A construction site safety early warning method and system based on multi-dimensional data mining

Abstract

The application discloses a kind of based on multi-dimensional data mining's construction site safety early warning method and system, it is related to construction site early warning technical field, including, acquisition multi-dimensional data and pre-processing, form multi-dimensional dataset;Multi-dimensional dataset is extracted using convolutional neural network feature, and through time series information is fused by space-time dynamic weighting mechanism, generates multi-dimensional risk feature set;Based on multi-dimensional risk feature set, risk feature is converted into node, the correlation between risk features is converted into edge, form the causal association structure of node and edge, construct dynamic risk atlas;The correlation intensity between risk factors is analyzed using dynamic risk atlas, overall risk level is evaluated, and risk event list is generated;Risk event in risk event list is trend forecasted based on time series, and risk trend index is obtained;According to risk trend index, dynamic early warning is carried out, and safety response is triggered.The application realizes the intellectualization of construction site safety early warning by multi-dimensional data mining and analysis.

Description

Technical Field

[0001] This invention relates to the field of construction site early warning technology, and in particular to a construction site safety early warning method and system based on multidimensional data mining. Background Technology

[0002] In traditional construction site safety management, risk early warning mainly relies on manual experience and static monitoring methods. This results in insufficient predictive ability for risk events and an inability to achieve dynamic and real-time risk trend analysis and early warning. Traditional methods are usually based on single-dimensional data, such as environmental data or equipment status, lacking the comprehensive utilization of multi-dimensional data. Furthermore, these methods cannot effectively capture the correlation between risk factors and their dynamic characteristics over time, leading to inaccurate risk assessment results and difficulty in timely identifying potential high-risk areas.

[0003] Especially for risk trend prediction in time series, traditional methods often employ simple statistical analysis or fixed models, which cannot handle the dynamic changes of multiple risk events in complex scenarios. The characteristic values ​​of risk events are easily affected by various factors, and traditional models struggle to capture their inherent spatiotemporal dependencies, resulting in low accuracy in predicting future risk trends and delayed early warning mechanisms. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides a construction site safety early warning method based on multidimensional data mining to solve the problem of the lack of comprehensive analysis of multidimensional data and accurate prediction of future risk trends in traditional construction site safety management.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0007] In a first aspect, the present invention provides a construction site safety early warning method based on multidimensional data mining, which includes collecting multidimensional data and preprocessing it to form a multidimensional dataset;

[0008] Convolutional neural networks are used to extract features from multidimensional datasets, and time series information is fused through a spatiotemporal dynamic weighting mechanism to generate a multidimensional risk feature set.

[0009] Based on a multidimensional risk feature set, risk features are transformed into nodes, and the relationships between risk features are transformed into edges, forming a causal relationship structure between nodes and edges, and constructing a dynamic risk graph.

[0010] The dynamic risk graph is used to analyze the correlation strength between risk factors, assess the overall risk level, and generate a list of risk events.

[0011] Based on time series analysis, the risk event list is used to predict trends and obtain a risk trend index.

[0012] Dynamic warnings are issued based on risk trend indices, and safety responses are triggered.

[0013] As a preferred embodiment of the construction site safety early warning method based on multidimensional data mining described in this invention, the data collection is carried out using a drone and sensors in a collaborative manner.

[0014] The multidimensional data includes environmental data, worker data, equipment data, risk events, time, and coordinates;

[0015] The environmental data includes temperature and humidity, dust concentration, nitrogen dioxide concentration, noise intensity, and wind speed;

[0016] The worker data includes the worker's heart rate, location, and behavior;

[0017] The equipment data includes the equipment's vibration, pressure, and displacement;

[0018] The risk events mentioned include worker fatigue, worker heatstroke, equipment failure, and environmental degradation.

[0019] The preprocessing includes data cleaning, normalization, and spatiotemporal alignment.

[0020] As a preferred embodiment of the construction site safety early warning method based on multidimensional data mining described in this invention, the method involves: extracting features from a multidimensional dataset using a convolutional neural network, and fusing time-series information through a spatiotemporal dynamic weighting mechanism to generate a multidimensional risk feature set. The specific steps are as follows:

[0021] A multidimensional dataset is input into a convolutional neural network, and the convolutional layers of the convolutional neural network are used to extract features from the multidimensional data to obtain spatial features.

[0022] Pooling layers of convolutional neural networks are used to reduce the dimensionality and denoise spatial features, and key spatial features are extracted.

