An Internet of Things-based intelligent construction safety monitoring system and method

Abstract

The present invention discloses an intelligent construction safety monitoring system and method based on the Internet of Things, which relates to the technical field of construction safety monitoring, including a monitoring historical data acquisition module, a warning condition setting and planning module, a construction safety monitoring module, and a soil heaping abnormal warning module. The monitoring historical data acquisition module collects historical construction monitoring data and historical soil heaping data at the construction site. The warning condition setting and planning module preprocesses the collected data to screen out reference historical data, analyzes the reference historical data, and establishes a warning reference model for planning soil heaping abnormal warning conditions. The construction safety monitoring module conducts safety monitoring on the current construction environment and collects monitoring data. When the current monitoring data meets the soil heaping abnormal warning conditions, the soil heaping abnormal warning module sends a soil heaping abnormal alarm signal to the monitoring terminal, effectively reducing the construction safety risks under different foundation pit environments and improving construction safety.

Description

Technical Field

[0001] The present invention relates to the technical field of construction safety monitoring, and specifically to an intelligent construction safety monitoring system and method based on the Internet of Things. Background Technique

[0002] Performing safety monitoring on the construction site can effectively ensure safety during the construction process, prevent accidents and injuries. By monitoring each area of the construction site in real time, it can help to promptly detect abnormal situations at the construction site for timely warning, reminding relevant personnel to resolve abnormal situations in a timely manner, and further ensuring the smooth and safe progress of construction operations;

[0003] At the construction site, it is generally stipulated that no soil should be piled within one meter around the foundation pit. Otherwise, there will be a risk of landslide during construction. Monitoring the abnormal soil piling phenomenon around the foundation pit can effectively avoid the risk of landslide during construction. However, in actual situations, due to the influence of the construction environment, the safe distance for soil piling will change. For some construction sites, if soil is piled outside one meter around the foundation pit, there will also be construction safety risks. Therefore, the safe distance for soil piling needs to be adaptively adjusted according to environmental changes. In the prior art, most relevant technical personnel make the analysis and regulations on the safe distance for soil piling by themselves according to the construction site environment, without making an automated adaptive plan for the soil piling warning conditions according to the actual construction environment and promptly warning the abnormal soil piling phenomenon beside the foundation pit at the construction site, and cannot effectively reduce the construction safety risks under different foundation pit environments to improve construction safety. Summary of the Invention

[0004] The purpose of the present invention is to provide an intelligent construction safety monitoring system and method based on the Internet of Things to solve the problems raised in the prior art.

[0005] To achieve the above purpose, the present invention provides the following technical solution: An intelligent construction safety monitoring system based on the Internet of Things, the system includes a monitoring historical data acquisition module, a warning condition setting and planning module, a construction safety monitoring module, and a soil piling abnormal warning module;

[0006] Collect historical construction monitoring data and historical soil piling data at the construction site through the monitoring historical data acquisition module;

[0007] After preprocessing the collected data through the warning condition setting and planning module, screen out the reference historical data, analyze the reference historical data and establish a warning reference model for planning the abnormal soil piling warning conditions;

[0008] Monitor the safety of the current construction environment through the construction safety monitoring module and collect monitoring data;

[0009] When the current monitoring data meets the earth filling abnormal warning conditions, the earth filling abnormal warning module sends an earth filling abnormal alarm signal to the monitoring terminal.

[0010] Preferably, the monitoring historical data acquisition module includes a construction data acquisition unit and a historical earth filling data acquisition unit;

[0011] The construction data acquisition unit acquires the foundation pit trench depth data of past construction sites, and the foundation pit trench depth is detected by an inclinometer;

[0012] The historical earth filling data acquisition unit acquires the historical earth filling data around the foundation pits with different trench depths at the construction sites where no abnormal phenomena occurred in the past. The historical earth filling data refers to the distance data from the waste soil piled around the foundation pit to the edge of the foundation pit. The abnormal phenomenon refers to the foundation pit collapse phenomenon caused by improper earth filling distance around the foundation pit.

[0013] Preferably, the warning condition setting and planning module includes a critical distance analysis unit and a warning reference model establishment unit;

[0014] The critical distance analysis unit retrieves the historical earth filling data, analyzes the reference index of the historical earth filling data for the earth filling abnormal warning condition planning, screens out the reference historical earth filling data according to the reference index, and analyzes the critical earth filling distance around the foundation pits with different trench depths based on the reference historical earth filling data;

[0015] The warning reference model establishment unit combines the trench depth data and the critical earth filling distance data around the foundation pit corresponding to the trench depth into training data, and after fitting the training data, establishes a warning reference model for the earth filling abnormal warning condition planning.

