Tunnel advanced geological forecasting method and system based on decision-making layer multi-source data fusion

Abstract

The invention provides a tunnel advanced geological forecasting method and system based on decision-making layer multi-source data fusion, and relates to a data processing method and system for a forecasting purpose, original index data is obtained through field exploration of various advanced geological forecasting methods, indexes are assigned according to an assignment rule and the index data, and a prediction result is obtained. The method comprises the steps of determining the importance degree of indexes for unfavorable geological problems, setting sections in a forecast mileage range and the unfavorable geological problems, obtaining corresponding indexes, calculating corresponding index weights by using an analytic hierarchy process and a Huber weighting method according to the importance degree of the corresponding indexes, and calculating the unfavorable geological problems according to the weights. And calculating the corresponding index weight and index assignment through a plurality of data fusion algorithms to obtain a risk index corresponding to each data fusion algorithm, and calculating an average value of all the risk indexes to obtain an unfavorable geological problem risk index of a set section, thereby solving the problem that the existing advanced geological forecast is inaccurate, and improving the prediction accuracy of the unfavorable geological problem. The method is suitable for advanced geological forecast.

Description

Technical Field

[0001] The present invention relates to a data processing method and system for prediction purposes, and in particular to a tunnel advanced geological prediction method and system based on multi-source data fusion at a decision-making layer. Background Art

[0002] In the field of tunnel engineering, the complex and unknown geological conditions in front of the tunnel face often cause major disasters such as surrounding rock collapse, rock burst, and sudden water and mud during tunnel excavation, which seriously threaten the safety of tunnel construction. The application of advanced geological prediction technology to detect unfavorable geological bodies in front of the tunnel face is a necessary work before tunnel excavation.

[0003] Existing advanced geological prediction methods, such as geological analysis method, advanced pilot pit method, advanced drilling method and geophysical exploration method, the geophysical exploration method includes seismic wave method, high-density resistivity method, transient electromagnetic method, geological radar method, etc., are usually used alone. However, a single advanced geological prediction method has disadvantages such as low resolution, short prediction distance and poor anti-interference ability on the one hand; on the other hand, the types of adverse geological bodies targeted are not comprehensive enough, which may lead to misjudgment and omission of certain adverse geological problems. Summary of the invention

[0004] Technical problem solved by the present invention: The present invention provides a method and system for tunnel advanced geological prediction based on multi-source data fusion at the decision-making layer, which solves the problem of inaccurate existing advanced geological prediction.

[0005] The present invention solves the above technical problems by adopting a technical solution: a tunnel advanced geological prediction method based on multi-source data fusion at the decision layer, comprising the following steps: S1. Conduct field surveys using a variety of advanced geological prediction methods to obtain original indicator data and clarify the mapping relationship between various indicators and adverse geological problems corresponding to each section within the forecast mileage range; S2. Setting the value assignment rules for each indicator, assigning values ​​to the indicators according to the value assignment rules and indicator data, and determining the importance level of the indicators for each adverse geological problem; S3, setting the sections and unfavorable geological problems within the forecast mileage range, obtaining corresponding indicators according to the mapping relationship, and calculating corresponding indicator weights using the hierarchical analysis method and the Huber weighting method according to the importance level of the corresponding indicators; S4. Calculate the corresponding indicator weights and indicator values ​​through a variety of data fusion algorithms to obtain the risk index corresponding to each data fusion algorithm, and calculate the average value of all risk indexes to obtain the risk index of adverse geological problems in the set section.

[0006] Furthermore, the advanced geological prediction method includes geological analysis method, advanced drilling method, advanced pilot pit method and geophysical exploration method, and the geophysical exploration method includes seismic method, high-density electrical method, transient electromagnetic method and ground penetrating radar method.

[0007] Furthermore, the assignment rule is to assign a value within the range of 0-1 according to the risk of occurrence of adverse geological problems corresponding to the real data of the indicator.

[0008] Furthermore, the assignment rules include node-type assignment and fitting-type assignment; the node-type assignment is to divide the data by clustering method to obtain discrete nodes, and assign the data in the range of 0 to 1 according to the nodes; the fitting-type assignment is to form a fitting curve of data and indicator assignment using a linear fitting method for continuous data, and assign the data in the range of 0 to 1 according to the fitting curve.

