Construction operator safety risk assessment early warning method, device and equipment

Abstract

The invention relates to the field of artificial intelligence, and discloses a construction worker safety risk assessment early warning method, device and equipment, and the method comprises the steps: obtaining the attribute data of a construction worker and the behavior data of a construction site, taking the data as input data, inputting the input data into a construction worker safety risk assessment model, the method comprises the following steps: inputting an input variable to obtain a multi-dimensional characteristic value of the input variable, carrying out clustering analysis on the multi-dimensional characteristic value to obtain a construction worker safety risk classification, and generating a construction worker safety risk early warning indication according to the construction worker safety risk classification. By comprehensively analyzing multiple dimensions of the constructors and inputting data into the safety risk assessment model, the system can mine data features of different dimensions and identify potential safety hazards of the constructors.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence, and in particular to a method, device and equipment for evaluating and warning safety risks of construction workers. Background Art

[0002] With the continuous increase in the scale and number of construction projects, the safety management of construction sites has gradually become an important research field. In order to scientifically evaluate the safety capabilities of construction workers, there are many research practices in the existing technology. For example, the safety capabilities of construction workers should match the job requirements, and a capability evaluation system has been designed; based on the "scenario-task-capability" analysis model, an accident emergency preparedness capability evaluation index system is proposed; starting from the principle of matching ability with position, a safety capability evaluation system for power enterprise operators is constructed to ensure the matching of employee characteristics with their job task requirements. In addition, with the development of intelligent technology, technical means such as the Internet of Things, big data and artificial intelligence have been introduced to try to achieve real-time dynamic warning at the job site. At the same time, the impact of environmental factors on safety behavior has been considered, and an environmental dynamic warning system has been developed. A construction information platform is built through the Internet of Things technology to achieve real-time collection, transmission and judgment of environmental data.

[0003] Although some research progress has been made in the safety risk assessment of construction workers, the existing technology still has the following technical problems:

[0004] First, the assessment dimensions are not comprehensive enough. Existing technologies often focus on the assessment of security knowledge and skills, ignoring key factors such as security awareness and behavioral habits;

[0005] Second, the evaluation standards lack uniformity and quantification, which affects the fairness and accuracy of the evaluation results;

[0006] 3. The evaluation method is relatively traditional, relying on questionnaires or on-site observations. The operation is complex and costly, making it difficult to promote and apply on a large scale on construction sites.

[0007] In addition, the safety supervision of many construction projects currently still adopts a manual management model. Construction personnel only need to register simply to enter the site, and subsequent operations lack effective supervision, which makes accidents caused by operational errors prone to occur. Summary of the invention

[0008] In view of this, the present application provides a method for safety risk assessment and early warning for construction workers, which solves the technical problem that in the prior art, each substation mainly obtains a numerical value through a line loss rate calculation formula, and then analyzes it according to a reasonable line loss rate set manually, which lacks a scientific basis, has a high error, and is difficult to achieve refined management.

[0009] According to a first aspect of the present application, a construction worker safety risk assessment and early warning method is provided, comprising:

[0010] Obtaining attribute data of construction workers and construction site behavior data as input data;

[0011] Inputting the input data into the construction worker safety risk assessment model to obtain multi-dimensional characteristic values ​​of the input variables;

[0012] Cluster analysis is performed on multidimensional eigenvalues ​​to obtain the safety risk classification of construction workers;

[0013] Based on the safety risk classification of construction workers, safety risk warning indicators for construction workers are generated.

[0014] According to a second aspect of the present application, a construction worker safety risk assessment and early warning device is provided, comprising:

[0015] The acquisition module is used to obtain the attribute data of construction workers and the behavior data of the construction site as input data;

[0016] A calculation module, used for inputting input data into the construction worker safety risk assessment model to obtain multi-dimensional characteristic values ​​of the input variables;

[0017] The classification module is used to perform cluster analysis on multi-dimensional feature values ​​to obtain the safety risk classification of construction workers;

[0018] The alarm module is used to generate safety risk warning indications for construction workers based on the safety risk classification of construction workers.

[0019] According to the third aspect of the present application, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, and when the processor executes the computer program, the steps of the above-mentioned construction worker safety risk assessment and early warning method are implemented.

