessary to initiate a primary early warning); Subsequently, the risk feature matching module 240 retrieves the comprehensive risk value output by the risk quantification module 230 , and determines the risk level after comparing it with the preset threshold. For example it is determined as low risk when it is determined as medium risk when it is determined as high risk when

[0050] Specifically, the risk feature matching module 240 synchronizes the risk level and the matching hidden danger type to the hierarchical danger early warning unit 300 through the internal interface. At the same time, it marks the risk level with different colors at the construction management terminal (low risk in green, medium risk in yellow, high risk in red), and generates a "hidden danger matching log" in the system background, recording the matching time, risk level, membership data and hidden danger type, which is convenient for subsequent hidden danger rectification tracking and review analysis.

[0051] The hierarchical danger early warning unit 300 receives the risk level result of the dynamic risk analysis unit 200, constructs a multi-dimensional early warning decision model by combining the spatial distribution parameters of personnel in the construction area, the physical state parameters of equipment operation, and the physical diffusion trend parameters of hidden dangers, dynamically adjusts the early warning reach range based on the physical influence range of the risk, adjusts the early warning intensity based on the physical harm degree of the risk, and at the same time pushes early warning information including the physical location of the hidden danger, the physical type of the risk, and the physical influence priority to the construction management terminal; In this embodiment, the hierarchical danger early warning unit 300 includes a parameter fusion processing module 310. The parameter fusion processing module 310 receives the risk level result of the dynamic risk analysis unit 200, combines the spatial distribution parameters of personnel in the construction area, the physical state parameters of equipment operation output by the multi-dimensional data automatic acquisition unit 100, and the physical diffusion trend parameters of hidden dangers output by the dynamic risk analysis unit 200, and performs collaborative fusion processing on multi-source physical parameters to provide standardized parameter input for the multi-dimensional early warning decision model.

[0052] Specifically, the parameter fusion processing module 310 synchronously receives three types of core parameters through its internal data interface: risk level results (such as "high risk", "medium risk", "low risk") and physical diffusion trend parameters of hidden dangers output by the dynamic risk analysis unit 200 (such as the expected diffusion rate of slope collapse of 0.5 m / min and the expected spread range of flooding of 10 m² / min); spatial distribution parameters of personnel in the construction area output by the personnel sensing module 130 in the multi-dimensional data automatic acquisition unit 100 (including the real-time coordinates (X, Y, Z) of each person and their work group, with a sampling period of 5 seconds / time); and physical status parameters of equipment operation output by the structure sensing module 110 in the multi-dimensional data automatic acquisition unit 100 (such as the real-time rotation angle of the tower crane, the operating speed of the construction elevator, and the working current of the welding machine, with a sampling period of 10 seconds / time).

[0053] Specifically, the parameter fusion processing module 310 performs three levels of verification on the received parameters: First, it verifies the data integrity (such as whether the personnel coordinates include Z-axis elevation information, and whether the equipment parameters are missing key indicators), marking missing parameters as "invalid" and replacing them with the historical average of the most recent three periods; second, it verifies the data timeliness (such as whether the timestamp of the hidden danger spread trend parameter is more than 1 minute away from the present), and automatically triggers the dynamic risk analysis unit 200 to re-push out timed-out data; finally, it verifies the parameter rationality (such as whether the personnel coordinates fall within the effective range of the electronic map of the construction area, and whether the equipment current exceeds 150% of the rated value), marking unreasonable parameters as "abnormal" and retaining the original values ​​for subsequent traceability.

[0054] Specifically, after verification, the parameter fusion processing module 310 converts multi-source parameters into a standardized format: personnel spatial distribution parameters are converted into a "construction area grid matrix" (the construction area is divided into 10m×10m grids, with each grid labeled with the number and density of personnel); equipment operation parameters are converted into "equipment status codes" (e.g., using values ​​from 0-100 to represent equipment normality, where 0 represents shutdown and 100 represents complete normality); and hazard physical diffusion trend parameters are converted into a "diffusion rate matrix" (including diffusion velocity and diffusion acceleration in the X and Y axes). Simultaneously, all parameters undergo spatiotemporal alignment processing, unifying the time reference to the system's millisecond-level clock and the spatial reference to the Gaussian coordinate system of the construction area, ensuring that parameters from different sources can be analyzed collaboratively in the same spatiotemporal dimension. The fused standardized parameters are output to the early warning range and intensity adjustment module 320 via the internal bus, and simultaneously cached locally in the parameter fusion processing module 310 (retaining data from the most recent 30 minutes for easy anomaly tracing).

