Construction safety management system based on big data

Abstract

The invention discloses a construction safety management system based on big data, and relates to the technical field of construction safety management. The trained construction risk prediction model predicts the current construction risk and obtains the corresponding risk prediction data; the operation status data of the intelligent construction equipment and the physical index data of the construction workers are monitored in real time. If there is abnormal data, an abnormal prompt instruction is issued, and the construction safety training program library outputs the construction safety training program. If a construction safety accident occurs in the construction area, a path planning algorithm plans a safe evacuation path for the construction workers; after collecting construction feedback data, a construction feedback data set is constructed, and a feedback coefficient is generated by the construction feedback data set. If the feedback coefficient does not exceed the expectation, the current construction management strategy is adjusted. If the physical health indicators of the construction workers frequently produce abnormalities, the construction risk is reduced after adaptive adjustment.

Description

Technical Field

[0001] The present invention relates to the technical field of construction safety management, and in particular to a construction safety management system based on big data. Background Art

[0002] Construction safety management refers to the important management activities during the implementation of construction projects, which aims to prevent and control the occurrence of various safety accidents at the construction site, protect the safety and health of construction personnel and the surrounding environment, and ensure the smooth progress of the project by formulating and implementing a series of safety regulations, operating procedures and emergency measures. It covers safety education and training, on-site safety inspections, hidden danger investigation and treatment, the installation and maintenance of safety protection facilities, special operations personnel management, accident emergency preparation and response, etc., and is an indispensable part of construction project management.

[0003] In the Chinese invention patent with application publication number CN117610918A, the power plant project construction safety management system, its camera module is used for all-round monitoring and real-time recording of the construction site; the sensor module is used for real-time monitoring and collection of the environment of the construction site; the smart wearable device is used to monitor the physiological indicators and activity trajectories of the construction personnel; the cloud platform is used to analyze and store the received video data, environmental data and personnel status data. The camera module and the sensor module are used to realize all-round monitoring and real-time collection of the construction site, and the smart wearable device locates the activity trajectory of the construction personnel in real time, so that the safety officer can view the activities of the construction personnel in the high-risk area in real time, and improve the safety and reliability of the construction site. Furthermore, by setting the body temperature module and the pulse module on the smart wearable device, the physiological health status of the construction personnel can also be detected, and the phenomenon of heat stroke caused by high temperature can be reduced.

[0004] Combining the above applications and the prior art:

[0005] When the construction area has entered the construction state, construction safety is the primary issue to be faced. Once a construction accident occurs, the construction period may be affected. In order to reduce construction risks, construction management personnel usually conduct regular inspections in the construction area to ensure that all construction operations are in compliance with regulations.

[0006] However, when encountering extreme weather conditions, the entire construction project will be affected to a certain extent, especially when more intelligent construction equipment is used in the construction area, such as autonomous driving and transport vehicles. The scene complexity in the construction area is relatively high, and the operation of intelligent construction equipment may also be disturbed, resulting in greater construction risks. In such a complex construction scenario, the existing construction safety management process usually continues to execute the current construction tasks, but due to the lack of attention to high-risk factors and prediction of construction risks, the construction risks in the actual construction process are still relatively high.

[0007] To this end, the present invention provides a construction safety management system based on big data. Summary of the invention

[0008] 1. Technical issues to be resolved

[0009] In view of the deficiencies of the prior art, the present invention provides a construction safety management system based on big data, which predicts the current construction risk through a trained construction risk prediction model and obtains corresponding risk prediction data; monitors the operating status data of intelligent construction equipment and the physical index data of construction workers in real time, and issues an abnormal prompt instruction if there is abnormal data, and outputs the corresponding construction safety training program from the safety training program library; if a construction safety accident occurs in the construction area, a path planning algorithm plans a safe evacuation path for the construction workers, and constructs a construction feedback data set after collecting construction feedback data, and generates a feedback coefficient from the construction feedback data set, and if the feedback coefficient does not exceed the feedback threshold, the current construction management strategy is adjusted. If the physical health indicators of construction workers frequently produce abnormalities, after adaptive adjustment, the construction risk is reduced, thereby solving the technical problems raised in the background technology.