[0016] Preferably, the construction safety monitoring module includes a foundation pit environment monitoring unit, an image recognition and detection unit, and a shortest distance analysis unit;

[0017] The foundation pit environment monitoring unit monitors the foundation pit environment of the current construction site: takes images of a circular area with the center of the foundation pit as the center and a radius of r. The circular area covers the foundation pit area, and at the same time uses an inclinometer to detect the foundation pit trench depth of the current construction site;

[0018] When the image recognition and detection unit monitors the occurrence of earth filling in the circular area, it uses OpenCV technology to perform image contour detection on the captured image to obtain the foundation pit contour and the contour of the piled waste soil;

[0019] Detect the shortest distance between the foundation pit contour and the contour of the dumped soil through the shortest distance analysis unit: By traversing all points on the foundation pit contour and the contour of the dumped soil and calculating the Euclidean distance between each pair of points, find the minimum value among all distances, and this minimum value is the shortest distance between the foundation pit contour and the contour of the dumped soil. Any point on the foundation pit contour and any point on the contour of the dumped soil form a pair of points. After converting the obtained shortest distance according to the scale, the actual soil stacking distance at the current construction site is obtained. For example, if the obtained shortest distance is q and the scale of the image is 1:b, the actual soil stacking distance p at the current construction site is calculated by p = q * b. The image scale is the scale of the image, and the historical soil stacking data is also obtained in the same way as the method for obtaining the actual soil pushing distance at the current construction site.

[0020] Preferably, the soil stacking abnormal warning module includes a monitoring data input unit, a current critical distance analysis unit, a data comparison unit, and a soil stacking abnormal alarm unit;

[0021] Input the monitored foundation pit trench depth data of the current construction site into the warning reference model through the monitoring data input unit to obtain model output data;

[0022] Use the model output data as the current critical soil stacking distance around the foundation pit through the current critical distance analysis unit;

[0023] Compare the actual soil stacking distance at the current construction site with the current critical soil stacking distance around the foundation pit through the data comparison unit and obtain the comparison result;

[0024] If the comparison result obtained through the soil stacking abnormal alarm unit is that the actual soil stacking distance at the current construction site is greater than or equal to the current critical soil stacking distance around the foundation pit, no soil stacking abnormal alarm is issued; if the comparison result is that the actual soil stacking distance at the current construction site is less than the current critical soil stacking distance around the foundation pit, then send a soil stacking abnormal alarm signal to the monitoring terminal to prompt to transfer the soil stacking position around the current foundation pit.

[0025] An Internet of Things-based construction safety intelligent monitoring method includes the following steps:

[0026] S1-1: Collect historical construction monitoring data and historical soil stacking data at the construction site;

[0027] S1-2: After preprocessing the collected data, screen out the reference historical data, analyze the reference historical data, and establish a warning reference model for planning soil stacking abnormal warning conditions;

[0028] S1-3: Conduct safety monitoring on the current construction environment and collect monitoring data;

[0029] S1-4: When the current monitoring data meets the soil pile abnormality warning condition, a soil pile abnormality alarm signal is sent to the monitoring terminal.

[0030] Preferably, in step S1-1: the depth set of foundation pits collected from previous construction sites is H={H1, H2, ...H i ,...H n}, n represents the number of foundation pits. The depths of the foundation pits in the set H are different. The distance data from the spoil piled around the foundation pits with different pit depths to the edge of the foundation pit are collected at the construction sites where no abnormal phenomena have occurred in the past. The pit depth H is retrieved from the data. i The distance set from the spoil piled around the foundation pit to the edge of the foundation pit is L={L1, L2, ...L m}, m means there are m construction sites with a pit depth of H i .

