Intelligent construction site area building personnel safety work management system and method

Abstract

The invention discloses an intelligent construction site area building personnel safety work management system and method. The system comprises an intelligent sensing module, an intelligent control module and an early warning response module. The intelligent sensing module is a multi-dimensional sensing network comprising video monitoring equipment, wearable safety equipment and environment monitoring equipment, and is used for acquiring multivariate data and transmitting the data to the intelligent control module; the intelligent control module comprises a behavior analysis sub-module and a risk prediction sub-module; the behavior analysis sub-module adopts a deep learning algorithm model and is used for identifying and analyzing construction site safety behaviors; the risk prediction sub-module is used for performing deep mining and correlation analysis through big data fusion and by applying a machine learning algorithm, and constructing a construction site safety risk assessment model to realize risk prediction; and the early warning response module is used for realizing immediate early warning and emergency response when the behavior analysis sub-module identifies unsafe behaviors or the risk prediction sub-module predicts an imminent risk.

Description

Technical Field

[0001] The present invention relates to the technical field of safety management, and in particular to a smart construction site area construction personnel safety work management system and method. Background Art

[0002] Modern construction sites have complex environments and a large number of people. Traditional safety supervision methods are difficult to achieve real-time and comprehensive monitoring, which can easily lead to safety hazards not being discovered in time. In recent years, although some construction sites have introduced video surveillance and basic Internet of Things technologies, they lack effective intelligent analysis methods and cannot automatically identify unsafe behaviors, and the early warning efficiency and accuracy are limited. The defects or shortcomings of existing technologies in the field of smart construction site personnel safety management are mainly reflected in the following aspects:

[0003] Monitoring limitations: Traditional video surveillance systems usually rely on manual monitoring, which is not only inefficient, but also prone to missing important security events due to human negligence. Even if some systems have detection functions, they often generate a large number of false alarms due to low recognition accuracy and cannot accurately distinguish between normal operations and unsafe behaviors.

[0004] Lack of intelligent analysis: In existing technologies, the judgment of personnel safety behavior mostly relies on rule presets and simple logical judgments, and lacks the ability to understand complex scenarios. This makes it difficult for the system to identify unsafe behaviors such as not wearing safety equipment correctly and operating heavy machinery in violation of regulations, and it is also unable to dynamically adjust safety standards according to environmental changes.

[0005] Early warning lag and passive response: Existing systems often trigger alarms only after an accident occurs, lacking an effective early warning mechanism. Even if there is an early warning, it is usually a simple judgment based on a fixed threshold, and cannot intelligently predict risks based on historical data and real-time situations, making it difficult to achieve active prevention.

[0006] Lack of personalized management and continuous optimization: For individual worker safety education and performance evaluation, the existing management system often adopts a one-size-fits-all approach, lacking targeted training and incentives. At the same time, the system itself lacks self-learning and optimization mechanisms, making it difficult to dynamically adjust the algorithm model according to actual conditions and improve the accuracy of identification and early warning.

[0007] User experience and operational complexity: Many existing systems have unfriendly interfaces and are complex to operate, making them difficult for frontline managers and workers with limited technical skills to quickly understand and use, which affects the popularization of the system and the maximization of its utility. Summary of the invention

[0008] In order to solve the problems raised in the above background technology, the technical solution adopted by the present invention is:

[0009] A smart construction site area construction personnel safety work management system, including an intelligent perception module, an intelligent control module and an early warning response module;

[0010] The intelligent sensing module is a multi-dimensional sensing network including video surveillance equipment, wearable safety equipment and environmental monitoring equipment, which is used to obtain multi-dimensional data and transmit the data to the intelligent control module;

[0011] The intelligent control module includes a behavior analysis submodule and a risk prediction submodule;

[0012] The behavior analysis submodule adopts a deep learning algorithm model to identify and analyze safety behaviors on construction sites;

[0013] The risk prediction submodule is used to build a construction site safety risk assessment model through big data fusion and deep mining and correlation analysis using machine learning algorithms to achieve risk prediction;

[0014] The early warning response module is used to implement immediate early warning and emergency response when the behavior analysis submodule identifies unsafe behavior or the risk prediction submodule predicts an impending risk.