[0010] (II) Technical solution

[0011] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0012] The construction safety management method based on big data includes collecting the operation data of intelligent construction equipment in the construction area under unconventional weather conditions, generating the scene complexity Coj during construction from the operation data, and if the scene complexity Coj exceeds the complexity threshold, collecting data in the construction area by the sensor network, and generating a construction environment data set after aggregation;

[0013] High-risk factors are screened out from a number of environmental factors, and the impact Yxt of environmental conditions on construction accident risks is generated based on the impact values ​​of the high-risk factors. If the impact Yxt exceeds the impact threshold, the trained construction risk prediction model is used to predict the current construction risk and obtain the corresponding risk prediction data.

[0014] Monitor the operating status data of intelligent construction equipment and the physical index data of construction workers in real time. If there is abnormal data, issue an abnormal prompt instruction. Generate a prompt density Moj based on the distribution state of the abnormal prompt instruction. If the prompt density Moj exceeds the density threshold, issue an early warning instruction to the outside.

[0015] After receiving the early warning instruction, the safety training program library outputs the construction safety training program. If a construction safety accident occurs in the construction area, a safe evacuation route is planned for the construction workers and marked on the electronic map;

[0016] After collecting the construction feedback data, a construction feedback data set is constructed, and a feedback coefficient Kts is generated from the construction feedback data set. If the feedback coefficient Kts does not exceed the feedback threshold, the current construction management strategy is adjusted.

[0017] Furthermore, construction data of intelligent construction equipment is collected in the construction area, including the number, location, operation status data and weather condition data of intelligent construction equipment, and the data are summarized to generate an equipment operation data set. Under dimensionless conditions, the scene complexity Coj during construction is generated by the equipment operation data set. If the obtained scene complexity Coj exceeds the complexity threshold, a sensor deployment instruction is issued to the outside.

[0018] Furthermore, the method of generating the scene complexity Coj during construction from the equipment operation data set is as follows:

[0019]

[0020] Among them, k and p are the number of time nodes and equipment activated, respectively. During the duration of the storm, C i is the number of intelligent construction equipment enabled at the i-th time node, C a is the corresponding mean, is the variance of the number of smart construction equipment enabled; Cov(YF) is the covariance between precipitation and wind speed during the construction period, Y i is the rainfall at the i-th time node, F i is the wind speed at the i-th time node, F b is the acceptable value of wind speed;

[0021] Furthermore, after receiving the sensor deployment instruction, a sensor network is deployed in the construction area so that the data collection frequency of the sensor network is positively correlated with the scene complexity Coj; the sensor network collects data in the construction area, and the collected data is aggregated to generate a construction environment data set.

[0022] Furthermore, after collecting historical accident data in the construction area, data visualization tools are used to perform visual analysis on the historical accident data, and through data recognition and feature engineering, areas and time periods where accidents frequently occur are identified; after obtaining historical accident data and corresponding construction environment data, multiple linear regression analysis is used to identify environmental conditions that have a greater correlation with accidents than expected as high-risk factors, and the corresponding regression coefficients are used as impact values.

[0023] Furthermore, the influence of environmental conditions on the risk of construction accidents is generated by the influence values ​​of several high-risk factors. Under dimensionless conditions, the following formula is used:

[0024]

[0025] Where n is the number of high risk factors, Xt i is the impact value of the ith high risk factor, Xt a is the average value of the influence value, Xt b is the acceptable value of the impact value; weight coefficient: 0≤S1≤1, 0≤S2≤1, and S1+S2=1;

[0026] If the obtained impact degree Yxt exceeds the preset impact threshold, a risk prediction instruction is issued to the outside.

[0027] Furthermore, after receiving the risk prediction instruction, historical accident data and corresponding environmental data are collected as sample data, and the neural convolution network is trained by the sample data to obtain the trained construction risk prediction model. The real-time construction environment data is used as input, and the trained construction risk prediction model predicts the current construction risk, obtains the corresponding risk prediction data, and issues monitoring instructions to the outside.

[0028] Furthermore, after obtaining the monitoring instructions, monitoring devices are installed in the construction area to collect worker activity data, intelligent construction equipment operation status data, and worker physical index data, and summarize them to generate a real-time monitoring data set; using the real-time monitoring data as input, the trained anomaly recognition model identifies abnormal data, and if abnormal data exists, the current situation is regarded as an abnormal situation, and an abnormal prompt instruction is issued to the outside.

[0029] Furthermore, after obtaining prompt instructions several times continuously within the prompt cycle, the time node when the abnormal prompt instruction is issued is recorded and used as the abnormal node; the prompt density Moj is generated according to the distribution state of the abnormal node, where:

[0030]

[0031] Among them, w is the number of abnormal nodes, Z ijis the time distance from the i-th abnormal node to the j-th abnormal node, Z a is the time distance average.