[0020] By means of the above technical scheme, the present application provides a method, device and equipment for safety risk assessment and early warning of construction workers, which obtains the attribute data and construction site behavior data of construction workers as input data, inputs the input data into the safety risk assessment model of construction workers to obtain the multi-dimensional characteristic values ​​of the input variables, performs cluster analysis on the multi-dimensional characteristic values ​​to obtain the safety risk classification of construction workers, and generates safety risk early warning indications for construction workers according to the safety risk classification of construction workers. By comprehensively analyzing multiple dimensions of construction workers and inputting the data into the safety risk assessment model, the system can mine data features of different dimensions and identify potential safety hazards of construction workers.

[0021] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application and to implement it in accordance with the contents of the specification, and to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0023] Figure 1 A schematic diagram of an application scenario of a construction worker safety risk assessment and early warning method provided in an embodiment of the present application is shown;

[0024] Figure 2 A schematic diagram of a process for evaluating and warning the safety risks of construction workers provided in an embodiment of the present application is shown;

[0025] Figure 3 A schematic diagram of the process of the improved DBSCAN density clustering algorithm provided in the embodiment of the present application is shown;

[0026] Figure 4 A structural schematic diagram of a construction worker safety risk assessment and early warning device provided in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0027] The specific implementation of the present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments. It should be noted that the embodiments and features in the embodiments of the present application can be combined with each other without conflict.

[0028] The embodiment of the present invention provides a construction worker safety risk assessment and early warning method, which can be applied to Figure 1 In the scenario of construction, managers need to understand the safety status of each construction worker and their behavior records in real time to ensure construction safety and prevent potential accidents. To achieve this goal, the system collects the following data sources to evaluate and warn of safety risks:

[0029] Construction site behavior data: Real-time collection of construction workers' on-site behavior data, such as equipment operation records, card swiping records for entering and leaving the construction site, helmet wearing conditions, real-time location tracking, etc. Through these data, the current working status of construction workers can be understood.

[0030] Construction site historical behavior database: This database stores the past behavior records of construction personnel, including past equipment operations, violation records, safety training participation, past assessment results, etc. These historical data provide a reference for subsequent safety assessments and help analyze whether construction personnel have safety hazards.

[0031] Construction worker attribute database: This database contains the basic information and attributes of construction workers, such as age, certificate qualification, type of work, length of service, health status, etc. By analyzing these attribute data, the system can determine whether each worker meets the job requirements or has health and safety risks.

[0032] Server: As the core hub of data processing, the server receives real-time and historical data from different data sources, and inputs these data into a pre-built safety risk assessment model. The method for safety risk assessment and early warning of construction workers provided by the present invention is implemented by obtaining the attribute data and construction site behavior data of the construction workers as input data, and inputting the input data into the safety risk assessment model of construction workers to obtain the multi-dimensional characteristic values ​​of the input variables, and performing cluster analysis on the multi-dimensional characteristic values ​​to obtain the safety risk classification of the construction workers. According to the safety risk classification of the construction workers, a safety risk early warning indication for the construction workers is generated. By comprehensively analyzing the multiple dimensions of the construction workers and inputting the data into the safety risk assessment model, the system can mine the data features of different dimensions and identify the potential safety hazards of the construction workers.

[0033] The present invention is described in detail below through specific embodiments.

[0034] Embodiment 1:

[0035] like Figure 2 As shown, a construction worker safety risk assessment and early warning method provided in an embodiment of the present invention includes:

[0036] Step 201: Acquire attribute data of construction workers and construction site behavior data as input data;

[0037] Step 202: input the input data into the construction worker safety risk assessment model to obtain multi-dimensional characteristic values ​​of the input variables;

[0038] Step 203: Perform cluster analysis on the multi-dimensional feature values ​​to obtain a safety risk classification of construction workers;

[0039] Step 204: Generate a safety risk warning indication for the construction workers based on the safety risk classification of the construction workers.

[0040] The invention provides a construction worker safety risk assessment and early warning method, which obtains the attribute data and construction site behavior data of the construction workers as input data, inputs the input data into the construction worker safety risk assessment model to obtain the multidimensional characteristic value of the input variable, performs cluster analysis on the multidimensional characteristic value to obtain the construction worker safety risk classification, and generates a construction worker safety risk early warning indication according to the construction worker safety risk classification. By comprehensively analyzing the multiple dimensions of the construction workers and inputting the data into the safety risk assessment model, the system can mine the data features of different dimensions and identify the potential safety hazards of the construction workers.