[0031] Preferably, in step S1-2: the distance value with the most occurrences in the set L is counted, and the distance value with the most occurrences appears q times in total. The pit depth H is calculated according to the following formula: i The reference index W for the distance data from the spoil piled around the foundation pit to the edge of the foundation pit i :

[0032] W i =q / m+1 / [∑ m j=1 (L j -∑ m j=1 (L j ) / m) 2 ] 1 / 2 ;

[0033] Among them, L j Indicates that the depth of the j-th foundation pit in the past is H i The distance from the spoil piled around the foundation pit to the edge of the foundation pit at the construction site is obtained, and the reference index set of the distance data from the spoil piled around the foundation pit to the edge of the foundation pit with a pit depth of the corresponding depth in the set H is W={W1, W2, ...W i ,...W n}, set the reference index threshold to (∑ n i=1 (W i )) / (2n / 3), compare W i and (∑ n i=1 (W i )) / (2n / 3): If W i ≥(∑ n i=1 (Wi )) / (2n / 3), select the distance data within set L as the reference historical soil stacking data, and obtain the pit depth as H i The critical soil stacking distance around the foundation pit with depth H is G i , G i = ∑ m j=1 (L j ) / m; If W i < (∑ n i=1 (W i )) / (2n / 3), the distance data within set L is not used as the reference historical soil stacking data. Compare the reference indices within set W with the reference index thresholds one by one, screen out the reference historical soil stacking data, and obtain the set of critical soil stacking distances around the foundation pits with different pit depths based on the screened reference historical soil stacking data as {G1, G2,... G k}, and the corresponding set of pit depths of the foundation pits is h = {h1, h2,... h k}, G1 represents the critical soil stacking distance around the foundation pit with a pit depth of h1 obtained based on the screened reference historical soil stacking data, k represents the number of data items within the screened set of critical soil stacking distances, and form the training data {(h1, G1), (h2, G2),... (h k , G k ). After performing linear fitting on the training data, establish an early warning reference model: y = θ1 * x + θ2, where θ1 and θ2 represent the fitting coefficients, x represents the variable representing the pit depth in the early warning reference model, and y represents the variable representing the critical soil stacking distance in the early warning reference model;

[0034] By collecting historical construction site data, the soil stacking phenomenon around the foundation pits at past construction sites is analyzed. The data of construction sites where no risk of collapse is caused by improper soil stacking distance is used as reference data. Using actual data as reference data for setting soil stacking anomaly warning conditions greatly improves the accuracy and reliability of setting soil stacking anomaly warning conditions. Considering that different depths of the foundation pit trenches will affect the stability of the foundation pit slopes, setting appropriate safe soil stacking distances for foundation pits with different slope stabilities helps to timely monitor and judge abnormal soil stacking phenomena. Before training the reference data, the reference data is screened. Data with large differences in soil stacking distance values in the historical soil stacking distance data, that is, data with relatively small reference indices, is screened out and then reorganized into training data for planning warning conditions. Considering that data with large differences has no reference value, the reliability of setting soil stacking anomaly warning conditions is further improved. After training the remaining data after screening, an early warning reference model is established. After substituting the monitoring data into the model, it is possible to judge whether the monitored phenomenon is an abnormal phenomenon based on the early warning reference model, realizing automatic adaptive planning of soil stacking warning conditions according to the actual construction environment.

[0035] Preferably, in step S1-3: Monitor the foundation pit environment of the current construction site: Take images of a circular area with the center of the foundation pit as the center and a radius of r. Use an inclinometer to detect that the depth of the foundation pit trench at the current construction site is D. When a soil stacking phenomenon is monitored in the circular area, use OpenCV technology to perform image contour detection on the captured images to obtain the contour of the foundation pit and the contour of the dumped soil. By traversing all points on the contour of the foundation pit and the contour of the dumped soil and calculating the Euclidean distance between each pair of points, find the minimum value among all distances. The minimum value is the shortest distance between the contour of the foundation pit and the contour of the dumped soil. Any point on the contour of the foundation pit and any point on the contour of the dumped soil form a pair of points. After converting the obtained shortest distance by the scale, the actual soil stacking distance at the current construction site is obtained, and the actual soil stacking distance at the current construction site is F.

[0036] Preferably, in step S1-4: Input the data of the depth of the foundation pit trench at the current construction site monitored into the early warning reference model: Let x = D, and the critical soil stacking distance around the current foundation pit is output as: θ1*D + θ2. Compare θ1*D + θ2 and F: If F ≥ θ1*D + θ2, no soil stacking anomaly alarm is made; if F < θ1*D + θ2, send a soil stacking anomaly alarm signal to the monitoring terminal to prompt to transfer the soil stacking position around the current foundation pit.

[0037] Substitute the actual monitored data of the current construction site, specifically the depth of the foundation pit trench, into the early warning reference model to obtain the safe soil stacking distance, that is, the critical soil stacking distance, for the foundation pit corresponding to the pit trench depth. When it is monitored that the actual soil stacking distance at the current construction site is less than the critical soil stacking distance, it is determined that the soil stacking phenomenon at the current construction site is abnormal, there is a risk of causing the foundation pit to collapse, and a soil stacking abnormal alarm signal is sent to the monitoring terminal and a prompt is given to transfer the soil stacking position, effectively reducing the construction safety risks in different foundation pit environments and improving construction safety.

