Shield construction parameter optimization boundary determination method and system based on large model

Abstract

The application provides a shield construction parameter optimization boundary determination method and system based on a large model, and relates to the technical field of shield engineering. In view of the poor precision and adaptability of the optimization boundary in the prior art, the settlement deformation value, the tunneling parameter and the geological parameter are extracted according to different extraction rules; two-by-two rank correlation analysis is performed on the extracted parameters, the correlation degree of each parameter with the settlement deformation value is determined, the importance of each parameter is sorted, the parameters are screened according to the importance, the explanation and prediction performance of the settlement deformation is verified based on the screened parameters, and the optimal parameter combination suitable for the settlement analysis is determined; the basic boundary conditions of each parameter are determined in combination with the engineering measured data, the equipment performance constraint and the settlement control standard; and the dynamic boundary generation system is constructed according to the heat map correlation analysis model based on the basic boundary conditions of each parameter, and the dynamic optimization boundary of each parameter is generated. The optimization boundary has high precision and adaptability.

Description

Technical Field

[0001] This invention relates to the field of tunnel boring machine (TBM) engineering technology, and in particular to a method and system for determining the boundary of TBM construction parameters based on a large model. Background Technology

[0002] The optimization boundary of tunnel boring machine (TBM) construction parameters refers to the reasonable range and constraint boundaries of various construction and geological parameters during TBM tunneling, provided that they meet the requirements of engineering settlement control standards, equipment performance limitations, and geological condition adaptability. Clearly defining the optimization boundary of TBM construction parameters effectively defines the parameter adjustment range, providing safe parameter optimization space for the intelligent TBM tunneling system, and is a key foundation for achieving efficient, safe, and precise management and control of TBM construction.

[0003] In existing technologies, the determination of optimal boundaries for tunnel boring machine (TBM) construction parameters largely relies on the construction experience of engineering technicians, supplemented by simple geological survey data and analogies from historical construction cases. Some technologies delineate static value ranges after performing a single statistical analysis on a small number of construction parameters. This reliance on experience to define fixed parameter boundaries fails to quantitatively define them in conjunction with actual engineering measurement data, equipment performance constraints, and settlement control standards. This approach only generates static parameter optimization boundaries, making it impossible to dynamically adjust the boundary range based on real-time construction geological data. Ultimately, this results in low accuracy and poor adaptability of the determined parameter optimization boundaries, failing to provide reliable support for TBM parameter optimization.

[0004] Therefore, developing a method and system for determining the optimization boundary of shield tunneling construction parameters based on a large model is of great significance for improving the accuracy and adaptability of the optimization boundary of shield tunneling construction parameters. Summary of the Invention

[0005] To address the issues of poor accuracy and adaptability of optimization boundaries in existing technologies, this invention proposes a method for determining the optimization boundaries of shield tunneling construction parameters based on a large model, specifically including the following steps: S1. Settlement deformation values, tunneling parameters, and geological parameters are extracted from the shield tunneling construction data according to different extraction rules. Among them, settlement deformation values ​​are extracted using automated extraction rules, tunneling parameters are extracted using shield ring number matching extraction rules, and geological parameters are extracted using composite stratum correction factor quantification extraction rules. S2. Perform pairwise rank correlation analysis on the extracted settlement deformation values, tunneling parameters and geological parameters to obtain the correlation coefficient matrix between each parameter, and determine the degree of correlation between each parameter and the settlement deformation value based on the correlation coefficient matrix. S3. Based on the correlation between each parameter and the settlement deformation value, the importance of each parameter is ranked, the parameters are screened according to the importance ranking, and the interpretation and prediction performance of settlement deformation is verified based on the screened parameters to determine the optimal parameter combination suitable for settlement analysis. S4. Based on the optimal parameter combination, combined with engineering measured data, equipment performance constraints and settlement control standards, determine the basic boundary conditions of each parameter; S5. Based on the basic boundary conditions of each parameter, and according to the heat map correlation analysis model, a dynamic boundary generation system is constructed. The optimal parameter combination, the basic boundary conditions of each parameter, and real-time construction geological data are input into the dynamic boundary generation system to generate the dynamic optimized boundary of each parameter.

[0006] Furthermore, in S1, the settlement deformation value is extracted using automated extraction rules, including: establishing data format parsing rules for shield tunneling deformation monitoring logs; configuring multi-dimensional extraction and retrieval conditions for monitoring type, spatial location, and construction stage; based on the retrieval conditions and parsing rules, traversing the target monitoring log file, automatically extracting settlement monitoring data, and removing invalid data and verifying the rationality of the extracted settlement monitoring data to obtain standardized settlement deformation values.

[0007] Furthermore, in step S2, pairwise rank correlation analysis is performed on the extracted settlement deformation values, tunneling parameters, and geological parameters to obtain the correlation coefficient matrix between each parameter. The degree of correlation between each parameter and the settlement deformation value is determined based on the correlation coefficient matrix. This includes: using Spearman correlation analysis to sort the parameters by numerical value and assign ranks; calculating the Spearman correlation coefficient between each pair of parameters and integrating them to form a correlation coefficient matrix; and accurately determining the degree of correlation between each parameter and the settlement deformation value based on the correlation coefficient matrix.

[0008] Furthermore, the degree of correlation between each parameter and the settlement deformation value was classified into three types: strong correlation, weak correlation, and moderate correlation.

