A risk assessment system for long tunnel construction across fault fracture zones

Abstract

This invention relates to the field of tunnel construction management technology and discloses a risk assessment system for long tunnels crossing fault fracture zones, including a data acquisition module and a risk assessment module. The system collects datasets through the data acquisition module, comprehensively collecting all possible risk factors, resulting in more accurate assessment results from high-quality data. The risk assessment module is equipped with an LSTM algorithm model, which analyzes and generates assessment data sets, tunnel progress data sets, and risk indices. The risk assessment module also features a schedule ladder, with each ladder corresponding to different construction content. The risk assessment module determines the schedule ladder at the current time point based on the tunnel progress data sets. If the tunnel progress data sets jump from one ladder to the next, the LSTM algorithm model is trained and updated accordingly. The risk assessment module compares the risk index with risk thresholds and outputs a corresponding report, significantly improving the safety of on-site tunnel construction. The intelligent update system exhibits high robustness.

Description

Technical Field

[0001] This invention relates to the field of tunnel construction management technology, specifically a risk assessment system for long tunnels crossing fault fracture zones. Background Technology

[0002] Constructing long tunnels through fault fracture zones is an extremely challenging engineering task. Fault fracture zones typically consist of abundant rock fragments, soil, and water. Due to the unstable geological conditions of these zones, various unforeseen problems may arise during construction. Therefore, the construction of long tunnels through fault fracture zones requires specialized construction techniques and methods to ensure the safety and smooth progress of the project. First, a detailed geological survey is necessary before construction. This includes a thorough investigation and analysis of the geological structure, rock type, and groundwater level of the fault fracture zone. This information is crucial for determining the construction plan and methods. For example, if the fault fracture zone is primarily composed of soft rock, special construction methods such as the shield tunneling method or the New Austrian Tunneling Method (NATM) may be required. Second, effective support measures are necessary during construction. Due to the unstable geological conditions of fault fracture zones, problems such as ground subsidence and collapse may occur during construction. Therefore, sufficient support structures need to be installed inside the tunnel to prevent these problems. These support structures may include steel arches, concrete linings, and anchor bolts. Furthermore, the construction speed must be strictly controlled during construction. Excessive construction speed may lead to ground subsidence and collapse. Strict monitoring is required during construction. This includes real-time monitoring of the tunnel's internal geological conditions, the stability of the support structure, and ground settlement. Any problems discovered must be addressed immediately. Finally, after construction is completed, a rigorous acceptance inspection is necessary. This involves a comprehensive examination and evaluation of the tunnel's structural stability and functionality. Only by passing this inspection can the tunnel's safety and normal operation be guaranteed.

[0003] Currently, traditional risk assessment systems for long tunnels crossing fault fracture zones are prone to incomplete, outdated, or inaccurate geological data due to the complexity and variability of geological conditions. This can lead to misleading results in subsequent risk analyses. Furthermore, as construction progresses and new geological information is acquired, the risk assessment model requires manual updates and maintenance. Failure to update in a timely manner will reduce the system's robustness. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a risk assessment system for long tunnel construction through fault fracture zones. This system boasts advantages such as higher accuracy of high-quality data assessment results and robustness of the intelligent update system. It solves the problems of unstable data quality and low system robustness in traditional risk assessment systems for long tunnel construction through fault fracture zones.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] A risk assessment system for the construction of long tunnels traversing fault fracture zones includes:

[0007] Data acquisition module and risk assessment module;

[0008] The data acquisition module includes a construction data unit, a geological data unit, and a meteorological data unit. The construction data unit collects construction datasets in real time via a network connection to a database and arranges them in chronological order from morning to evening. The construction dataset includes construction data for the long tunnel at all time points. The geological data unit collects geological datasets in real time via a network connection to a geological monitoring device and arranges them in chronological order from morning to evening. The geological dataset includes geological data for the area where the long tunnel is located. The meteorological data unit collects meteorological datasets via a network connection to a big data platform and arranges them in chronological order from morning to evening. The meteorological dataset includes meteorological and hydrological data for the area where the long tunnel is located. The data acquisition module transmits the construction dataset, geological dataset, and meteorological dataset to the risk assessment module via the network.