[0055] In this embodiment, the graded hazard warning unit 300 also includes a warning range and intensity adjustment module 320. The warning range and intensity adjustment module 320 dynamically adjusts the warning reach range and warning intensity based on the fusion parameters output by the parameter fusion processing module 310, combined with the physical impact range characteristics and physical hazard degree characteristics of the risk. Based on the mechanical parameters of structural hazards or the diffusion parameters of environmental hazards, the spatial range that the hazard may affect is determined as the warning reach range, and then the corresponding sound and light output and terminal prompt warning intensity form are matched according to the risk level.

[0056] Specifically, after receiving the fused parameters output by the parameter fusion processing module 310, the warning range and intensity adjustment module 320 first determines the physical impact range calculation model based on the hazard type: For structural hazards (such as slope collapse and steel structure deformation), a "mechanical parameter-range mapping model" is adopted. The slope displacement rate and steel structure stress deviation output by the dynamic risk analysis unit 200 are used as inputs. Combined with the "safety distance requirements for foundation pit engineering" in the "Standard for Safety Inspection of Building Construction" (JGJ59-2011), the spatial range that the hazard may affect is calculated (e.g., a slope displacement rate of 0.2 mm / min corresponds to an influence radius of 15 m, and a rate of 0.5 mm / min corresponds to an influence radius of 30 m). For environmental hazards (such as flooding due to rainfall and the impact of strong winds), the "diffusion parameter-range mapping model" is adopted. The rainfall intensity and wind speed are used as inputs, and the topographic slope data of the construction area (entered during system initialization) are combined to calculate the diffusion range (e.g., when the rainfall is 15mm / h and the slope is 5°, the flooding range within 10 minutes is the area around the foundation pit within 20m).

[0057] Specifically, the warning reach area is dynamically expanded based on the physical influence area: when the personnel spatial distribution parameters in the fusion parameters indicate the presence of personnel within the physical influence area, the warning reach area is defined as "influence area + 5m safety buffer zone"; when large equipment (such as tower cranes or concrete pump trucks) is present, the warning reach area is further expanded by 1.5 times the equipment's operating radius (e.g., if the tower crane's operating radius is 30m, it is expanded to 45m). The defined warning reach area is stored in the form of an "electronic fence polygon coordinate set," containing the Gaussian coordinates (X,Y) and elevation Z value of the boundary points.

[0058] Specifically, the warning intensity is matched according to the risk level output by the dynamic risk analysis unit 200: low risk corresponds to Level 1 warning intensity, using "text prompts on the construction management terminal" at a frequency of once every 5 minutes; medium risk corresponds to Level 2 warning intensity, using "text prompts + on-site audible and visual alarms" at a frequency of once per minute; high risk corresponds to Level 3 warning intensity, using "text prompts + on-site audible and visual alarms + telephone notifications to management personnel" at a real-time push frequency (updating the location change of the hazard every 10 seconds). Simultaneously, the warning range and intensity adjustment module 320 outputs the "warning reach range coordinate set" and "warning intensity level" to the warning information generation and push module 330, and also synchronizes them to the electronic map display of the construction area (the warning range is marked with different colored dashed boxes, with red boxes representing Level 3 warnings and yellow boxes representing Level 2 warnings).

[0059] In this embodiment, the graded hazard warning unit 300 also includes a warning information generation and push module 330. The warning information generation and push module 330 generates warning information containing the physical location of the hazard, the physical type of the risk, and the priority based on the physical impact, based on the warning reach range and warning intensity output by the warning range and intensity adjustment module 320, and combines the results of the physical location of the hazard and the physical type of the risk from the dynamic risk analysis unit 200, and pushes it accurately to the construction management terminal.

