Construction safety supervision method and system based on data analysis

Abstract

This invention provides a data analysis-based construction safety supervision method and system, belonging to the technical field of construction safety management. The data analysis-based construction safety supervision method includes acquiring multi-source safety data during the construction process; preprocessing the multi-source safety data to construct a construction safety database; calculating the comprehensive risk value of each stage based on the construction safety database, and identifying high-risk stages based on the comprehensive risk value; predicting the development trend of the high-risk stages based on the comprehensive risk value; and pushing early warning information based on the comprehensive risk value and the development trend of the high-risk stages. The data analysis-based construction safety supervision method provided by this invention, by acquiring multi-source safety data during the construction process, can break through the limitations of single-dimensional data collection in traditional supervision methods, comprehensively covering various safety-related information in the construction scenario, and eliminating blind spots in safety supervision due to data gaps.

Description

Technical Field

[0001] The present invention belongs to the technical field of construction safety management. More specifically, it relates to a construction safety supervision method and system based on data analysis. Background Art

[0002] Construction sites are characterized by complex working environments, large personnel flow, and intensive large-scale equipment. During the construction process, various safety risks are intertwined and superimposed. Once a safety accident occurs, it will directly threaten the lives of construction workers, cause significant property losses and adverse social impacts. Therefore, construction safety supervision is the core link to ensure the smooth progress of the project. Currently, safety supervision in the industry mainly relies on traditional methods such as manual inspections, fixed-point video monitoring, and paper ledger records. Although such methods have formed a certain application basis in supervision practice, with the expansion of project scale and the increase in complexity, their limitations have become increasingly prominent and are difficult to meet the precise requirements of modern construction safety control.

[0003] The safety supervision in the prior art has the following defects: It is mostly limited to the collection and management of static basic information such as personnel attendance, equipment registration, and paper records of potential hazards, and fails to construct a full-dimensional data collection system for the dynamic characteristics of the construction scenario. This limitation in data collection directly leads to incomplete coverage of safety supervision elements, inability to completely capture dynamic risk sources during the construction process, difficulty in timely identifying high-risk links and predicting risks. Summary of the Invention

[0004] The purpose of the present invention is to provide a construction safety supervision method and system based on data analysis, aiming to solve the problems in the prior art that the coverage of safety supervision elements is incomplete, unable to completely capture dynamic risk sources during the construction process, difficult to timely identify high-risk links and predict risks.

[0005] To achieve the above object, the technical solution adopted by the present invention is: In the first aspect, a construction safety supervision method based on data analysis is provided, including: Obtain multi-source safety data during the construction process; Preprocess the multi-source safety data to construct a construction safety database; Based on the construction safety database, calculate the comprehensive risk value of each link, and locate high-risk links according to the comprehensive risk value; Predict the development trend of the high-risk links according to the comprehensive risk value; Push warning information according to the comprehensive risk value and the development trend of the high-risk links.

[0006] In one possible implementation, the multi-source safety data includes: personnel data, equipment data, environmental data, and operational behavior data; the personnel data includes personnel compliance rate, the equipment data includes equipment health index, the environmental data includes environmental early warning index, and the operational behavior data includes historical risk frequency.

[0007] In one possible implementation, the preprocessing of the multi-source safety data to construct a construction safety database includes: The multi-source security data is cleaned by filtering out noise and removing redundant and erroneous data to obtain cleaned data. The heterogeneity of the cleaned data is eliminated and fused to construct a multi-factor associated fused dataset; The fused dataset is standardized and the data field format is unified to construct a construction safety database and a standardized data dictionary.

[0008] In one possible implementation, eliminating the heterogeneity of the cleaned data and fusing it to construct a multi-factor associated fused dataset includes: The timestamp format of the cleaned data is standardized, and a unique identifier is established for each type of data. Using timestamps as the association key, different types of data at the same time are associated and matched to generate a fused wide table that includes multiple elements; Vertical fusion of multi-source data for the same security element is performed, and a single feature value is calculated after fusion to obtain a fused dataset with multi-element association.