[0015] In some embodiments, in the smart sensing module:

[0016] The video surveillance equipment uses a 5G camera with night vision function to ensure all-weather and all-scenario coverage. The video data collected by the video surveillance equipment is compressed and encoded and transmitted to the intelligent control module via a wireless network;

[0017] The wearable safety equipment includes a smart helmet and a positioning bracelet, and the wearable safety equipment has multiple types of sensors built in to monitor the worker's position, movement status and physiological indicators in real time;

[0018] The environmental monitoring equipment includes a thermometer and a gas detector, which are used to monitor the environmental parameters of the construction site, including temperature, humidity and concentration of harmful gases;

[0019] The data collected by the wearable safety device and the environmental monitoring device are uploaded to the intelligent control module through the IoT protocol.

[0020] In some embodiments, the intelligent control module is configured in an Internet of Things cloud platform and / or a data processing center.

[0021] In some embodiments, the behavior analysis submodule adopts a deep learning algorithm model, which establishes a video data set through multiple pre-collected and labeled construction site safety behavior video samples, and is trained using the deep learning framework TensorFlow or PyTorch. The model is used to automatically identify unsafe behaviors, including not wearing a safety helmet, not wearing protective clothing, operating machinery in violation of regulations, and entering restricted areas.

[0022] In some embodiments, in the risk prediction submodule, the multivariate data collected by the intelligent perception module, the analysis result data of the behavior analysis submodule and the historical accident record data are integrated through big data fusion, and a construction site safety risk assessment model is constructed using a machine learning algorithm to comprehensively consider personnel behavior, environmental factors, and historical accident records, to conduct a quantitative assessment of the overall safety situation of the construction site, and to predict future risk probabilities.

[0023] In some embodiments, the early warning response module is specifically used to immediately send early warning information to the construction site manager, safety supervisor and workers' personal smart devices through the intelligent control module when the risk prediction submodule identifies unsafe behavior or predicts an impending risk, and triggers a preset emergency response process, including but not limited to on-site warnings, personnel evacuation, and emergency rescue.

[0024] In some embodiments, a personalized safety training and performance management module is also included, which is used to provide customized online safety training courses and establish a safety performance evaluation system based on the safety behavior records of each worker analyzed by the system.

[0025] The present invention provides a method for managing the safety work of construction personnel in a smart construction site area, which adopts the above-mentioned smart construction site area construction personnel safety work management system and includes the following steps:

[0026] S1, acquiring multivariate data through an intelligent sensing module and transmitting the data to the intelligent control module;

[0027] S2. In the behavior analysis submodule of the intelligent control module, a deep learning algorithm model is used to realize the identification and analysis of safety behaviors on the construction site;

[0028] S3. In the risk prediction submodule of the intelligent control module, through big data fusion and deep mining and correlation analysis using machine learning algorithms, a construction site safety risk assessment model is constructed, and risk prediction is achieved using the construction site safety risk assessment model;

[0029] S4. When the behavior analysis submodule identifies unsafe behavior, or the risk prediction submodule predicts an impending risk, immediate warning and emergency response are achieved through the early warning response module.

[0030] In some embodiments, in step S3, the following steps are specifically included:

[0031] S31. Integrate the multivariate data collected by the intelligent perception module, the analysis result data of the behavior analysis submodule and the historical accident record data through big data fusion;

[0032] S32. When constructing a construction site safety risk assessment model, select m evaluation objects and n safety indicators from the integrated data set, determine the weight of each safety indicator, and construct a decision matrix X = [x ij ] m×n , where 1≤i≤m, 1≤j≤n, x ij represents the jth safety index data of the i-th evaluation object;

[0033] S33, normalization is performed by the following formula:

[0034]

[0035] Among them, max(x j )、min(x j ) represent the maximum and minimum values ​​of the j-th safety index respectively;