[0009] Furthermore, the analytic hierarchy process includes the geometric mean method, the arithmetic mean method, the eigenvector method and the least squares method.

[0010] Furthermore, the calculation formula of the corresponding indicator weight is: ,in represents the corresponding indicator weight, represents the number of analytic hierarchy process, Indicates The weights are obtained by analytic hierarchy process.

[0011] Furthermore, the data fusion algorithm includes weighted averaging method, fuzzy fusion theory, Laplace fusion theory, Bayesian averaging method and robust statistics method.

[0012] Furthermore, the calculation formula for the risk index of adverse geological problems in the section is set as: ,in Indicates the risk index of adverse geological problems in the set section. represents the number of data fusion algorithms, Indicates The risk index is obtained by a data fusion algorithm.

[0013] The present invention also provides a tunnel advanced geological prediction system based on decision-making layer multi-source data fusion, which implements the tunnel advanced geological prediction method based on decision-making layer multi-source data fusion as mentioned above, and the system includes a data input module, an indicator system module, an indicator weight calculation module and a risk index calculation module; The data input module is used to input the original data of the bad geological problems, the sections of the forecast mileage range and the indicators corresponding to the bad geological problems; The indicator system module is used to store the indicator system, which includes the sections within the forecast mileage range, the original indicator data, the adverse geological problems corresponding to the indicators, the indicator values ​​and the corresponding indicator importance levels, and obtains the corresponding indicator importance levels and indicator values ​​according to the adverse geological problems, the sections within the forecast mileage range and the original indicator data corresponding to the adverse geological problems input by the input module; The indicator weight calculation module is used to calculate the corresponding indicator weight by using the hierarchical analysis method and the Huber weighting method according to the importance level of the corresponding indicator; The risk index calculation module is used to calculate the corresponding indicator weights and indicator values ​​through multiple data fusion algorithms, obtain the risk index corresponding to each data fusion algorithm, and calculate the average value of all risk indices to obtain the risk index of adverse geological problems in the set section.

[0014] Beneficial effects of the present invention: The present invention provides a method and system for tunnel advanced geological prediction based on decision-making layer multi-source data fusion, obtains original index data through on-site exploration of multiple advanced geological prediction methods, clarifies the mapping relationship between various indicators and adverse geological problems corresponding to various sections within the prediction mileage range, assigns indicators according to assignment rules and indicator data, determines the importance of indicators to adverse geological problems, sets sections and adverse geological problems within the prediction mileage range, obtains corresponding indicators, calculates corresponding indicator weights according to the importance levels of the corresponding indicators using hierarchical analysis method and Huber weighting method, calculates corresponding indicator weights and indicator assignments through multiple data fusion algorithms, obtains risk indexes corresponding to each data fusion algorithm, calculates the average value of all risk indices, and obtains the adverse geological problem risk index of the set section, thereby solving the problem of inaccurate existing advanced geological prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is a flow chart of a method for advanced geological prediction of tunnels based on multi-source data fusion at the decision-making layer provided by the present invention. DETAILED DESCRIPTION

[0016] The present invention aims at the existing advanced geological prediction which usually adopts a prediction method, such as geological analysis method, advanced pilot pit method, advanced drilling method, seismic wave method, high-density resistivity method, transient electromagnetic method and geological radar method. On the one hand, it has disadvantages such as low resolution, short prediction distance and poor anti-interference performance; on the other hand, the types of adverse geological bodies targeted are not comprehensive enough, which may lead to misjudgment and omission of certain adverse geological problems. A tunnel advanced geological prediction method based on multi-source data fusion at the decision-making layer is proposed. Through on-site exploration of multiple advanced geological prediction methods, original indicator data is obtained, and the mapping relationship between various indicators and adverse geological problems corresponding to each section within the prediction mileage range is clarified. The indicators are fused according to the assignment rules and indicator data. Assign values, determine the importance of indicators for adverse geological problems, set sections and adverse geological problems within the forecast mileage, obtain corresponding indicators, calculate corresponding indicator weights using the analytic hierarchy process and the Huber weighted method according to the importance levels of the corresponding indicators, calculate the corresponding indicator weights and indicator assignments through a variety of data fusion algorithms, obtain the risk index corresponding to each data fusion algorithm, and calculate the average value of all risk indices to obtain the risk index of adverse geological problems in the set section. In this way, a variety of advanced geological prediction methods are integrated to improve the prediction accuracy of adverse geological problems. In addition, the types of adverse geology are broadened, so that the method can adapt to the prediction of a variety of adverse geological problems.

