Building construction safety monitoring method and system based on artificial intelligence

Abstract

The invention relates to the technical field of safety monitoring, in particular to a building construction safety monitoring method and system based on artificial intelligence, and the method comprises the steps: obtaining multi-source heterogeneous data of a construction site, carrying out the distributed feature extraction of the multi-source heterogeneous data through employing a federal learning framework, and generating time-space correlated construction site state representation data; based on a preset dynamic risk prediction model, risk prediction is carried out by using the construction site state representation data, a multi-level risk prediction result is output, and the preset dynamic risk prediction model is constructed based on a construction safety knowledge graph and a space-time diagram neural network; and triggering an adaptive feedback mechanism according to the risk level corresponding to the prediction result, generating visual early warning information and an equipment control instruction, and linking a construction site control system to execute emergency response operation. The problems that a traditional monitoring method is tedious in data processing, insufficient in real-time performance, high in cost, lack of prediction capacity and the like are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of safety monitoring, and in particular to an artificial intelligence-based construction safety monitoring method and system. Background Art

[0002] In modern building and infrastructure construction, construction safety monitoring is a critical step in ensuring the smooth progress of projects, protecting workers' lives, and reducing accident risks. Construction safety monitoring encompasses multiple aspects, including but not limited to environmental monitoring, structural health monitoring, and human behavior monitoring.

[0003] Traditional methods rely on manual inspections, sensor data collection, and manual data analysis. While these methods are effective, they also have many limitations:

[0004] Data processing is cumbersome: Large amounts of sensor data require manual screening and analysis, which is time-consuming and prone to errors.

[0005] Lack of real-time performance: Manual inspections cannot achieve real-time monitoring around the clock, which may result in safety hazards not being discovered in time.

[0006] High cost: Frequent manual inspections and complex data analysis increase the operating costs of the project.

[0007] Lack of predictive capabilities: Traditional methods make it difficult to provide early warning of potential risks, resulting in measures being taken only after an accident occurs.

[0008] Therefore, the present invention provides a construction safety monitoring method and system based on artificial intelligence to solve the above problems. Summary of the Invention

[0009] In view of the above situation, in order to overcome the shortcomings of the existing technology, the present invention provides a construction safety monitoring method and system based on artificial intelligence, which solves the problems of traditional monitoring methods such as cumbersome data processing, lack of real-time performance, high cost and lack of predictive ability.

[0010] In order to achieve the above object, the technical solution adopted by the present invention is:

[0011] In the first aspect, a construction safety monitoring method based on artificial intelligence is provided, the method comprising: acquiring multi-source heterogeneous data from a construction site, performing distributed feature extraction on the multi-source heterogeneous data using a federated learning framework, and generating spatiotemporally correlated construction site status representation data, wherein the multi-source heterogeneous data includes video images, environmental monitoring data, equipment status data, and personnel positioning data; performing risk prediction using the construction site status representation data based on a preset dynamic risk prediction model, and outputting multi-level risk prediction results, wherein the preset dynamic risk prediction model is constructed based on a construction safety knowledge graph and a spatiotemporal graph neural network; triggering an adaptive feedback mechanism according to the risk level corresponding to the prediction result, generating visual warning information and equipment control instructions, and linking the construction site control system to perform emergency response operations.

[0012] A further improvement of the present application is that the distributed feature extraction of multi-source heterogeneous data using a federated learning framework to generate spatiotemporally correlated construction site status representation data includes: obtaining local feature model parameters of each data source, wherein video data is subjected to a 3D convolutional neural network to extract spatial-motion features, environmental monitoring data is subjected to an LSTM network to extract temporal dependency features, equipment status data is subjected to an autoencoder to extract abnormal pattern features, and personnel positioning data is subjected to a graph attention network to extract group interaction features; a spatiotemporal fusion module is deployed in the federated learning server, the spatiotemporal resolution of multi-source features is aligned through deformable convolution, and a cross-modal attention mechanism is used to establish dynamic associations between features; feature parameters are encrypted and aggregated based on differential privacy technology to generate a global spatiotemporal correlation matrix as construction site status representation data.