[0060] Specifically, the early warning information generation and push module 330 receives the early warning reach range and early warning intensity output by the early warning range and intensity adjustment module 320, as well as the physical location of the hidden danger (accurate to specific coordinates, such as "northeast slope of the foundation pit, X=352.2m, Y=186.5m") and the physical type of risk (such as "slope collapse risk" and "personnel violation operation risk") output by the dynamic risk analysis unit 200. Then, it first extracts the core information elements: the physical location of the hidden danger is converted into "common name of the construction area + coordinates" (such as "northeast slope of foundation pit No. 3 (352.2, 186.5)"); the physical type of risk is associated with the corresponding "emergency response guidance code" (such as "immediately evacuate the affected area and start the slope monitoring encryption mode" for slope collapse risk); and the early warning reach range is associated with the "list of personnel and equipment within the range" (generated based on the personnel distribution and equipment status parameters of the parameter fusion processing module 310).

[0061] Specifically, the elements of the early warning information are prioritized based on their physical impact: the first priority is "physical location of the hazard + risk level" (e.g., "High risk: risk of collapse of the northeast slope of foundation pit No. 3"); the second priority is "personnel / equipment alerts within the affected area" (e.g., "There are 5 workers and 1 excavator within the affected area"); and the third priority is "emergency response guidelines" (e.g., "Please immediately organize personnel to evacuate to the safe area on the west side of the foundation pit and suspend surrounding machinery operations"). The sorted information is combined into a standardized early warning text, with a text length controlled within 200 characters (ensuring complete display on mobile devices).

[0062] Specifically, the targeted push notification employs a "multi-terminal collaboration" mechanism: for workers within the warning range, audio-visual alerts are pushed through their smart safety helmet terminals (matching the warning intensity; for a level 3 warning, the safety helmet vibrates and a red light flashes); the construction management terminal pushes complete warning information text and electronic map markings (displaying the location of the hazard and the warning range); for high-risk warnings, the project emergency command center's voice notification system is additionally triggered, automatically dialing preset management personnel numbers (such as the project manager and safety director) and playing the warning content in voice (e.g., "Emergency Notice: A high-risk slope collapse warning has occurred on the northeast side of foundation pit No. 3; please handle it immediately"). After the push is completed, the warning information generation and push module 330 records logs such as "push time, receiving terminal ID, and information open status," which are synchronized to the system backend to ensure that the warning information is traceable and verifiable.

[0063] The emergency linkage control unit 400 is linked with the graded hazard warning unit 300 based on physical signals. When the risk level reaches the medium or high level, it sends a shutdown physical command to the control system of special equipment in the construction area through the industrial bus and simultaneously triggers the audio signal output of the on-site emergency broadcast.

[0064] In this embodiment, the emergency response control unit 400 includes a signal receiving and judging module 410 and a device control execution module 420, wherein: The signal receiving and judgment module 410 is connected to the graded hazard warning unit 300 via a physical signal link. It receives the risk level results output by the graded hazard warning unit 300 in real time. When the risk level is identified as medium or high, it sends a linkage trigger signal to the equipment control execution module 420. Specifically, the signal receiving and judgment module 410 and the graded hazard warning unit 300 employ a dual-link physical connection to ensure reliability: the main link is an industrial Ethernet (based on TCP / IP protocol, transmission rate 100Mbps, communication cycle 100ms), used to receive risk level results and associated physical hazard type information in real time; the backup link is an RS485 bus (baud rate 9600bps, using Modbus-RTU protocol), which automatically switches only when the main link is interrupted to ensure no signal loss. Both links are laid with shielded twisted-pair cable, maintaining a distance of more than 50cm from power cables to avoid electromagnetic interference.

[0065] Specifically, the signal receiving and judgment module 410 has a built-in "risk level monitoring thread" that parses the received risk level data every 50ms. First, it verifies the data frame format (using a custom protocol, including a frame header "0xAA", risk level code (1 byte, 0x01 represents low risk, 0x02 represents medium risk, and 0x03 represents high risk), hazard type code (2 bytes, such as 0x0001 corresponding to slope collapse risk), check bit (CRC16), and frame tail "0x55"). Data with incorrect format is discarded and logged. Second, it compares the current risk level with the preset trigger threshold (medium risk and above, i.e., code ≥ 0x02). When the level meets the standard for three consecutive communication cycles (cumulative 300ms), it is determined to be a valid trigger condition (to avoid malfunctions caused by momentary interference).