[0009] In one possible implementation, the step of calculating the comprehensive risk value of each stage based on the construction safety database, and identifying high-risk stages based on the comprehensive risk value, includes: Construct the input feature set; Build a risk identification model; The comprehensive risk value of each stage is calculated and output based on the input feature set and the risk identification model. High-risk stages are identified based on the comprehensive risk value of each stage.

[0010] In one possible implementation, constructing the input feature set includes: The personnel compliance rate, the equipment health index, the environmental early warning index, and the historical risk frequency are used as core input features; Calculate the correlation between each feature and the risk event label; The input feature set is constructed by filtering features whose correlation degree is not lower than the correlation degree threshold.

[0011] In a possible implementation manner, predicting the development trend of the high-risk link according to the comprehensive risk value includes: Performing smoothing processing on the comprehensive risk values of each link; Constructing a risk trend prediction model; Analyzing the time evolution law of the comprehensive risk values of each link by using the risk trend prediction model to predict the development trend of the high-risk link.

[0012] In a possible implementation manner, pushing warning information according to the comprehensive risk value and the development trend of the high-risk link includes: Establishing a warning level system and a hierarchical pushing mechanism; Determining the warning level and the pushing strategy according to the comprehensive risk value and the development trend of the high-risk link, and giving a warning.

[0013] In a possible implementation manner, the warning level system includes level one, level two, level three and level four.

[0014] In a second aspect, a construction safety supervision system based on data analysis is provided, which is applied to the construction safety supervision method based on data analysis as described in the first aspect, and includes: A data acquisition unit, configured to acquire multi-source safety data during the construction process; A construction safety database construction unit, configured to preprocess the multi-source safety data and construct a construction safety database; A comprehensive risk value calculation and high-risk link positioning unit, configured to calculate the comprehensive risk value of each link based on the construction safety database, and locate the high-risk link according to the comprehensive risk value; A high-risk link development trend prediction unit, configured to predict the development trend of the high-risk link according to the comprehensive risk value; A warning unit, configured to push warning information according to the comprehensive risk value and the development trend of the high-risk link.

[0015] The beneficial effect of the construction safety supervision method based on data analysis provided by the present invention is that: compared with the prior art, the construction safety supervision method based on data analysis of the present invention can break through the limitation of single data collection dimension under the traditional supervision method by acquiring multi-source safety data during the construction process, comprehensively cover various safety-related information in the construction scenario, and no longer rely only on static basic information, so as to more completely capture the dynamic risk sources during the construction process. The acquisition of multi-source data provides a rich and comprehensive basis for subsequent safety analysis, making the safety supervision no longer have blind spots due to data missing.

[0016] Preprocessing multi-source safety data and constructing a construction safety database can effectively solve the problem of data disorder from different sources, filtering out noise, redundancy, and errors, while eliminating data heterogeneity and achieving fusion, and unifying data field formats. This process transforms the originally scattered and irregular data into standardized, orderly, and correlated data, laying a reliable data foundation for the subsequent accurate calculation of comprehensive risk values ​​and avoiding the impact of data quality issues on risk assessment results.

[0017] By calculating the comprehensive risk value of each stage based on a construction safety database and identifying high-risk stages accordingly, abstract safety conditions can be transformed into quantifiable indicators. By comprehensively considering various safety factors reflected in multi-source data, the risk level of each construction stage can be accurately assessed, thereby precisely identifying stages with high safety hazards. This changes the traditional regulatory situation where it is difficult to accurately determine the key risks, allowing regulators to clearly define their regulatory focus and avoid blind inspections.

[0018] By predicting the development trend of high-risk segments based on comprehensive risk values, it is possible to anticipate the direction of risk changes and potential safety situations in these segments. This goes beyond simply identifying current risks; by analyzing historical and current data, it allows for the prediction of future risk evolution, shifting safety supervision from a reactive to a proactive approach. This buys time for early intervention and reduces the likelihood of safety accidents.