[0023] Based on the time series information of the collected data, a time weight is assigned to each key spatial feature in an exponentially decaying manner.

[0024] A weighted fusion method is used to combine time series information with spatial features extracted by a convolutional neural network to generate a multidimensional risk feature set.

[0025] As a preferred embodiment of the construction site safety early warning method based on multidimensional data mining described in this invention, the method involves: based on a multidimensional risk feature set, transforming risk features into nodes, transforming the relationships between risk features into edges, forming a causal relationship structure between nodes and edges, and constructing a dynamic risk graph. The specific steps are as follows.

[0026] Risk factors are extracted from a multidimensional risk feature set, and each risk factor is used as a node.

[0027] Each node contains the characteristic values ​​of the risk factors and spatiotemporal information;

[0028] Analyze the relationships between risk factors and transform them into edges between nodes;

[0029] Using Granger causality analysis, edges with low relevance are removed to obtain causal inference results;

[0030] The causal inference results are combined with time series information to generate a dynamic risk map that changes over time.

[0031] As a preferred embodiment of the construction site safety early warning method based on multidimensional data mining described in this invention, the following steps are taken: Dynamic risk graph analysis is used to assess the correlation strength between risk factors, evaluate the overall risk level, and generate a list of risk events.

[0032] Based on the risk factors and their correlations in the dynamic risk map, specific high-risk areas and potential risk events are identified.

[0033] Based on the correlation between nodes and edges in the dynamic risk graph, the risk weight of each node is calculated.

[0034] The overall risk level of the construction site is calculated based on the number and risk weight of risk factors in high-risk areas.

[0035] The risk factors that constitute risk events in high-risk areas are arranged according to their probability of occurrence to generate a list of risk events.

[0036] As a preferred embodiment of the construction site safety early warning method based on multidimensional data mining described in this invention, the method involves: predicting the trend of risk events in a risk event list based on time series data to obtain a risk trend index. The specific steps are as follows:

[0037] The risk event list and its corresponding time series data are divided according to time windows to form training sets, test sets, and validation sets;

[0038] As a preferred embodiment of the construction site safety early warning method based on multidimensional data mining described in this invention, the method involves: dynamically issuing early warnings based on a risk trend index and triggering a safety response. The specific steps are as follows:

[0039] Analyze the risk trend index to determine the low-risk and medium-risk thresholds;

[0040] When the risk trend index is less than or equal to the low-risk threshold, a low-level warning is issued to remind workers to pay attention.

[0041] When the risk trend index is greater than the low-risk threshold and less than or equal to the medium-risk threshold, a medium-level warning is issued, related operations are suspended, and the equipment and environmental status are checked.

[0042] When the risk trend index exceeds the medium-risk threshold, a high-level warning is issued, personnel are urgently evacuated, drones are used to inspect high-risk areas, and risk changes are monitored in real time.

[0043] Secondly, the present invention provides a construction site safety early warning system based on multidimensional data mining, including a data acquisition module, a feature set module, a risk map module, a risk assessment module, a trend prediction module, and an early warning module;

[0044] The data acquisition module is used to collect multidimensional data and preprocess it to form a multidimensional dataset;

[0045] The feature set module is used to extract features from a multidimensional dataset using a convolutional neural network and to fuse time series information through a spatiotemporal dynamic weighting mechanism to generate a multidimensional risk feature set.

[0046] The risk graph module is used to transform risk features into nodes and the relationships between risk features into edges based on a multi-dimensional risk feature set, forming a causal relationship structure between nodes and edges, and constructing a dynamic risk graph.

[0047] The risk assessment module is used to analyze the correlation strength between risk factors using a dynamic risk graph, assess the overall risk level, and generate a list of risk events.

[0048] The trend prediction module is used to predict the trend of risk events in the risk event list based on time series data to obtain a risk trend index.

[0049] The early warning module is used to provide dynamic early warnings based on the risk trend index and trigger a safety response.

[0050] Thirdly, the present invention provides a computer device, including a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, it implements any step of the construction site safety early warning method based on multidimensional data mining as described in the first aspect of the present invention.

[0051] Fourthly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the construction site safety early warning method based on multidimensional data mining as described in the first aspect of the present invention.