[0013] A further improvement of the present application is that the construction of a preset dynamic risk prediction model based on the construction safety knowledge graph and the spatiotemporal graph neural network includes: constructing a construction safety knowledge graph, wherein the construction safety knowledge graph contains the association between construction element entities, safety specification constraints and accident cases, and mapping the entity relationship into a low-dimensional vector through graph embedding technology; determining the spatiotemporal graph neural network, and fusing real-time construction site status data with knowledge graph semantic information through a gated spatiotemporal propagation mechanism; establishing a risk propagation dynamics equation, combining the causal analysis results of historical accident data, and establishing a diffusion model of risk probability in the spatiotemporal dimension; integrating a cellular automaton simulator to deduce the risk propagation path based on neighborhood weights and risk accumulation effects.

[0014] A further improvement of the present application is that the expression of the dynamic risk prediction model is:

[0015] (1), in expression (1), Represents the risk probability value at the spatiotemporal position (t,s), and its value range is [0,1]. Represents a spatiotemporal graph neural network, which is used to output the hidden state of the fusion of real-time data and knowledge graph. represents a cellular automaton simulator used to predict risk propagation paths, Represents a multi-granularity risk perception module, which is used to comprehensively output device-level, regional-level, and global-level situation prediction results.

[0016] A further improvement of the present application is that the adaptive feedback mechanism includes: constructing a mapping rule base between risk levels and response strategies, the rule base containing three levels of risk thresholds; constructing a distributed control instruction verification mechanism, verifying the legitimacy of the instructions through blockchain smart contracts, and then synchronously issuing them to the tower crane control system, power safety module and personnel evacuation guidance system; updating the emergency response log in real time, the log including the risk trigger time, instruction execution status and operator identification, and storing the log data through a distributed database.

[0017] A further improvement of the present application is that the three-level risk threshold of the rule base includes: when the predicted risk value exceeds the first threshold, an augmented reality visual warning is triggered, the risk area is highlighted in the construction worker's smart helmet, and a voice prompt is pushed; when the predicted risk value exceeds the second threshold, an equipment speed limit instruction and an area isolation instruction are generated to limit the operating speed of equipment in the dangerous area and isolate high-risk workstations through electronic fences; when the predicted risk value exceeds the third threshold, an equipment shutdown instruction and a personnel evacuation instruction are generated, the power supply to the dangerous equipment is simultaneously cut off, and the sound and light alarm of the evacuation guidance system is activated;

[0018] A further improvement of this application is to verify the legitimacy of control instructions through blockchain smart contracts, including: generating smart contract code based on predefined security policy templates and deploying it to private chain nodes; using asymmetric encryption technology to verify the instruction signature to ensure that the instruction source is credible and has not been tampered with; and synchronously sending the verified instructions to the tower crane control system, power safety module and personnel evacuation guidance system through a cross-chain protocol.

[0019] A further improvement of the present application is that the federated learning framework adopts a dynamic weighted aggregation strategy to calculate the dynamic fusion weight of each node parameter according to the spatiotemporal coverage completeness, feature update frequency and noise level of the data source.

[0020] Secondly, an artificial intelligence-based construction safety monitoring system is used to implement the above-mentioned method. The monitoring system includes: a multi-source data acquisition module for acquiring video images, environmental monitoring data, equipment status data and personnel positioning data; a federated learning feature extraction module for distributed generation of spatiotemporal-related construction site status representation data; a dynamic risk prediction module that integrates construction safety knowledge graphs, spatiotemporal graph neural networks and cellular automaton simulators to output multi-level risk prediction results; and an adaptive feedback control module for generating visual warning information and equipment control instructions, and jointly executing emergency response operations.