[0017] The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.

[0018] The invention provides a tunnel advanced geological prediction method based on decision-making layer multi-source data fusion, such as Figure 1 As shown, the following steps are included: S1. Conduct field surveys using a variety of advanced geological prediction methods to obtain original indicator data and clarify the mapping relationship between various indicators and adverse geological problems corresponding to each section within the forecast mileage range; Specifically, the advanced geological prediction methods include geological analysis, advanced drilling, advanced pilot pit and geophysical exploration, and the geophysical exploration methods include seismic method, high-density electrical method, transient electromagnetic method and ground penetrating radar method; adverse geological problems include but are not limited to high-stress hard rock rock bursts and large deformation of soft rocks, fault and fracture zone collapse, water gushing, karst caves, sudden water and mud, deformation and collapse of weak rock zones, high geothermal heat damage, toxic and harmful gas hazards and radioactive hazards; the mapping relationship between indicators and specific adverse geological problems is connected through adverse geological bodies, that is, the indicators show the existence of adverse geological bodies, and the adverse geological bodies confirm the risk of adverse geological problems. For example, in the TSP method, when the Vp / Vs ratio, Vs value or Vp value becomes smaller, or the inversion interface density is larger, it indicates that the surrounding rock integrity is poor and the rock mass is broken, and then it is judged that there may be a greater risk of collapse in fault zones and fracture zones. The sections predicted by the indicators corresponding to different advanced geological prediction methods are not necessarily the same. Therefore, the corresponding indicators are determined based on the sections to achieve the principle of indicator uniformity.

[0019] S2. Setting the value assignment rules for each indicator, assigning values ​​to the indicators according to the value assignment rules and indicator data, and determining the importance level of the indicators for each adverse geological problem; Specifically, the assignment rule is to assign values ​​within the range of 0-1 based on the risk of occurrence of adverse geological problems corresponding to the real data of the indicator. The indicator assignment rule embodies the core content of multi-source data fusion at the decision-making level, namely dimensionless processing and data normalization. 1 represents that the adverse geological problem is bound to occur, and 0 in the indicator assignment represents that the adverse geological problem does not occur. The assignment rules can be divided into node-type assignment and fitting-type assignment; the assignment rules include node-type assignment and fitting-type assignment; the node-type assignment is to divide the data by clustering method to obtain discrete nodes, and assign the data according to the nodes within the range of 0 to 1; the fitting-type assignment is to form a fitting curve of data and indicator assignment by a linear fitting method for continuous data, and assign the data within the range of 0 to 1 according to the fitting curve. The importance levels of the indicators for adverse geological problems include four levels: A, B, C and D, which qualitatively represent extremely important, relatively important, less important and extremely unimportant. When determining the importance level of indicators for adverse geological problems, the relative importance relationship between indicators should also be considered. For example, for the radioactive rock hazard problem, the previous geological survey and analysis data showed that there was radioactive rock in the section, and the surrounding rock classification index of the geological analysis method established a strong correlation with the radioactive hazard. However, since the section also adopted the sensor radioactive detection method, the sensor radioactive detection method most directly reflects whether the radioactive hazard really exists. In determining the importance level classification, the importance level of the surrounding rock classification index is set to B, and the importance level of the sensor radioactive monitoring index is set to A. The assignment of indicators and the determination of the importance level of indicators should follow the principle of being based on geological information and drilling results, supplemented by information such as geophysical exploration, monitoring and test results. While fully considering the sensitivity and accuracy of different forecasting methods for various adverse geological problems, it is ensured that the geological survey and drilling results are the prerequisites for data fusion calculation, so that the establishment standard of the indicator system is more targeted.

