Tunnel construction safety early warning method and system based on multi-dimensional data processing

Abstract

The invention discloses a tunnel construction safety early warning method and system based on multi-dimensional data processing, and the method comprises the steps: obtaining a historical disaster situation data set composed of various disaster type data, and training a disaster type recognition model according to the historical disaster situation data set; a plurality of sensors used for acquiring information in a tunnel in real time are arranged in the tunnel, a tunnel BIM model is constructed, and the information in the tunnel is mapped into the tunnel BIM model; inputting information, obtained in real time, in the tunnel into the disaster type identification model, identifying the types of disasters possibly occurring in the tunnel, and mapping the types of disasters to the tunnel BIM model; and calculating a real-time risk value of each position in the tunnel according to the information in the tunnel, marking and connecting the positions corresponding to the real-time risk values meeting a preset condition to form an escape path, mapping the escape path to the tunnel BIM model, and displaying the tunnel BIM model on a virtual reality device in real time.

Description

Technical Field

[0001] This invention belongs to the field of tunnel construction safety early warning technology, and more specifically, relates to a tunnel construction safety early warning method and system based on multi-dimensional data processing. Background Technology

[0002] Various types of disasters can occur inside tunnels, mainly including water and mud inrushes, collapses, rock bursts, gas leaks, fires, structural cracks and deformations, and lining detachment. These disasters can occur during the construction phase or accumulate and be triggered during operation, characterized by their suddenness, difficulty in early warning, and severe consequences. Water and mud inrushes are usually caused by high permeability of the surrounding rock, high groundwater pressure, and inadequate sealing of the tunnel face, easily leading to equipment damage and casualties. Collapses are mostly related to poor stability of the surrounding rock, support failure, or disturbance, and are one of the most common tunnel disasters. Fires can be caused by mechanical and electrical failures or flammable transported materials, spreading rapidly and making escape difficult. Currently, there is no technical solution that can effectively provide early warning for the above-mentioned tunnel disasters. Summary of the Invention

[0003] To address the above technical problems, this invention proposes a tunnel construction safety early warning method based on multidimensional data processing, comprising:

[0004] Obtain a historical disaster dataset consisting of data on various disaster types, and train a disaster type identification model based on the historical disaster dataset;

[0005] Multiple sensors are installed inside the tunnel to acquire information in real time, and a tunnel BIM model is constructed to map the information inside the tunnel into the tunnel BIM model.

[0006] The real-time information acquired inside the tunnel is input into the disaster type identification model to identify the types of disasters that may occur inside the tunnel and map them into the tunnel BIM model;

[0007] Based on the information inside the tunnel, the real-time risk value of each location inside the tunnel is calculated. Locations corresponding to the real-time risk values ​​that meet preset conditions are marked and connected to form an escape path, which is then mapped into the tunnel BIM model. The tunnel BIM model is then displayed in real time on a virtual reality device.

[0008] Furthermore, the disaster type identification model adopts an LSTM or Transformer structure, and the historical disaster data corresponding to each disaster type is input into the LSTM or Transformer structure to complete the training of the disaster type identification model.

[0009] Furthermore, the disaster type identification model is as follows:

[0010]

[0011] Among them, ∏ k (t) represents the predicted value of the k-th type of disaster occurring at time t, where n is the number of information variables, and ω i，k Let σ be the weight of the influence of the i-th information variable on the k-th type of disaster. i (t) represents the i-th information variable in the tunnel information at time t, γ i，k Let χ be the response factor of the i-th information variable in the k-th type of disaster. i (t) is the state evolution factor of the i-th information variable at time t, λ i，k Let δ be the anomalous response suppression coefficient of the i-th information variable to the k-th type of disaster. i (t) represents the offset of the i-th information variable at time t from the corresponding historical average information variable.

[0012] Furthermore, it also includes: obtaining the corresponding emergency plan based on the disaster type, extracting entities from the emergency plan, constructing a knowledge graph triple consisting of (disaster type, recommended measures, applicable conditions), and displaying it in real time on a virtual reality device.

[0013] Furthermore, the preset conditions include: setting a risk threshold, marking and connecting the positions corresponding to the real-time risk values ​​that are less than the risk threshold to form an escape path.

[0014] Furthermore, the location of construction workers is monitored in real time. When a construction worker moves to the next location, the real-time risk value of all other locations is updated and compared with the risk threshold. Locations with real-time risk values ​​exceeding the risk threshold are excluded, and escape routes are reconstructed. This process is repeated until the construction worker escapes from the tunnel.