[0038] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0039] The present invention analyzes the soil stacking phenomenon around the foundation pits of past construction sites by collecting historical construction site data, and uses the data of construction sites where there is no risk of collapse due to improper soil stacking distance as reference data. Using actual data as reference data for setting soil stacking abnormal early warning conditions greatly improves the accuracy and reliability of setting soil stacking abnormal early warning conditions. Considering that different depths of the foundation pit trench will affect the stability of the foundation pit slope, and setting appropriate safe soil stacking distances for foundation pits with different slope stabilities is conducive to helping to monitor and judge abnormal soil stacking phenomena in a timely manner. Before training the reference data, the reference data is screened and processed. The data with a relatively large difference in soil stacking distance values in the historical soil stacking distance data, that is, the data with a relatively small reference index, is screened out and then combined to form training data for planning early warning conditions. Considering that the data with a relatively large difference does not have reference value, the reliability of setting soil stacking abnormal early warning conditions is further improved. After training the remaining data after screening and processing, an early warning reference model is established. After substituting the monitored data into the model, it is possible to judge whether the monitored phenomenon is an abnormal phenomenon based on the early warning reference model, realizing the automatic adaptive planning of soil stacking early warning conditions according to the actual construction environment;

[0040] Substitute the actual monitored data of the current construction site, specifically the depth of the foundation pit trench, into the early warning reference model to obtain the safe soil stacking distance, that is, the critical soil stacking distance, for the foundation pit corresponding to the pit trench depth. When it is monitored that the actual soil stacking distance at the current construction site is less than the critical soil stacking distance, it is determined that the soil stacking phenomenon at the current construction site is abnormal, there is a risk of causing the foundation pit to collapse, and a soil stacking abnormal alarm signal is sent to the monitoring terminal and a prompt is given to transfer the soil stacking position, effectively reducing the construction safety risks in different foundation pit environments and improving construction safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 It is a schematic structural diagram of an intelligent construction safety monitoring system based on the Internet of Things according to the present invention;

[0042] Figure 2It is a schematic flow chart of a construction safety intelligent monitoring method based on the Internet of Things according to the present invention. Specific embodiments

[0043] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative work shall fall within the protection scope of the present invention. Embodiment

[0044] As Figure 1 shown, this embodiment provides a construction safety intelligent monitoring system based on the Internet of Things. The system includes: a monitoring historical data collection module, a warning condition setting and planning module, a construction safety monitoring module, and a soil stacking abnormal warning module; the historical construction monitoring data and the historical soil stacking data of the construction site are collected through the monitoring historical data collection module; the collected data is preprocessed through the warning condition setting and planning module to screen out the reference historical data, the reference historical data is analyzed, and a warning reference model is established for planning the soil stacking abnormal warning conditions; the current construction environment is safely monitored and the monitoring data is collected through the construction safety monitoring module; when the current monitoring data meets the soil stacking abnormal warning conditions, a soil stacking abnormal alarm signal is sent to the monitoring terminal through the soil stacking abnormal warning module.

[0045] The monitoring historical data collection module includes a construction data collection unit and a historical soil stacking data collection unit; the foundation pit trench depth data of the past construction site is collected through the construction data collection unit, and the foundation pit trench depth is detected by an inclinometer; the historical soil stacking data around the foundation pits with different trench depths in the past construction sites without abnormal phenomena is collected through the historical soil stacking data collection unit. The historical soil stacking data refers to the distance data from the waste soil piled around the foundation pit to the edge of the foundation pit, and the abnormal phenomenon refers to the foundation pit collapse phenomenon caused by the improper distance of the soil stacking around the foundation pit.

[0046] The warning condition setting and planning module includes a critical distance analysis unit and a warning reference model establishment unit; the historical soil stacking data is retrieved through the critical distance analysis unit, the reference index of the historical soil stacking data for the planning of the soil stacking abnormal warning conditions is analyzed, the reference historical soil stacking data is screened out according to the reference index, and the critical soil stacking distance around the foundation pits with different trench depths is analyzed according to the reference historical soil stacking data; the warning reference model establishment unit combines the trench depth data and the critical soil stacking distance data around the foundation pit corresponding to the trench depth into training data, and after fitting the training data, a warning reference model for planning the soil stacking abnormal warning conditions is established.