[0066] Specifically, upon meeting the triggering conditions, the signal receiving and judging module 410 generates a linkage trigger signal: the signal uses a 24V DC level signal (active high), which is output to the signal input terminal of the equipment control execution module 420 through an opto-isolated relay. Simultaneously, a hazard type code is embedded in the signal (achieved through pulse width modulation, such as a 100ms pulse representing slope collapse risk and a 200ms pulse representing strong wind impact risk). The signal receiving and judging module 410 locally stores trigger records (including trigger time, risk level, hazard type, and link status) for 90 days, and supports querying historical data through the construction management platform.

[0067] After receiving the linkage trigger signal, the equipment control execution module 420 sends a standardized shutdown physical command to the control system of special equipment in the construction area through the industrial bus. At the same time, it triggers the audio signal output of the on-site emergency broadcast system through the audio control interface. The audio signal content is pre-stored in the built-in storage component of the emergency broadcast system based on the risk physical type.

[0068] Specifically, the equipment control execution module 420 is connected to the control system of special equipment in the construction area via a Profibus-DP industrial bus (transmission rate 1.5Mbps, maximum transmission distance 1000m, extended to cover the entire construction area via repeaters). The connected special equipment includes tower cranes, construction elevators, concrete pumps, and suspended platforms for high-altitude operations. Each device is assigned a unique bus address (within the range of 1-126), and the module stores an "equipment address-type mapping table" (e.g., address 1 corresponds to tower crane #1, and address 5 corresponds to construction elevator #1).

[0069] Specifically, upon receiving the linkage trigger signal, the equipment control execution module 420 first determines the scope of equipment to be controlled based on the embedded hazard type code: structural hazards (such as slope collapse, steel structure deformation) trigger the shutdown of all tower cranes, construction elevators, and high-altitude work platforms; environmental hazards (such as strong winds, flooding due to precipitation) additionally trigger the shutdown of concrete pumps (to avoid operations in the rain). Subsequently, a standardized shutdown physical command is generated. The command format follows the control signal requirements in the "Safety Technical Regulations for the Use of Construction Machinery" (JGJ33), including: equipment address (1 byte), command type (e.g., 0x01 represents emergency shutdown), safety code (2 bytes, matching the equipment's preset code to prevent erroneous commands), and execution time limit (1 byte, e.g., 0x0A represents shutdown within 10 seconds). The command is broadcast via the industrial bus. Upon receiving the command, the target equipment immediately executes the shutdown action (cuts off the drive power and activates the mechanical brake) and returns a "shutdown complete" confirmation signal (including the equipment address and shutdown status code) within 5 seconds.

[0070] Meanwhile, the equipment control execution module 420 triggers on-site broadcasting through the audio control interface (using RS232 protocol, connected to the main control component of the on-site emergency broadcasting system): First, it calls the pre-stored audio signal content (audio file format is MP3, sampling rate is 44.1kHz, stored in the SD card built into the broadcasting system) according to the hazard type code. For example, the slope collapse risk corresponds to "Please note: There is a risk of slope collapse in the No. 3 foundation pit area. Please evacuate to the safe area immediately. The equipment has been shut down in an emergency"); Second, it sends playback control instructions (including the playback area code, which is consistent with the warning coverage area. For example, code 0x03 corresponds to the area around the foundation pit), volume level (70% volume for medium risk, 100% volume for high risk) and number of loops (3 loops for medium risk, continuous playback for high risk until manually stopped).

[0071] Specifically, the equipment control execution module 420 monitors the execution result of the shutdown command and the broadcast status in real time: if no confirmation signal is received from the equipment within 5 seconds, the command is automatically resent (up to 3 times). If it still fails, an alarm "Equipment shutdown failed, please handle manually" is pushed to the construction management terminal. After the broadcast system returns the "playback completed" signal, the module records the execution result (including the number of shutdown equipment and the broadcast coverage area) and synchronizes it to the graded hazard warning unit 300 to form an emergency linkage closed loop.

[0072] Those skilled in the art will understand that the process of implementing all or part of the steps of the above embodiments can be carried out by hardware or by a program instructing the relevant hardware.