[0032] Further, corresponding risk prediction data is obtained, and feature engineering is performed on the risk prediction data to obtain corresponding risk features; a number of construction safety training programs are obtained, and according to the correspondence between the risk features and the training programs, the corresponding construction safety training programs are output from the safety training program library;

[0033] After determining the accident coverage, it is regarded as the accident area, and an electronic map covering the construction area is constructed. The accident area is marked on the electronic map, and the intelligent construction equipment is suspended to obtain the corresponding location information and the location of obstacles in the construction area.

[0034] Furthermore, the construction feedback data in each sub-area is collected during the observation period and summarized to generate a construction feedback data set; the feedback coefficient Kts is generated from the feedback data in the construction feedback data set, wherein the continuous safe construction time Ap and the training completion rate Pv are linearly normalized, and the corresponding data values ​​are mapped to the interval [0,1] in the following manner:

[0035]

[0036] Among them, q is the number of sub-regions, Ap i is the continuous safe construction time of the ith sub-area, Ap a is the average duration of continuous safe construction, Pv i is the training completion rate in the ith sub-region, Pv a is the average value of the training completion rate, weight coefficient: 0≤K1≤1, 0≤K2≤1, and K1+K2=1.

[0037] The construction safety management system based on big data includes a construction scene analysis unit, which collects the operation data of intelligent construction equipment in the construction area under unconventional weather conditions, and generates the scene complexity Coj during construction from the operation data. If the scene complexity Coj exceeds the complexity threshold, the sensor network collects data in the construction area, and generates a construction environment data set after aggregation;

[0038] The abnormal factor identification unit selects high-risk factors from a number of environmental conditions and generates the influence degree Yxt of environmental conditions on the risk of construction accidents based on the influence value of the high-risk factors. If the influence degree Yxt exceeds the influence threshold, the trained construction risk prediction model predicts the current construction risk and obtains the corresponding risk prediction data.

[0039] The abnormal warning unit monitors the operating status data of the intelligent construction equipment and the physical index data of the construction workers in real time. If there is abnormal data, it will issue an abnormal prompt instruction, and generate a prompt density Moj based on the distribution state of the abnormal prompt instruction. If the prompt density Moj exceeds the density threshold, it will issue a warning instruction to the outside.

[0040] After receiving the early warning instruction, the emergency handling unit will output the corresponding construction safety training plan from the safety training plan library. If a construction safety accident occurs in the construction area, the path planning algorithm will plan a safe evacuation path for the construction workers and mark it on the electronic map;

[0041] The feedback optimization unit collects the construction feedback data and builds a construction feedback data set, and generates a feedback coefficient Kts from the construction feedback data set. If the feedback coefficient Kts does not exceed the feedback threshold, the current construction management strategy is adjusted.

[0042] (III) Beneficial effects

[0043] The present invention provides a construction safety management system based on big data, which has the following beneficial effects:

[0044] 1. Use the sensor network to collect data in the construction area to monitor the current construction environment. When the current construction environment does not meet the construction conditions, adjust the construction status in time. By making the data collection frequency positively correlated with the complexity of the scene, avoid omissions in data collection in complex construction scenes.

[0045] 2. Through multiple linear regression analysis, high-risk factors can be screened out from several environmental factors, so that high-risk factors can be dealt with in a targeted manner during construction; the overall impact of external environmental factors on current construction safety can be judged. If the impact of external environmental factors on construction safety is large, the construction progress and construction tasks can be adjusted in construction scenarios with higher complexity to ensure construction safety.

[0046] 3. After collecting real-time construction environment data, the trained construction risk prediction model can predict the current construction risks, so that after predicting the possible construction risks, it can be dealt with in a timely manner to ensure construction safety; data monitoring and collection are carried out on construction workers in the construction state. When the health indicators of construction workers are abnormal, abnormal prompts are issued to remind the construction workers, which can reduce the risk of accidents.

[0047] 4. Describe the current situation of abnormal prompts to the outside based on the prompt density Moj, and determine the feedback of the construction workers under the construction status. If the physical health indicators of the construction workers frequently become abnormal, it means that they are not suitable for construction at present, or the current construction strategy is insufficient. After adaptive adjustments, the construction risk can be reduced.