[0047] The construction safety monitoring module includes a foundation pit environment monitoring unit, an image recognition and detection unit, and a shortest distance analysis unit. The foundation pit environment monitoring unit monitors the foundation pit environment at the current construction site: takes images of a circular area with the center of the foundation pit as the center and a radius of r, where the circular area covers the foundation pit area, and at the same time uses an inclinometer to detect the depth of the foundation pit trench at the current construction site. When the image recognition and detection unit monitors the occurrence of soil stacking in the circular area, it uses OpenCV technology to perform image contour detection on the captured images to obtain the contour of the foundation pit and the contour of the stacked soil. The shortest distance analysis unit detects the shortest distance between the contour of the foundation pit and the contour of the stacked soil: by traversing all points on the contour of the foundation pit and the contour of the stacked soil and calculating the Euclidean distance between each pair of points, the minimum value among all distances is found, and this minimum value is the shortest distance between the contour of the foundation pit and the contour of the stacked soil. Any point on the contour of the foundation pit and any point on the contour of the stacked soil form a pair of points. The obtained shortest distance is converted by the scale to obtain the actual soil stacking distance at the current construction site. For example, if the obtained shortest distance is q and the scale of the image is 1:b, the actual soil stacking distance at the current construction site is calculated as p = q * b. The image scale is the scale of the image, and the historical soil stacking data is also obtained in the same way as the method for obtaining the actual soil stacking distance at the current construction site.

[0048] The soil stacking abnormal warning module includes a monitoring data input unit, a current critical distance analysis unit, a data comparison unit, and a soil stacking abnormal alarm unit. The monitoring data input unit inputs the data of the depth of the foundation pit trench at the current construction site obtained by monitoring into the warning reference model to obtain the model output data. The current critical distance analysis unit uses the model output data as the critical soil stacking distance around the current foundation pit. The data comparison unit compares the actual soil stacking distance at the current construction site with the critical soil stacking distance around the current foundation pit and obtains the comparison result. The soil stacking abnormal alarm unit: if the comparison result is that the actual soil stacking distance at the current construction site is greater than or equal to the critical soil stacking distance around the current foundation pit, no soil stacking abnormal alarm is issued; if the comparison result is that the actual soil stacking distance at the current construction site is less than the critical soil stacking distance around the current foundation pit, a soil stacking abnormal alarm signal is sent to the monitoring terminal to prompt the transfer of the soil stacking position around the current foundation pit. Embodiment

[0049] As Figure 2 shown, this embodiment provides an Internet of Things-based construction safety intelligent monitoring method, which is implemented based on the intelligent monitoring system in the embodiment, and specifically includes the following steps:

[0050] S1-1: Collect historical construction monitoring data and historical soil stacking data at the construction site: In step S1-1: The set of foundation pit depths at the previous construction site is collected as H = {H1, H2,...H i ,...H n}, where n represents the number of foundation pits, and the depths of the foundation pits in the set H are all different. Collect the distances from the waste soil piled around the foundation pits with different depths to the edge of the foundation pit at the construction sites where no abnormal phenomena occurred in the past. Retrieve the set of distances from the waste soil piled around the foundation pit with a depth of H i to the edge of the foundation pit as L = {L1, L2,...L m}, where m represents that there were m construction sites with a foundation pit depth of H i in the past;

[0051] S1-2: After preprocessing the collected data, screen out the reference historical data, analyze the reference historical data, and establish an early warning reference model for planning soil stacking abnormal warning conditions: Statistically, the distance value that appears most frequently in the set L is obtained. The distance value that appears most frequently appears q times in total. Calculate the reference index W i of the distance data from the waste soil piled around the foundation pit with a depth of H i to the edge of the foundation pit according to the following formula:

[0052] W i = q / m + 1 / [(∑ m j=1 (L j - ∑ m j=1 (L j ) / m) 2 ) / m] 1 / 2 ;