[0041] Embodiment 2:

[0042] The training process of the construction worker safety risk assessment model before step 202 includes:

[0043] Step 202-1, obtaining attribute data of construction workers and historical behavior data of the construction site, and creating a data set;

[0044] The attribute data S of the construction workers and the historical behavior data X of the construction site are obtained. The attribute data of the construction workers include the qualification score of the construction workers' certificates, the score of the construction workers, the score of the safety training of the construction workers entering the site, and the score of the illegal behavior of the construction workers. The historical behavior data of the construction site include the construction time period, the identification of the construction site, and the construction type.

[0045] Based on the attribute data S of the construction workers and the historical behavior data X of the construction site, for a single sample (S:X), an N-dimensional tensor is constructed, where N = N s +N x , N s is the dimension of the attribute data of construction workers, N x It is the dimension of the historical behavior data of the construction site.

[0046] Step 202-2, preprocessing the data set to divide it into a training set and a test set;

[0047] Step 202-3: construct a deep convolutional neural network model, obtain multidimensional feature values ​​of input variables, and perform cluster analysis on the multidimensional feature values ​​to obtain corresponding classification outputs;

[0048] Among them, a deep convolutional neural network model is constructed, and a multi-layer convolution kernel is used to extract the attribute characteristics of construction personnel, the periodic change characteristics of construction behavior, and the characteristics of construction violation behaviors. The DBSCAN density clustering algorithm is used to perform cluster analysis on the characteristic values ​​to identify the safety risk classification of construction workers.

[0049] Step 202-4: Use the created dataset to train and test the deep convolutional neural network model.

[0050] Among them, the deep convolutional neural network model is trained and optimized by the stochastic gradient descent algorithm SGD, the loss function adopts the cross entropy loss function, and the network parameters are adjusted during the training process to improve the model classification accuracy.

[0051] In the second embodiment of the present invention, the DBSCAN density clustering algorithm is introduced to perform cluster analysis on the eigenvalues. It is not necessary to preset the number of clusters, and clusters can be automatically formed according to the density of the data. By analyzing the multidimensional eigenvalues ​​of the construction personnel (such as violation records, equipment operation frequency, working time data, etc.), the algorithm can automatically identify personnel with higher safety risks (such as personnel who often violate regulations or are in high-risk areas for a long time) and classify them as high-risk personnel. Compared with traditional methods, the automated clustering process reduces human intervention and improves the accuracy of identification. The data at the construction site usually have some noise or abnormal points (such as abnormal records caused by equipment failure or behavioral data of temporary personnel). DBSCAN can automatically identify and exclude noise points by setting a reasonable minimum number of points MinPts and radius ε. The method can adaptively adjust the classification criteria in different construction scenarios (such as different types of work, different project scales) to identify construction personnel with similar risk characteristics. The method has a wide range of applicability and can meet the safety assessment needs of different projects.

[0052] Embodiment three:

[0053] The behavior data at the construction site often contains a lot of noise and anomalies, such as abnormal records caused by equipment failure, irregular operation behaviors, etc. Traditional DBSCAN is prone to misclassifying noise points as normal behavior points in dense data, resulting in inaccurate clustering results, such as Figure 3 As shown, the improved DBSCAN density clustering algorithm is introduced in the third embodiment of the present invention. Accordingly, the step of using the DBSCAN density clustering algorithm to perform clustering analysis on the eigenvalues ​​in step 202-3 includes:

[0054] Step 202-31, initializing the first distance threshold ε1, the second distance threshold ε2 and the minimum number of neighbors MinPts;

[0055] Among them, ε1 is less than ε2, ε1 corresponds to the first neighborhood and high-density core points, ε2 corresponds to the second neighborhood and low-density core points, and the calculation steps of the double threshold parameters and the minimum number of neighbors MinPts include:

[0056] Perform statistics on the eigenvalues, generate a histogram, and record the frequency of each eigenvalue;

[0057] According to the frequency of occurrence of eigenvalues, the eigenvalues ​​are divided into high-density areas and low-density areas;

[0058] A first distance threshold ε1 is set for a high-density area, and a second distance threshold ε2 is set for a low-density area;

[0059] A first minimum number of neighbors MinPts1 is set for a high-density area, and a second minimum number of neighbors MinPts2 is set for a low-density area.