[0048] 5. Based on the construction risk prediction data, the safety training program library outputs the corresponding construction safety training program. Through construction safety training, the construction risk can be reduced when there may be construction risks; by planning safe evacuation routes for construction workers, the evacuation of construction personnel can be accelerated to ensure the safety of construction personnel.

[0049] 6. Based on the feedback coefficient Kts, a comprehensive judgment can be made on the current construction process and construction feedback. If the construction feedback effect exceeds expectations, it means that the current construction process is normal. If not, the current construction management strategy should be adjusted to achieve construction management in complex scenarios and reduce construction risks and safety risks for construction personnel as a whole. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 It is a flowchart of the construction safety management method based on big data of the present invention;

[0051] Figure 2 It is a schematic diagram of the structure of the construction safety management system based on big data of the present invention. DETAILED DESCRIPTION

[0052] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions 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.

[0053] See also Figure 1 The present invention provides a construction safety management method based on big data, including:

[0054] Step 1: Under unconventional weather conditions, the operation data of intelligent construction equipment is collected in the construction area, and the scene complexity Coj during construction is generated from the operation data. If the scene complexity Coj exceeds the complexity threshold, the sensor network collects data in the construction area and generates a construction environment data set after aggregation;

[0055] The step 1 includes the following contents:

[0056] Step 101: When encountering extreme weather such as storms during construction, collect construction data of intelligent construction equipment in the construction area, including the number, location, operation status data and weather condition data of the intelligent construction equipment, and generate an equipment operation data set after aggregation;

[0057] Under dimensionless conditions, the scene complexity Coj during construction is generated by the equipment operation data set, where:

[0058]

[0059] Among them, k and p are the number of time nodes and equipment activated, respectively. During the duration of the storm, C i is the number of intelligent construction equipment enabled at the i-th time node, C a is the corresponding mean, is the variance of the number of smart construction equipment enabled; Cov(YF) is the covariance between precipitation and wind speed during the construction period, Y i is the rainfall at the i-th time node, F i is the wind speed at the i-th time node, F b is the acceptable value of wind speed;

[0060] Based on historical data and expectations for construction scene management, a complexity threshold is pre-set; if the obtained scene complexity Coj exceeds the complexity threshold, it means that the current construction environment is relatively complex and the construction risk is relatively high. At this time, a sensor deployment instruction is issued to the outside;

[0061] When extreme weather occurs during construction, by calculating the scene complexity Coj, it is possible to describe the current possible construction risks in advance, and suspend construction if necessary, thus providing a certain guarantee for construction safety.

[0062] Step 102: after receiving the sensor deployment instruction, a sensor network is deployed in the construction area, including sensors such as temperature, humidity, gas concentration, and vibration, and a data collection frequency is set for the sensor network so that the data collection frequency is positively correlated with the scene complexity Coj; the sensor network collects data in the construction area, such as temperature data, humidity data, vibration data, sound data, equipment location data, and light data, and the collected data is aggregated to generate a construction environment data set;

[0063] When using, combine the contents in steps 101 and 102:

[0064] By deploying a sensor network in the construction area and using the sensor network to collect data in the construction area, the current construction environment can be monitored based on the collected construction environment data. When the current construction environment does not meet the construction conditions, the construction status can be adjusted in time. At the same time, by making the data collection frequency positively correlated with the scene complexity, it can avoid many omissions in data collection in complex construction scenarios.

[0065] However, when encountering extreme weather conditions, the entire construction project will be affected to a certain extent, especially when more intelligent construction equipment is used in the construction area, such as autonomous driving and transport vehicles. The scene complexity in the construction area is relatively high, and the operation of intelligent construction equipment may also be disturbed, resulting in greater construction risks. In such a complex construction scenario, the existing construction safety management process usually continues to execute the current construction tasks, but due to the lack of attention to high-risk factors and prediction of construction risks, the construction risks in the actual construction process are still relatively high.

[0066] Step 2: Select high-risk factors from a number of environmental factors, and generate the impact of environmental conditions on construction accident risks Yxt according to the impact values ​​of the high-risk factors. If the impact Yxt exceeds the impact threshold, the trained construction risk prediction model is used to predict the current construction risk and obtain the corresponding risk prediction data.

