Intelligent construction site safety situation assessment method and device based on big data, medium and equipment

Abstract

The invention discloses an intelligent construction site safety situation assessment method and device based on big data, a medium and equipment. The method comprises the steps that historical sensing data information and corresponding safety risk type information in an intelligent construction site are acquired; inputting the historical data information into a gradient boosting decision tree model for iterative training, and determining the weight of each feature in each tree in the gradient boosting decision tree; screening out a key feature subset according to the weight of each feature in each tree in the gradient boosting decision tree; inputting the key feature subset and the security risk type information into a random forest model for iterative training, and determining an optimal random forest model; and determining the current security situation level of the construction site according to the current sensing data and the optimal random forest model. According to the invention, real-time accurate assessment of the security situation of the intelligent construction site is realized, manual intervention is not needed, and the labor cost is greatly reduced.

Description

Technical Field

[0001] The present invention relates to the field of big data technology, and in particular to a method, device, medium and equipment for evaluating the safety situation of a smart construction site based on big data. Background Art

[0002] Smart safety situation assessment is an important part of ensuring construction site safety and the smooth progress of the project. Traditional safety situation assessment methods mainly rely on manual inspections and empirical judgments, which are highly subjective and inefficient. With the rise of big data technology, more and more fields are beginning to try to use big data for situation assessment and prediction. Summary of the invention

[0003] The purpose of the present invention is to overcome the deficiencies in the prior art and provide a method, device, medium and equipment for smart construction site safety situation assessment based on big data. To achieve the above purpose, the present invention is implemented by adopting the following technical solutions: In a first aspect, the present invention provides a method for assessing safety situation of a smart construction site based on big data, comprising: Obtain historical sensor data information and corresponding safety risk type information in the smart construction site, the sensor data information includes but is not limited to construction site environment data, equipment status data, personnel behavior data, the number of high-risk operation areas, and the number of current unrectified risks; Inputting the historical data information into a gradient boosting decision tree model for iterative training, and determining the weight of each feature in each tree in the gradient boosting decision tree; Filtering out a subset of key features according to the weight of each feature in each tree in the gradient boosting decision tree; Inputting the key feature subset and the security risk category information into a random forest model for iterative training to determine an optimal random forest model; The current safety situation level of the construction site is determined according to the current sensor data and the optimal random forest model, where the safety situation level includes severe danger, danger, and safety.

[0004] In combination with the first aspect, optionally, screening out a subset of key features according to the weight of each feature in each tree in the gradient boosting decision tree includes: According to the weight of each feature in each tree in the gradient boosting decision tree, the importance score and the current training accuracy corresponding to each feature are calculated; Remove the feature corresponding to the minimum importance score in the feature set to obtain an updated feature set, and record the current training accuracy and the updated feature set; Training the gradient boosting decision tree according to the updated feature set, determining the weight of each feature, and then calculating the importance score and current training accuracy corresponding to each feature, deleting the feature corresponding to the minimum importance score, recording the current training accuracy and the updated feature set, and performing the next round of training until there are no features in the updated feature set; The feature subset corresponding to the highest training accuracy value is determined as the key feature subset.

[0005] In combination with the first aspect, optionally, calculating the importance score corresponding to each feature according to the weight of each feature in each tree in the gradient boosting decision tree includes: The Gini index of each node in each tree is calculated. The calculation formula of the Gini index is: , where k represents the feature number, a represents the node number, is the weight value of feature k on node a, is the Gini index of feature k at node a; The Gini index change of feature k at node a is calculated. The calculation formula of the Gini index change is: ,in and are the Gini indexes of the left and right child nodes respectively, is the Gini index of the current node; Calculate the importance score of feature k at node a. The calculation formula of the importance score is: ,in Indicates the proportion of the sample size of node a to the total sample size, is the change in the Gini index of feature k at node a; According to the importance score of the feature k at node a, the importance scores of the feature k in all trees are calculated. The calculation formula for the importance scores of the feature k in all trees is: ,in is the change in the Gini index of feature k at node a, is the number of trees, is the number of nodes in each tree.

[0006] In combination with the first aspect, optionally, determining the current safety situation level of the construction site according to the current sensor data and the optimal random forest model includes: Screening feature data corresponding to key features in the current sensor data to determine them as current key feature data; Inputting the current key feature data into the optimal random forest model to obtain the voting score corresponding to each current security risk type; Calculate the current security situation value according to the voting score corresponding to each current security risk type and the severity weight value corresponding to each security risk type; The current security situation level is determined based on the current security situation value and a preset security situation quantification mapping table.