[0009] The risk assessment module includes a quality analysis unit, a construction analysis unit, and a risk analysis unit. The risk assessment module is equipped with an LSTM algorithm model. The quality analysis unit inputs the construction dataset, geological dataset, and meteorological dataset into the LSTM algorithm model to generate an assessment data set Pgsj. The construction analysis unit inputs the construction dataset and the assessment data set Pgsj into the LSTM algorithm model to generate a tunnel progress data set Sdjd. The risk analysis unit inputs the tunnel progress data set Sdjd and the assessment data set Pgsj into the LSTM algorithm model to generate a risk index Fxzs. The risk assessment module sets a construction period tier based on the construction dataset. Based on the tunnel progress data set Sdjd, the risk assessment module determines the construction period tier at the current time point and updates the LSTM algorithm model accordingly. The risk assessment module sets a risk threshold and compares the risk index Fxzs with the risk threshold, outputting a corresponding assessment report or safety management report.

[0010] Furthermore, the expression for the construction dataset is {S1} t S2 t S3 t ... Sn t}, S1 t To Sn t The data corresponds to the construction data of the long tunnel at each point in time. The construction data includes the construction scale, design scheme and construction method. t represents the time point when the database updates the construction data, and 1 to n represent the number of sets of long tunnel construction data updated in the database.

[0011] Furthermore, the expression for the geological dataset is {D1} m D2 m D3 m ... Dz m}, D1 m To Dz m The geological data of the area where the long tunnel is located are monitored by the geological monitoring device in sequence. The geological data includes fault activity, rock burst data and groundwater data. m represents the time point when the geological monitoring device monitors the geological data, and 1 to z represent the number of sets of geological data monitored by the geological monitoring device.

[0012] Furthermore, the expression for the meteorological dataset is {Q1}. s Q2 s Q3 s ... Qx s}, Q1 s To Qx s The data corresponds to the meteorological and hydrological data provided by the big data platform each time. s represents the time point when the big data platform provides the meteorological and hydrological data, and 1 to x represent the number of sets of meteorological and hydrological data provided by the big data platform.

[0013] Furthermore, the calculation formula for the evaluation data set Pgsj is as follows:

[0014]

[0015] Pgsj=∑ LSTM [(Sn t -Sn-1 t )+DZ+Qx s ]

[0016] In the formula, Pgsj represents the evaluation data set, and BD represents the fixed parameters used by the LSTM algorithm model to determine whether geological data is anomaly. This indicates that the LSTM algorithm model filters out geological data DZ and Sn-1 from the geological dataset that contain anomalies in chronological order. t Sn represents the construction data of a long tunnel compared to the previous time point. t -Sn-1 t This represents the difference in long tunnel construction data between two points in time, which is the latest updated long tunnel construction content at the current time point. ∑ LSTM [(Sn t -Sn-1 t )+DZ+Qx sThe ] indicates that the LSTM algorithm model filters out the latest updated long tunnel construction content, anomaly-prone geological data, and latest meteorological and hydrological data from the construction dataset, geological dataset, and meteorological dataset in chronological order.

[0017] Furthermore, the calculation formula for the tunnel progress data group Sdjd is as follows:

[0018]

[0019] In the formula, Sdjd represents the tunnel progress data group, and ZS represents the construction data set for the design scheme of long tunnels. This indicates that the LSTM algorithm model analyzes the latest construction content at the current time to determine the proportion of the completed part of the project relative to the planned progress of the long tunnel design.

[0020] Furthermore, the risk index Fxzs is calculated using the following formula:

[0021]

[0022] In the formula, Fxzs represents the risk index, α represents the weight of the tunnel progress data group in the risk index calculation formula, and β represents the weight of the evaluation data group in the risk index calculation formula. This indicates that the LSTM algorithm calculates the weighted average of the tunnel progress data set and the evaluation data set according to the α and β weights, which is the risk index corresponding to the construction of a long tunnel crossing a fault fracture zone at the current time point.