[0036] S34. Determine the weight of each safety indicator and create a decision matrix Y = [y ij ] m×n ,in:

[0037]

[0038] w j represents the weight of the jth security indicator;

[0039] S35. Determine the positive ideal solution Y of the decision matrix by the following formula + and the negative ideal solution Y - :

[0040]

[0041] S36. Calculate the Euclidean distance of each evaluation object to the positive and negative ideal solutions:

[0042]

[0043] in, Respectively represent the Euclidean distance from the i-th evaluation object to the positive and negative ideal solutions;

[0044] S37. Calculate the relative fit by the following formula:

[0045]

[0046] The relative fit T i Indicates the degree to which the i-th evaluation object is close to the positive ideal solution and away from the negative ideal solution;

[0047] S38. Using the current relative fit T i The value of the risk prediction is realized, T i The larger the value, the smaller the risk.

[0048] Compared with the prior art, the present invention has the following beneficial effects:

[0049] The smart construction site area construction personnel safety work management system and method provided by the present invention integrates the Internet of Things, big data analysis, cloud computing and artificial intelligence technologies to solve the problems of low monitoring efficiency, inaccurate identification, weak risk prediction ability, lack of data support for management decisions, insufficient personalized safety training, etc. in traditional safety management systems. It realizes real-time and accurate monitoring of the safety behavior of construction site personnel, effectively prevents and reduces the occurrence of safety accidents, and improves the overall safety management level of the construction site. At the same time, it also provides intelligent and data-driven decision-making support for construction site safety management, and helps to make construction site safety management intelligent, refined and efficient, which is conducive to improving the operating efficiency of the park, optimizing resource allocation, and enhancing safety management and control capabilities. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 A schematic diagram of the smart construction site area construction personnel safety work management system provided by the present invention;

[0051] Figure 2 A schematic flow chart of the method for managing safe work of construction personnel in a smart construction site area provided by the present invention. DETAILED DESCRIPTION

[0052] In order to make the technical means, creative features, objectives and effects achieved by the present invention easy to understand, the following further describes how the present invention is implemented in conjunction with the accompanying drawings and specific implementation methods.

[0053] Reference Figure 1As shown, the present invention provides a smart construction site area construction personnel safety work management system, including an intelligent perception module 1, an intelligent control module 2 and an early warning response module 3; the intelligent perception module 1 is a multi-dimensional perception network including video surveillance equipment, wearable safety equipment and environmental monitoring equipment, which is used to obtain multivariate data and transmit the data to the intelligent control module 2; the intelligent control module 2 includes a behavior analysis submodule and a risk prediction submodule; the behavior analysis submodule adopts a deep learning algorithm model for identifying and analyzing safety behaviors on the construction site; the risk prediction submodule is used to fuse big data and use machine learning algorithms for deep mining and correlation analysis to construct a construction site safety risk assessment model to achieve risk prediction; the early warning response module 3 is used to achieve immediate early warning and emergency response when the behavior analysis submodule identifies unsafe behaviors or the risk prediction submodule predicts impending risks.

[0054] Preferably, in the intelligent sensing module 1: the video surveillance equipment adopts a 5G camera with night vision function to ensure all-weather and full-scene coverage, and the video data collected by the video surveillance equipment is compressed and encoded, and then transmitted to the intelligent control module 2 through the wireless network; the wearable safety equipment includes a smart safety helmet and a positioning bracelet, and the wearable safety equipment has multiple types of sensors built in to monitor the workers' position, movement status and physiological indicators in real time; the environmental monitoring equipment includes a thermometer and a gas detector to monitor the environmental parameters of the construction site, including temperature, humidity and harmful gas concentration; the data collected by the wearable safety equipment and the environmental monitoring equipment are uploaded to the intelligent control module 2 through the IoT protocol.

[0055] Preferably, the intelligent control module 2 is configured in the IoT cloud platform and / or data processing center. According to the actual situation on site, a centralized or distributed data processing center can be established to store, process and analyze the collected data; the cloud platform can provide computing resources to support big data analysis, AI model training and operation.