[0052] The beneficial effects of this invention are as follows: This invention achieves intelligent and dynamic monitoring and early warning of construction site safety through multi-dimensional data mining and time series analysis. It comprehensively covers safety risk factors at construction sites by combining multi-dimensional data collected by drones and sensors, including environmental, worker, and equipment data. Features are extracted using convolutional neural networks and spatiotemporal dynamic weighting mechanisms, and a dynamic risk map is constructed using Granger causal analysis. This map can intuitively present the correlation between risk factors and their changes over time, providing support for accurate risk assessment. The introduction of the ARIMA time series model for trend prediction of risk events generates a risk trend index, which can capture the dynamic changes of risk over time, providing a scientific basis for risk early warning. Based on the risk trend index, low, medium, and high-level dynamic early warnings are issued, triggering corresponding safety response measures, improving the initiative and effectiveness of construction site safety management, and reducing the harm of risk events to personnel and equipment. Attached Figure Description

[0053] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. 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.

[0054] Figure 1 This is a flowchart of the construction site safety early warning method based on multidimensional data mining in Example 1.

[0055] Figure 2 This is a module diagram of the construction site safety early warning system based on multidimensional data mining in Example 1. Detailed Implementation

[0056] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0057] Example 1, referring to Figure 1 and Figure 2 This is the first embodiment of the present invention, which provides a construction site safety early warning method based on multidimensional data mining, including the following steps:

[0058] S1. Collect multidimensional data and preprocess it to form a multidimensional dataset, including the following steps.

[0059] The system uses drones and sensors to collaboratively collect multidimensional data, including environmental data, worker data, equipment data, risk events, time, and coordinates. Environmental data includes temperature and humidity, dust concentration, nitrogen dioxide concentration, noise intensity, and wind speed. Worker data includes workers' heart rate, location, and behavior. Equipment data includes equipment vibration, pressure, and displacement. Risk events include worker fatigue, worker heatstroke, equipment failure, and environmental degradation. Preprocessing includes data cleaning, normalization, and spatiotemporal alignment.

[0060] Specifically, drones are used to collect large-area environmental data, including temperature and humidity, dust concentration, nitrogen dioxide concentration, noise intensity, and wind speed. Sensors are deployed in key areas of the construction site to monitor equipment status (vibration, pressure, displacement) and worker status (heart rate, location, behavior) in real time. Drones and sensors work together via a wireless network to collect data periodically or on demand, covering the entire construction site and key areas. The data collection frequency can be adjusted as needed, for example, once every 5 minutes, or with increased frequency during critical processes.

[0061] Data cleaning involves checking the integrity of collected data, removing data with severe missing values, handling outliers (e.g., removing data outside a reasonable range according to statistical rules), eliminating duplicate data, and ensuring that each data point is unique.

[0062] Normalization is a process that converts data of different dimensions (such as temperature and vibration) into dimensionless values ​​using normalization methods (such as min-max normalization or z-score normalization), ensuring that all data are within the same range and facilitating subsequent model processing.

[0063] Spatiotemporal alignment aligns data based on timestamps, ensuring that data from different dimensions match at the same point in time. It also involves spatial clustering of data based on coordinates, integrating multidimensional data from the same location, and interpolating to fill in missing data in time and space (e.g., linear interpolation). Data undergoes initial screening via edge computing devices before being uploaded to the data center.

[0064] It should be noted that the collected multidimensional data comprehensively covers various risk factors such as environment, workers, and equipment, reflecting the overall safety situation of the construction site, and establishing a spatiotemporal dynamic model through time and coordinate information. The comprehensiveness and accuracy of the multidimensional data provide a solid data foundation for subsequent feature extraction, risk assessment, and trend prediction.

[0065] S2. Feature extraction from a multidimensional dataset is performed using a convolutional neural network, and time-series information is fused through a spatiotemporal dynamic weighting mechanism to generate a multidimensional risk feature set. This includes the following steps:

[0066] A multidimensional dataset is input into a convolutional neural network (CNN). The convolutional layers of the CNN extract features from the multidimensional data to obtain spatial features. Specifically, the multidimensional dataset is first organized by arranging environmental data, worker data, equipment data, etc., according to time and spatial dimensions to form a structured multidimensional input tensor.

[0067] Convolutional kernels are used to perform convolution operations on multidimensional data to extract local spatial information. Convolutional layers move across the input tensor through a sliding window to capture feature patterns at different locations. Each convolutional kernel corresponds to a specific feature pattern, such as temperature and humidity change patterns in environmental data or feature patterns of worker behavior.