[0053] Among them, L j represents the distance from the waste soil piled around the foundation pit at the construction site of the jth foundation pit with a depth of H i to the edge of the foundation pit. The set of reference indices of the distance data from the waste soil piled around the foundation pits corresponding to the depths in the set H to the edge of the foundation pit is W = {W1, W2,...W i ,...W n}. Set the reference index threshold as (∑ n i=1 (W i )) / (2n / 3), and compare W i and (∑ n i=1 (W i )) / (2n / 3): If W i ≥ (∑ n i=1(W i )) / (2n / 3), select the distance data within the set L as the reference historical soil stacking data, and obtain the pit depth as H i The critical soil stacking distance around the foundation pit with is G i , G i = ∑ m j=1 (L j ) / m; If W i < (∑ n i=1 (W i )) / (2n / 3), the distance data within the set L is not used as the reference historical soil stacking data. Compare the reference indices within the set W with the reference index thresholds one by one, screen out the reference historical soil stacking data, and obtain the set of critical soil stacking distances around the foundation pits with different pit depths based on the screened reference historical soil stacking data as {G1, G2,... G k}; the corresponding set of pit depths of the foundation pits is h = {h1, h2,... h k}; G1 represents the critical soil stacking distance around the foundation pit with a pit depth of h1 obtained based on the screened reference historical soil stacking data, k represents the number of data items within the screened set of critical soil stacking distances, and form the training data {(h1, G1), (h2, G2),... (h k , G k )}; after linearly fitting the training data, establish an early warning reference model: y = θ1 * x + θ2, where θ1 and θ2 represent the fitting coefficients, x represents the variable representing the pit depth in the early warning reference model, and y represents the variable representing the critical soil stacking distance in the early warning reference model;

[0054] For example: Collect the distance data from the waste soil piled around the foundation pits with different pit depths to the edge of the foundation pit at the construction site where no abnormal phenomena have occurred in the past. Retrieve the set of distances from the waste soil piled around the foundation pit with a pit depth of 5 meters to the edge of the foundation pit as L = {L1, L2, L3, L4, L5, L6, L7, L8, L9, L 10} = {1.8, 1.5, 2.0, 2.2, 2.2, 1.9, 1.7, 2.2, 2.2, 2.2}, unit: meter. Statistically, the distance value that appears most frequently within the set L is 2.2, and it is obtained that the distance value that appears most frequently appears 5 times. Calculate the reference index of the distance data from the waste soil piled around the foundation pit with a pit depth of 5 meters to the edge of the foundation pit as: W i ≈ 4.62. Set the reference index threshold as 3.22, 4.62 > 3.22, select the distance data within the set L as the reference historical soil stacking data, and obtain the critical soil stacking distance around the foundation pit with a pit depth of 5 meters as G i , Gi = ∑ m j=1 (L j ) / m = 1.99, and the data (5, 1.99) is taken as one piece of training data in the training dataset.

[0055] S1 - 3: Conduct safety monitoring on the current construction environment and collect monitoring data: Monitor the foundation pit environment at the current construction site: Take images of a circular area with the center of the foundation pit as the center and a radius of r. Use an inclinometer to detect that the depth of the foundation pit trench at the current construction site is D. When it is monitored that there is soil stacking in the circular area, use OpenCV technology to perform image contour detection on the captured images, obtain the contour of the foundation pit and the contour of the stacked soil. By traversing all points on the contour of the foundation pit and the contour of the stacked soil and calculating the Euclidean distance between each pair of points, find the minimum value among all distances. The minimum value is the shortest distance between the contour of the foundation pit and the contour of the stacked soil. Any point on the contour of the foundation pit and any point on the contour of the stacked soil form a pair of points. After converting the obtained shortest distance through a scale, obtain the actual soil stacking distance at the current construction site, and the actual soil stacking distance at the current construction site is F;

[0056] S1 - 4: Send a soil stacking abnormal alarm signal to the monitoring terminal when the current monitoring data meets the soil stacking abnormal warning condition: Input the depth data of the foundation pit trench at the current construction site obtained from the monitoring into the warning reference model: Let x = D, and the output is the critical soil stacking distance around the current foundation pit: θ1 * D + θ2. Compare θ1 * D + θ2 and F: If F ≥ θ1 * D + θ2, no soil stacking abnormal alarm is made; if F < θ1 * D + θ2, send a soil stacking abnormal alarm signal to the monitoring terminal, prompting to transfer the soil stacking position around the current foundation pit.

[0057] For those skilled in the art, it is obvious that the present invention is not limited to the details of the above - mentioned exemplary embodiments, and can be implemented in other specific forms without departing from the spirit or basic characteristics of the present invention. Therefore, from any point of view, the embodiments should be regarded as exemplary and non - restrictive. The scope of the present invention is defined by the appended claims rather than the above description. Therefore, all changes falling within the meaning and scope of the equivalent elements of the claims are intended to be embraced within the present invention. Any reference signs in the claims should not be regarded as limiting the claimed rights.