[0019] By sending early warning information based on comprehensive risk values ​​and the development trends of high-risk areas, regulatory personnel can obtain timely alerts of critical safety risks. The early warning information is based on quantitative risk assessment and trend prediction, making it targeted and timely. This allows regulatory personnel to react quickly to the warning content, take corresponding preventative measures, effectively curb the transformation of high-risk areas into safety accidents, protect the lives and property of construction workers, and promote the smooth progress of the project. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 A schematic diagram illustrating the main steps of the data analysis-based construction safety supervision method provided in this embodiment of the invention; Figure 2 This is a flowchart illustrating a data analysis-based construction safety supervision method provided in an embodiment of the present invention. Detailed Implementation

[0022] To make the technical problems, solutions, and beneficial effects of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0023] Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without inventive effort are within the scope of protection of this application.

[0024] It should be further noted that the accompanying drawings and embodiments of the present invention mainly describe the concept of the present invention. Based on this concept, some specific forms and arrangements of connection relationships, positional relationships, power mechanisms, power supply systems, hydraulic systems and control systems may not be fully described. However, under the premise that those skilled in the art understand the concept of the present invention, they can implement the above-mentioned specific forms and arrangements in a well-known manner.

[0025] When a component is referred to as being "fixed to" or "set on" another component, it can be directly on or indirectly on that other component. When a component is referred to as being "connected to" another component, it can be directly connected to or indirectly connected to that other component.

[0026] In the description of this invention, "a plurality of" means two or more, and "several" means one or more, unless otherwise explicitly specified.

[0027] The directional terms "inner" and "outer" refer to the inner and outer contours relative to the outline of each component itself; the term "length"... The orientations or positional relationships indicated by terms such as “width,” “up,” “down,” “front,” “back,” “left,” “right,” “vertical,” “horizontal,” “top,” “bottom,” “inner,” and “outer” are based on the orientations or positional relationships shown in the accompanying drawings and are only for the purpose of facilitating the description of the present invention and simplifying the description. They are not intended to indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as limiting the present invention.

[0028] For ease of description, spatial relative terms such as "above," "on top of," "on the upper surface of," and "above" are used here to describe the spatial positional relationship between a device or feature and other devices or features, as shown in the figure. It should be understood that spatial relative terms are intended to... The invention includes different orientations of the device in use or operation, in addition to those described in the figures. For example, if a device in the figures is inverted, a device described as "above" or "on top of" other devices or structures will be positioned as "below" or "under" other devices or structures. Thus, the exemplary term "above" can include both "above" and "below". The device may also be positioned in other different ways, and the spatial relative descriptions used herein are interpreted accordingly. The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of the invention, "a plurality of" means two or more, and "a number" means one or more, unless otherwise explicitly specified.

[0029] Reference Figures 1 to 2 The present invention will now describe the construction safety supervision method and system based on data analysis.

[0030] Firstly, a data-driven construction safety supervision method is provided, including: S100. Acquire multi-source safety data during the construction process.

[0031] In one possible implementation, the multi-source safety data in step S100 includes: personnel data, equipment data, environmental data, and operational behavior data; personnel data includes personnel compliance rate, equipment data includes equipment health index, environmental data includes environmental early warning index, and operational behavior data includes historical risk frequency.

[0032] Specifically, personnel data also includes personnel ID, attendance time, real-time location coordinates, heart rate, and body temperature. Personnel data can be obtained by deploying facial recognition terminals at construction entrances and UWB positioning base stations in the work area to collect personnel identity information, attendance status, real-time location, and work duration; physiological data such as heart rate and body temperature can also be collected through sensors built into smart safety helmets.

[0033] Equipment data also includes: equipment ID, operating parameters, fault codes, runtime, vibration acceleration, and component temperature. Equipment data can be acquired by connecting to the controllers of large machinery such as tower cranes, construction elevators, and excavators via an IoT module to collect data such as equipment operating parameters, fault codes, and runtime; vibration sensors and temperature sensors can be installed in key parts of the equipment to collect vibration acceleration and temperature data of critical components.