[0068] Pooling layers in convolutional neural networks are used to reduce the dimensionality and denoise spatial features, extracting key spatial features. Specifically, max pooling or average pooling operations are used to reduce the dimensionality of the feature maps output by the convolutional layers, extracting the most salient features (such as maximum or average values) within each pooling region. This reduces redundant features, removes noise and local outliers, enhances the model's ability to express key features, and generates simplified and more compact spatial features, providing a foundation for subsequent weighted fusion processing.

[0069] Based on the time-series information of the collected data, a time weight is assigned to each key spatial feature, decreasing exponentially. Specifically, each spatial feature is assigned an initial time weight based on its corresponding timestamp. Using the exponential decay function, features closer to the current time point are assigned higher weights, and features farther from the current time point are assigned lower weights. The importance of spatial features is adjusted in conjunction with the time weights to ensure that the model focuses on current and recent key information.

[0070] A weighted fusion method is used to combine time-series information with spatial features extracted by a convolutional neural network to generate a multidimensional risk feature set. Specifically, each spatial feature is multiplied by its corresponding time weight to generate a weighted feature. These weighted features are stacked in chronological order to form a spatiotemporal feature matrix. A fully connected layer is used to perform a nonlinear transformation on the spatiotemporal feature matrix to further fuse time and spatial information, outputting a multidimensional risk feature set containing spatiotemporal dynamic information of multidimensional features such as environment, workers, and equipment.

[0071] It should be noted that the spatial feature extraction capability of convolutional neural networks, combined with a dynamic time weighting mechanism, can effectively capture the spatiotemporal correlation characteristics of multidimensional data. By allocating time weights through an exponential decay function, the model's ability to focus on recent data is enhanced, while reducing the interference of long-term data on risk prediction. The multidimensional risk feature set provides accurate and efficient foundational data for subsequent causal analysis and the construction of dynamic risk maps.

[0072] S3. Based on a multi-dimensional risk feature set, risk features are transformed into nodes, and the relationships between risk features are transformed into edges, forming a causal relationship structure between nodes and edges, thus constructing a dynamic risk graph. This includes the following steps.

[0073] Risk factors are extracted from a multidimensional risk feature set, and each risk factor is treated as a node. Specifically, each feature in the multidimensional risk feature set is classified to identify risk factors in categories such as environment, workers, and equipment. Each risk factor, such as temperature and humidity, worker heart rate, and equipment vibration, is treated as an independent node. The attributes of the node include the feature value, category, and association weight of the risk factor.

[0074] Each node contains the feature values ​​and spatiotemporal information of the risk factor. Specifically, the node stores the current feature value and trend of the corresponding risk factor, includes the timestamp information of the risk factor for dynamically updating the graph, records the spatial coordinate information of the node, and identifies the specific location of the risk factor.

[0075] Analyze the relationships between risk factors and transform them into edges between nodes. Specifically, calculate the correlation between risk factors, for example, using the Pearson correlation coefficient or mutual information method, connect risk factors with high correlation as edges, and assign edge weights. The attributes of the edges include correlation strength and direction, with direction reflecting causal relationships.

[0076] Granger causality analysis is used to remove edges with low correlation to obtain causal inference results. Specifically, Granger causality analysis is used to detect whether the causal relationship between risk factors is significant. By selecting typical risk factors and setting thresholds for edges, edges with insignificant causal relationships or low correlation are removed, and the optimized causal association structure is output.

[0077] The causal inference results are combined with time series information to generate a dynamic risk map that changes over time. Specifically, the nodes and edges in the causal relationship structure are dynamically updated, and the changes in the relationships between risk factors are reflected by combining time series information. The dynamic risk map contains nodes, edges, and their attribute values ​​that change over time. This map can display the risk distribution and causal relationships at the construction site in real time.

[0078] It should be noted that the dynamic risk map, through causal analysis and time series updates, can intuitively reflect the strength of the correlation and the trend of change between risk factors. This map provides an important basis for accurately identifying high-risk areas and potential risk events, and provides spatiotemporal dynamic support for subsequent risk assessment and early warning.

[0079] S4. Utilize dynamic risk mapping to analyze the correlation strength between risk factors, assess the overall risk level, and generate a list of risk events, including the following steps:

[0080] Based on the risk factors and their correlations in the dynamic risk map, specific high-risk areas and potential risk events are identified. Specifically, nodes and edges with high correlation strength in the dynamic risk map are screened, and high-risk areas are identified by clustering based on the spatial coordinates of the nodes. Potential risk events, such as worker fatigue and equipment failure, are then extracted from these high-risk areas.

