A TBM tunnel disaster type advanced identification method and system based on multi-source information fusion, a storage medium and equipment

Abstract

The application discloses a kind of TBM tunnel disaster type advanced identification method, system, storage medium and equipment based on multi-source information fusion, it is related to TBM tunnel disaster monitoring and early warning technical field.Method includes: constructing TBM driving data-microseismic data-tunnel disaster knowledge base;Domain knowledge modeling is carried out to large language model, and large model of disaster mechanism analysis is obtained;Establish data stream sample library;Build time series advanced prediction model based on Transformer;Real-time data stream is collected;Real-time data stream is input into time series prediction model, and the data stream in front of working face is predicted;Real-time predicted data stream in front of working face is input into large model of disaster mechanism analysis for prediction, and advanced prediction result is obtained.The application can advance, real-time identify the surrounding rock failure type in front of working face by deeply fusing domain knowledge and large model, provide accurate disaster warning and decision support for TBM safe, efficient driving.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of TBM tunnel disaster monitoring and early warning, and particularly relates to a TBM tunnel disaster type advanced identification method and system based on multi-source information fusion, a storage medium and equipment. BACKGROUND

[0002] Traditional surrounding rock geological disaster early warning methods mainly rely on geological radar scanning, advanced drilling analysis and other geophysical exploration means. These conventional technical means usually need to interrupt the tunnel construction process in the actual application process, which seriously affects the engineering efficiency, and at the same time, due to the strong electromagnetic interference of large metal components in the TBM equipment, the effective detection distance and identification accuracy have obvious limitations.

[0003] In the existing TBM tunnel disaster identification technology, there are problems of insufficient knowledge fusion, low advanced prediction accuracy and uninterpretable results. For example, the method of "Layered rock mass underground cavern surrounding rock failure type prediction method in high ground stress area", which mainly relies on static geological parameters to make judgments through table lookup, lacks the response ability to dynamic geological changes, and therefore it is difficult to realize real-time and dynamic disaster early warning function; in "Rock burst prediction method for TBM tunneling", although artificial intelligence algorithm is introduced for analysis, microseismic real-time monitoring data stream is not fully integrated, resulting in limited timeliness and accuracy of the early warning; and "Rock burst risk early warning method based on TBM multi-source data fusion and hybrid algorithm", although it realizes the fusion of microseismic monitoring information at the data level, it mainly focuses on rock burst single disaster, and the disaster coverage type is relatively single, and has not reached the level of real advanced prediction, and cannot realize comprehensive coverage of multiple potential geological disasters. Overall, these methods generally have some common defects, including narrow disaster identification type coverage, insufficient real-time early warning and advanced prediction ability, shallow multi-source data fusion level, weak generalization ability of algorithm model, and limited applicability in actual complex engineering scenarios.

[0004] Therefore, a TBM tunnel disaster type advanced identification method, system, storage medium and equipment based on multi-source information fusion are proposed to solve the problems existing in the prior art, which is a problem urgently needed to be solved by those skilled in the art. SUMMARY

[0005] Therefore, the application provides a TBM tunnel disaster type advanced identification method, system, storage medium and equipment based on multi-source information fusion, which can deeply fuse TBM parameter data stream and microseismic data stream through a knowledge driving mechanism, and realize comprehensive identification of various surrounding rock damage types.

[0006] In order to achieve the above purpose, the application adopts the following technical solutions:

[0007] A TBM tunnel disaster type advanced identification method based on multi-source information fusion, comprising:

[0008] S1, constructing a TBM tunneling data-microseismic data-tunnel disaster knowledge base according to the mapping relationship between TBM tunneling parameter characteristics and microseismic data characteristics and tunnel disaster types;

[0009] S2, performing field knowledge modeling on a large language model based on the TBM tunneling data-microseismic data-tunnel disaster knowledge base, to obtain a disaster mechanism analysis large model for predicting tunnel disaster types according to TBM tunneling parameter characteristics and microseismic data characteristics;

[0010] S3, establishing a data stream sample library containing TBM tunneling parameter data stream and microseismic monitoring data stream of different types of tunnel disasters;

[0011] S4, constructing a time series advanced prediction model based on Transformer, training the time series advanced prediction model through the data stream sample library, and obtaining a trained time series advanced prediction model;