[0021] A further improvement of the present application is that the dynamic risk prediction module includes: a multi-granularity risk perception sub-module, which includes a device-level anomaly detection layer, a regional-level risk assessment layer, and a global-level situation prediction layer; a risk evolution deduction sub-module, which dynamically simulates the risk propagation path based on the principle of cellular automata, and displays the deduction results through a visual interface.

[0022] The beneficial effects of this invention include: by integrating multi-source heterogeneous data and utilizing a federated learning framework for distributed feature extraction, the efficiency and accuracy of data processing are effectively improved. A dynamic risk prediction model based on a construction safety knowledge graph and a spatiotemporal graph neural network not only enhances prediction capabilities but also provides timely warnings of potential risks, significantly reducing the likelihood of accidents. The introduction of an adaptive feedback mechanism enables the system to automatically initiate appropriate emergency response measures based on prediction results, ensuring the swiftness and effectiveness of emergency response. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 This is a schematic flow chart of a construction safety monitoring method based on artificial intelligence according to the present invention;

[0024] Figure 2 This is a schematic structural diagram of an artificial intelligence-based construction safety monitoring system of the present invention. DETAILED DESCRIPTION

[0025] The following will describe various embodiments of the present invention in detail with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only used to explain the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0026] Traditional methods rely on manual inspections, sensor data collection, and manual data analysis. While these methods are effective, they also have many limitations:

[0027] Data processing is cumbersome: Large amounts of sensor data require manual screening and analysis, which is time-consuming and prone to errors.

[0028] Lack of real-time performance: Manual inspections cannot achieve real-time monitoring around the clock, which may result in safety hazards not being discovered in time.

[0029] High cost: Frequent manual inspections and complex data analysis increase the operating costs of the project.

[0030] Lack of predictive capabilities: Traditional methods make it difficult to provide early warning of potential risks, resulting in measures being taken only after an accident occurs.

[0031] Therefore, the present invention provides an artificial intelligence-based construction safety monitoring method and system to solve the above problems.

[0032] The technical solution will be described in detail below in conjunction with specific embodiments.

[0033] Example

[0034] refer to Figure 1 , a construction safety monitoring method based on artificial intelligence, the method comprising the following steps S100-S300:

[0035] S100, acquiring multi-source heterogeneous data from the construction site, performing distributed feature extraction on the multi-source heterogeneous data using a federated learning framework, and generating spatiotemporally correlated construction site status representation data, wherein the multi-source heterogeneous data includes video images, environmental monitoring data, equipment status data, and personnel location data;

[0036] S200, based on a preset dynamic risk prediction model, using the construction site status representation data to perform risk prediction and output a multi-level risk prediction result, wherein the preset dynamic risk prediction model is constructed based on a construction safety knowledge graph and a spatiotemporal graph neural network;

[0037] S300. Triggering an adaptive feedback mechanism according to the risk level corresponding to the prediction result, generating visual warning information and equipment control instructions, and linking the construction site control system to execute emergency response operations.

[0038] In step S100, the distributed feature extraction of multi-source heterogeneous data using a federated learning framework to generate spatiotemporally correlated construction site status representation data includes:

[0039] S101. Obtain local feature model parameters for each data source, where video data is extracted using a 3D convolutional neural network to extract spatial-motion features, environmental monitoring data is extracted using an LSTM network to extract temporal dependency features, device status data is extracted using an autoencoder to extract abnormal pattern features, and personnel positioning data is extracted using a graph attention network to extract group interaction features;

[0040] S102. Deploy a spatiotemporal fusion module in the federated learning server to align the spatiotemporal resolution of multi-source features through deformable convolution and establish dynamic associations between features using a cross-modal attention mechanism.

[0041] S103. Encrypt and aggregate the characteristic parameters based on differential privacy technology to generate a global spatiotemporal correlation matrix as the construction site status representation data.

[0042] In step S100, the federated learning framework adopts a dynamic weighted aggregation strategy to calculate the dynamic fusion weight of each node parameter according to the spatiotemporal coverage completeness, feature update frequency and noise level of the data source.