[0060] Step 202-32, randomly select an unprocessed sample point P in the data set, and determine whether P is a high-density core point. If the number of sample points in the first neighborhood of P is greater than or equal to MinPts, mark the point as a high-density core point, and start cluster expansion; if the number of sample points in the first neighborhood of P is less than MinPts, determine whether P is a low-density core point. If the number of sample points in the second neighborhood of P is greater than or equal to MinPts, mark the point as a low-density core point, and start cluster expansion;

[0061] Step 202-33, according to the type of the core point, mark all points in the neighborhood of the core point as the same class, and for all core points in the neighborhood, continue to expand their neighborhoods and add unprocessed points to the corresponding class;

[0062] Step 202-34: For points that do not belong to the neighborhood of any core point, mark them as noise points.

[0063] In construction scenarios, some potential high-risk behaviors may not occur frequently but have a significant impact on safety. DBSCAN's dual threshold mechanism (using a smaller neighborhood ε1 in high-density areas and a larger neighborhood ε2 in low-density areas) can effectively identify low-frequency but important potential high-risk behaviors, such as occasional illegal operations or improper use of key equipment, and issue early warnings for these hidden dangers.

[0064] The second embodiment of the present invention reduces the overall calculation amount and improves the clustering efficiency through histogram-based distance calculation. It is suitable for large-scale construction behavior data. The dual threshold mechanism can distinguish between high-frequency and low-frequency risk behaviors, so that the model can more accurately identify core points and boundary points, thereby reducing errors. The improved method can process the data flow of the construction site more quickly, effectively identify risk behaviors and issue timely warnings.

[0065] Further, as Figures 2 to 3 In the specific implementation of the method, an embodiment of the present invention provides a construction worker safety risk assessment and early warning device, such as Figure 4 As shown, the device comprises:

[0066] The collection module 410 is used to obtain the attribute data of the construction workers and the behavior data of the construction site as input data;

[0067] A calculation module 420, for inputting input data into a construction worker safety risk assessment model to obtain multi-dimensional characteristic values ​​of input variables;

[0068] A classification module 430 is used to perform cluster analysis on the multi-dimensional feature values ​​to obtain a safety risk classification of construction workers;

[0069] The alarm module 440 is used to generate a safety risk warning indication for construction workers according to the safety risk classification of construction workers.

[0070] In an embodiment of the present invention, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of implementing the construction worker safety risk assessment and early warning method include:

[0071] Obtaining attribute data of construction workers and construction site behavior data as input data;

[0072] Inputting the input data into the construction worker safety risk assessment model to obtain multi-dimensional characteristic values ​​of the input variables;

[0073] Cluster analysis is performed on multidimensional eigenvalues ​​to obtain the safety risk classification of construction workers;

[0074] Based on the safety risk classification of construction workers, safety risk warning indicators for construction workers are generated.

[0075] It should be noted that the above embodiments only use the safety risk assessment of construction workers to illustrate the principles and implementation steps of the embodiments of the present invention, without specific limitation on the actual application scenarios. The invention is also applicable to other scenarios that require real-time safety monitoring and risk management, such as industrial production safety management, safety management in the electricity and energy industries, logistics and warehousing management, etc. For the functions or steps that can be implemented by computer-readable storage media or computer devices, please refer to the aforementioned method embodiments. To avoid repetition, they will not be described one by one here.

[0076] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0077] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0078] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A construction worker safety risk assessment and early warning method, characterized in that: include: Obtaining attribute data of construction workers and construction site behavior data as input data; Inputting the input data into the construction worker safety risk assessment model to obtain multi-dimensional characteristic values ​​of the input variables; Performing cluster analysis on the multidimensional feature values ​​to obtain a safety risk classification of construction workers; Based on the safety risk classification of construction workers, safety risk warning indicators for construction workers are generated.

2. The construction worker safety risk assessment and early warning method according to claim 1 is characterized in that: Before the step of inputting the input data into the construction worker safety risk assessment model to obtain the multi-dimensional characteristic values ​​of the input variables, the method includes: Obtain the attribute data of construction workers and historical behavior data of the construction site to create a data set; Preprocess the data set and divide it into training set and test set; Constructing a deep convolutional neural network model, obtaining multidimensional feature values ​​of input variables, and performing cluster analysis on the multidimensional feature values ​​to obtain corresponding classification outputs; The deep convolutional neural network model is trained and tested using the created dataset.