[0067] The step 2 includes the following contents:

[0068] Step 201: After collecting historical accident data in the construction area, use data visualization tools to perform visualization analysis on the historical accident data, and identify areas and time periods where accidents frequently occur through data recognition and feature engineering;

[0069] After obtaining historical accident data and corresponding construction environment data, multiple linear regression analysis is used to identify environmental factors that are more relevant to accidents than expected, such as temperature data, humidity data, vibration data, sound data, and weather conditions, as high-risk factors, and the corresponding regression coefficients are used as impact values;

[0070] When in use, after identifying the areas and time periods where accidents frequently occur, multiple linear regression analysis is used to screen out high-risk factors from a number of environmental conditions, so that high-risk factors can be dealt with in a targeted manner during construction;

[0071] Step 202: Generate the influence degree Yxt of environmental conditions on the risk of construction accidents from the influence values ​​of several high-risk factors, under dimensionless conditions, according to the following formula:

[0072]

[0073] Where n is the number of high risk factors, Xt i is the impact value of the ith high risk factor, Xt a is the average value of the influence value, Xt b is the acceptable value of the impact value; weight coefficient: 0≤S1≤1, 0≤S2≤1, and S1+S2=1; the weight coefficient can be obtained by referring to the hierarchical analysis method;

[0074] If the obtained influence degree Yxt exceeds the preset influence threshold, it means that the external environmental conditions have a greater impact on the current construction conditions. At this time, a risk prediction instruction is issued to the outside;

[0075] When in use, after screening out several high-risk factors, the impact degree Yxt is analyzed and obtained. This can be used to judge the overall degree of influence of external environmental conditions on current construction safety. If the external environmental conditions have a greater impact on construction safety, the construction progress and construction tasks can be adjusted in construction scenarios with higher complexity to ensure construction safety.

[0076] Step 203: after receiving the risk prediction instruction, historical accident data and corresponding environmental data are collected as sample data, a neural convolutional network is trained by the sample data, a trained construction risk prediction model is obtained, and real-time construction environment data is used as input. The trained construction risk prediction model predicts the current construction risk, obtains corresponding risk prediction data, and issues a monitoring instruction to the outside;

[0077] When using, combine the contents in steps 201 to 203:

[0078] A construction risk prediction model is trained based on the historical data. After collecting real-time construction environment data, the trained construction risk prediction model can predict the current construction risks. After predicting possible construction risks, they can be dealt with in a timely manner to ensure construction safety.

[0079] Step 3: monitor the operation status data of intelligent construction equipment and the physical index data of construction workers in real time. If there is abnormal data, issue an abnormal prompt instruction, generate a prompt density Moj according to the distribution state of the abnormal prompt instruction, and issue an early warning instruction to the outside if the prompt density Moj exceeds the density threshold;

[0080] The step three includes the following contents:

[0081] Step 301: After obtaining the monitoring instruction, a monitoring device is installed in the construction area to collect worker activity data, operation status data of intelligent construction equipment, and physical index data of workers, and summarize and generate a real-time monitoring data set;

[0082] The abnormality recognition model is obtained by training sample data, and the real-time monitoring data is used as input. The trained abnormality recognition model identifies abnormal data. If there is abnormal data, such as excessive heart rate and high temperature of workers, the current situation is regarded as an abnormal situation, and an abnormal prompt instruction is issued to the outside;

[0083] When in use, in complex construction scenarios, data of construction workers in construction status is monitored and collected. After abnormal data identification, when the health indicators of construction workers become abnormal, abnormal prompts are issued to remind the construction workers, which can reduce the risk of accidents.

[0084] Step 302: pre-set a prompt cycle, and after obtaining prompt instructions several times continuously within the prompt cycle, record the time node when the abnormal prompt instruction is issued, and use it as an abnormal node; generate a prompt density Moj according to the distribution state of the abnormal node, where:

[0085]

[0086] Among them, w is the number of abnormal nodes, Z ij is the time distance from the i-th abnormal node to the j-th abnormal node, Z a is the time distance average;

[0087] Based on historical data and the degree of acceptance of abnormal situations, a density threshold is pre-set; if the density of abnormal situations exceeds the density threshold, it means that the physical and feedback status of the construction workers are poor in the current construction environment and may no longer be suitable for continuing construction. At this time, an early warning instruction is issued to the outside, such as through SMS, email or APP push;

[0088] When used, combine the contents in steps 301 and 302:

[0089] The prompt density Moj is constructed based on the continuous receipt of several abnormal prompt instructions. The current situation of issuing abnormal prompts to the outside is described according to the prompt density Moj to determine the feedback of the construction workers in the construction state. If the physical health indicators of the construction workers frequently become abnormal, it means that they are not suitable for construction at present, or the current construction strategy is insufficient. After adaptive adjustment, the construction risk is reduced.