[0056] Preferably, the behavior analysis submodule adopts a deep learning algorithm model, which establishes a video data set through multiple pre-collected and labeled construction site safety behavior video samples. The video samples include positive examples (safe operations) and negative examples (unsafe behaviors). The model is trained using the deep learning framework TensorFlow or PyTorch. The model is used to automatically identify unsafe behaviors, including not wearing a safety helmet, not wearing protective clothing, operating machinery in violation of regulations, and entering restricted areas.

[0057] Preferably, in the risk prediction submodule, the multivariate data collected by the intelligent perception module 1, the analysis result data of the behavior analysis submodule and the historical accident record data are integrated through big data fusion, and a construction site safety risk assessment model is constructed using a machine learning algorithm to comprehensively consider personnel behavior, environmental factors, and historical accident records, to conduct a quantitative assessment of the overall safety situation of the construction site, and to predict future risk probabilities.

[0058] Preferably, the early warning response module 3 is specifically used to send early warning information to the construction site manager, safety supervisor and workers' personal smart devices immediately through the intelligent control module 2 when the risk prediction submodule identifies unsafe behavior or predicts an impending risk, and trigger the preset emergency handling process at the same time, including but not limited to on-site warnings, personnel evacuation, emergency rescue, and relevant personnel can be notified through text messages, APP push, on-site broadcasts, etc., and according to the early warning level, the corresponding emergency plan is initiated, such as emergency evacuation, equipment shutdown inspection, medical assistance, etc.

[0059] Preferably, it also includes a personalized safety training and performance management module, which is used to provide customized online safety training courses and establish a safety performance evaluation system based on the safety behavior records of each worker analyzed by the system.

[0060] It can be seen that the present invention realizes intelligent real-time monitoring. By integrating deep learning and Internet of Things technology, a set of intelligent monitoring systems that can automatically and accurately identify unsafe behaviors of construction site personnel are constructed, which significantly improves monitoring efficiency and accuracy and reduces the burden of human resources. The present invention also establishes a comprehensive risk prediction model, integrates multi-source data, including environmental parameters, personnel behavior, etc., and uses big data analysis and machine learning algorithms to establish a comprehensive risk assessment system to achieve early warning and intervention of potential safety issues.

[0061] In addition, it can provide managers with an intuitive Web or APP interface to display real-time monitoring images, safety reports, warning logs, etc. By providing a visual management platform and data analysis reports, it helps management quickly understand the safety status of the construction site, formulate more scientific and reasonable safety management measures based on data-driven decision support, and optimize safety management decisions.

[0062] Reference Figure 2 As shown, another aspect of the present invention provides a method for managing the safety work of construction personnel in a smart construction site area, which adopts the above-mentioned smart construction site area construction personnel safety work management system and includes the following steps:

[0063] S1, acquiring multivariate data through the intelligent sensing module 1, and transmitting the data to the intelligent control module 2;

[0064] S2. In the behavior analysis submodule of the intelligent control module 2, a deep learning algorithm model is used to realize the identification and analysis of safety behaviors on the construction site;

[0065] S3. In the risk prediction submodule of the intelligent control module 2, a construction site safety risk assessment model is constructed by integrating big data and using a machine learning algorithm for deep mining and correlation analysis, and risk prediction is achieved using the construction site safety risk assessment model;

[0066] S4. When the behavior analysis submodule identifies unsafe behavior, or the risk prediction submodule predicts an impending risk, immediate warning and emergency response are implemented through the warning response module 3.