[0043] In step S200, the construction of a preset dynamic risk prediction model based on the construction safety knowledge graph and the spatiotemporal graph neural network includes:

[0044] S201. Construct a construction safety knowledge graph, wherein the construction safety knowledge graph includes associations between construction element entities, safety specification constraints, and accident cases, and maps entity relationships into low-dimensional vectors using graph embedding technology;

[0045] S202. Determine a spatiotemporal graph neural network and fuse real-time construction site status data with knowledge graph semantic information through a gated spatiotemporal propagation mechanism;

[0046] S203. Establish a risk propagation dynamics equation, combine the causal analysis results of historical accident data, and establish a risk probability diffusion model in the time and space dimensions;

[0047] S204. Integrate the cellular automaton simulator to deduce the risk propagation path based on neighborhood weights and risk accumulation effects.

[0048] In one embodiment of the present application, the expression of the dynamic risk prediction model is:

[0049] (1),

[0050] In expression (1), Represents the risk probability value at the spatiotemporal position (t,s), and its value range is [0,1]. Represents a spatiotemporal graph neural network, which is used to output the hidden state of the fusion of real-time data and knowledge graph. represents a cellular automaton simulator used to predict risk propagation paths, Represents a multi-granularity risk perception module, which is used to comprehensively output device-level, regional-level, and global-level situation prediction results.

[0051] Specifically, the dynamic inference layer of the spatiotemporal graph neural network consists of gated spatiotemporal convolution and multi-headed spatiotemporal Transformer, and its expression is:

[0052] (2),

[0053] In expression (2), represents the state representation of the construction site at time t (from the spatiotemporal association matrix of federated learning), represents the knowledge graph embedding vector (generated by the TransE algorithm), Represents gated spatiotemporal convolution, the formula is , where ⊙ represents element-by-element multiplication, which is used to dynamically adjust the feature importance; represents a multi-head spatiotemporal attention encoder, whose attention weights are calculated as: ,in, They are generated by spatiotemporal position encoding respectively; represents the activation function (such as ReLU), is the trainable weight matrix;

[0054] Among them, the state transition rule of cellular automaton is defined as follows:

[0055] (3),

[0056] In expression (3), represents the risk value of cell s at time t, represents the neighborhood of cell s (such as adjacent equipment or workstations), Represents the neighborhood weight, which is dynamically calculated by the entity association strength in the knowledge graph: , where Sim is the cosine similarity, represents the bias term, which is used to adjust the sensitivity of risk accumulation;

[0057] Multi-granularity risk perception module , output risk results in three levels:

[0058] (4),

[0059] In expression (4), device-level anomaly detection is:

[0060] ;

[0061] The regional level risk assessment is:

[0062] ;

[0063] The global situation forecast is:

[0064] ;

[0065] Based on the partial differential equation to quantify risk diffusion, the expression of the risk propagation dynamics equation is:

[0066] (5),

[0067] In expression (5), represents the spatial diffusion coefficient (determined by the layout of the construction area), represents the risk self-enhancement factor (derived from the chain reaction analysis of historical accidents), Indicates the environmental load threshold (matching the security specification constraint), Represents the risk attenuation coefficient (reflecting the efficiency of emergency response).

[0068] In step S300, the adaptive feedback mechanism includes:

[0069] S301, building a mapping rule base between risk levels and response strategies, wherein the rule base includes three levels of risk thresholds;

[0070] S302: Build a distributed control command verification mechanism. After verifying the legitimacy of the command through the blockchain smart contract, it will be simultaneously issued to the tower crane control system, power safety module, and personnel evacuation guidance system;

[0071] S303 updates the emergency response log in real time. The log includes the risk triggering time, instruction execution status and operator identification, and stores the log data through a distributed database.