Claims

1. A construction safety intelligent monitoring method based on the Internet of Things, characterized by: The following steps are involved: S1-1: Collect historical construction monitoring data and historical soil accumulation data of the construction site; S1-2: Pre-process the collected data and filter out reference historical data, analyze the reference historical data and establish an early warning reference model for planning abnormal soil pile warning conditions; S1-3: Conduct safety monitoring of the current construction environment and collect monitoring data; S1-4: sending a soil pile abnormality alarm signal to the monitoring terminal when the current monitoring data meets the soil pile abnormality warning condition; In step S1-1: the depth set of foundation pits collected from previous construction sites is H={H1, H2, ...H i ,...H n }, n represents the number of foundation pits. The distance data from the spoil piled around the foundation pits with different pit depths to the edge of the foundation pit are collected at construction sites where no abnormal phenomena have occurred in the past. The pit depth H is retrieved from the data. i The distance set from the spoil piled around the foundation pit to the edge of the foundation pit is L={L1, L2, ...L m }, m means there are m construction sites with a pit depth of H i ; In step S1-2: the distance value with the most occurrences in the set L is counted, and the distance value with the most occurrences appears q times in total. The pit depth H is calculated according to the following formula: i The reference index W for the distance data from the spoil piled around the foundation pit to the edge of the foundation pit i : W i =q / m+1 / [∑ m j=1 (L j -∑ m j=1 (L j ) / m) 2 ] 1 / 2 ; Among them, L j Indicates that the depth of the j-th foundation pit in the past is H i The distance from the spoil piled around the foundation pit to the edge of the foundation pit at the construction site is obtained, and the reference index set of the distance data from the spoil piled around the foundation pit to the edge of the foundation pit with a pit depth of the corresponding depth in the set H is W={W1, W2, ...W i ,...W n }, set the reference index threshold to (∑ n i=1 (W i )) / (2n / 3), compare W i and (∑ n i=1 (W i )) / (2n / 3): If W i ≥(∑ n i=1 (W i )) / (2n / 3), select the distance data in the set L as the reference historical soil data, and get the pit depth H i The critical distance of soil pile around the foundation pit is G i , G i =∑ m j=1 (L j ) / m; if W i <(∑ n i=1 (W i )) / (2n / 3), the distance data in the set L are not used as the reference historical soil filling data, and the reference indexes in the set W are compared with the reference index thresholds one by one to screen out the reference historical soil filling data. The critical soil filling distance set around the foundation pit with different pit depths obtained based on the screened reference historical soil filling data is {G1, G2, ...G k }, the corresponding pit depth set is h={h1, h2, ...h k }, G1 represents the critical soil stacking distance around the foundation pit with a pit depth of h1 obtained based on the filtered reference historical soil stacking data, and k represents the number of data items in the filtered critical soil stacking distance set, which constitutes the training data {(h1, G1), (h2, G2), ... (h k , G k )}, a warning reference model is established after performing a straight-line fitting on the training data: y=θ1*x+θ2, where θ1 and θ2 represent fitting coefficients, x represents the variable representing the pit depth in the warning reference model, and y represents the variable representing the critical soil pile distance in the warning reference model.

2. The method for intelligent construction safety monitoring based on the Internet of Things according to claim 1, characterized in that: In step S1-3: the foundation pit environment of the current construction site is monitored: an image of a circular area with the center of the foundation pit as the center and a radius of r is captured, and the depth of the foundation pit groove of the current construction site is detected by using an inclinometer as D. When soil piling is detected in the circular area, the captured image is subjected to image contour detection using OpenCV technology to obtain the foundation pit contour and the contour of the piled spoil. By traversing all points on the foundation pit contour and the contour of the piled spoil and calculating the Euclidean distance between each pair of points, the minimum value of all distances is found. The minimum value is the shortest distance between the foundation pit contour and the contour of the piled spoil. Any point on the foundation pit contour and any point on the contour of the piled spoil form a pair of points. The obtained shortest distance is converted to a scale to obtain the actual soil piling distance of the current construction site, and the actual soil piling distance of the current construction site is obtained as F.

3. The method for intelligent construction safety monitoring based on the Internet of Things according to claim 2, characterized in that: In step S1-4: the foundation pit depth data of the current construction site obtained by monitoring is input into the early warning reference model: let x=D, output the critical soil pile distance around the current foundation pit as: θ1*D+θ2, compare θ1*D+θ2 and F: if F≥θ1*D+θ2, no soil pile abnormality alarm is issued; if F<θ1*D+θ2, send a soil pile abnormality alarm signal to the monitoring terminal, prompting the soil pile position around the current foundation pit to be moved.