[0067] In a specific embodiment, in step S3, the following steps are specifically included:

[0068] S31, integrating the multivariate data collected by the intelligent perception module 1, the analysis result data of the behavior analysis submodule and the historical accident record data through big data fusion;

[0069] S32. When constructing a construction site safety risk assessment model, select m evaluation objects and n safety indicators from the integrated data set, determine the weight of each safety indicator, and construct a decision matrix X = [x ij ] m×n , where 1≤i≤m, 1≤j≤n, x ij represents the jth safety index data of the i-th evaluation object;

[0070] S33, normalization is performed by the following formula:

[0071]

[0072] Among them, max(x j )、min(x j ) represent the maximum and minimum values ​​of the j-th safety index respectively;

[0073] S34. Determine the weight of each safety indicator and create a decision matrix Y = [y ij ] m×n ,in:

[0074]

[0075] w j represents the weight of the jth security indicator;

[0076] S35. Determine the positive ideal solution Y of the decision matrix by the following formula + and the negative ideal solution Y - :

[0077]

[0078] S36. Calculate the Euclidean distance of each evaluation object to the positive and negative ideal solutions:

[0079]

[0080] in, Respectively represent the Euclidean distance from the i-th evaluation object to the positive and negative ideal solutions;

[0081] S37. Calculate the relative fit by the following formula:

[0082]

[0083] The relative fit T i Indicates the degree to which the i-th evaluation object is close to the positive ideal solution and away from the negative ideal solution;

[0084] S38. Using the current relative fit T i The value of the risk prediction is realized, T i The larger the value of , the smaller the risk. Multiple thresholds can be set to correspond to different risk levels, thereby triggering different degrees of early warning responses in step S4.

[0085] In summary, the smart construction site area construction personnel safety work management system and method provided by the present invention integrates the Internet of Things, big data analysis, cloud computing and artificial intelligence technologies to solve the problems of low monitoring efficiency, inaccurate identification, weak risk prediction ability, lack of data support for management decisions, insufficient personalized safety training, etc. in traditional safety management systems. It realizes real-time and accurate monitoring of the safety behavior of construction site personnel, effectively prevents and reduces the occurrence of safety accidents, and improves the overall safety management level of the construction site. At the same time, it also provides intelligent and data-driven decision-making support for construction site safety management, and helps to make construction site safety management intelligent, refined and efficient, which is conducive to improving the operating efficiency of the park, optimizing resource allocation, and enhancing safety management and control capabilities.

[0086] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the purpose and scope of the technical solution of the present invention, which should be included in the scope of the claims of the present invention.

Claims

1. A smart construction site area construction personnel safety work management system, characterized in that: It comprises an intelligent sensing module (1), an intelligent control module (2) and an early warning response module (3); The intelligent sensing module (1) is a multi-dimensional sensing network including video surveillance equipment, wearable safety equipment and environmental monitoring equipment, and is used to obtain multi-dimensional data and transmit the data to the intelligent control module (2); The intelligent control module (2) comprises a behavior analysis submodule and a risk prediction submodule; The behavior analysis submodule adopts a deep learning algorithm model to identify and analyze safety behaviors on construction sites; The risk prediction submodule is used to build a construction site safety risk assessment model through big data fusion and deep mining and correlation analysis using machine learning algorithms to achieve risk prediction; The early warning response module (3) is used to implement immediate early warning and emergency response when the behavior analysis submodule identifies an unsafe behavior or the risk prediction submodule predicts an impending risk.

2. The smart construction site area construction personnel safety work management system according to claim 1 is characterized in that: In the intelligent sensing module (1): The video surveillance device uses a 5G camera with night vision function to ensure all-weather and all-scenario coverage. The video data collected by the video surveillance device is compressed and encoded and then transmitted to the intelligent control module (2) via a wireless network; The wearable safety equipment includes a smart helmet and a positioning bracelet, and the wearable safety equipment has multiple types of sensors built in to monitor the worker's position, movement status and physiological indicators in real time; The environmental monitoring equipment includes a thermometer and a gas detector, which are used to monitor the environmental parameters of the construction site, including temperature, humidity and concentration of harmful gases; The data collected by the wearable safety device and the environmental monitoring device are uploaded to the intelligent control module (2) through the IoT protocol.

3. The smart construction site area construction personnel safety work management system according to claim 1 is characterized in that: The intelligent control module (2) is configured on an Internet of Things cloud platform and / or a data processing center.