[0015] This invention also proposes a tunnel construction safety early warning system based on multidimensional data processing, comprising:

[0016] The model training module is used to acquire a historical disaster dataset consisting of data on various disaster types, and to train a disaster type identification model based on the historical disaster dataset.

[0017] A BIM model module is constructed to install various sensors inside the tunnel for real-time acquisition of tunnel information, and to construct a tunnel BIM model, mapping the tunnel information into the tunnel BIM model.

[0018] The disaster type identification module is used to input real-time information about the tunnel into the disaster type identification model, identify the types of disasters that may occur in the tunnel, and map them into the tunnel BIM model.

[0019] The risk avoidance module is used to calculate the real-time risk value of each location in the tunnel based on the information inside the tunnel, mark and connect the locations corresponding to the real-time risk values ​​that meet preset conditions to form an escape path, map it into the tunnel BIM model, and display the tunnel BIM model in real time on a virtual reality device.

[0020] Furthermore, the disaster type identification model adopts an LSTM or Transformer structure, and the historical disaster data corresponding to each disaster type is input into the LSTM or Transformer structure to complete the training of the disaster type identification model.

[0021] Furthermore, the disaster type identification model is as follows:

[0022]

[0023] Among them, Π k (t) represents the predicted value of the k-th type of disaster occurring at time t, where n is the number of information variables, and ω i，k Let σ be the weight of the influence of the i-th information variable on the k-th type of disaster. i (t) represents the i-th information variable in the tunnel information at time t, γ i，k Let χ be the response factor of the i-th information variable in the k-th type of disaster. i (t) is the state evolution factor of the i-th information variable at time t, λ i，k Let δ be the anomalous response suppression coefficient of the i-th information variable to the k-th type of disaster. i (t) represents the offset of the i-th information variable at time t from the corresponding historical average information variable.

[0024] Furthermore, it also includes: obtaining the corresponding emergency plan based on the disaster type, extracting entities from the emergency plan, constructing a knowledge graph triple consisting of (disaster type, recommended measures, applicable conditions), and displaying it in real time on a virtual reality device.

[0025] In summary, the technical solutions conceived by this invention have the following beneficial effects compared with the prior art:

[0026] Through the above technical solutions, this invention can accurately predict the types of disasters that may occur in tunnels, provide emergency plans, and provide escape routes for construction workers. Construction workers can view relevant information in real time through virtual reality devices, thereby improving the safety of construction in tunnels. Attached Figure Description

[0027] Figure 1 This is a flowchart of the method in Embodiment 1 of the present invention;

[0028] Figure 2 This is a system structure diagram of Embodiment 2 of the present invention. Detailed Implementation

[0029] To better understand the above technical solutions, the following will provide a detailed explanation of the technical solutions in conjunction with the accompanying drawings and specific implementation methods.

[0030] The method provided by this invention can be implemented in a terminal environment that may include one or more of the following components: a processor, a storage medium, and a display screen. The storage medium stores at least one instruction, which is loaded and executed by the processor to implement the method described in the following embodiments.

[0031] A processor may include one or more processing cores. The processor uses various interfaces and lines to connect various parts of the terminal, and performs various functions and processes data by running or executing instructions, programs, code sets or instruction sets stored in the storage medium, and by calling data stored in the storage medium.

[0032] Storage media can include random access memory (RAM) or read-only memory (ROM). Storage media can be used to store instructions, programs, code, code sets, or instructions.

[0033] The display screen is used to show the user interface of each application.

[0034] In addition, those skilled in the art will understand that the structure of the terminal described above does not constitute a limitation on the terminal. The terminal may include more or fewer components, or combine certain components, or have different component arrangements. For example, the terminal may also include radio frequency circuits, input units, sensors, audio circuits, power supplies, and other components, which will not be described in detail here.

[0035] Example 1

[0036] like Figure 1 As shown, this embodiment proposes a tunnel construction safety early warning method based on multi-dimensional data processing, including:

[0037] Step 101: Obtain a historical disaster dataset consisting of data on various disaster types, and train a disaster type identification model based on the historical disaster dataset;

[0038] Preferably, the historical disaster dataset can be text, image and / or structured data, covering typical types such as sudden water and mud inrush, surrounding rock collapse, and fire;

[0039] Specifically, the disaster type identification model adopts an LSTM or Transformer structure, and the historical disaster data corresponding to each disaster type is input into the LSTM or Transformer structure to complete the training of the disaster type identification model.