[0072] In step S301, the three-level risk threshold of the rule base includes:

[0073] When the predicted risk value exceeds the first threshold, an augmented reality visual warning is triggered, the risk area is highlighted in the construction worker's smart helmet, and a voice prompt is pushed;

[0074] When the predicted risk value exceeds the second threshold, a speed limit instruction and an area isolation instruction are generated to limit the operating speed of equipment in the dangerous area and isolate high-risk workstations through electronic fences.

[0075] When the predicted risk value exceeds the third threshold, an equipment shutdown instruction and a personnel evacuation instruction are generated, the power supply to the dangerous equipment is cut off simultaneously, and the sound and light alarm of the evacuation guidance system is activated;

[0076] In step 302, the legitimacy of the control instruction is verified through the blockchain smart contract, including:

[0077] Generate smart contract code based on predefined security policy templates and deploy it to private chain nodes;

[0078] Use asymmetric encryption technology to verify the instruction signature to ensure that the instruction source is credible and has not been tampered with;

[0079] Through the cross-chain protocol, the verified instructions are synchronously sent to the tower crane control system, power safety module and personnel evacuation guidance system.

[0080] To facilitate understanding of the above method, an example is given below:

[0081] At a construction site, in the tower crane operating area (labeled as cell s0), sensors detected abnormal vibrations in the crane's boom, exceeding a preset safety threshold. The system acquired this equipment status data in real time through a multi-source data acquisition module and initiated a dynamic risk prediction process.

[0082] 1. Data Flow and Preprocessing

[0083] Data collection:

[0084] Equipment status data: The tower crane vibration sensor recorded a vibration frequency of 25 Hz (the safety threshold is 20 Hz), and the health score dropped to 0.6 (out of a maximum score of 1).

[0085] Video data: The camera captures an abnormal increase in the swing amplitude of the tower crane hook. The spatial-motion features are extracted through a 3D convolutional neural network, and the event label "lifting imbalance" is generated.

[0086] Environmental monitoring data: The wind speed sensor shows that the instantaneous wind speed is 8m / s (the safety upper limit is 6m / s).

[0087] Preprocessing:

[0088] Spatiotemporal alignment: All data are aligned by timestamp, and features are extracted through the local feature model of the federated learning framework (e.g., abnormal pattern features of vibration data are extracted through autoencoders).

[0089] Normalizing Input: Generating a Spatiotemporal Correlation Matrix , including the fusion representation of device status, environmental data and video features.

[0090] 2. Dynamic Risk Prediction Model Operation

[0091] (a) Spatiotemporal Graph Neural Network Reasoning

[0092] enter:

[0093] -Real-time site status representation (From Federated Learning).

[0094] -Knowledge Graph Embedding : Query the related entities of "tower crane overload accident" in the knowledge graph, including the causal chain of historical accidents (such as "overload → structural fatigue → fracture"), and map them into low-dimensional vectors.

[0095] Calculation process:

[0096] -Gated spatiotemporal convolution: Fusion of vibration data with the “overload risk” model in the knowledge graph to output hidden states .

[0097] -Multi-head spatiotemporal Transformer: Captures the long-range dependencies between the tower crane area and surrounding workstations (such as the impact of wind speed on lifting stability) and generates an attention weight matrix.

[0098] - Comprehensive output: STGN outputs the hidden risk status at the current moment =0.8 (equipment-level risk probability).

[0099] (b) Cellular Automata (CA) risk propagation simulation

[0100] -Neighborhood definition: the neighborhood of cell s0 Including the adjacent steel bar processing area (s1) and personnel passage (s2).

[0101] -Weight calculation:

[0102] -Calculate neighborhood weights based on the strength of entity associations in the knowledge graph:

[0103] (Tower crane is associated with equipment in the steel bar processing area) (Relevance of personnel passage safety)

[0104] - Bias term b = 0.1 (reflecting the additional risk of the current wind speed).

[0105] -State transfer:

[0106]

[0107] This indicates that the risk value in the tower crane area has slightly dropped from 0.8 to 0.75, but is still in the high-risk range.