[0034] Calculate the device's health value using the following formula:

[0035] in, For device health values; The number of key operating parameters of the equipment; For the first Real-time values ​​of each parameter; For the first The rated values ​​of each parameter; For the first The weights of each parameter.

[0036] The health status of an equipment can be determined by its health value. When the health value is not lower than 0.8, the equipment is in a healthy state; when the health value is lower than 0.8 but not lower than 0.5, the equipment is in a sub-healthy state; when the health value is lower than 0.5, the equipment is in an abnormal state.

[0037] Environmental data includes the deployment locations of environmental monitoring stations, sensor range parameters, and geological hazard risk thresholds. Environmental data is acquired by deploying environmental monitoring stations at 20m x 20m intervals in the construction area to collect data such as wind speed, wind direction, temperature, humidity, PM2.5 concentration, and noise levels (decibels); displacement sensors are deployed in deep foundation pits and high slope areas to collect slope settlement and displacement data.

[0038] The environmental early warning index is calculated using the following formula:

[0039] in, This is an environmental early warning index; Real-time wind speed; The safe wind speed threshold; This refers to the real-time PM2.5 concentration. The PM2.5 safety threshold; Real-time slope displacement; This refers to the safety threshold for slope displacement. This is real-time noise; This is the noise safety threshold.

[0040] The environmental early warning index can be used to determine whether the current construction environment is safe. When the environmental early warning index is below 1, it is safe; when the environmental early warning index is not lower than 1 but lower than 1.5, there is a risk in the construction environment and an early warning is required; when the environmental early warning index is not lower than 1.5, the construction environment is dangerous and an alarm is required.

[0041] Work behavior data includes: work process ID, behavior type, number of compliant behaviors, total number of behaviors, compliance rate, and description of violations. Work behavior data is collected by deploying high-definition cameras in key construction areas and using AI visual recognition algorithms to collect data on whether workers are wearing safety helmets and safety belts, whether there are violations of cross-operation regulations, and whether machinery is operating in violation of regulations. Combined with construction log data, data such as process connection time and work flow execution status are also collected.

[0042] The compliance rate is calculated using the following formula:

[0043] in, For compliance rate; To monitor the number of compliant operations during the monitoring period; To monitor the total number of operations within a monitoring period.

[0044] S200. Preprocess multi-source safety data to construct a construction safety database.

[0045] In one possible implementation, step S200 involves preprocessing the multi-source safety data to construct a construction safety database, including: S210. Clean the multi-source security data, filter out noise in the multi-source security data, remove redundant and erroneous data in the multi-source security data, and obtain cleaned data.

[0046] S220. Eliminate the heterogeneity of the cleaned data and fuse them to construct a multi-factor associated fused dataset.

[0047] In one possible implementation, step S220. Eliminating the heterogeneity of the cleaned data and fusing it to construct a multi-factor associated fused dataset, including: S221. Standardize the timestamp format of the cleaned data and establish a unique identifier for each type of data.

[0048] S222. Using timestamps as the association key, perform association matching on different types of data at the same time to generate a fused wide table that includes multiple elements.

[0049] S223. Perform vertical fusion of multi-source data for the same security element and calculate the single feature value after fusion to obtain a fused dataset of multi-element association.

[0050] The fused single feature value is calculated using the following formula:

[0051]

[0052] in, The single feature value after fusion; The number of data sources for the same security element; For the first The weight of each data source; For the first Feature values ​​of each data source.

[0053] S230. Standardize the fused dataset and unify the data field format to build a construction safety database and a standardized data dictionary.

[0054] The data in the fused dataset is standardized using the following formula, mapping the fused data to the [0,1] interval:

[0055] in, These are the standardized values; These are the original fused data values; This is the minimum value for that data dimension; This represents the maximum value for this data dimension.

[0056] S300. Based on the construction safety database, calculate the comprehensive risk value of each stage, and locate high-risk stages according to the comprehensive risk value.