[0081] Based on the correlation between nodes and edges in the dynamic risk graph, the risk weight of each node is calculated. Specifically, the initial weight of the risk factor of each node is calculated according to its feature values ​​(such as worker heart rate and equipment vibration), and the risk weight of the node is adjusted by taking into account the correlation strength of the edges, and the final risk weight of each node is output.

[0082] The overall risk level of the construction site is calculated based on the number and risk weight of risk factors in high-risk areas. Specifically, the number of risk factors in high-risk areas is counted, the weighted average of the risk weights of all nodes in the high-risk areas is calculated, and the overall risk level of the construction site is determined according to the preset risk level classification standards.

[0083] The risk factors that constitute risk events in high-risk areas are arranged according to their probability of occurrence, generating a list of risk events. Specifically, based on the weight and time trend of the risk factors, their probability of occurrence is predicted. High-probability risk factors are prioritized to generate a list of risk events, outputting a list containing information such as the type, location, and probability of occurrence of each risk event.

[0084] It should be noted that by analyzing the nodes and edges in the dynamic risk map, high-risk areas can be accurately located and the overall risk level of the construction site can be quantified. The risk event list provides decision-makers at the construction site with a clear risk priority ranking, which facilitates the implementation of targeted safety response measures.

[0085] S5. Based on time series analysis, predict the trends of risk events in the risk event list to obtain a risk trend index, including the following steps.

[0086] The list of risk events and their corresponding time-series data are divided into training, testing, and validation sets according to time windows. Specifically, time windows are divided, and the historical data for each risk event is divided into training, testing, and validation sets to ensure that the training set data covers a sufficient time range. The testing and validation sets are used for model evaluation.

[0087] The time-series data for each risk event is represented as follows:

[0088] {Y(t)|t=1,2,…,T w};

[0089] Where Y(t) is the characteristic value of the risk event at time t, t is the index of the time point, and T wFor the Tth w A specific point in time.

[0090] For each risk event, stationarity testing and differencing are performed on the time series data to construct the input features of the time series model. The differencing formula is expressed as follows:

[0091] ΔY(t)=Y(t)-Y(t-1),t>1;

[0092] Where ΔY(t) is the difference between the characteristic value of the risk event at time t and the characteristic value of the risk event at time t-1, Y(t) is the characteristic value of the risk event at time t, and Y(t-1) is the characteristic value of the risk event at time t-1.

[0093] An ARIMA time series analysis model is trained using the training, test, and validation sets to obtain a risk trend model, denoted as follows:

[0094]

[0095] Where T is the risk trend index, n is the total number of risk events, c is the fixed base value of the trend, i is the index of the risk event number, p is the order of the autoregressive term, j is the counter index of the autoregressive term, k is the counter index of the moving average term, ∈ is the random error, and Y i (t) represents the characteristic value of the i-th risk event at time t, R i As the weight of risk events, φ j For autoregressive coefficients, θ k This is the moving average coefficient.

[0096] It should be noted that by using the ARIMA model to capture the temporal dynamic characteristics of risk events, it is possible to accurately predict future risk trends. The comprehensive trend index provides a quantitative basis for construction site safety early warning, which facilitates further graded response.

[0097] S6. Issue dynamic warnings based on the risk trend index and trigger a safety response, including the following steps:

[0098] The risk trend index is analyzed to determine low-risk and medium-risk thresholds. Specifically, the risk trend index is divided into three risk levels: low, medium, and high.

[0099] When the risk trend index is less than or equal to the low-risk threshold, a low-level warning is issued to remind workers to be cautious. Specifically, a low-level warning message is sent to construction site staff, advising workers to strengthen their safety awareness and pay attention to changes in the surrounding environment.

[0100] When the risk trend index is greater than the low-risk threshold but less than or equal to the medium-risk threshold, a medium-level warning is issued, related operations are suspended, and equipment and environmental conditions are inspected. Specifically, work activities in high-risk areas are halted, safety inspectors are dispatched to check equipment and environmental conditions, and drones are deployed to conduct patrols of high-risk areas.

[0101] When the risk trend index exceeds the medium-risk threshold, a high-level warning is issued, personnel are urgently evacuated, and drones are deployed to patrol high-risk areas and monitor risk changes in real time. Specifically, all personnel in high-risk areas are urgently evacuated, drones are activated to monitor high-risk areas in real time, and emergency management departments are notified to take further measures.