[0108] (c) Risk propagation dynamics equation

[0109] -Parameter settings:

[0110] =0.2 (the tower crane area is an open space, and the risk spreads quickly);

[0111] =0.5 (historical data shows that overloading accidents are likely to trigger chain reactions);

[0112] =0.8 (safety regulations limit the maximum risk bearing value of the tower crane area);

[0113] =0.3 (moderate emergency response efficiency).

[0114] - Diffusion prediction:

[0115] By solving the partial differential equation , it is predicted that the risk will spread to the steel bar processing area within 10 minutes (the s1 risk value rises to 0.65).

[0116] 3. Multi-granularity risk perception and early warning triggering

[0117] -Device-level anomaly detection:

[0118] 0.8 (abnormal vibration), triggering a device-level alarm.

[0119] -Regional level risk assessment:

[0120] Extract the highest risk value in the cellular automation output through the maximum pooling operation 0.75, the tower crane area is judged to be high risk.

[0121] -Global situation forecast:

[0122] The LSTM model integrates device and regional risks to predict the global risk probability within the next 15 minutes. 0.7.

[0123] Warning trigger logic:

[0124] -First threshold (0.6): The AR helmet highlights the crane area and pushes a voice prompt: "Pay attention to hoisting safety."

[0125] - Second threshold (0.7):

[0126] -Issue equipment speed limit instructions to reduce the tower crane's operating speed to 50%.

[0127] -Activate the electronic fence to isolate the steel bar processing area (S1).

[0128] -Third threshold (0.8): Not triggered, but the system continues to monitor.

[0129] 4. Adaptive Feedback and Instruction Execution

[0130] - Blockchain smart contract verification:

[0131] -After the control instruction (speed limit, isolation) is generated, the legality of the instruction format is verified through the predefined smart contract template.

[0132] - Use RSA asymmetric encryption to sign the instructions to ensure that the instruction source is a trusted server.

[0133] -After verification, the instructions are synchronized to the tower crane control system and electronic fence module through the cross-chain protocol.

[0134] - Emergency Response Log Updates:

[0135] -Record event time and command execution status (such as "the crane speed has dropped to 50%").

[0136] -Log data is stored in a distributed database and marked as incremental learning samples.

[0137] 5. Model Optimization and Closed-Loop Feedback

[0138] -Incremental learning:

[0139] The event data (abnormal vibration, excessive wind speed, and warning results) are stored in the memory playback buffer.

[0140] Model parameters are updated through the elastic weight consolidation (EWC) algorithm, retaining key weights of historical accident patterns (such as overload risk).

[0141] -Dynamic adjustment of risk thresholds:

[0142] Based on the emergency response delay time in the log (for example, it takes 2 seconds to send a command), the second threshold is automatically optimized to 0.68 to trigger isolation measures in advance.

[0143] Compared with the existing technology, this application proposes an artificial intelligence-based construction safety monitoring method that can realize the automation of risk prediction and emergency response, and significantly improve the safety management level of construction sites. By integrating real-time data and knowledge graphs through spatiotemporal graph neural networks, the system can accurately capture changes in site status and predict potential risks. Cellular automata simulate risk propagation paths, further enhancing the accuracy and foresight of early warnings. The multi-granularity risk perception module comprehensively outputs device-level, regional-level, and global-level situation prediction results, providing managers with a comprehensive risk view. The adaptive feedback mechanism triggers response strategies at different levels based on the predicted risk value, verifies the legitimacy of instructions through blockchain smart contracts, and ensures the reliable execution of instructions. This method not only improves the efficiency and accuracy of safety monitoring, but also promotes the digital transformation and intelligent upgrading of the construction industry.

[0144] refer to Figure 2 , an artificial intelligence-based construction safety monitoring system for implementing the above-mentioned method, the monitoring system comprises:

[0145] Multi-source data acquisition module, used to obtain video images, environmental monitoring data, equipment status data and personnel positioning data;

[0146] Federated learning feature extraction module, used for distributed generation of spatiotemporally correlated construction site status representation data;

[0147] Dynamic risk prediction module, integrating construction safety knowledge graph, spatiotemporal graph neural network and cellular automaton simulator, outputs multi-level risk prediction results;

[0148] The adaptive feedback control module is used to generate visual warning information and equipment control instructions, and to jointly execute emergency response operations.