[0007] In combination with the first aspect, optionally, the preset security situation quantification mapping table includes: security situation level entries and security situation value interval entries, and the security situation levels and security situation value intervals are in one-to-one correspondence.

[0008] In combination with the first aspect, optionally, the safety risk type information includes: personnel safety risk, machinery and equipment safety risk, material safety risk, environmental safety risk, and system safety risk.

[0009] In a second aspect, the present invention provides a smart construction site safety situation assessment device based on big data, comprising: Training data acquisition module: obtains historical sensor data information and corresponding safety risk type information in the smart construction site, the sensor data information includes but is not limited to construction site environment data, equipment status data, personnel behavior data, the number of high-risk operation areas, and the number of current unrectified risks; Gradient boosting decision tree training module: inputting the historical data information into a gradient boosting decision tree model for iterative training, and determining the weight of each feature in each tree in the gradient boosting decision tree; Key feature subset determination module: screen out key feature subsets according to the weight of each feature in each tree in the gradient boosting decision tree; Optimal random forest model training module: inputting the key feature subset and the security risk category information into the random forest model for iterative training to determine the optimal random forest model; Current safety situation assessment module: determines the current safety situation level of the construction site based on the current sensor data and the optimal random forest model, and the safety situation level includes severe danger, danger, and safety.

[0010] In a third aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the computer program implements the big data-based smart construction site safety situation assessment method as described in any one of the first aspects.

[0011] In a fourth aspect, the present invention provides a device, comprising: A memory for storing instructions; The processor is used to execute the instructions so that the device implements the smart construction site safety situation assessment method based on big data as described in any one of the first aspects.

[0012] Compared with the prior art, the beneficial effects achieved by the method, device, medium and equipment for smart construction site safety situation assessment based on big data provided by the embodiments of the present invention include: The present invention obtains historical sensor data information and corresponding safety risk type information in a smart construction site, wherein the sensor data information includes but is not limited to construction site environment data, equipment status data, personnel behavior data, the number of high-risk operation areas, and the number of current unrectified risks; the historical data information is input into a gradient boosting decision tree model for iterative training to determine the weight of each feature in each tree in the gradient boosting decision tree; according to the weight of each feature in each tree in the gradient boosting decision tree, a key feature subset is screened out; the key feature subset and the safety risk type information are input into a random forest model for iterative training to determine the optimal random forest model; according to the current sensor data and the optimal random forest model, the current safety situation level of the construction site is determined, and the safety situation level includes severe danger, danger, and safety. The pre-trained neural network model is obtained. The present invention realizes real-time and accurate assessment of the safety situation of smart construction sites without manual intervention, which greatly reduces labor costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 A flowchart of a method for evaluating the safety situation of a smart construction site based on big data is provided in Example 1 of the present invention. DETAILED DESCRIPTION

[0014] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of this embodiment can be combined with each other. The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0015] Embodiment 1: like Figure 1 As shown, an embodiment of the present invention provides a method for assessing safety situation of a smart construction site based on big data, comprising: S1: Obtain historical sensor data information and corresponding safety risk type information in the smart construction site, wherein the sensor data information includes but is not limited to construction site environment data, equipment status data, personnel behavior data, the number of high-risk operation areas, and the number of currently unrectified risks; Specifically, the safety risk type information includes: personnel safety risk, machinery and equipment safety risk, material safety risk, environmental safety risk, and system safety risk.

[0016] S2: Inputting the historical data information into a gradient boosting decision tree model for iterative training, and determining the weight of each feature in each tree in the gradient boosting decision tree; S3: Screening out a subset of key features according to the weight of each feature in each tree in the gradient boosting decision tree; Specifically, the key feature subset is screened out according to the weight of each feature in each tree in the gradient boosting decision tree, including: S3-1: Calculate the importance score and current training accuracy of each feature according to the weight of each feature in each tree in the gradient boosting decision tree; Specifically, the step of calculating the importance score corresponding to each feature according to the weight of each feature in each tree in the gradient boosting decision tree includes: S3-1-1: Calculate the Gini index of each node in each tree. The calculation formula of the Gini index is: , where k represents the feature number, a represents the node number, is the weight value of feature k on node a, is the Gini index of feature k at node a; S3-1-2: Calculate the change in the Gini index of feature k at node a. The calculation formula for the change in the Gini index is: ,in and are the Gini indexes of the left and right child nodes respectively, is the Gini index of the current node; S3-1-3: Calculate the importance score of feature k at node a. The calculation formula of the importance score is: ,in Indicates the proportion of the sample size of node a to the total sample size, is the change in the Gini index of feature k at node a; S3-1-4: According to the importance score of the feature k at node a, calculate the importance score of the feature k in all trees. The calculation formula for the importance score of the feature k in all trees is: ,in is the change in the Gini index of feature k at node a, is the number of trees, is the number of nodes in each tree.