[0073] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely preferred examples and are not intended to limit the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A construction safety monitoring system with automatic detection and hazard warning functions, characterized in that, include: The multi-dimensional data automatic acquisition unit (100) collects building structure stability parameters, construction environment interference parameters and personnel operation position parameters in real time through physical parameter sensors deployed in key construction areas, and transmits the collected physical parameter data to the dynamic risk analysis unit (200) through an industrial-grade wireless encrypted transmission link. The dynamic risk analysis unit (200) is based on the real-time physical parameters output by the multi-dimensional data automatic acquisition unit (100) and uses an improved weighted fuzzy comprehensive evaluation model to quantitatively analyze construction safety risks. The improved weighted fuzzy comprehensive evaluation model dynamically adjusts the parameter weights according to the structural mechanical characteristics and environmental interference characteristics of the construction stage to determine whether there are any safety hazards and the risk level of the safety hazards. The graded hazard warning unit (300) receives the risk level results from the dynamic risk analysis unit (200), and constructs a multi-dimensional warning decision model by combining the spatial distribution parameters of personnel in the construction area, the physical state parameters of equipment operation, and the physical diffusion trend parameters of hidden dangers. It dynamically adjusts the warning reach range based on the physical impact range of the risk, adjusts the warning intensity based on the physical hazard degree of the risk, and pushes warning information containing the physical location of hidden dangers, physical type of risk, and priority based on physical impact to the construction management terminal. Emergency linkage control unit (400) is linked with graded hazard warning unit (300) based on physical signals. When the risk level reaches medium or high level, it sends a shutdown physical command to the control system of special equipment in the construction area through industrial bus and simultaneously triggers the audio signal output of on-site emergency broadcast.

2. The construction safety monitoring system with automatic detection and hazard warning functions according to claim 1, characterized in that, The multi-dimensional data automatic acquisition unit (100) includes a structure sensing module (110), an environment sensing module (120), and a personnel sensing module (130), wherein: The structural sensing module (110) is deployed in the key construction area and uses slope displacement sensors, scaffolding upright tilt sensors and steel structure stress sensors to collect building structure stability parameters. The environmental sensing module (120) is distributed in the construction area and uses harmful gas concentration sensors, environmental wind speed sensors and precipitation sensors to collect construction environmental interference parameters. The personnel sensing module (130) relies on the smart safety helmet and positioning base station, and uses UWB positioning sensors and contact status sensors to collect personnel working position and safety helmet wearing status parameters.

3. The construction safety monitoring system with automatic detection and hazard warning functions according to claim 2, characterized in that, The multi-dimensional data automatic acquisition unit (100) also includes a data validity verification module (140). The data validity verification module (140) judges the validity of the raw data collected by the structure sensing module (110), environment sensing module (120), and personnel sensing module (130) based on the physical measurement principle of sensors. When the data is determined to be abnormal or the transmission is interrupted, a sensor fault prompt is triggered, and the abnormal data is temporarily supplemented by the physical parameters of the adjacent valid period to ensure the continuity of data transmission to the dynamic risk analysis unit (200).

4. The construction safety monitoring system with automatic detection and hazard warning functions according to claim 3, characterized in that, The dynamic risk analysis unit (200) includes a construction stage identification module (210), which identifies the current construction stage based on the physical condition parameters output by the multi-dimensional data automatic acquisition unit (100). The construction stage identification module (210) performs the following steps: S210.

1. Real-time acquisition of physical condition parameters of the construction area, including the excavation depth of the foundation pit, through sensor data interface. Elevation of steel structure components and total building height ; S210.2, Preset construction stage judgment threshold: when and At that time, it was determined to be the foundation pit excavation stage; when and When, it is determined to be the main structure pouring stage; when At that time, it was determined to be in the decoration and renovation stage; S210.3, Real-time data collection , Compare with the preset threshold and output the current construction stage identifier.

5. The construction safety monitoring system with automatic detection and hazard warning functions according to claim 4, characterized in that, The dynamic risk analysis unit (200) further includes a weight dynamic adjustment module (220), which dynamically adjusts the weight ratio of each physical parameter based on the construction stage identifier output by the construction stage identification module (210); the weight dynamic adjustment module (220) performs the following steps to dynamically adjust the weight: S220.