[0023] Furthermore, the risk assessment module sets up a construction period ladder based on the long tunnel design scheme in the construction dataset. The construction period ladder includes a first-level ladder, a second-level ladder, a third-level ladder, and a fourth-level ladder, and each ladder corresponds to different construction content. The risk assessment module determines the construction period ladder at the current time point based on the tunnel progress data group Sdjd. If the tunnel progress data group Sdjd jumps from the previous ladder to the next ladder, the LSTM algorithm model is trained and updated based on the construction dataset, geological dataset, and meteorological dataset at all time points.

[0024] Furthermore, when the risk index Fxzs exceeds the risk threshold, the risk assessment module determines that there are anomalies in the construction data, geological data, and meteorological and hydrological data of the long tunnel, and outputs a high-risk assessment report.

[0025] Furthermore, when the risk index Fxzs does not exceed the risk threshold, the risk assessment module determines that there are no abnormalities in the construction data, geological data, and meteorological and hydrological data of the long tunnel, and outputs a safety management report.

[0026] Compared with the prior art, the present invention provides a risk assessment system for the construction of long tunnels crossing fault fracture zones, which has the following beneficial effects:

[0027] 1. This invention sets up construction data units, geological data units, and meteorological data units through a data acquisition module. The construction data unit collects construction datasets in real time through a network connection to a database and arranges them in chronological order from early to late. The construction dataset includes construction data for the long tunnel at all time points. The geological data unit collects geological datasets in real time through a network connection to geological monitoring devices and arranges them in chronological order from early to late. The geological monitoring devices include seismometers, strain monitors, and groundwater monitors. The geological dataset includes geological data of the area where the long tunnel is located. The meteorological data unit collects meteorological datasets in real time through a network connection to geological monitoring devices and arranges them in chronological order from early to late. The meteorological dataset includes meteorological and hydrological data of the area where the long tunnel is located. This comprehensive collection of all possible risk factors ensures strong data traceability and more accurate high-quality data assessment results.

[0028] 2. This invention sets up a quality analysis unit, a construction analysis unit, and a risk analysis unit through a risk assessment module. The risk assessment module is equipped with an LSTM algorithm model. The quality analysis unit analyzes and generates an assessment data set Pgsj, filtering out the latest updated long tunnel construction content, anomaly geological data, and the latest meteorological and hydrological data from the construction dataset, geological dataset, and meteorological dataset. The construction analysis unit analyzes and generates a tunnel progress data set Sdjd, analyzing the proportion of completed work to the planned progress of the long tunnel design, facilitating rapid determination of the construction period tier and targeted management, thus improving the efficiency and accuracy of risk assessment. The risk analysis unit analyzes and generates a risk index Fxzs, digitally assessing multiple aspects of construction risks. The risk assessment module is set up with a construction period tier. The system is structured in a stepped manner, with each step corresponding to different construction tasks. The risk assessment module determines the current construction period step based on the tunnel progress data set Sdjd. If the tunnel progress data set Sdjd jumps from the previous step to the next step, the LSTM algorithm model is trained and updated based on the construction dataset, geological dataset, and meteorological dataset at all time points. The risk assessment module sets a risk threshold and compares the risk index Fxzs with the risk threshold. When the risk index Fxzs exceeds the risk threshold, the risk assessment module determines that there are anomalies in the construction data, geological data, and meteorological and hydrological data of the long tunnel and outputs a high-risk assessment report. The well-defined risk response strategies are transformed into specific construction plans and operation guidelines, which significantly improves the safety of on-site construction of long tunnels. The intelligent update system has high robustness. Attached Figure Description

[0029] Figure 1 This is a schematic diagram of the system flow of the present invention. Detailed Implementation

[0030] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0031] Traditional risk assessment systems for long tunnels crossing fault fracture zones are prone to incomplete, outdated, or inaccurate geological data due to the complexity and variability of geological conditions. This can lead to misleading results in subsequent risk analyses. Furthermore, as construction progresses and new geological information is acquired, the risk assessment model requires manual updates and maintenance. Failure to update in a timely manner will reduce the system's robustness. Therefore, this paper presents a risk assessment system for long tunnels crossing fault fracture zones. Please refer to [link / reference]. Figure 1 The system includes a data acquisition module and a risk assessment module;