[0090] Step 4: After receiving the warning instruction, the safety training program library outputs the corresponding construction safety training program. If a construction safety accident occurs in the construction area, the path planning algorithm plans a safe evacuation path for the construction workers and marks it on the electronic map;

[0091] The step 4 includes the following contents:

[0092] Step 401: Obtain corresponding risk prediction data, and perform feature engineering on the risk prediction data to obtain corresponding risk features; obtain several construction safety training programs through online retrieval or offline formulation and collection, including training courses on construction safety knowledge, emergency handling capabilities, etc., and generate a safety training program library after aggregation;

[0093] After receiving the early warning instruction, the safety training program library outputs the corresponding construction safety training program based on the correspondence between the risk characteristics and the training program;

[0094] When in use, after predicting the next construction risk and pausing the current construction status, the safety training program library outputs the corresponding construction safety training program based on the construction risk prediction data. By conducting construction safety training, it is possible to reduce construction risks when there may be construction risks.

[0095] Step 402: After a construction safety accident occurs in the construction area, determine the accident coverage area and use it as the accident area; construct an electronic map covering the construction area, and mark the accident area on the electronic map; after suspending the work of the intelligent construction equipment, obtain the corresponding location information and the location of obstacles in the construction area, and use the path planning algorithm to plan a safe evacuation path for the construction workers, and mark it on the electronic map;

[0096] When using, combine the contents in steps 401 and 402:

[0097] When a construction safety accident occurs in a complex construction scenario, planning a safe evacuation route for construction workers can speed up the evacuation of construction workers and ensure their safety.

[0098] Step 5: After collecting the construction feedback data, a construction feedback data set is constructed, and a feedback coefficient Kts is generated from the construction feedback data set. If the feedback coefficient Kts does not exceed the feedback threshold, the current construction management strategy is adjusted;

[0099] The step five includes the following contents:

[0100] Step 501: pre-set a construction safety observation period, collect construction feedback data in each sub-area during the observation period, such as accident rate, continuous safe construction time, hidden danger rectification rate, employee training completion rate, etc., and aggregate the collected feedback data to generate a construction feedback data set;

[0101] The feedback coefficient Kts is generated from the feedback data in the construction feedback data set, where the continuous safe construction time Ap and the training completion rate Pv are linearly normalized and the corresponding data values ​​are mapped to the interval [0,1] in the following manner:

[0102]

[0103] Among them, q is the number of sub-regions, Ap i is the continuous safe construction time of the ith sub-area, Ap a is the average duration of continuous safe construction, Pv i is the training completion rate in the ith sub-region, Pv a is the average value of the training completion rate, weight coefficient: 0≤K1≤1, 0≤K2≤1, and K1+K2=1; the weight coefficient can be obtained by referring to the hierarchical analysis method;

[0104] Step 502: pre-set a feedback threshold based on historical data and management expectations for construction feedback; if the feedback coefficient Kts does not exceed the feedback threshold, it means that the current construction strategy in the construction area may be insufficient. At this time, a strategy optimization instruction is issued to the outside; after receiving the strategy optimization instruction, the current construction management strategy is adjusted using a multi-objective optimization algorithm to obtain the adjusted construction management strategy;

[0105] When using, combine the contents in steps 501 and 502:

[0106] After completing the phased construction and dealing with possible construction risks, construction feedback data is collected and the feedback coefficient Kts is constructed. Based on the feedback coefficient Kts, a comprehensive judgment can be made on the current construction process and construction feedback. If the construction feedback effect exceeds expectations, it means that the current construction process is normal and in line with expectations. If the opposite is true, the current construction management strategy is adjusted to achieve construction management in complex scenarios and reduce construction risks and safety risks for construction personnel as a whole.

[0107] The Analytic Hierarchy Process (AHP) is a decision-making method that decomposes decision-related elements into levels such as goals, criteria, and plans, and conducts qualitative and quantitative analysis on this basis. It is particularly suitable for dealing with target systems with hierarchical and staggered evaluation indicators, and when the target value is difficult to describe quantitatively, the AHP is an effective decision-making tool.

[0108] The core of the hierarchical analysis method is to decompose the decision problem into multiple levels to form a hierarchical structure, which usually includes the target level, the criterion level, the sub-criterion level and the solution level. By solving the eigenvector of the judgment matrix, the priority weight of each element of each level to an element of the previous level is obtained, and the final weight of each alternative solution to the total goal is recursively merged by the weighted sum method, so as to find the optimal solution.