[0040] Alternatively, this embodiment also designs a disaster type identification model, specifically as follows:

[0041]

[0042] Among them, Π k (t) represents the predicted value (between 0 and 1) for the occurrence of the k-th type of disaster at time t, where n is the number of information variables, and ω i，k Let σ be the weight of the influence of the i-th information variable on the k-th type of disaster. i (t) represents the i-th information variable in the tunnel information at time t, γ i，k Let χ be the response factor of the i-th information variable in the k-th type of disaster. i (t) is the state evolution factor of the i-th information variable at time t, λ i，k Let δ be the anomalous response suppression coefficient of the i-th information variable to the k-th type of disaster. i (t) represents the offset of the i-th information variable at time t from the corresponding historical average information variable.

[0043] Preferably, γ i，k λ is used to control the sensitivity threshold of variables to disasters. For example, in a fire, the temperature must rise to a critical point before the corresponding disaster potential is activated. i，k It is used to control whether a variable is just a short-term fluctuation. For example, a sudden small fluctuation in water pressure is not the same as a sudden water outburst. It must deviate continuously to activate the corresponding disaster potential.

[0044] Preferably, the disaster type identification model is trained using the historical disaster dataset, and the response factor γ of the i-th information variable in the k-th type of disaster is determined based on the actual disaster type corresponding to the historical disaster dataset. i，k The weight ω of the influence of the i-th information variable on the k-th type of disaster i,k and the anomalous response suppression coefficient λ of the i-th information variable to the k-th type of disaster i，k The fitting process can be performed using either the ant colony algorithm or the gradient descent method.

[0045] Preferably, this embodiment provides the following calculation method to calculate the state evolution factor χ of the i-th information variable at time t. i (t), as shown below:

[0046]

[0047] Where β1 is the first weight of the state evolution factor, ρ1 is the first adjustment factor of the state evolution factor, β2 is the second weight of the state evolution factor, and Var(σ) i (t-θ: t)) represents the variance of the i-th information variable over the past time θ, ρ2 is the second adjustment factor of the state evolution factor, β3 is the third weight of the state evolution factor, and ρ3 is the third adjustment factor of the state evolution factor.

[0048] Regarding the meaning of each part in the above formula, this embodiment provides subscripts for explanation:

[0049]

[0050] Similarly, the state evolution factor χ of the i-th information variable at time t is calculated. i Substituting the formula (t) into the disaster type identification model above, we can calculate the state evolution factor χ of the i-th information variable at time t. i The formula for (t) is fitted with each weight and adjustment factor.

[0051] Step 102: Install various sensors inside the tunnel to acquire information inside the tunnel in real time, and construct a tunnel BIM model to map the information inside the tunnel into the tunnel BIM model;

[0052] Preferably, regarding the sensor setup, this embodiment designs stress / temperature / water pressure / smoke detection points at the arch, tunnel face, floor, and ventilation openings, respectively, deploys sensor data access gateways, and synchronizes them in a unified time. The above sensor setup is just an example given in this embodiment. For example, the sensor setup can be based on distance steps, and this embodiment does not impose any restrictions. In addition, this embodiment normalizes all information within the tunnel.

[0053] Step 103: Input the real-time acquired information inside the tunnel into the disaster type identification model to identify the types of disasters that may occur inside the tunnel and map them into the tunnel BIM model;

[0054] To make the disaster type identification model clearer, this embodiment provides the following example of the predicted value ∏ for a disaster of type k occurring at time t. k (t) is explained as follows:

[0055] Assuming the monitored information within the tunnel includes: σ1(t): deformation rate of the surrounding rock at time t, σ2(t): seepage pressure at time t, σ3(t): temperature and humidity changes at time t, and σ4(t): gas content at time t, then:

[0056]

[0057] Step 104: Based on the information inside the tunnel, calculate the real-time risk value of each location inside the tunnel, mark and connect the locations corresponding to the real-time risk values ​​that meet the preset conditions to form an escape path, map it into the tunnel BIM model, and display the tunnel BIM model in real time on a virtual reality device.

[0058] Preferably, in this embodiment, the real-time risk value at each location within the tunnel is calculated using the following formula:

[0059]

[0060] Where γ′(t) is the real-time risk value at the current position and time t, n′ is the number of state evolution factors, and ρ′ k′ χ is the weight of the rate of change of the k′-th state evolution factor. k′ (t) is the k′-th state evolution factor at time t, K k′ χ is the weight of the integral of the k′-th state evolution factor. k′ (τ) is the k′-th state evolution factor at time τ, Θ′ c This is the disaster trigger threshold.