[0012] S5, collecting and preprocessing TBM tunneling parameter data stream and microseismic monitoring data stream in real time;

[0013] S6, inputting the real-time collected TBM tunneling parameter data stream and microseismic monitoring data stream into the time series prediction model based on Transformer to predict TBM tunneling parameter data stream and microseismic monitoring data stream within a certain distance in front of the working face;

[0014] S7, inputting the real-time predicted TBM tunneling parameter data stream and microseismic monitoring data stream in front of the working face into the disaster mechanism analysis large model for prediction, to obtain an advanced prediction result of the tunnel disaster type in front of the working face containing an inference process.

[0015] The method, optionally, in S1, the tunnel disaster type includes: strain type rock burst, strain-structure surface slip type rock burst, fracture type rock burst, stress type collapse, structure type collapse and stress-structure type collapse.

[0016] The method, optionally, in S2, the field knowledge modeling on the large language model includes: injecting the field knowledge in the TBM tunneling data-microseismic data-tunnel disaster knowledge base into the open source large language model through the RAG method.

[0017] The method, optionally, in S3, the TBM tunneling parameter data stream and the microseismic monitoring data stream are preprocessed, and a data stream sample library is established.

[0018] The data preprocessing includes: outlier rejection, data standardization and time synchronization.

[0019] The method, optionally, in S4, the time series advance prediction model based on the Transformer is constructed, and specifically:

[0020] The time series advance prediction model based on the Transformer is constructed by taking the data stream sample library containing the TBM tunneling parameter data stream and the microseismic monitoring data stream of different types of tunnel disasters as training samples and taking the Transformer decoder as the core feature learning network.

[0021] The method, optionally, the preprocessed TBM tunneling parameter data stream and the microseismic monitoring data stream have the same format as the TBM tunneling parameter data stream and the microseismic monitoring data stream sample containing different types of tunnel disasters in the data stream sample library.

[0022] A TBM tunnel disaster type advance identification system based on multi-source information fusion executes any one of the TBM tunnel disaster type advance identification methods based on multi-source information fusion, and includes:

[0023] The TBM tunneling data-microseismic data-tunnel disaster knowledge base module is used for storing the mapping relationship between the TBM tunneling parameter features and the microseismic data features and the tunnel disaster type.

[0024] The multi-source data acquisition and preprocessing module is used for acquiring the TBM tunneling parameter data stream and the microseismic monitoring data stream in real time and pre-processing.

[0025] The advance prediction module is used for deploying the time series advance prediction model based on the Transformer to advance predict the TBM tunneling parameter data stream and the microseismic monitoring data stream within a certain distance in front of the tunnel face.

[0026] Disaster mechanism analysis large model module: used for enhancing the domain knowledge of a large language model based on TBM tunneling data-microseismic data-tunnel disaster knowledge base, realizing professional field data feature analysis and knowledge association reasoning;

[0027] Recognition result output module: used for outputting an advanced prediction result of the tunnel disaster type in front of the working face.

[0028] A computer-readable storage medium, the computer-readable storage medium stores executable instructions, the instructions are executed by a processor to implement the method of any one of the preceding claims.

[0029] A TBM tunnel disaster type advanced identification device based on multi-source information fusion, comprising a processor, a memory and a communication interface.

[0030] The memory stores executable instructions of the processor;

[0031] The processor executes the instructions to implement the method of any one of the preceding claims.

[0032] Via the technical solutions described above, compared with the prior art, the present application provides a TBM tunnel disaster type advanced identification method, system, storage medium and device based on multi-source information fusion, which has the following beneficial effects: the present application provides high-quality training data for a time series advanced prediction model based on a Transformer decoder by using preprocessed TBM region excavation and microseismic data stream sample libraries, which can effectively capture the time sequence dependence of surrounding rock data streams and realize accurate advanced prediction of related data streams within a certain distance in front of the working face; relying on a structured "data-disaster" knowledge base containing retrievable knowledge units, the present application integrates domain knowledge into an open-source large language model through the RAG method to form a disaster mechanism analysis large model with professional data analysis and knowledge reasoning capabilities, combined with real-time multi-source data preprocessing in the same format as the sample library, the present application can advanced identify multiple disasters such as strain rock burst and stress collapse, while outputting results containing reasoning processes and key features, which significantly improves the disaster identification accuracy and interpretability, and provides reliable protection for TBM tunnel construction safety. BRIEF DESCRIPTION OF DRAWINGS

[0033] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are only embodiments of the present application, and for those skilled in the art, other drawings can be obtained without creative labor based on the provided drawings.