[0032] The data acquisition module includes a construction data unit, a geological data unit, and a meteorological data unit. The construction data unit connects to a database via a network to collect construction datasets in real time, arranged chronologically from earliest to latest. The construction dataset includes construction data for the long tunnel at all time points, and the expression for the construction dataset is {S1}. t S2 t S3 t ... Sn t}, S1 t To Sn t The database contains construction data for each long tunnel at each point in time. The construction data includes the construction scale, design scheme, and construction method. The construction scale includes the tunnel's length, width, and height, as well as the expected start and end times of construction. The design scheme includes the design structure of the long tunnel, and the construction method includes methods such as shield tunneling, drill-and-blast method, and tunneling machine method. t represents the time point when the database updates the construction data. Whenever the construction of a long tunnel begins, the database will synchronously store the new construction data. 1 to n indicates that there are n sets of long tunnel construction data updated in the database.

[0033] The geological data unit collects geological datasets in real time via a network connection to geological monitoring devices, arranged chronologically from earliest to latest. These monitoring devices include seismometers, strain monitors, and groundwater monitors. The geological datasets include geological data from the area where the long tunnel is located. The expression for the geological dataset is {D1}.m D2 m D3 m ... Dz m}, D1 m To Dz m The geological data of the long tunnel area monitored by the geological monitoring device each time are sequentially represented. The geological data includes fault activity, rock burst data and groundwater data. Specifically, it includes the geological structure of the fault fracture zone, rock type and groundwater conditions. m represents the time point when the geological monitoring device monitors the geological data, and 1 to z represent the number of sets of geological data monitored by the geological monitoring device.

[0034] The meteorological data unit collects meteorological datasets via a network connection to a big data platform, arranging them chronologically from morning to evening. The meteorological datasets include meteorological and hydrological data for the area where the long tunnel is located. The expression for the meteorological dataset is {Q1}. s Q2 s Q3 s ... Qx s}, Q1 s To Qx s The data corresponds to the meteorological and hydrological data provided by the big data platform each time. The meteorological and hydrological data includes the average temperature, precipitation, humidity and evaporation in different seasons in the region. s represents the time point when the big data platform provides the meteorological and hydrological data, and 1 to x represent the x sets of meteorological and hydrological data provided by the big data platform. In the construction of long tunnels, meteorological and hydrological data are crucial to ensuring construction safety and quality. These data provide detailed information about the climate and hydrological conditions of the tunnel area, which helps engineers better plan the construction schedule and deal with possible environmental risks.

[0035] The data acquisition module transmits construction datasets, geological datasets, and meteorological datasets to the risk assessment module via the network, comprehensively collecting all possible risk factors. The data has strong traceability, and the high-quality data assessment results are more accurate.

[0036] The risk assessment module includes a quality analysis unit, a construction analysis unit, and a risk analysis unit. The risk assessment module is equipped with an LSTM algorithm model. The quality analysis unit inputs the construction dataset, geological dataset, and meteorological dataset into the LSTM algorithm model to generate the assessment dataset JPgsj. The calculation formula is as follows:

[0037]

[0038] JPgsj=∑ LsTM [(Sn t -Sn-1 t )+DZ+Qx s ]

[0039] In the formula, Pgsj represents the evaluation data set, and BD represents the fixed parameters used by the LSTM algorithm model to determine whether geological data is abnormal. These fixed parameters specifically refer to geological fault activity, rockburst data, and groundwater data within a safe range. If the actual geological data exceeds these fixed parameters, it indicates that a geological hazard may occur at the current time point, posing a threat to the safety of long tunnel construction. This indicates that the LSTM algorithm model filters out geological data DZ and Sn-1 from the geological dataset that contain anomalies in chronological order. t Sn represents the construction data of a long tunnel compared to the previous time point. t -Sn-1 t This represents the difference in long tunnel construction data between two points in time, which is the latest updated long tunnel construction content at the current time point. ∑ LSTM [(Sn t -Sn-1 t )+DZ+Qx s The ] indicates that the LSTM algorithm model filters out the latest updated long tunnel construction content, anomaly-prone geological data, and latest time-point meteorological and hydrological data from the construction dataset, geological dataset, and meteorological dataset in chronological order;