1. Based on the construction stage identifier output by the construction stage identification module (210), retrieve the basic weight matrix of the corresponding construction stage. ; S220.2 Retrieve the mechanical influence factor matrix corresponding to the construction stage identifier. ; S220.3, Combining the basic weight matrix With the mechanical influence factor matrix The actual weight matrix of each physical parameter is calculated. S220.4 Normalize the actual weight matrix to ensure that the sum of the actual weights of each physical parameter is 1, and output the dynamic weights for subsequent risk quantification.

6. The construction safety monitoring system with automatic detection and hazard warning functions according to claim 5, characterized in that, The dynamic risk analysis unit (200) further includes a risk quantification module (230) and a risk feature matching module (240), wherein: The risk quantification module (230) calculates a comprehensive risk value based on the dynamic weights output by the weight dynamic adjustment module (220) and the measured values ​​of physical parameters output by the multi-dimensional data automatic acquisition unit (100). Specifically, it includes obtaining the single-parameter deviation by comparing the measured values ​​of physical parameters with the corresponding safety thresholds. The deviation acceleration, which characterizes the rate of deterioration of parameters, is calculated based on the continuous periodic deviation variation. The single-parameter risk membership degree is obtained by combining the deviation degree and the deviation acceleration. The comprehensive risk value is then obtained by weighting the dynamic weights and the single-parameter risk membership. ; The risk feature matching module (240) has a built-in risk physical feature library, which is used to classify risk levels and match specific safety hazard types, specifically including: based on comprehensive risk value The risk level is determined by comparing the risk with a preset threshold; the risk membership of each single parameter is then assigned. The combined features are compared and matched with the various types of hazard feature patterns stored in the risk physical feature library, and the specific safety hazard type that best matches the real-time parameter features is output.

7. The construction safety monitoring system with automatic detection and hazard warning functions according to claim 6, characterized in that, The graded hazard early warning unit (300) includes a parameter fusion processing module (310). The parameter fusion processing module (310) receives the risk level results from the dynamic risk analysis unit (200), combines the spatial distribution parameters of personnel in the construction area and the physical state parameters of equipment operation output by the multi-dimensional data automatic acquisition unit (100), and the physical diffusion trend parameters of hidden dangers output by the dynamic risk analysis unit (200), and performs collaborative fusion processing on the multi-source physical parameters to provide standardized parameter input for the multi-dimensional early warning decision model.

8. The construction safety monitoring system with automatic detection and hazard warning functions according to claim 7, characterized in that, The graded hazard warning unit (300) also includes a warning range and intensity adjustment module (320). The warning range and intensity adjustment module (320) dynamically adjusts the warning reach range and warning intensity based on the fusion parameters output by the parameter fusion processing module (310), combined with the physical impact range characteristics and physical hazard degree characteristics of the risk. Based on the mechanical parameters of structural hazards or the diffusion parameters of environmental hazards, the spatial range that the hazard may affect is determined as the warning reach range, and then the warning intensity form of the corresponding sound and light output and terminal prompt is matched according to the risk level.

9. The construction safety monitoring system with automatic detection and hazard warning functions according to claim 8, characterized in that, The graded hazard warning unit (300) also includes a warning information generation and push module (330). The warning information generation and push module (330) generates warning information containing the physical location of the hazard, the physical type of the risk, and the priority of physical impact based on the warning reach range and warning intensity output by the warning range and intensity adjustment module (320), combined with the results of the physical location of the hazard and the physical type of the risk from the dynamic risk analysis unit (200), and pushes it accurately to the construction management terminal.

10. The construction safety monitoring system with automatic detection and hazard warning functions according to claim 9, characterized in that, The emergency response control unit (400) includes a signal receiving and judging module (410) and an equipment control execution module (420), wherein: The signal receiving and judgment module (410) is connected to the graded hazard warning unit (300) through a physical signal link. It receives the risk level result output by the graded hazard warning unit (300) in real time. When the risk level is identified as medium risk or high risk, it sends a linkage trigger signal to the equipment control execution module (420). After receiving the linkage trigger signal, the equipment control execution module (420) sends a standardized shutdown physical command to the control system of special equipment in the construction area through the industrial bus. At the same time, it triggers the audio signal output of the on-site emergency broadcast system through the audio control interface. The audio signal content is pre-stored in the built-in storage component of the emergency broadcast system based on the risk physical type.

Images