[0034] Figure 1 A flow chart of a TBM tunnel disaster type advanced identification method based on multi-source information fusion provided by the present application is provided.

[0035] Figure 2 A structural diagram of a TBM tunnel disaster type advanced identification system based on multi-source information fusion provided by the present application is provided.

[0036] Figure 3 A geological distribution map of the K237+280~290 region in a specific embodiment of the TBM tunnel disaster type advanced identification method based on multi-source information fusion provided by the present application is provided.

[0037] Figure 4 An input data stream and a predicted data stream in a specific embodiment of the TBM tunnel disaster type advanced identification method based on multi-source information fusion provided by the present application is provided. DETAILED DESCRIPTION

[0038] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present application.

[0039] In the present application, the relationship terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations, the term “comprises”, “comprises” or any other variant thereof is intended to cover non-exclusive inclusion, so that the process, method, article or equipment including a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such process, method, article or equipment. Without more limitations, the element defined by the statement “comprises a” does not exclude the presence of additional identical elements in the process, method, article or equipment including the element.

[0040] Reference Figure 1 As shown in the figure, the present application discloses a TBM tunnel disaster type advanced identification method based on multi-source information fusion, comprising:

[0041] S1, according to the mapping relationship between the TBM tunneling parameter characteristics, the microseismic data characteristics and the tunnel disaster type, a TBM tunneling data-microseismic data-tunnel disaster knowledge base is constructed;

[0042] S2, based on TBM tunneling data-microseismic data-tunnel disaster knowledge base, the field knowledge of the large language model is modeled, and a disaster mechanism analysis large model for predicting the tunnel disaster type according to the TBM tunneling parameter characteristics and the microseismic data characteristics is obtained;

[0043] S3, a data stream sample library containing TBM tunneling parameter data stream and microseismic monitoring data stream of different types of tunnel disasters is established;

[0044] S4, a time series advance prediction model based on Transformer is constructed, and the time series advance prediction model is trained through the data stream sample library to obtain a trained time series advance prediction model;

[0045] S5, real-time acquisition and preprocessing of TBM tunneling parameter data stream and microseismic monitoring data stream are performed;

[0046] S6, the real-time acquisition TBM tunneling parameter data stream and microseismic monitoring data stream are input into the time series prediction model based on Transformer to predict the TBM tunneling parameter data stream and microseismic monitoring data stream within a certain distance in front of the tunnel face;

[0047] S7, the real-time predicted TBM tunneling parameter data stream and microseismic monitoring data stream in front of the tunnel face are input into the disaster mechanism analysis large model for prediction to obtain an advance prediction result of the tunnel disaster type in front of the tunnel face containing the reasoning process.

[0048] Further, in S1, the TBM tunneling parameter characteristics and the microseismic data characteristics include but are not limited to TBM tunneling parameter long-term and short-term trends, microseismic event generation rate, and microseismic release energy change rate; the disaster types include but are not limited to strain rockburst, strain-structure surface slip rockburst, stress collapse, and structure collapse;

[0049] Further, the knowledge base is constructed by structured storage and contains retrievable knowledge units, each unit containing feature description, trend analysis, and disaster correlation rule.

[0050] Further, in S2, the field knowledge of the large language model is modeled, including: injecting the field knowledge in the TBM tunneling data-microseismic data-tunnel disaster knowledge base into the open source large language model through the RAG method;

[0051] Further, the RAG method includes knowledge retrieval layer construction, retrieval mechanism, and knowledge injection. The retrieval layer construction converts the knowledge units of the "data-disaster" knowledge base into vector representation and stores them in a vector database; the retrieval mechanism retrieves Top-K related knowledge units through cosine similarity calculation for the input data stream characteristics; the knowledge injection splices the retrieved knowledge units as context with the input data characteristics to form the model input sequence.