[0061] Specifically, the preset conditions include: setting a risk threshold, marking and connecting the positions corresponding to the real-time risk values ​​that are less than the risk threshold to form an escape path.

[0062] Specifically, the location of construction workers is monitored in real time. When a construction worker moves to the next location, the real-time risk value of all other locations is updated and compared with the risk threshold. Locations with real-time risk values ​​exceeding the risk threshold are excluded, and escape routes are reconstructed. This process is repeated until the construction worker escapes from the tunnel.

[0063] Specifically, it also includes: obtaining the corresponding emergency plan based on the disaster type, extracting entities from the emergency plan, constructing a knowledge graph triple consisting of (disaster type, recommended measures, applicable conditions), and displaying it in real time on a virtual reality device.

[0064] Example 2

[0065] like Figure 2 As shown, this embodiment proposes a tunnel construction safety early warning system based on multi-dimensional data processing, including:

[0066] The model training module is used to acquire a historical disaster dataset consisting of data on various disaster types, and to train a disaster type identification model based on the historical disaster dataset.

[0067] Preferably, the historical disaster dataset can be text, image and / or structured data, covering typical types such as sudden water and mud inrush, surrounding rock collapse, and fire;

[0068] Specifically, the disaster type identification model adopts an LSTM or Transformer structure, and the historical disaster data corresponding to each disaster type is input into the LSTM or Transformer structure to complete the training of the disaster type identification model.

[0069] Alternatively, this embodiment also designs a disaster type identification model, specifically as follows:

[0070]

[0071] Among them, Π k (t) represents the predicted value (between 0 and 1) for the occurrence of the k-th type of disaster at time t, where n is the number of information variables, and ω i，k Let σ be the weight of the influence of the i-th information variable on the k-th type of disaster. i (t) represents the i-th information variable in the tunnel information at time t, γ i，k Let X be the response factor of the i-th information variable in the k-th type of disaster. i (t) is the state evolution factor of the i-th information variable at time t, λ i，k Let δ be the anomalous response suppression coefficient of the i-th information variable to the k-th type of disaster. i (t) represents the offset of the i-th information variable at time t from the corresponding historical average information variable.

[0072] Preferably, γ i，k λ is used to control the sensitivity threshold of variables to disasters. For example, in a fire, the temperature must rise to a critical point before the corresponding disaster potential is activated. i，k It is used to control whether a variable is just a short-term fluctuation. For example, a sudden small fluctuation in water pressure is not the same as a sudden water outburst. It must deviate continuously to activate the corresponding disaster potential.

[0073] Preferably, the disaster type identification model is trained using the historical disaster dataset, and the response factor γ of the i-th information variable in the k-th type of disaster is determined based on the actual disaster type corresponding to the historical disaster dataset. i，k The weight ω of the influence of the i-th information variable on the k-th type of disaster i，k and the anomalous response suppression coefficient λ of the i-th information variable to the k-th type of disaster i，k The fitting process can be performed using either the ant colony algorithm or the gradient descent method.

[0074] Preferably, this embodiment provides the following calculation method to calculate the state evolution factor X of the i-th information variable at time t. i(t), as shown below:

[0075]

[0076] Where β1 is the first weight of the state evolution factor, ρ1 is the first adjustment factor of the state evolution factor, β2 is the second weight of the state evolution factor, and Var(σ) i (t-θ: t)) represents the variance of the i-th information variable over the past time θ, ρ2 is the second adjustment factor of the state evolution factor, β3 is the third weight of the state evolution factor, and ρ3 is the third adjustment factor of the state evolution factor.

[0077] Similarly, the state evolution factor χ of the i-th information variable at time t is calculated. i Substituting the formula (t) into the disaster type identification model above, we can calculate the state evolution factor χ of the i-th information variable at time t. i The formula for (t) is fitted with each weight and adjustment factor.

[0078] A BIM model module is constructed to install various sensors inside the tunnel for real-time acquisition of tunnel information, and to construct a tunnel BIM model, mapping the tunnel information into the tunnel BIM model.

[0079] Preferably, regarding the sensor setup, this embodiment designs stress / temperature / water pressure / smoke detection points at the arch, tunnel face, floor, and ventilation openings, respectively, deploys sensor data access gateways, and synchronizes them in a unified time. The above sensor setup is just an example given in this embodiment. For example, the sensor setup can be based on distance steps, and this embodiment does not impose any restrictions. In addition, this embodiment normalizes all information within the tunnel.