4. An Internet of Things-based construction safety intelligent monitoring system, applied to the Internet of Things-based construction safety intelligent monitoring method according to claim 1, characterized in that: It includes monitoring historical data collection module, warning condition setting planning module, construction safety monitoring module and soil pile abnormality warning module; Collect historical construction monitoring data and historical soil pile data of the construction site through the monitoring historical data collection module; The warning condition setting planning module pre-processes the collected data and filters out reference historical data, analyzes the reference historical data and establishes a warning reference model for planning abnormal soil pile warning conditions; The construction safety monitoring module is used to monitor the current construction environment and collect monitoring data; The soil pile abnormality warning module sends a soil pile abnormality alarm signal to the monitoring terminal when the current monitoring data meets the soil pile abnormality warning condition.

5. The construction safety intelligent monitoring system based on the Internet of Things according to claim 4 is characterized by: The monitoring history data acquisition module includes a construction data acquisition unit and a historical soil pile data acquisition unit; The construction data acquisition unit collects foundation pit depth data of previous construction sites, where the foundation pit depth is measured by an inclinometer; The historical soil pile data collection unit collects historical soil pile data around foundation pits with different pit depths at construction sites where no abnormal phenomena have occurred in the past. The historical soil pile data refers to the distance data from the waste soil piled around the foundation pit to the edge of the foundation pit.

6. The construction safety intelligent monitoring system based on the Internet of Things according to claim 5 is characterized by: The warning condition setting planning module includes a critical distance analysis unit and a warning reference model establishment unit; Retrieving historical soil pile data through the critical distance analysis unit, analyzing the reference index of the historical soil pile data for soil pile abnormality warning condition planning, screening reference historical soil pile data based on the reference index, and analyzing the critical soil pile distances around foundation pits with different pit depths based on the reference historical soil pile data; The warning reference model establishment unit combines the pit depth data and the critical soil pile distance data around the foundation pit corresponding to the pit depth into training data, and establishes a warning reference model for soil pile abnormality warning condition planning after fitting the training data.

7. The construction safety intelligent monitoring system based on the Internet of Things according to claim 6 is characterized by: The construction safety monitoring module includes a foundation pit environment monitoring unit, an image recognition detection unit and a shortest distance analysis unit; The foundation pit environment monitoring unit monitors the foundation pit environment of the current construction site by capturing an image of a circular area with a radius r and a center of the foundation pit, and simultaneously detecting the depth of the foundation pit at the current construction site using an inclinometer; When the image recognition detection unit monitors the occurrence of soil piling in the circular area, it uses OpenCV technology to perform image contour detection on the captured image to obtain the outline of the foundation pit and the outline of the piled soil; The shortest distance between the foundation pit outline and the outline of the piled spoil is detected by the shortest distance analysis unit: by traversing all points on the foundation pit outline and the outline of the piled spoil and calculating the Euclidean distance between each pair of points, the minimum value of all distances is found, and the minimum value is the shortest distance between the foundation pit outline and the outline of the piled spoil. Any point on the foundation pit outline and any point on the outline of the piled spoil form a pair of points. The obtained shortest distance is converted to a scale to obtain the actual soil pile distance of the current construction site.

8. The construction safety intelligent monitoring system based on the Internet of Things according to claim 7 is characterized by: The soil pile abnormality early warning module includes a monitoring data input unit, a current critical distance analysis unit, a data comparison unit and a soil pile abnormality alarm unit; Inputting the foundation pit depth data of the current construction site obtained through monitoring into the early warning reference model through the monitoring data input unit to obtain model output data; The model output data is used as the critical earth pile distance around the current foundation pit by the current critical distance analysis unit; The data comparison unit compares the actual earth pile distance of the current construction site with the critical earth pile distance around the current foundation pit and obtains a comparison result; If the comparison result of the soil pile abnormality alarm unit is that the actual soil pile distance of the current construction site is greater than or equal to the critical soil pile distance around the current foundation pit, no soil pile abnormality alarm is issued; if the comparison result is that the actual soil pile distance of the current construction site is less than the critical soil pile distance around the current foundation pit, a soil pile abnormality alarm signal is sent to the monitoring terminal, prompting the monitoring terminal to move the soil pile position around the current foundation pit.

Images