[0052] Further, in S3, the TBM tunneling parameter data stream and the microseismic monitoring data stream are preprocessed, and a data stream sample library is established.

[0053] The data preprocessing includes: outlier rejection, data standardization, and time synchronization.

[0054] Further, in S4, a time series advance prediction model based on Transformer is constructed, specifically:

[0055] The time series advance prediction model based on Transformer is constructed by taking the data stream sample library containing TBM tunneling parameter data streams and microseismic monitoring data streams of different types of tunnel disasters as training samples and taking the Transformer decoder as the core feature learning network, for capturing the time sequence dependence of the excavated surrounding rock data stream behind the tunnel face and predicting the data stream within a certain range in front of the tunnel face.

[0056] Further, the preprocessed TBM tunneling parameter data stream and microseismic monitoring data stream have the same format as the TBM tunneling parameter data stream and microseismic monitoring data stream samples containing different types of tunnel disasters in the data stream sample library.

[0057] Referring to Figure 2 Fig. 1, a TBM tunnel disaster type advance identification system based on multi-source information fusion is executed according to any one of the TBM tunnel disaster type advance identification methods based on multi-source information fusion, and includes:

[0058] The TBM tunneling data-microseismic data-tunnel disaster knowledge base module is used to store the mapping relationship between TBM tunneling parameter features, microseismic data features, and tunnel disaster types.

[0059] The multi-source data acquisition and preprocessing module is used to acquire TBM tunneling parameter data streams and microseismic monitoring data streams in real time and perform preprocessing.

[0060] The advance prediction module is used to deploy the time series advance prediction model based on Transformer to advance predict TBM tunneling parameter data streams and microseismic monitoring data streams within a certain distance in front of the tunnel face.

[0061] The disaster mechanism analysis large model module is used to enhance the domain knowledge of the large language model based on the TBM tunneling data-microseismic data-tunnel disaster knowledge base, to realize professional field data feature analysis and knowledge association reasoning.

[0062] The identification result output module is used to output the advance prediction result of the tunnel disaster type in front of the tunnel face containing the reasoning process.

[0063] A computer-readable storage medium, the computer-readable storage medium has stored executable instructions, the instructions are executed by a processor to implement any one of the above-mentioned multi-source information fusion-based TBM tunnel disaster type advanced identification method.

[0064] A multi-source information fusion-based TBM tunnel disaster type advanced identification device, comprising: a processor, a memory and a communication interface.

[0065] The memory stores executable instructions of the processor.

[0066] The processor executes the instructions to implement any one of the above-mentioned multi-source information fusion-based TBM tunnel disaster type advanced identification method.

[0067] In one specific embodiment, the real-time requirement configuration core device:

[0068] TBM tunneling parameter acquisition device: TBM tunneling parameters are acquired through the TBM self-provided PLC system; the parameters acquired in this embodiment are total thrust, cutterhead torque and penetration;

[0069] Microseismic monitoring device: an 8-channel microseismic monitoring system is deployed to acquire microseismic event occurrence time, energy and apparent volume;

[0070] Computing and storage device: edge computing gateway (for real-time data preprocessing), GPU server (NVIDIA A6000, 36GB of video memory, for deploying Transformer prediction model and disaster mechanism analysis large model), Milvus vector database (for storing knowledge base vectors), MySQL database (for storing structured knowledge base and sample data);

[0071] The historical data of a 15km range of the tunnel has been selected, including 85 times of strain-type rockburst, 200 times of strain-structure surface slip-type rockburst, 50 times of stress-type collapse, and 200 times of structure-type collapse tunneling and microseismic data;

[0072] The collected data is subsequently preprocessed, the 3σ rule is used to eliminate abnormal values in the thrust and torque, all parameters are mapped to the [-1, 1] interval through Z-score standardization, the TBM parameters and microseismic data are synchronized to 1s time step based on the GPS timestamp, and finally a dataset containing 10,000 valid samples is formed;