[0057] In one possible implementation, S300. Based on the construction safety database, calculate the comprehensive risk value of each stage, and identify high-risk stages based on the comprehensive risk value, including: S310. Construct the input feature set.

[0058] In one possible implementation, step S310, constructing the input feature set, includes: S311. The core input features are personnel compliance rate, equipment health value, environmental early warning index, and historical risk frequency.

[0059] S312. Calculate the correlation between each feature and the risk event label.

[0060]

[0061] in, Features Risk event tags The degree of correlation between them; The features to be evaluated; Label the risk event; for Pick and Pick The probability of; for Pick The marginal probability; for Pick The marginal probability.

[0062] S313. Select features with a correlation degree not lower than the correlation degree threshold to construct the input feature set.

[0063] S320. Construct a risk identification model.

[0064] The risk identification model employs a random forest model, with 100 decision trees, a maximum depth of 10 for each tree, a minimum number of splits of 5, and the number of random features selected being the square root of the feature set size. The feature set is divided into training and testing sets in a 7:3 ratio. Historical risk event data (marked as 1 for occurrence and 0 for non-occurrence) is used as labels. The model is trained using the training set and validated using the testing set. Accuracy is used as the evaluation metric. If the model accuracy is <90%, parameters are adjusted using a grid search method: the number of decision trees ranges from 50 to 200, with a step size of 25 (i.e., candidate values ​​of 50, 75, 100, 125, 150, 175, 200); the maximum depth ranges from 5 to 15, with a step size of 2 (i.e., candidate values ​​of 5, 7, 9, 11, 13, 15). The optimal solution is selected by traversing all parameter combinations until the model accuracy is ≥90%.

[0065] S330. Calculate and output the comprehensive risk value of each stage based on the input feature set and risk identification model.

[0066] The overall risk value for each stage is calculated based on the following formula:

[0067] in, This represents the overall risk value during the construction phase. , , and These are the weighting coefficients; For compliance rate; This is an environmental early warning index; For device health values; This represents the historical risk frequency.

[0068] S340. Identify high-risk stages based on the comprehensive risk value of each stage.

[0069] S400. Predict the development trend of high-risk segments based on comprehensive risk values.

[0070] In one possible implementation, step S400, predicting the development trend of high-risk segments based on the comprehensive risk value, includes: S410. The overall risk value of each stage is handled smoothly.

[0071] The ADF test was used to examine the stationarity of the historical risk value time series for each stage. If the ADF test p-value > 0.05, the data was determined to be non-stationary. Stationarity was then achieved through d-order differencing, where d ranges from 0 to 2.

[0072] S420. Construct a risk trend prediction model.

[0073] The risk trend prediction model is based on ARIMA(p,d,q), which takes the historical risk values ​​(time series data) of each stage as input and determines the model parameters p (autoregressive order), d (difference order), and q (moving average order) through the AIC criterion to predict the risk value change trend in the next 7 days.

[0074] S430. Analyze the time evolution of the comprehensive risk value of each link using a risk trend prediction model in order to predict the development trend of high-risk links.

[0075] The following formula is used to predict the development trend of high-risk segments:

[0076] in, for Risk value at any given moment; for Order difference operator; For constant terms; These are the autoregressive coefficients; The moving average coefficient; for A sequence of white noise at each time step; The order of autoregression; It is the difference order; This represents the order of the moving average.

[0077] S500. Pushes early warning information based on the comprehensive risk value and the development trend of high-risk links.

[0078] In one possible implementation, S500 pushes early warning information based on the comprehensive risk value and the development trend of high-risk links, including: S510. Establish an early warning level system and a graded notification mechanism.

[0079] In one possible implementation, the early warning level system in step S510 includes Level 1, Level 2, Level 3 and Level 4.

[0080] S520. Determine the early warning level and push strategy based on the comprehensive risk value and the development trend of high-risk links, and issue early warnings.

[0081] The early warning level system is as follows:

[0082] Among them, is the current comprehensive risk value; is the future risk value output by the risk trend prediction model.