[0149] In one embodiment of the present application, the dynamic risk prediction module includes:

[0150] Multi-granularity risk perception submodule, including device-level anomaly detection layer, regional-level risk assessment layer, and global-level situation prediction layer;

[0151] The risk evolution deduction sub-module dynamically simulates the risk propagation path based on the principle of cellular automaton and displays the deduction results through a visual interface.

[0152] Various embodiments of the systems and techniques described above can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), system-on-chip systems (SOCs), programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include being implemented in one or more computer programs that are executable and / or interpreted on a programmable system that includes at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0153] The program code for implementing the method of the present disclosure can be written in any combination of one or more programming languages. These program codes can be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device so that when the program code is executed by the processor or controller, the functions / operations specified in the flow chart and / or block diagram are implemented. The program code can be executed entirely on the machine, partially on the machine, as a stand-alone software package, partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0154] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of machine-readable storage media may include an electrical connection based on one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), optical fibers, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0155] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0156] The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network (LAN), a wide area network (WAN), and the Internet.

[0157] A computer system may include a client and a server. The client and server are generally remote from each other and typically interact through a communication network. The client-server relationship arises through computer programs running on the respective computers and having a client-server relationship with each other. The server may be a cloud server, a server in a distributed system, or a server integrated with a blockchain.

[0158] It should be understood that the various forms of the processes shown above can be used to reorder, add, or delete steps. For example, the steps described in this disclosure can be performed in parallel, sequentially, or in a different order, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved. This is not a limitation herein.

[0159] The above specific embodiments do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions may be made based on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure shall be included within the scope of protection of this disclosure.

Claims

1. A construction safety monitoring method based on artificial intelligence, characterized in that: The method comprises: Acquire multi-source heterogeneous data from the construction site, including video images, environmental monitoring data, equipment status data, and personnel location data, and perform distributed feature extraction on the data using a federated learning framework to generate spatiotemporally correlated site status representation data. Based on a preset dynamic risk prediction model, the construction site status representation data is used to perform risk prediction and output a multi-level risk prediction result, wherein the preset dynamic risk prediction model is constructed based on a construction safety knowledge graph and a spatiotemporal graph neural network; An adaptive feedback mechanism is triggered according to the risk level corresponding to the prediction result, visual warning information and equipment control instructions are generated, and the construction site control system is linked to perform emergency response operations.

2. The artificial intelligence-based construction safety monitoring method according to claim 1 is characterized in that: The federated learning framework is used to perform distributed feature extraction on multi-source heterogeneous data to generate spatiotemporally correlated construction site status representation data, including: Obtain local feature model parameters for each data source. For video data, use a 3D convolutional neural network to extract spatial-motion features. For environmental monitoring data, use an LSTM network to extract temporal dependency features. For device status data, use an autoencoder to extract abnormal pattern features. For personnel positioning data, use a graph attention network to extract group interaction features. Deploy a spatiotemporal fusion module in the federated learning server, align the spatiotemporal resolution of multi-source features through deformable convolution, and use a cross-modal attention mechanism to establish dynamic associations between features; Based on differential privacy technology, the feature parameters are encrypted and aggregated to generate a global spatiotemporal correlation matrix as the site status representation data.

3. The artificial intelligence-based construction safety monitoring method according to claim 1 is characterized in that: The preset dynamic risk prediction model based on the construction safety knowledge graph and spatiotemporal graph neural network includes: Construct a construction safety knowledge graph that includes the relationships between construction element entities, safety regulation constraints, and accident cases, and maps entity relationships into low-dimensional vectors using graph embedding technology; Determine the spatiotemporal graph neural network and fuse real-time construction site status data with knowledge graph semantic information through a gated spatiotemporal propagation mechanism; Establish a risk propagation dynamics equation, combine the causal analysis results of historical accident data, and establish a risk probability diffusion model in the time and space dimensions; An integrated cellular automaton simulator is used to deduce the risk propagation path based on neighborhood weights and risk accumulation effects.