[0073] The Transformer architecture containing 6 layers of decoder (attention head number 8, hidden layer dimension 512) is designed, the data stream 2000s behind the tunnel face is taken as input, the data stream 500s in front of the tunnel face is taken as output, the MAE loss function is adopted, 80% of the samples are taken as the training set and 20% of the samples are taken as the verification set, the model is trained for 200 rounds on the GPU server, and the prediction error of the verification set is stabilized below 4.8%, and the model converges;

[0074] In this embodiment, the mapping relationship between the TBM tunneling parameter characteristics and the microseismic data characteristics and the tunnel disaster type is constructed by analyzing and summarizing historical disaster cases to build a knowledge base;

[0075] The text description of the knowledge unit is converted into a vector by using a text2vec model, and is stored in a ChromaDB vector database, and the domain knowledge is integrated into an open source large language model by using an RAG method to form a disaster mechanism analysis large model with professional data analysis and knowledge reasoning capabilities; in this embodiment, an open source large language model ChatGLM3-6B is selected as a basic large model;

[0076] Referring to Figure 3 As shown in the figure, the system gives a disaster warning when the tunnel face is located at K237+284, and the warning disaster type is strain-structure plane slip type rock burst, which is consistent with the actual occurrence;

[0077] Taking the total thrust and cutter torque as an example, after data preprocessing, the input data stream and the predicted data stream are input, as shown in Figure 4 The red dashed line is the predicted data stream of the model;

[0078] The output result of the disaster mechanism analysis large model is:

[0079] (1) TBM tunneling parameter trend analysis

[0080]

[0081] (2) Mechanism reasoning

[0082] Crack initiation and development stage: the fluctuation of TBM thrust and torque and the rise of microseismic frequency are the signals that the hidden structure plane (such as joints and fissures) in the surrounding rock in front of the tunnel face begins to “initially slip” under the disturbance of excavation, at this time, the structure plane only has small amplitude shear slip, the resistance of the surrounding rock to the cutter head is “unstable fluctuation”, the micro-fissure expansion causes a small amount of microseismic events, and the energy accumulates slowly;

[0083] Crack accumulation and slight slip stage: the expansion of the torque fluctuation amplitude is the performance that the structure plane slip enters the “accelerating stage”, with the continuous advancement of the TBM, the normal stress of the structure plane increases, the slip speed accelerates after overcoming the friction, and the shear resistance of the surrounding rock to the cutter head increases sharply;

[0084] Rock burst occurrence stage: the microseismic energy release increases several times, and the frequency maintains a high level, which is the key sign of the structure surface slip entering the "energy release stage". At this time, the structure surface has occurred large-scale shear failure, the strain energy in the surrounding rock is concentratedly released, microseismic events occur densely, and the energy condition required for rock burst has been met, which will cause strain-structure surface slip type rock burst when the working face advances to this area;

[0085] This prediction passes through three-level verification of "parameter trend analysis-knowledge base matching-mechanism reasoning", confirms that the area in front of the working face meets the core characteristics of strain-structure surface slip type rock burst, and makes correct early warning in time. Subsequent measures such as reducing the excavation rate, strengthening monitoring and personnel evacuation can be taken to reduce the harm of rock burst.

[0086] Each of the above-described embodiments is described in a progressive manner, and the same or similar parts between each embodiment can be referred to each other. Each embodiment mainly describes the difference from other embodiments. In particular, for the system or system embodiment, since it is basically similar to the method embodiment, it is described more simply, and the related part can be referred to the part of the method embodiment. The above-described system and system embodiment are only illustrative, and the units described as separate components can be or can not be physically separated, and the components displayed as units can be or can not be physical units, that is, they can be located in one place or distributed on multiple network units. Part or all of the modules can be selected to achieve the purpose of the embodiment according to the actual needs. Those skilled in the art can understand and implement it without creative labor.