3. The construction worker safety risk assessment and early warning method according to claim 2 is characterized in that: The step of obtaining the attribute data of the construction workers and the historical behavior data of the construction site and creating a data set includes: Obtain the attribute data S of the construction workers and the historical behavior data X of the construction site, wherein the attribute data of the construction workers include the qualification score of the construction workers' certificates, the score of the construction workers, the score of the safety training for the construction workers entering the site, and the score of the illegal behavior of the construction workers, and the historical behavior data of the construction site include the construction time period, the identification of the construction site, and the construction type; Based on the attribute data S of the construction workers and the historical behavior data X of the construction site, for a single sample (S:X), an N-dimensional tensor is constructed, where N = N s +N x , N s is the dimension of the attribute data of construction workers, N x It is the dimension of the historical behavior data of the construction site.

4. The construction worker safety risk assessment and early warning method according to claim 2 is characterized in that: The step of constructing a deep convolutional neural network model, obtaining multidimensional feature values ​​of input variables, and performing cluster analysis on the multidimensional feature values ​​to obtain corresponding classification outputs includes: A deep convolutional neural network model was constructed, and multi-layer convolution kernels were used to extract the attribute characteristics of construction personnel, the periodic change characteristics of construction behavior, and the characteristics of construction violation behaviors. The DBSCAN density clustering algorithm was used to perform cluster analysis on the feature values ​​to identify the safety risk classification of construction workers.

5. The construction worker safety risk assessment and early warning method according to claim 2 is characterized in that: The step of using the created data set to train and test the deep convolutional neural network model comprises: The deep convolutional neural network model is trained and optimized by the stochastic gradient descent algorithm SGD. The loss function adopts the cross entropy loss function. The network parameters are adjusted during the training process to improve the classification accuracy of the model.

6. The construction worker safety risk assessment and early warning method according to claim 4 is characterized in that: The step of using the DBSCAN density clustering algorithm to perform cluster analysis on the eigenvalues ​​includes: Initialize the first distance threshold ε1, the second distance threshold ε2 and the minimum number of neighbors MinPts, where ε1 is smaller than ε2, ε1 corresponds to the first neighborhood and high-density core points, and ε2 corresponds to the second neighborhood and low-density core points; Randomly select an unprocessed sample point P in the data set and determine whether P is a high-density core point. If the number of sample points in the first neighborhood of P is greater than or equal to MinPts, mark the point as a high-density core point and start cluster expansion. If the number of sample points in the first neighborhood of P is less than MinPts, determine whether P is a low-density core point. If the number of sample points in the second neighborhood of P is greater than or equal to MinPts, mark the point as a low-density core point and start cluster expansion. According to the type of core point, all points in the neighborhood of the core point are marked as the same class. For all core points in the neighborhood, their neighborhoods are further expanded and unprocessed points are added to the corresponding class. Points that do not belong to the neighborhood of any core point are marked as noise points.

7. The construction worker safety risk assessment and early warning method according to claim 6 is characterized in that: The steps of initializing the first distance threshold ε1, the second distance threshold ε2 and the minimum number of neighbors MinPts include: Perform statistics on the eigenvalues, generate a histogram, and record the frequency of each eigenvalue; According to the frequency of occurrence of eigenvalues, the eigenvalues ​​are divided into high-density areas and low-density areas; A first distance threshold ε1 is set for a high-density area, and a second distance threshold ε2 is set for a low-density area; A first minimum number of neighbors MinPts1 is set for a high-density area, and a second minimum number of neighbors MinPts2 is set for a low-density area.

8. A construction worker safety risk assessment and early warning device, characterized in that: include: The acquisition module is used to obtain the attribute data of construction workers and the behavior data of the construction site as input data; A calculation module, used for inputting input data into the construction worker safety risk assessment model to obtain multi-dimensional characteristic values ​​of the input variables; A classification module, used for performing cluster analysis on the multi-dimensional feature values ​​to obtain a safety risk classification of construction workers; The alarm module is used to generate safety risk warning indications for construction workers based on the safety risk classification of construction workers.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the construction worker safety risk assessment and early warning method as described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by the processor, the steps of the construction worker safety risk assessment and early warning method as described in any one of claims 1 to 7 are implemented.

Images