4. The smart construction site area construction personnel safety work management system according to claim 1 is characterized in that: The behavior analysis submodule adopts a deep learning algorithm model, which establishes a video dataset through multiple pre-collected and labeled construction site safety behavior video samples, and is trained using the deep learning framework TensorFlow or PyTorch. The model is used to automatically identify unsafe behaviors, including not wearing a safety helmet, not wearing protective clothing, operating machinery in violation of regulations, and entering prohibited areas.

5. The smart construction site area construction personnel safety work management system according to claim 1 is characterized in that: In the risk prediction submodule, the multivariate data collected by the intelligent perception module (1), the analysis result data of the behavior analysis submodule and the historical accident record data are integrated through big data fusion, and a construction site safety risk assessment model is constructed using a machine learning algorithm to comprehensively consider personnel behavior, environmental factors, and historical accident records, conduct a quantitative assessment of the overall safety situation of the construction site, and predict future risk probabilities.

6. The smart construction site area construction personnel safety work management system according to claim 1 is characterized in that: The early warning response module (3) is specifically used to immediately send early warning information to the construction site manager, safety supervisor and workers' personal smart devices through the intelligent control module (2) when the risk prediction submodule identifies unsafe behavior or predicts an impending risk, and simultaneously trigger a preset emergency response process, including but not limited to on-site warnings, personnel evacuation, and emergency rescue.

7. The smart construction site area construction personnel safety work management system according to claim 1 is characterized in that: It also includes a personalized safety training and performance management module, which is used to provide customized online safety training courses and establish a safety performance evaluation system based on the safety behavior records of each worker analyzed by the system.

8. A method for managing the safety of construction personnel in a smart construction site, characterized in that: The smart construction site area construction personnel safety work management system according to any one of claims 1 to 7 is adopted, and comprises the following steps: S1, acquiring multivariate data through the intelligent sensing module (1), and transmitting the data to the intelligent control module (2); S2. In the behavior analysis submodule of the intelligent control module (2), a deep learning algorithm model is used to realize the identification and analysis of safety behaviors on the construction site; S3. In the risk prediction submodule of the intelligent control module (2), a construction site safety risk assessment model is constructed by integrating big data and using a machine learning algorithm for deep mining and correlation analysis, and risk prediction is achieved using the construction site safety risk assessment model; S4. When the behavior analysis submodule identifies an unsafe behavior, or the risk prediction submodule predicts an impending risk, an immediate warning and emergency response are implemented through the warning response module (3).

9. The method for managing safe work of construction personnel in a smart construction site according to claim 8, characterized in that: In step S3, the following steps are specifically included: S31, integrating the multivariate data collected by the intelligent perception module (1), the analysis result data of the behavior analysis submodule and the historical accident record data through big data fusion; S32. When constructing a construction site safety risk assessment model, select m evaluation objects and n safety indicators from the integrated data set, determine the weight of each safety indicator, and construct a decision matrix X = [x ij ] m×n , where 1≤i≤m, 1≤j≤n, x ij represents the jth safety index data of the i-th evaluation object; S33, normalization is performed by the following formula: Among them, max(x j )、min(x j ) represent the maximum and minimum values ​​of the j-th safety index respectively; S34. Determine the weight of each safety indicator and create a decision matrix Y = [y ij ] m×n ,in: w j represents the weight of the jth security indicator; S35. Determine the positive ideal solution Y of the decision matrix by the following formula + and the negative ideal solution Y - : S36. Calculate the Euclidean distance of each evaluation object to the positive and negative ideal solutions: in, Respectively represent the Euclidean distance from the i-th evaluation object to the positive and negative ideal solutions; S37. Calculate the relative fit by the following formula: The relative fit T i Indicates the degree to which the i-th evaluation object is close to the positive ideal solution and away from the negative ideal solution; S38. Using the current relative fit T i The value of the risk prediction is realized, T i The larger the value, the smaller the risk.

Images