[0087] The above description of the disclosed embodiments enables a person skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for TBM tunnel disaster type advanced identification based on multi-source information fusion, characterized in that, Comprise: S1, according to the mapping relationship between TBM tunneling parameter characteristics and microseismic data characteristics and tunnel disaster type, construct TBM tunneling data-microseismic data-tunnel disaster knowledge base; S2, based on TBM tunneling data-microseismic data-tunnel disaster knowledge base, domain knowledge modeling is carried out on large language model, and disaster mechanism analysis large model for predicting tunnel disaster type according to TBM tunneling parameter characteristics and microseismic data characteristics is obtained; S3, a data stream sample library containing TBM tunneling parameter data stream and microseismic monitoring data stream of different types of tunnel disasters is established; S4, a time series advance prediction model based on Transformer is constructed, and the time series advance prediction model is trained through the data stream sample library to obtain the trained time series advance prediction model; S5, real-time acquisition and preprocessing of TBM tunneling parameter data stream and microseismic monitoring data stream are carried out; S6, the real-time acquisition TBM tunneling parameter data stream and microseismic monitoring data stream are input into the time series prediction model based on Transformer to advance predict the TBM tunneling parameter data stream and microseismic monitoring data stream within a certain distance in front of the tunnel face; S7, the real-time predicted TBM tunneling parameter data stream and microseismic monitoring data stream in front of the tunnel face are input into the disaster mechanism analysis large model for prediction to obtain the advance prediction result of the tunnel disaster type in front of the tunnel face containing the reasoning process.

2. The TBM tunnel disaster type advance identification method based on multi-source information fusion according to claim 1, wherein the tunnel disaster type in S1 comprises: strain rockburst, strain-structure surface slip rockburst, fracture rockburst, stress collapse, structure collapse and stress-structure collapse.

3. The TBM tunnel disaster type advance identification method based on multi-source information fusion according to claim 1, wherein in S2, the domain knowledge modeling is carried out on the large language model, comprising: injecting the domain knowledge in the TBM tunneling data-microseismic data-tunnel disaster knowledge base into the open source large language model through the RAG method.

4. The TBM tunnel disaster type advance identification method based on multi-source information fusion according to claim 1, wherein in S3, the TBM tunneling parameter data stream and microseismic monitoring data stream of the excavated tunnel are preprocessed to establish the data stream sample library; The data preprocessing comprises: outlier rejection, data standardization and time synchronization.

5. The TBM tunnel disaster type advance identification method based on multi-source information fusion according to claim 1, wherein in S4, the time series advance prediction model based on Transformer is constructed, specifically: Taking the data stream sample library containing TBM tunneling parameter data stream and microseismic monitoring data stream of different types of tunnel disasters as training samples, and taking the Transformer decoder as the core feature learning network, the time series advance prediction model based on Transformer is constructed.

6. The TBM tunnel disaster type advance identification method based on multi-source information fusion according to claim 1, wherein ​ ​ ​ ​ The pretreated TBM tunneling parameter data stream and microseismic monitoring data stream have the same format as the TBM tunneling parameter data stream and microseismic monitoring data stream samples of different types of tunnel disasters in the data stream sample library.

7. A multi-source information fusion based TBM tunnel disaster type advanced identification system, which executes a multi-source information fusion based TBM tunnel disaster type advanced identification method according to any one of claims 1-6, characterized in that, Comprise: A TBM tunneling data-microseismic data-tunnel disaster knowledge base module for storing the mapping relationship between TBM tunneling parameter features, microseismic data features and tunnel disaster types; A multi-source data acquisition and preprocessing module for real-time acquisition and preprocessing of TBM tunneling parameter data stream and microseismic monitoring data stream; An advanced prediction module for deploying a Transformer-based time series advanced prediction model to predict TBM tunneling parameter data stream and microseismic monitoring data stream within a certain distance in front of the tunnel face; A disaster mechanism analysis large model module for enhancing the domain knowledge of a large language model based on the TBM tunneling data-microseismic data-tunnel disaster knowledge base, realizing professional field data feature analysis and knowledge association reasoning; An identification result output module for outputting the advanced prediction result of the tunnel disaster type in front of the tunnel face containing the reasoning process.

8. A computer-readable storage medium, characterized in that, A computer readable storage medium stores executable instructions, and the instructions are executed by a processor to implement the TBM tunnel disaster type advanced identification method based on multi-source information fusion according to any one of claims 1-6.

9. A TBM tunnel disaster type advanced identification device based on multi-source information fusion, characterized in that, Comprise: A processor, a memory and a communication interface; The memory stores executable instructions of the processor; The processor executes the instructions to implement the TBM tunnel disaster type advanced identification method based on multi-source information fusion according to any one of claims 1-6.

Images