[0017] S3-2: Remove the feature corresponding to the minimum importance score in the feature set to obtain an updated feature set, and record the current training accuracy and the updated feature set; S3-3: training the gradient boosting decision tree according to the updated feature set, determining the weight of each feature, and then calculating the importance score and current training accuracy corresponding to each feature, deleting the feature corresponding to the minimum importance score, recording the current training accuracy and the updated feature set, and performing the next round of training until there are no features in the updated feature set; S3-4: Determine the feature subset corresponding to the highest training accuracy value as the key feature subset.

[0018] S4: Inputting the key feature subset and the security risk category information into a random forest model for iterative training to determine the optimal random forest model; S5: Determine the current safety situation level of the construction site based on the current sensor data and the optimal random forest model, where the safety situation level includes severe danger, danger, and safety.

[0019] Specifically, determining the current safety situation level of the construction site according to the current sensor data and the optimal random forest model includes: S5-1: Filtering feature data corresponding to the key feature in the current sensing data to determine as current key feature data; S5-2: Input the current key feature data into the optimal random forest model to obtain the voting score corresponding to each current security risk type; S5-3: Calculate the current security situation value according to the voting score corresponding to each current security risk type and the severity weight value corresponding to each security risk type; S5-4: Determine the current security situation level according to the current security situation value and a preset security situation quantification mapping table.

[0020] Specifically, the preset security situation quantification mapping table includes: security situation level entries and security situation value interval entries, and the security situation levels and security situation value intervals are in one-to-one correspondence.

[0021] The big data-based smart construction site safety situation assessment method provided by the present invention realizes real-time and accurate assessment of the safety situation of smart construction sites without the need for human intervention, thus greatly reducing labor costs.

[0022] Embodiment 2: The embodiment of the present invention provides a smart construction site safety situation assessment device based on big data, comprising: Training data acquisition module: obtains historical sensor data information and corresponding safety risk type information in the smart construction site, the sensor data information includes but is not limited to construction site environment data, equipment status data, personnel behavior data, the number of high-risk operation areas, and the number of current unrectified risks; Gradient boosting decision tree training module: inputting the historical data information into a gradient boosting decision tree model for iterative training, and determining the weight of each feature in each tree in the gradient boosting decision tree; Key feature subset determination module: screen out key feature subsets according to the weight of each feature in each tree in the gradient boosting decision tree; Optimal random forest model training module: inputting the key feature subset and the security risk category information into the random forest model for iterative training to determine the optimal random forest model; Current safety situation assessment module: determines the current safety situation level of the construction site based on the current sensor data and the optimal random forest model, and the safety situation level includes severe danger, danger, and safety.

[0023] The smart construction site safety situation assessment device based on big data provided by the present invention realizes real-time and accurate assessment of the safety situation of the smart construction site without the need for human intervention, thus greatly reducing labor costs.

[0024] Embodiment three: An embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, a smart construction site safety situation assessment method based on big data as described in Embodiment 1 is implemented.

[0025] Embodiment 4: An embodiment of the present invention provides a device, including: A memory for storing instructions; The processor is used to execute the instructions so that the device executes a smart construction site safety situation assessment method based on big data as described in Example 1.

[0026] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0027] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0028] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0029] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0030] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A method for assessing the safety situation of a smart construction site based on big data, characterized in that: include: Obtain historical sensor data information and corresponding safety risk type information in the smart construction site, the sensor data information includes but is not limited to construction site environment data, equipment status data, personnel behavior data, the number of high-risk operation areas, and the number of current unrectified risks; Inputting the historical data information into a gradient boosting decision tree model for iterative training, and determining the weight of each feature in each tree in the gradient boosting decision tree; Filtering out a subset of key features according to the weight of each feature in each tree in the gradient boosting decision tree; Inputting the key feature subset and the security risk category information into a random forest model for iterative training to determine an optimal random forest model; The current safety situation level of the construction site is determined according to the current sensor data and the optimal random forest model, where the safety situation level includes severe danger, danger, and safety.