[0109] See also Figure 2 The present invention provides a construction safety management system based on big data, including:

[0110] The construction scene analysis unit collects the operation data of the intelligent construction equipment in the construction area under unconventional weather conditions, and generates the scene complexity Coj during construction from the operation data. If the scene complexity Coj exceeds the complexity threshold, the sensor network collects data in the construction area and generates a construction environment data set after aggregation;

[0111] The abnormal factor identification unit selects high-risk factors from a number of environmental conditions and generates the influence degree Yxt of environmental conditions on the risk of construction accidents based on the influence value of the high-risk factors. If the influence degree Yxt exceeds the influence threshold, the trained construction risk prediction model predicts the current construction risk and obtains the corresponding risk prediction data.

[0112] The abnormal warning unit monitors the operating status data of the intelligent construction equipment and the physical index data of the construction workers in real time. If there is abnormal data, it will issue an abnormal prompt instruction, and generate a prompt density Moj based on the distribution state of the abnormal prompt instruction. If the prompt density Moj exceeds the density threshold, it will issue a warning instruction to the outside.

[0113] After receiving the early warning instruction, the emergency handling unit will output the corresponding construction safety training plan from the safety training plan library. If a construction safety accident occurs in the construction area, the path planning algorithm will plan a safe evacuation path for the construction workers and mark it on the electronic map;

[0114] The feedback optimization unit collects the construction feedback data and builds a construction feedback data set, and generates a feedback coefficient Kts from the construction feedback data set. If the feedback coefficient Kts does not exceed the feedback threshold, the current construction management strategy is adjusted.

[0115] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0116] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0117] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0118] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only for some logical functions. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0119] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0120] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0121] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROM), random access memories (RAM), magnetic disks or optical disks.

[0122] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A construction safety management method based on big data, characterized by: include, Under unconventional weather conditions, the operation data of intelligent construction equipment is collected in the construction area, and the scene complexity during construction is generated from the operation data. , if the scene complexity When the complexity threshold is exceeded, the sensor network collects data in the construction area and generates a construction environment data set after aggregation; the scene complexity during construction is generated by the equipment operation data set. The way is as follows: ； in, and The number of time nodes and devices enabled during the duration of the storm, is the number of intelligent construction equipment enabled at the i-th time node, is the corresponding mean, Variance in the number of enabled smart construction equipment; is the covariance between precipitation and wind speed during the construction period, is the rainfall at the i-th time node, is the wind speed at the i-th time node, is the acceptable value of wind speed; High-risk factors are selected from a number of environmental factors, and the impact of environmental conditions on the risk of construction accidents is generated based on the impact values ​​of high-risk factors. , if the influence When the impact threshold is exceeded, the trained construction risk prediction model predicts the current construction risk and obtains the corresponding risk prediction data; the impact values ​​of several high-risk factors are used to generate the impact of environmental conditions on the risk of construction accidents. , under dimensionless conditions, according to the following formula: ； in, is the number of high risk factors, It is The impact value of high risk factors, is the average value of the influence value, is the acceptable value of the impact value; weight coefficient: , ,and ; If the influence obtained If the impact threshold is exceeded, a risk prediction instruction will be issued to the outside world; Monitor the operating status data of intelligent construction equipment and the physical index data of construction workers in real time. If there is abnormal data, issue abnormal prompt instructions and generate prompt density according to the distribution status of abnormal prompt instructions. , if the prompt density When the density threshold is exceeded, an early warning instruction is issued to the outside. After obtaining several prompt instructions continuously within the prompt cycle, the time node of the abnormal prompt instruction is recorded and used as the abnormal node. The prompt density is generated according to the distribution state of the abnormal node. ,in: ； in, is the number of abnormal nodes, It is The abnormal node to the The time distance of abnormal nodes, is the time distance average; After receiving the early warning instruction, the safety training program library outputs the construction safety training program. If a construction safety accident occurs in the construction area, a safe evacuation route is planned for the construction workers and marked on the electronic map; After collecting the construction feedback data, a construction feedback data set is constructed, and the feedback coefficient is generated from the construction feedback data set. , if the feedback coefficient If the feedback threshold is not exceeded, the current construction management strategy is adjusted, wherein the construction feedback data in each sub-area is collected during the observation period and a construction feedback data set is generated; Generate feedback coefficients from feedback data in the construction feedback data set , among which, the duration of continuous safe construction and training completion rate Perform linear normalization and map the corresponding data values ​​to the interval In the following way: ； in, is the number of sub-regions, It is The duration of continuous safe construction in each sub-area, is the average duration of continuous safe construction. It is The training completion rate in each sub-region, is the average value of training completion rate, weight coefficient: , ,and .