4. The artificial intelligence-based construction safety monitoring method according to claim 1 is characterized in that: The expression of the dynamic risk prediction model is: （1）， In expression (1), Represents the risk probability value at the spatiotemporal position (t,s), and its value range is [0,1]. Represents a spatiotemporal graph neural network, which is used to output the hidden state of the fusion of real-time data and knowledge graph. represents a cellular automaton simulator used to predict risk propagation paths, Represents a multi-granularity risk perception module, which is used to comprehensively output device-level, regional-level, and global-level situation prediction results.

5. The construction safety monitoring method based on artificial intelligence according to claim 1 is characterized in that: The adaptive feedback mechanism includes: Building a mapping rule base between risk levels and response strategies, the rule base including three levels of risk thresholds; Build a distributed control command verification mechanism. After verifying the legitimacy of the command through blockchain smart contracts, it will be simultaneously issued to the tower crane control system, power safety module, and personnel evacuation guidance system. The emergency response log is updated in real time. The log includes the risk trigger time, instruction execution status and operator identification, and the log data is stored in a distributed database.

6. The artificial intelligence-based construction safety monitoring method according to claim 5 is characterized in that: The three-level risk thresholds of the rule base include: When the predicted risk value exceeds the first threshold, an augmented reality visual warning is triggered, the risk area is highlighted in the construction worker's smart helmet, and a voice prompt is pushed; When the predicted risk value exceeds the second threshold, a speed limit instruction and an area isolation instruction are generated to limit the operating speed of equipment in the dangerous area and isolate high-risk workstations through electronic fences. When the predicted risk value exceeds the third threshold, an equipment shutdown instruction and a personnel evacuation instruction are generated, the power supply to the dangerous equipment is cut off simultaneously, and the sound and light alarm of the evacuation guidance system is activated.

7. The artificial intelligence-based construction safety monitoring method according to claim 5 is characterized in that: Verify the legitimacy of control instructions through blockchain smart contracts, including: Generate smart contract code based on predefined security policy templates and deploy it to private chain nodes; Use asymmetric encryption technology to verify the instruction signature to ensure that the instruction source is credible and has not been tampered with; Through the cross-chain protocol, the verified instructions are synchronously sent to the tower crane control system, power safety module and personnel evacuation guidance system.

8. The artificial intelligence-based construction safety monitoring method according to claim 2 is characterized in that: The federated learning framework adopts a dynamic weighted aggregation strategy to calculate the dynamic fusion weight of each node parameter according to the spatiotemporal coverage completeness, feature update frequency and noise level of the data source.

9. An artificial intelligence-based construction safety monitoring system for implementing the method according to any one of claims 1 to 8, characterized in that: The monitoring system comprises: Multi-source data acquisition module, used to obtain video images, environmental monitoring data, equipment status data and personnel positioning data; Federated learning feature extraction module, used for distributed generation of spatiotemporally correlated construction site status representation data; Dynamic risk prediction module, integrating construction safety knowledge graph, spatiotemporal graph neural network and cellular automaton simulator, outputs multi-level risk prediction results; The adaptive feedback control module is used to generate visual warning information and equipment control instructions, and to jointly execute emergency response operations.

10. The artificial intelligence-based construction safety monitoring system according to claim 9 is characterized in that: The dynamic risk prediction module includes: Multi-granularity risk perception submodule, including device-level anomaly detection layer, regional-level risk assessment layer, and global-level situation prediction layer; The risk evolution deduction sub-module dynamically simulates the risk propagation path based on the principle of cellular automaton and displays the deduction results through a visual interface.

Images