[0040] The construction analysis unit substitutes the construction dataset and the evaluation data set Pgsj into the LSTM algorithm model to generate the tunnel progress data set Sdjd, whose calculation formula is as follows:

[0041]

[0042] In the formula, Sdjd represents the tunnel progress data group, and ZS represents the construction data set for the design scheme of long tunnels. This means that the LSTM algorithm model analyzes the latest construction content at the current time to determine the proportion of the completed part of the project to the planned progress of the long tunnel design, which facilitates the rapid determination of the construction period level, targeted management, and higher efficiency and accuracy of risk assessment.

[0043] The risk analysis unit substitutes the tunnel progress data set Sdjd and the evaluation data set Pgsj into the LSTM algorithm model to generate the risk index Fxzs, which is calculated as follows:

[0044]

[0045] In the formula, Fxzs represents the risk index, α represents the weight of the tunnel progress data group in the risk index calculation formula, and β represents the weight of the evaluation data group in the risk index calculation formula. α and β are constants, and α + β = 1. The LSTM algorithm model calculates the weighted average of the tunnel progress data set and the evaluation data set according to the α weight and β weight, which is the risk index corresponding to the construction of a long tunnel crossing a fault fracture zone at the current time point, and digitally assesses the construction risks in multiple aspects.

[0046] The risk assessment module sets up a construction period ladder based on the long tunnel design scheme in the construction dataset. This ladder includes four levels: Level 1, Level 2, Level 3, and Level 4, each corresponding to different construction tasks. For example, Level 1 corresponds to the construction period of the initial tunnel section, Level 2 to the middle section, Level 3 to the final section, and Level 4 to the acceptance period after tunnel completion. The risk assessment module determines the current construction period ladder based on the tunnel progress data set Sdjd. If the tunnel progress data set Sdjd jumps from the previous level to the next, the LSTM algorithm model is trained and updated based on the construction dataset, geological dataset, and meteorological dataset from all time points. The risk assessment module sets a risk threshold and assigns risk indicators... By comparing the risk index Fxzs with the risk threshold, if the risk index Fxzs exceeds the risk threshold, the risk assessment module determines that there are anomalies in the construction data, geological data, and meteorological and hydrological data of long tunnels, and outputs a high-risk assessment report. This transforms the established risk response strategies into specific construction plans and operation guidelines, significantly improving the safety of on-site construction of long tunnels. If the risk index Fxzs does not exceed the risk threshold, the risk assessment module determines that there are no anomalies in the construction data, geological data, and meteorological and hydrological data of long tunnels, and outputs a safety management report. Regular review and assessment effectively enhance the prevention of geological disasters, reduce project delays, and effectively protect construction equipment and tunnel structures from damage. The LSTM algorithm model has higher accuracy in risk assessment, and the intelligent update system has high robustness.