[0080] The disaster type identification module is used to input real-time information about the tunnel into the disaster type identification model, identify the types of disasters that may occur in the tunnel, and map them into the tunnel BIM model.

[0081] To make the disaster type identification model clearer, this embodiment provides the following example of the predicted value Π for a disaster of type k occurring at time t. k (t) is explained as follows:

[0082] The risk avoidance module is used to calculate the real-time risk value of each location in the tunnel based on the information inside the tunnel, mark and connect the locations corresponding to the real-time risk values ​​that meet preset conditions to form an escape path, map it into the tunnel BIM model, and display the tunnel BIM model in real time on a virtual reality device.

[0083] Preferably, in this embodiment, the real-time risk value at each location within the tunnel is calculated using the following formula:

[0084]

[0085] Where γ′(t) is the real-time risk value at the current position and time t, n′ is the number of state evolution factors, and ρ′ k′ χ is the weight of the rate of change of the k′-th state evolution factor. k′ (t) is the k′-th state evolution factor at time t, K k′ The weight of the integral of the k′-th state evolution factor, x k′ (τ) is the k′-th state evolution factor at time τ, Θ′ c This is the disaster trigger threshold.

[0086] Specifically, the preset conditions include: setting a risk threshold, marking and connecting the positions corresponding to the real-time risk values ​​that are less than the risk threshold to form an escape path.

[0087] Specifically, the location of construction workers is monitored in real time. When a construction worker moves to the next location, the real-time risk value of all other locations is updated and compared with the risk threshold. Locations with real-time risk values ​​exceeding the risk threshold are excluded, and escape routes are reconstructed. This process is repeated until the construction worker escapes from the tunnel.

[0088] Specifically, it also includes: obtaining the corresponding emergency plan based on the disaster type, extracting entities from the emergency plan, constructing a knowledge graph triple consisting of (disaster type, recommended measures, applicable conditions), and displaying it in real time on a virtual reality device.

[0089] Example 3

[0090] This invention also proposes a storage medium storing multiple instructions, which are used to implement the tunnel construction safety early warning method based on multidimensional data processing.

[0091] Optionally, in this embodiment, the storage medium may be located in any computer terminal in a group of computer terminals in a computer network, or in any mobile terminal in a group of mobile terminals.

[0092] Optionally, in this embodiment, the storage medium is configured to store program code for performing the method steps of Embodiment 1.

[0093] Example 4

[0094] This invention also proposes an electronic device, including a processor and a storage medium connected to the processor. The storage medium stores multiple instructions, which can be loaded and executed by the processor to enable the processor to execute the tunnel construction safety early warning method based on multidimensional data processing.

[0095] Specifically, the electronic device in this embodiment can be a computer terminal, which may include one or more processors and a storage medium.

[0096] The storage medium can be used to store software programs and modules, such as the tunnel construction safety early warning method based on multi-dimensional data processing in this embodiment of the invention. The corresponding program instructions / modules allow the processor to execute various functional applications and data processing by running the software programs and modules stored in the storage medium, thus realizing the aforementioned tunnel construction safety early warning method based on multi-dimensional data processing. The storage medium may include high-speed random access storage media, and may also include non-volatile storage media, such as one or more magnetic storage systems, flash memory, or other non-volatile solid-state storage media. In some instances, the storage medium may further include storage media remotely configured relative to the processor, which can be connected to the terminal via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0097] The processor can execute the method steps of Embodiment 1 by calling the information and application stored in the storage medium through the transmission system.

[0098] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0099] In the above embodiments of the present invention, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0100] In the several embodiments provided by this invention, it should be understood that the disclosed technical content can be implemented in other ways. The system embodiments described above are merely illustrative; for example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces, indirect coupling or communication connection between units or modules, and may be electrical or other forms.

[0101] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0102] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0103] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes: USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, optical disks, and other media capable of storing program code.

[0104] Obviously, the above embodiments are merely illustrative examples for clear explanation and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations here. However, obvious variations or modifications derived therefrom are still within the scope of protection of this invention.

Claims

1. A tunnel construction safety early warning method based on multidimensional data processing, characterized in that, include: Obtain a historical disaster dataset consisting of data on various disaster types, and train a disaster type identification model based on the historical disaster dataset; Multiple sensors are installed inside the tunnel to acquire information in real time, and a tunnel BIM model is constructed to map the information inside the tunnel into the tunnel BIM model. The real-time information acquired inside the tunnel is input into the disaster type identification model to identify the types of disasters that may occur inside the tunnel and map them into the tunnel BIM model; Based on the information inside the tunnel, the real-time risk value of each location inside the tunnel is calculated. Locations corresponding to the real-time risk values ​​that meet preset conditions are marked and connected to form an escape path, which is then mapped into the tunnel BIM model. The tunnel BIM model is then displayed in real time on a virtual reality device.