2. According to the big data-based smart construction site safety situation assessment method of claim 1, it is characterized in that: The method of screening out a subset of key features according to the weight of each feature in each tree in the gradient boosting decision tree includes: According to the weight of each feature in each tree in the gradient boosting decision tree, the importance score and the current training accuracy corresponding to each feature are calculated; Remove the feature corresponding to the minimum importance score in the feature set to obtain an updated feature set, and record the current training accuracy and the updated feature set; Training the gradient boosting decision tree according to the updated feature set, determining the weight of each feature, and then calculating the importance score and current training accuracy corresponding to each feature, deleting the feature corresponding to the minimum importance score, recording the current training accuracy and the updated feature set, and performing the next round of training until there are no features in the updated feature set; The feature subset corresponding to the highest training accuracy value is determined as the key feature subset.

3. According to the big data-based smart construction site safety situation assessment method of claim 2, it is characterized in that: The step of calculating the importance score corresponding to each feature according to the weight of each feature in each tree in the gradient boosting decision tree includes: The Gini index of each node in each tree is calculated. The calculation formula of the Gini index is: , where k represents the feature number, a represents the node number, is the weight value of feature k on node a, is the Gini index of feature k at node a; The Gini index change of feature k at node a is calculated. The calculation formula of the Gini index change is: ,in and are the Gini indexes of the left and right child nodes respectively, is the Gini index of the current node; Calculate the importance score of feature k at node a. The calculation formula of the importance score is: ,in Indicates the proportion of the sample size of node a to the total sample size, is the change in the Gini index of feature k at node a; According to the importance score of the feature k at node a, the importance scores of the feature k in all trees are calculated. The calculation formula for the importance scores of the feature k in all trees is: ,in is the change in the Gini index of feature k at node a, is the number of trees, is the number of nodes in each tree.

4. According to the big data-based smart construction site safety situation assessment method of claim 1, it is characterized in that: Determining the current safety situation level of the construction site according to the current sensor data and the optimal random forest model includes: Screening feature data corresponding to key features in the current sensor data to determine them as current key feature data; Inputting the current key feature data into the optimal random forest model to obtain the voting score corresponding to each current security risk type; Calculate the current security situation value according to the voting score corresponding to each current security risk type and the severity weight value corresponding to each security risk type; The current security situation level is determined based on the current security situation value and a preset security situation quantification mapping table.

5. The method for evaluating safety situation of smart construction sites based on big data according to claim 4 is characterized in that: The preset security situation quantification mapping table includes: security situation level entries and security situation value interval entries, and the security situation level and the security situation value interval are in one-to-one correspondence.

6. The method for evaluating safety situation of smart construction sites based on big data according to claim 1 is characterized in that: The safety risk types information includes: personnel safety risks, machinery and equipment safety risks, material safety risks, environmental safety risks, and system safety risks.

7. A smart construction site safety situation assessment device based on big data, characterized in that: include: Training data acquisition module: obtains historical sensor data information and corresponding safety risk type information in the smart construction site, the sensor data information includes but is not limited to construction site environment data, equipment status data, personnel behavior data, the number of high-risk operation areas, and the number of current unrectified risks; Gradient boosting decision tree training module: inputting the historical data information into a gradient boosting decision tree model for iterative training, and determining the weight of each feature in each tree in the gradient boosting decision tree; Key feature subset determination module: screen out key feature subsets according to the weight of each feature in each tree in the gradient boosting decision tree; Optimal random forest model training module: inputting the key feature subset and the security risk category information into the random forest model for iterative training to determine the optimal random forest model; Current safety situation assessment module: determines the current safety situation level of the construction site based on the current sensor data and the optimal random forest model, and the safety situation level includes severe danger, danger, and safety.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, it implements the big data-based smart construction site safety situation assessment method as described in any one of claims 1-6.

9. A device, characterized in that: include: A memory for storing instructions; A processor is used to execute the instructions so that the device implements the big data-based smart construction site safety situation assessment method as described in any one of claims 1-6.

Images