[0020] S3, setting the sections and unfavorable geological problems within the forecast mileage range, obtaining corresponding indicators according to the mapping relationship, and calculating corresponding indicator weights using the hierarchical analysis method and the Huber weighting method according to the importance level of the corresponding indicators; Specifically, the combination of the analytic hierarchy process and the Huber weighting algorithm can make up for the lack of scientificity of the traditional subjective weighting method. The analytic hierarchy process is highly subjective. Therefore, in actual calculation, the four commonly used weight calculation methods of the analytic hierarchy process: geometric mean method, arithmetic mean method, eigenvector method and least square method are combined to determine the corresponding indicator weights as accurately and objectively as possible. The calculation formula of the corresponding indicator weights is: ,in represents the corresponding indicator weight, represents the number of analytic hierarchy process, Indicates The weights obtained by the hierarchical analysis method are is 4, , , and They represent the indicator weights obtained by the geometric mean method, arithmetic mean method, eigenvector method and least squares method respectively.

[0021] S4, calculating the corresponding indicator weights and indicator values ​​through a variety of data fusion algorithms, obtaining the risk index corresponding to each data fusion algorithm, and calculating the average value of all risk indexes to obtain the risk index of adverse geological problems in the set section; Specifically, the data fusion algorithm includes weighted average method, fuzzy fusion theory, Laplace fusion theory, Bayesian averaging method and robust statistical method. The weighted average method is used to take the weight of each indicator into account, the fuzzy fusion theory deals with uncertainty and ambiguity, the Laplace fusion theory further optimizes the fusion process of multiple indicators, the Bayesian averaging rule performs fusion based on prior information, and the robust statistical method is used to deal with possible outliers or abnormal situations to ensure the robustness and reliability of the fusion results. The robust statistical method uses the Huber weight function and weighted average method to fuse data. The calculation formula for setting the risk index of adverse geological problems in the section is: ,in Indicates the risk index of adverse geological problems in the set section. represents the number of data fusion algorithms, Indicates The risk index is obtained by using a data fusion algorithm. In particular, when the above five data fusion algorithms are used in combination, the applicability of the fusion algorithms should be considered, and the fusion algorithms with abnormal results can be removed.

[0022] Furthermore, the tunnel advanced geological prediction method based on decision-making layer multi-source data fusion also includes: S5, fusing and imaging the risk index of adverse geological problems in each section within the forecast mileage range to obtain a risk visualization map corresponding to the forecast mileage range.

[0023] Specifically, S3 and S4 are repeated to obtain the risk index of adverse geological problems in all sections within the forecast mileage range, and the risk index results are represented by using color as a scale to obtain a risk visualization map corresponding to the forecast mileage range.

[0024] The present invention also provides a tunnel advanced geological prediction system based on decision-making layer multi-source data fusion, which implements the tunnel advanced geological prediction method based on decision-making layer multi-source data fusion as mentioned above, and the system includes a data input module, an indicator system module, an indicator weight calculation module and a risk index calculation module; The data input module is used to input the original data of the bad geological problems, the sections of the forecast mileage range and the indicators corresponding to the bad geological problems; The indicator system module is used to store the indicator system, which includes the sections within the forecast mileage range, the original indicator data, the adverse geological problems corresponding to the indicators, the indicator values ​​and the corresponding indicator importance levels, and obtains the corresponding indicator importance levels and indicator values ​​according to the adverse geological problems, the sections within the forecast mileage range and the original indicator data corresponding to the adverse geological problems input by the input module; The indicator weight calculation module is used to calculate the corresponding indicator weight by using the hierarchical analysis method and the Huber weighting method according to the importance level of the corresponding indicator; The risk index calculation module is used to calculate the corresponding indicator weights and indicator values ​​through multiple data fusion algorithms, obtain the risk index corresponding to each data fusion algorithm, and calculate the average value of all risk indices to obtain the risk index of adverse geological problems in the set section.

[0025] Furthermore, the tunnel advanced geological prediction system based on multi-source data fusion at the decision-making layer also includes a visualization interface, which is used to fuse and image the risk index of adverse geological problems in each section within the forecast mileage, specifically: using color as a scale to represent the risk index result.