2. The construction safety management method based on big data according to claim 1 is characterized by: Collect construction data of intelligent construction equipment in the construction area, including the number, location, operation status and weather condition data of intelligent construction equipment, and generate equipment operation data set after aggregation; under dimensionless conditions, generate the scene complexity during construction from the equipment operation data set , if the scene complexity obtained is When the complexity threshold is exceeded, a sensor placement instruction is issued to the outside.

3. The construction safety management method based on big data according to claim 2 is characterized by: After receiving the sensor deployment instruction, the sensor network is deployed in the construction area to make the data collection frequency of the sensor network and the scene complexity The sensor network collects data in the construction area, and aggregates the collected data to generate a construction environment data set.

4. The construction safety management method based on big data according to claim 3 is characterized by: After collecting historical accident data in the construction area, data visualization tools are used to perform visual analysis of the historical accident data, and through data recognition and feature engineering, the areas and time periods where accidents frequently occur are identified; after obtaining the historical accident data and the corresponding construction environment data, multiple linear regression analysis is used to identify environmental conditions that are more correlated with accidents than expected as high-risk factors, and the corresponding regression coefficients are used as impact values.

5. The construction safety management method based on big data according to claim 4 is characterized by: After receiving the risk prediction instruction, historical accident data and corresponding environmental data are collected as sample data, and the neural convolution network is trained by the sample data to obtain the trained construction risk prediction model. The real-time construction environment data is used as input, and the trained construction risk prediction model is used to predict the current construction risk, obtain the corresponding risk prediction data, and issue monitoring instructions to the outside.

6. The construction safety management method based on big data according to claim 5 is characterized by: After receiving the monitoring instructions, the monitoring device is installed in the construction area to collect the worker activity data, the operation status data of the intelligent construction equipment and the worker's physical index data, and summarize them to generate a real-time monitoring data set; With real-time monitoring data as input, the trained anomaly recognition model identifies abnormal data. If abnormal data exists, the current situation is regarded as an abnormal situation and an abnormal prompt instruction is issued to the outside.

7. The construction safety management method based on big data according to claim 6 is characterized by: Obtain the corresponding risk prediction data, and perform feature engineering on the risk prediction data to obtain the corresponding risk features; Obtain several construction safety training plans, and output corresponding construction safety training plans from the safety training plan library based on the correspondence between risk characteristics and training plans; After determining the accident coverage, it is regarded as the accident area, and an electronic map covering the construction area is constructed. The accident area is marked on the electronic map, and the intelligent construction equipment is suspended to obtain the corresponding location information and the location of obstacles in the construction area.

8. A construction safety management system based on big data, applying the construction safety management method according to any one of claims 1 to 7, characterized in that: include, The construction scene analysis unit collects the operation data of intelligent construction equipment in the construction area under unconventional weather conditions, and generates the scene complexity during construction from the operation data , if the scene complexity When the complexity threshold is exceeded, the sensor network collects data in the construction area and aggregates them to generate a construction environment data set; The abnormal factor identification unit selects high-risk factors from a number of environmental conditions and generates the impact of environmental conditions on the risk of construction accidents based on the impact values ​​of high-risk factors. , if the influence When the impact threshold is exceeded, the trained construction risk prediction model predicts the current construction risk and obtains the corresponding risk prediction data; The abnormal warning unit monitors the operating status data of intelligent construction equipment and the physical index data of construction workers in real time. If there is abnormal data, it will issue an abnormal prompt instruction and generate a prompt density based on the distribution status of the abnormal prompt instruction. , if the prompt density When the density threshold is exceeded, an early warning command is issued to the outside world; After receiving the early warning instruction, the emergency handling unit will output the corresponding construction safety training plan from the safety training plan library. If a construction safety accident occurs in the construction area, the path planning algorithm will plan a safe evacuation path for the construction workers and mark it on the electronic map; Feedback optimization unit, which collects construction feedback data and builds a construction feedback data set, and generates feedback coefficients from the construction feedback data set , if the feedback coefficient Without exceeding the feedback threshold, the current construction management strategy is adjusted.

Images