[0083] The hierarchical push mechanism is as follows: The first-level early warning is pushed to the project manager, safety director and the person in charge of the construction unit; the second-level early warning is pushed to the safety director and the construction team leader; the third-level early warning is pushed to the construction team leader and the safety officer; the fourth-level early warning is pushed to the safety officer.

[0084] The beneficial effect of the construction safety supervision method based on data analysis provided by the present invention is that: compared with the prior art, the construction safety supervision method based on data analysis in the present invention can break through the limitation of single data collection dimension under the traditional supervision method by obtaining multi-source safety data during the construction process, comprehensively cover various safety-related information in the construction scenario, and no longer rely solely on static basic information, so as to more completely capture the dynamic risk sources during the construction process. The acquisition of multi-source data provides a rich and comprehensive basis for subsequent safety analysis, making the safety supervision free from blind spots due to data lack.

[0085] Preprocessing the multi-source safety data and constructing a construction safety database can effectively solve the problem of clutter of data from different sources, filter out the noise, redundancy and error data therein, eliminate data heterogeneity and achieve fusion at the same time, and can also unify the data field format. This process makes the originally scattered and irregular data become regular, orderly and relevant, laying a reliable data foundation for accurately calculating the comprehensive risk value subsequently, and avoiding affecting the risk assessment result due to data quality problems.

[0086] Calculating the comprehensive risk value of each link based on the construction safety database and positioning the high-risk links according to this value can convert the abstract safety situation into a quantifiable index. By comprehensively considering various safety factors reflected by multi-source data, accurately evaluating the risk degree of each link in the construction, and thus accurately identifying the links with relatively high safety hazards. This changes the situation in traditional supervision where it is difficult to accurately judge the key points of risks, enabling the supervisors to clarify the supervision focus and avoid blind patrols.

[0087] Predicting the development trend of high-risk links based on the comprehensive risk value can grasp the risk change direction and possible safety situation of high-risk links in advance. It is no longer limited to the identification of current risks, but through the analysis of historical and current data, predicting the evolution of future risks, making the safety supervision shift from passive response to active prevention, striving for time to take intervention measures in advance, and reducing the possibility of safety accidents.

[0088] By sending early warning information based on comprehensive risk values ​​and the development trends of high-risk areas, regulatory personnel can obtain timely alerts of critical safety risks. The early warning information is based on quantitative risk assessment and trend prediction, making it targeted and timely. This allows regulatory personnel to react quickly to the warning content, take corresponding preventative measures, effectively curb the transformation of high-risk areas into safety accidents, protect the lives and property of construction workers, and promote the smooth progress of the project.

[0089] Secondly, a data analysis-based construction safety supervision system is provided, applied to the data analysis-based construction safety supervision method described in the first aspect, including: The data acquisition unit is used to acquire multi-source safety data during the construction process; The construction safety database construction unit is used to preprocess multi-source safety data and construct a construction safety database. The comprehensive risk value calculation and high-risk link location unit is used to calculate the comprehensive risk value of each link based on the construction safety database, and to locate high-risk links based on the comprehensive risk value; The high-risk segment development trend prediction unit is used to predict the development trend of high-risk segments based on the comprehensive risk value. The early warning unit is used to push early warning information based on the comprehensive risk value and the development trend of high-risk links.

[0090] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

[0091] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the exemplary embodiments according to this application. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0092] Unless otherwise specifically stated, the relative arrangement, numerical expressions, and values ​​of the components and steps described in these embodiments do not limit the scope of this application. It should also be understood that, for ease of description, the dimensions of the various parts shown in the drawings are not drawn to actual scale. Techniques, methods, and devices known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and devices should be considered part of the specification. In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limitations. Therefore, other examples of exemplary embodiments may have different values. It should be noted that similar reference numerals and letters in the following drawings denote similar items; therefore, once an item is defined in one drawing, it need not be further discussed in subsequent drawings.