[0047] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A risk assessment system for long tunnel construction through fault fracture zones, characterized in that: Includes a data acquisition module and a risk assessment module; The data acquisition module includes a construction data unit, a geological data unit, and a meteorological data unit. The construction data unit collects construction datasets in real time via a network connection to a database and arranges them in chronological order from morning to evening. The construction dataset includes construction data for the long tunnel at all time points. The geological data unit collects geological datasets in real time via a network connection to a geological monitoring device and arranges them in chronological order from morning to evening. The geological dataset includes geological data for the area where the long tunnel is located. The meteorological data unit collects meteorological datasets via a network connection to a big data platform and arranges them in chronological order from morning to evening. The meteorological dataset includes meteorological and hydrological data for the area where the long tunnel is located. The data acquisition module transmits the construction dataset, geological dataset, and meteorological dataset to the risk assessment module via the network. The risk assessment module includes a quality analysis unit, a construction analysis unit, and a risk analysis unit. The risk assessment module is equipped with an LSTM algorithm model. The quality analysis unit inputs construction datasets, geological datasets, and meteorological datasets into the LSTM algorithm model to analyze and generate assessment data sets. The construction analysis unit will combine the construction dataset and the evaluation data set. Substitute the data into the LSTM algorithm model to analyze and generate tunnel progress data sets. The risk analysis unit will group the tunnel progress data. and evaluation data group Substitute the data into the LSTM algorithm model to analyze and generate a risk index. The risk assessment module is configured with a construction period tier based on the construction dataset, and the risk assessment module is based on the tunnel progress data set. The system determines the current time step of the project schedule and updates the LSTM algorithm model accordingly. The risk assessment module sets a risk threshold and sets the risk index. Compare the risk thresholds and output the corresponding assessment report or safety management report; The expression for the construction dataset is as follows: , to The data corresponds to the construction data of the long tunnel at each point in time, including the construction scale, design scheme, and construction methods. Indicates the time point at which the database updated the construction data, from 1 to... The database update indicates the following long tunnel construction data: Group; The expression for the geological dataset is: , to Each time, a corresponding geological monitoring device monitors the geological data of the area where the long tunnel is located. The geological data includes fault activity, rock burst data, and groundwater data. Indicates the time point in time when the geological monitoring device monitors geological data, from 1 to... This indicates that the geological data monitored by the geological monitoring device includes Group; The expression for the meteorological dataset is: , to This corresponds sequentially to the meteorological and hydrological data provided by the big data platform each time. This indicates the time point at which the big data platform provides meteorological and hydrological data, from 1 to... This indicates that the meteorological and hydrological data provided by the big data platform have Group; The evaluation data set The calculation formula is as follows: In the formula, Indicates the evaluation data group, This represents the fixed parameters used by the LSTM algorithm model to determine whether geological data is abnormal. This indicates that the LSTM algorithm model filters out geological data with anomalies in the geological dataset according to time sequence. , This represents the construction data of the long tunnel compared to the previous time point. This represents the difference in long tunnel construction data between two points in time, which is the latest updated long tunnel construction information at the current time. This means that the LSTM algorithm model filters out the latest updated long tunnel construction content, anomaly-prone geological data, and latest time-point meteorological and hydrological data from the construction dataset, geological dataset, and meteorological dataset in chronological order. The tunnel progress data group The calculation formula is as follows: In the formula, This represents the tunnel progress data group. This indicates that the construction data is concentrated on the design scheme for long tunnels. This indicates that the LSTM algorithm model analyzes the latest construction content at the current time to determine the proportion of the completed part of the project to the planned progress of the long tunnel design. The risk index The calculation formula is as follows: In the formula, Indicates the risk index. This indicates the weight of the tunnel progress data set in the risk index calculation formula. This indicates the weight of the assessment data group in the risk index calculation formula. The LSTM algorithm model is described according to... and The weighted average of the tunnel progress data set and the evaluation data set is calculated, which is the risk index corresponding to the construction of a long tunnel crossing a fault fracture zone at the current time point.

2. The risk assessment system for long tunnel construction through fault fracture zones according to claim 1, characterized in that: The risk assessment module sets up a construction period tier based on the long tunnel design scheme in the construction dataset. This tier includes four levels: level one, level two, level three, and level four, with each level corresponding to different construction tasks. The risk assessment module then uses the tunnel progress data as a basis for its implementation. Determine the construction period step at the current time point, if the tunnel progress data group When moving from one step to the next, the LSTM algorithm model is trained and updated based on the construction dataset, geological dataset, and meteorological dataset at all time points.

3. The risk assessment system for long tunnel construction through fault fracture zones according to claim 2, characterized in that: The risk index When the risk threshold is exceeded, the risk assessment module determines that there are anomalies in the construction data, geological data, and meteorological and hydrological data of the long tunnel, and outputs a high-risk assessment report.

4. The risk assessment system for long tunnel construction through fault fracture zones according to claim 3, characterized in that: The risk index If the risk threshold is not exceeded, the risk assessment module determines that there are no abnormalities in the construction data, geological data, and meteorological and hydrological data of the long tunnel, and outputs a safety management report.

Images