2. The tunnel construction safety early warning method based on multidimensional data processing as described in claim 1, characterized in that, The disaster type identification model adopts an LSTM or Transformer structure, and the historical disaster data corresponding to each disaster type is input into the LSTM or Transformer structure to complete the training of the disaster type identification model.

3. The tunnel construction safety early warning method based on multidimensional data processing as described in claim 1, characterized in that, The disaster type identification model is as follows: Among them, Π k (t) represents the predicted value of the k-th type of disaster occurring at time t, where n is the number of information variables, and ω i，k Let σ be the weight of the influence of the i-th information variable on the k-th type of disaster. i (t) represents the i-th information variable in the tunnel information at time t, γ i,k Let χ be the response factor of the i-th information variable in the k-th type of disaster. i (t) is the state evolution factor of the i-th information variable at time t, λ i,k Let δ be the anomalous response suppression coefficient of the i-th information variable to the k-th type of disaster. i (t) represents the offset of the i-th information variable at time t from the corresponding historical average information variable.

4. The tunnel construction safety early warning method based on multidimensional data processing as described in claim 1, characterized in that, Also includes: Based on the disaster type, obtain the corresponding emergency plan, extract the entities in the emergency plan, construct a knowledge graph triple consisting of (disaster type, recommended measures, applicable conditions), and display it in real time on a virtual reality device.

5. The tunnel construction safety early warning method based on multidimensional data processing as described in claim 1, characterized in that, The preset conditions include: setting a risk threshold, marking and connecting the positions corresponding to the real-time risk values ​​that are less than the risk threshold to form an escape path.

6. The tunnel construction safety early warning method based on multidimensional data processing as described in claim 5, characterized in that, The system monitors the location of construction workers in real time. When a worker moves to the next location, the system updates the real-time risk value of all other locations and compares it with the risk threshold. Locations with real-time risk values ​​exceeding the risk threshold are excluded. Escape routes are then reconstructed and the process is repeated until the worker escapes from the tunnel.

7. A tunnel construction safety early warning system based on multidimensional data processing, characterized in that, include: The model training module is used to acquire a historical disaster dataset consisting of data on various disaster types, and to train a disaster type identification model based on the historical disaster dataset. A BIM model module is constructed to install various sensors inside the tunnel for real-time acquisition of tunnel information, and to construct a tunnel BIM model, mapping the tunnel information into the tunnel BIM model. The disaster type identification module is used to input real-time information about the tunnel into the disaster type identification model, identify the types of disasters that may occur in the tunnel, and map them into the tunnel BIM model. The risk avoidance module is used to calculate the real-time risk value of each location in the tunnel based on the information inside the tunnel, mark and connect the locations corresponding to the real-time risk values ​​that meet preset conditions to form an escape path, map it into the tunnel BIM model, and display the tunnel BIM model in real time on a virtual reality device.

8. A tunnel construction safety early warning system based on multidimensional data processing as described in claim 7, characterized in that, The disaster type identification model adopts an LSTM or Transformer structure, and the historical disaster data corresponding to each disaster type is input into the LSTM or Transformer structure to complete the training of the disaster type identification model.

9. A tunnel construction safety early warning system based on multidimensional data processing as described in claim 7, characterized in that, The disaster type identification model is as follows: Among them, ∏ k (t) represents the predicted value of the k-th type of disaster occurring at time t, where n is the number of information variables, and ω i，k Let σ be the weight of the influence of the i-th information variable on the k-th type of disaster. i (t) represents the i-th information variable in the tunnel information at time t, γ i,k Let χ be the response factor of the i-th information variable in the k-th type of disaster. i (t) is the state evolution factor of the i-th information variable at time t, λ i，k Let δ be the anomalous response suppression coefficient of the i-th information variable to the k-th type of disaster. i (t) represents the offset of the i-th information variable at time t from the corresponding historical average information variable.

10. A tunnel construction safety early warning system based on multidimensional data processing as described in claim 7, characterized in that, Also includes: Based on the disaster type, obtain the corresponding emergency plan, extract the entities in the emergency plan, construct a knowledge graph triple consisting of (disaster type, recommended measures, applicable conditions), and display it in real time on a virtual reality device.

Images