Claims

1. A construction safety supervision method based on data analysis, characterized in that, It includes: Obtain multi-source safety data during the construction process; Preprocess the multi-source safety data to construct a construction safety database; Based on the construction safety database, calculate the comprehensive risk value of each link, and locate high-risk links according to the comprehensive risk value; Predict the development trend of the high-risk links according to the comprehensive risk value; Push warning information according to the comprehensive risk value and the development trend of the high-risk links.

2. The construction safety supervision method based on data analysis as described in claim 1, characterized in that, The multi-source safety data includes: personnel data, equipment data, environmental data, and operation behavior data; the personnel data includes the personnel compliance rate, the equipment data includes the equipment health index, the environmental data includes the environmental warning index, and the operation behavior data includes the historical risk frequency.

3. The construction safety supervision method based on data analysis as described in claim 1, characterized in that, The preprocessing of the multi-source safety data to construct a construction safety database includes: Clean the multi-source safety data, filter the noise in the multi-source safety data, remove the redundant and incorrect data in the multi-source safety data, and obtain the cleaned data; Eliminate the heterogeneity of the cleaned data and perform fusion to construct a multi-factor associated fusion dataset; Standardize the fusion dataset and unified the data field format to construct a construction safety database and a standardized data dictionary.

4. The construction safety supervision method based on data analysis as described in claim 3, characterized in that, The elimination of the heterogeneity of the cleaned data and the fusion to construct a multi-factor associated fusion dataset includes: Unify the timestamp format of the cleaned data and establish a unique identifier for each type of data; Using the timestamp as the association key, perform association matching on different types of data at the same moment to generate a fusion wide table including multiple factors; Vertically fuse the multi-source data of the same safety factor and calculate the single eigenvalue after fusion to obtain a multi-factor associated fusion dataset.

5. The construction safety supervision method based on data analysis as described in claim 2, characterized in that, Based on the construction safety database, calculate the comprehensive risk value of each link, and locate high-risk links according to the comprehensive risk value, including: Construct an input feature set; Construct a risk identification model; Calculate and output the comprehensive risk value of each link based on the input feature set and the risk identification model; Locate high-risk links according to the comprehensive risk value of each link.

6. The construction safety supervision method based on data analysis as described in claim 5, characterized in that, The construction of the input feature set includes: Use the personnel compliance rate, the equipment health index, the environmental warning index, and the historical risk frequency as the core input features; Calculate the correlation degree between each feature and the risk event label; Screen the features with a correlation degree not lower than the correlation degree threshold to construct an input feature set.

7. The construction safety supervision method based on data analysis as described in claim 1, characterized in that, The prediction of the development trend of the high-risk links according to the comprehensive risk value includes: Perform a smoothing process on the comprehensive risk value of each link; Construct a risk trend prediction model; Use the risk trend prediction model to analyze the time evolution law of the comprehensive risk value of each link to predict the development trend of the high-risk links.

8. The construction safety supervision method based on data analysis as described in claim 1, characterized in that, The push of warning information according to the comprehensive risk value and the development trend of the high-risk links includes: Establish a warning level system and a hierarchical push mechanism; Determine the warning level and push strategy according to the comprehensive risk value and the development trend of the high-risk links, and issue a warning.

9. The construction safety supervision method based on data analysis as described in claim 8, characterized in that, The warning level system includes level one, level two, level three, and level four.

10. A data-based construction safety supervision system, applied to the data-based construction safety supervision method as described in any one of claims 1 to 9, characterized in that, It includes: The data acquisition unit is used to acquire multi-source safety data during the construction process; The construction safety database construction unit is used to preprocess the multi-source safety data and construct a construction safety database. The comprehensive risk value calculation and high-risk link location unit is used to calculate the comprehensive risk value of each link based on the construction safety database, and locate the high-risk link according to the comprehensive risk value; A high-risk segment development trend prediction unit is used to predict the development trend of the high-risk segment based on the comprehensive risk value. The early warning unit is used to push early warning information based on the comprehensive risk value and the development trend of the high-risk links.

Images