Claims

1. A tunnel advanced geological prediction method based on multi-source data fusion at the decision-making level, characterized in that: The following steps are involved: S1. Conduct field surveys using a variety of advanced geological prediction methods to obtain original indicator data and clarify the mapping relationship between various indicators and adverse geological problems corresponding to each section within the forecast mileage range; S2. Setting the value assignment rules for each indicator, assigning values ​​to the indicators according to the value assignment rules and indicator data, and determining the importance level of the indicators for each adverse geological problem; S3, setting the sections and unfavorable geological problems within the forecast mileage range, obtaining corresponding indicators according to the mapping relationship, and calculating corresponding indicator weights using the hierarchical analysis method and the Huber weighting method according to the importance level of the corresponding indicators; S4. Calculate the corresponding indicator weights and indicator values ​​through a variety of data fusion algorithms to obtain the risk index corresponding to each data fusion algorithm, and calculate the average value of all risk indexes to obtain the risk index of adverse geological problems in the set section.

2. The tunnel advanced geological prediction method based on decision-making layer multi-source data fusion according to claim 1 is characterized in that: The advanced geological prediction method includes geological analysis method, advanced drilling method, advanced pilot pit method and geophysical exploration method. The geophysical exploration method includes seismic method, high-density electrical method, transient electromagnetic method and ground penetrating radar method.

3. The method for advanced tunnel geological prediction based on decision-making layer multi-source data fusion according to claim 1 is characterized in that: The assignment rule is to assign a value within the range of 0-1 based on the risk of occurrence of adverse geological problems corresponding to the real data of the indicator.

4. The method for advanced tunnel geological prediction based on decision-making layer multi-source data fusion according to claim 1 is characterized in that: The assignment rules include node-type assignment and fitting-type assignment; the node-type assignment is to divide the data by clustering method to obtain discrete nodes, and assign values ​​to the data in the range of 0 to 1 according to the nodes; The fitting type assignment is to form a fitting curve of data and index assignment using a linear fitting method for continuous data, and assign values ​​to the data in the range of 0 to 1 according to the fitting curve.

5. The method for advanced tunnel geological prediction based on decision-making layer multi-source data fusion according to claim 1 is characterized in that: The hierarchical analysis method includes geometric mean method, arithmetic mean method, eigenvector method and least square method.

6. The method for advanced tunnel geological prediction based on decision-making layer multi-source data fusion according to claim 5 is characterized in that: The calculation formula of the corresponding indicator weight is: ,in represents the corresponding indicator weight, represents the number of analytic hierarchy process, Indicates The weights are obtained by analytic hierarchy process.

7. The method for advanced tunnel geological prediction based on decision-making layer multi-source data fusion according to claim 1 is characterized in that: The data fusion algorithm includes weighted average method, fuzzy fusion theory, Laplace fusion theory, Bayesian average method and robust statistical method.

8. The method for advanced tunnel geological prediction based on decision-making layer multi-source data fusion according to claim 7 is characterized in that: The calculation formula for the risk index of adverse geological problems in the set section is: ,in Indicates the risk index of adverse geological problems in the set section. represents the number of data fusion algorithms, Indicates The risk index is obtained by a data fusion algorithm.

9. The method for advanced tunnel geological prediction based on decision-making layer multi-source data fusion according to claim 1 is characterized in that: Also includes: S5. The risk index of adverse geological problems in each section within the forecast mileage range is fused and imaged to obtain a risk visualization map corresponding to the forecast mileage range.

10. The tunnel advanced geological prediction system based on multi-source data fusion at the decision-making level is characterized by: Implementing the tunnel advanced geological prediction method based on decision-making layer multi-source data fusion as described in any one of claims 1 to 8, the system includes a data input module, an indicator system module, an indicator weight calculation module and a risk index calculation module; The data input module is used to input the original data of the bad geological problems, the sections of the forecast mileage range and the indicators corresponding to the bad geological problems; The indicator system module is used to store the indicator system, which includes the sections within the forecast mileage range, the original indicator data, the adverse geological problems corresponding to the indicators, the indicator values ​​and the corresponding indicator importance levels, and obtains the corresponding indicator importance levels and indicator values ​​according to the adverse geological problems, the sections within the forecast mileage range and the original indicator data corresponding to the adverse geological problems input by the input module; The indicator weight calculation module is used to calculate the corresponding indicator weight by using the hierarchical analysis method and the Huber weighting method according to the importance level of the corresponding indicator; The risk index calculation module is used to calculate the corresponding indicator weights and indicator values ​​through multiple data fusion algorithms, obtain the risk index corresponding to each data fusion algorithm, and calculate the average value of all risk indices to obtain the risk index of adverse geological problems in the set section.

Images