[0102] It should be noted that the tiered early warning mechanism can dynamically adjust safety response measures based on the risk trend index. The low, medium, and high-level early warnings cover the entire life cycle management of different risk levels, effectively reducing the probability of construction site safety accidents.

[0103] This embodiment also provides a construction site safety early warning system based on multidimensional data mining, including: a data acquisition module, a feature set module, a risk map module, a risk assessment module, a trend prediction module, and an early warning module;

[0104] The data acquisition module is used to collect multidimensional data and preprocess it to form a multidimensional dataset;

[0105] The feature set module is used to extract features from multidimensional datasets using convolutional neural networks and to fuse time series information through a spatiotemporal dynamic weighting mechanism to generate a multidimensional risk feature set.

[0106] The risk graph module is used to transform risk features into nodes based on a multi-dimensional risk feature set, transform the relationships between risk features into edges, form a causal relationship structure between nodes and edges, and construct a dynamic risk graph.

[0107] The risk assessment module is used to analyze the correlation strength between risk factors using a dynamic risk graph, assess the overall risk level, and generate a list of risk events.

[0108] The trend prediction module is used to predict the trends of risk events in the risk event list based on time series data, and obtain the risk trend index.

[0109] The early warning module is used to provide dynamic early warnings based on risk trend indices and trigger safety responses.

[0110] This embodiment also provides a computer device applicable to the construction site safety early warning method based on multidimensional data mining, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to realize the construction site safety early warning method based on multidimensional data mining as proposed in the above embodiment.

[0111] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0112] This embodiment also provides a storage medium storing a computer program, which, when executed by a processor, implements the construction site safety early warning method based on multidimensional data mining as proposed in the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0113] In summary, this invention achieves intelligent and dynamic construction site safety monitoring and early warning through multi-dimensional data mining and time series analysis. It comprehensively covers construction site safety risk factors by combining multi-dimensional data collected by drones and sensors, including environmental, worker, and equipment data. Features are extracted using convolutional neural networks and spatiotemporal dynamic weighting mechanisms, and a dynamic risk map is constructed using Granger causality analysis. This map visually presents the correlations between risk factors and their changes over time, providing support for accurate risk assessment. The introduction of the ARIMA time series model for trend prediction of risk events generates a risk trend index, capturing the dynamic changes of risk over time and providing a scientific basis for risk early warning. Based on the risk trend index, low, medium, and high-level dynamic early warnings are issued, triggering corresponding safety response measures, improving the initiative and effectiveness of construction site safety management, and reducing the harm of risk events to personnel and equipment.

[0114] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A construction site safety early warning method based on multi-dimensional data mining, characterized in that: The application relates to a method for constructing a dynamic risk graph based on multi-dimensional data. The method comprises the following steps:

1. Collecting and preprocessing multi-dimensional data to form a multi-dimensional data set, wherein the multi-dimensional data comprises environmental data, worker data, equipment data, risk events, time and coordinates; 2. Extracting spatial features from the multi-dimensional data set by using a convolutional neural network, and generating a multi-dimensional risk feature set by fusing time sequence information through a spatio-temporal dynamic weighting mechanism; 3. Extracting risk factors from the multi-dimensional risk feature set, and taking each risk factor as a node; 4. Each node contains the characteristic value of the risk factor and the spatio-temporal information; 5. Analyzing the correlation between the risk factors and converting it into edges between the nodes; 6. Using Granger causality analysis method to remove edges with low correlation, and obtaining causality inference results; 7. Generating a dynamic risk graph that changes over time by combining the causality inference results with the time sequence information; 8. Analyzing the correlation strength between the risk factors by using the dynamic risk graph, evaluating the overall risk level, and generating a risk event list; 9. Trend forecasting of the risk events in the risk event list based on the time sequence, and obtaining a risk trend index; 2. The construction site safety early warning method based on multi-dimensional data mining according to claim 1, characterized in that:

10. Dynamic early warning according to the risk trend index, and triggering a safety response. The collection is achieved by using a UAV and a sensor in cooperation; The environmental data comprises temperature and humidity, dust concentration, nitrogen dioxide concentration, noise intensity and wind speed; The worker data comprises the heart rate, position and behavior of the worker; The equipment data comprises the vibration, pressure and displacement of the equipment; The risk events comprise worker fatigue, worker heatstroke, equipment failure and environmental deterioration; 3. The construction site safety early warning method based on multi-dimensional data mining according to claim 2, characterized in that: The preprocessing comprises data cleaning, normalization processing and spatio-temporal alignment. The specific steps are as follows:

1. Inputting the multi-dimensional data set into the convolutional neural network, and extracting spatial features from the multi-dimensional data by using the convolutional layer of the convolutional neural network; 2. Using the pooling layer of the convolutional neural network to reduce the dimension and denoise the spatial features, and extracting key spatial features; 3. According to the time sequence information of the collected data, assigning each key spatial feature with a time weight that decreases according to an exponential decay function; 4. The construction site safety early warning method based on multi-dimensional data mining according to claim 3, characterized in that:

4. Using a weighted fusion method to combine the time sequence information with the spatial features obtained by the convolutional neural network, and generating a multi-dimensional risk feature set. The specific steps are as follows:

1. According to the risk factors and correlation in the dynamic risk graph, identifying specific high-risk areas and potential risk events; 2. Calculating the risk weight of each node based on the correlation of the nodes and edges in the dynamic risk graph; 3. Calculating the risk level of the overall construction site according to the number of risk factors and the risk weight in the high-risk area; 5. The method of claim 4, wherein the method further comprises:

4. Arranging the risk factors belonging to the risk events in the high-risk area according to the occurrence probability, and generating a risk event list. The specific steps are as follows:

1. Dividing the risk event list and the corresponding time sequence data according to the time window to form a training set, a test set and a validation set; 2. Training a risk trend prediction model based on the training set and the test set; 3. Using the trained risk trend prediction model to predict the trend of the risk events in the risk event list based on the validation set, and obtaining a risk trend index. The time series data of each risk event is subjected to stationarity detection and difference processing to construct input features of the time series model; An ARIMA time series analysis model is trained using the training set, test set and validation set to obtain a risk trend model.

6. The construction site safety early warning method based on multi-dimensional data mining according to claim 5, characterized in that: According to the risk trend index, dynamic early warning is performed and safety response is triggered, and the specific steps are as follows, The risk trend index is analyzed to determine a low risk threshold and a medium risk threshold; When the risk trend index is less than or equal to the low risk threshold, low-level early warning is performed to remind workers; When the risk trend index is greater than the low risk threshold and less than or equal to the medium risk threshold, medium-level early warning is performed, the relevant operation is suspended, and the equipment and environment state are checked; When the risk trend index is greater than the medium risk threshold, high-level early warning is performed, personnel are evacuated in an emergency, a UAV is used to patrol a high-risk area, and real-time monitoring of risk changes is performed.

7. A construction site safety early warning system based on multi-dimensional data mining, based on the construction site safety early warning method based on multi-dimensional data mining of any one of claims 1-6, characterized in that: It includes a data acquisition module, a feature set module, a risk map module, a risk assessment module, a trend prediction module and an early warning module; The data acquisition module is used to acquire and preprocess multi-dimensional data to form a multi-dimensional data set, wherein the multi-dimensional data includes environmental data, worker data, equipment data, risk events, time and coordinates; The feature set module is used to extract spatial features from the multi-dimensional data set by using a convolutional neural network, and generate a multi-dimensional risk feature set by fusing time series information through a spatio-temporal dynamic weighting mechanism; The risk map module is used to extract risk factors from the multi-dimensional risk feature set, and take each risk factor as a node; Each node contains a characteristic value of a risk factor and spatio-temporal information; The correlation between risk factors is analyzed and converted into edges between nodes; Granger causality analysis method is used to remove edges with low correlation to obtain causality inference results; the causality inference results are combined with time series information to generate a dynamic risk map that changes over time; The risk assessment module is used to analyze the correlation strength between risk factors based on the dynamic risk map, assess the overall risk level, and generate a risk event list; The trend prediction module is used to predict the trend of risk events in the risk event list based on time series to obtain a risk trend index; The early warning module is used to perform dynamic early warning according to the risk trend index and trigger safety response.

8. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that: The processor executes the computer program to realize the steps of the construction site safety early warning method based on multi-dimensional data mining according to any one of claims 1-6.

9. A computer readable storage medium having stored thereon a computer program, characterized in that: The computer program is executed by the processor to realize the steps of the construction site safety early warning method based on multi-dimensional data mining according to any one of claims 1-6.

Images