[0009] Furthermore, in S3, based on the degree of correlation between each parameter and the settlement deformation value, the parameters are ranked in order of importance, and the parameters are screened according to the order of importance, including: taking the degree of correlation between each parameter and the settlement deformation value as the core basis, and combining the correlation coefficient matrix to reflect the correlation law between parameters, the parameters are ranked in order of importance gradient; low-importance redundant parameters are gradually eliminated in order of importance from low to high.

[0010] Furthermore, based on the screened parameters, the interpretive and predictive performance of settlement deformation is verified, and the optimal parameter combination suitable for settlement analysis is determined. This includes: verifying the interpretive and predictive performance of the remaining parameters after each screening on settlement deformation; stopping the screening when the interpretive and predictive performance of the screened parameters reaches the set conditions, and confirming that the current parameter combination is the optimal parameter combination suitable for settlement analysis.

[0011] Furthermore, in step S4, based on the optimal parameter combination, combined with engineering measured data, equipment performance constraints, and settlement control standards, the basic boundary conditions of each parameter are determined, including: eliminating abnormal data in the optimal parameter combination based on engineering measured data, defining the physical feasible boundary of each parameter; delineating the operational feasible boundary of each parameter based on equipment performance; clarifying the safety constraint boundary of each parameter with reference to settlement control standards; and integrating the physical feasible boundary, operational feasible boundary, and safety constraint boundary to form the basic boundary conditions of each parameter.

[0012] Furthermore, in S5, the heatmap correlation analysis model is a correlation coefficient matrix heatmap model constructed based on Spearman correlation analysis.

[0013] This invention also provides a system for determining the boundary of shield tunneling construction parameters based on a large model. The system is used to execute any of the above-mentioned methods for determining the boundary of shield tunneling construction parameters based on a large model. The system includes: The parameter extraction module is used to extract settlement deformation values, tunneling parameters, and geological parameters from shield tunneling construction data according to different extraction rules. Among them, the settlement deformation values ​​are extracted using automated extraction rules, the tunneling parameters are extracted using shield ring number matching extraction rules, and the geological parameters are extracted using composite stratum correction factor quantification extraction rules. The correlation analysis module is used to perform pairwise rank correlation analysis on the extracted and sorted settlement deformation values, tunneling parameters and geological parameters to obtain the correlation coefficient matrix between each parameter, and to determine the degree of correlation between each parameter and the settlement deformation value based on the correlation coefficient matrix. The filtering module is used to rank the parameters by importance based on the correlation coefficient matrix and the degree of correlation between each parameter and the settlement deformation value, filter the parameters according to the importance ranking, verify the interpretation and prediction performance of settlement deformation based on the filtered parameters, and determine the optimal parameter combination suitable for settlement analysis. The basic boundary determination module is used to determine the basic boundary conditions of each parameter based on the optimal parameter combination, combined with engineering measured data, equipment performance constraints and settlement control standards. The dynamic boundary generation module is used to construct a dynamic boundary generation system based on the basic boundary conditions of each parameter and combined with the heat map correlation analysis model. The optimal parameter combination, the basic boundary conditions of each parameter, and real-time construction geological data are input into the dynamic boundary generation system to generate dynamic optimized boundaries for each parameter.

[0014] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention develops specific extraction rules for different types of tunnel boring machine (TBM) construction parameters, enabling targeted adaptation to the data sources, structural characteristics, and engineering correlation attributes of settlement deformation values, tunneling parameters, and geological parameters. This achieves accurate, efficient, and standardized extraction of various parameters, effectively avoiding the problems of large errors and low efficiency associated with manual extraction. By conducting pairwise rank correlation analysis on all parameters and combining it with predictive performance verification to screen for optimal parameter combinations, the invention systematically and accurately quantifies the correlation between each parameter and settlement deformation values. This effectively identifies parameters that play a key role in settlement analysis, avoiding the problems of one-sided analysis and reliance on experience in traditional parameter selection, and ensuring the accuracy of the determination of the optimization boundary for TBM construction parameters. The basic boundary conditions determined by combining engineering measured data, equipment performance constraints, and settlement control standards provide a quantitative engineering basis for the parameter boundaries, improving the safety and rationality of the boundaries. By constructing a dynamic boundary generation system based on a heat map correlation analysis model and integrating real-time construction geological data to generate dynamic optimized boundaries, the system can intuitively and accurately uncover deep correlations between parameters. Furthermore, by incorporating real-time construction geological data into the boundary generation process, the optimized boundary can be adjusted in a timely manner according to the dynamic changes in geological conditions and construction processes. This significantly improves the accuracy, adaptability, and engineering feasibility of the optimized boundary, providing a compact and secure search space for shield tunneling parameter optimization algorithms. It can significantly improve the efficiency of shield tunneling parameter optimization and provide reliable support for intelligent and refined management and precise settlement control in shield tunneling construction. Attached Figure Description

[0015] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0016] Figure 1 This is a flowchart of the method for determining the boundary of shield tunneling construction parameters based on a large model, provided in an embodiment of the present invention. Figure 2 This is a schematic diagram of the structure of the shield tunneling parameter optimization boundary determination system based on a large model provided in an embodiment of the present invention. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below. Obviously, the described embodiments are only a part of the embodiments of this invention, and not all of them. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0018] The specific embodiments of the present invention will be described below.

[0019] To address the issues of poor accuracy and adaptability of optimized boundaries in existing technologies, this invention extracts settlement deformation values, tunneling parameters, and geological parameters according to different extraction rules. Pairwise rank correlation analysis is performed on the extracted parameters to determine the degree of correlation between each parameter and the settlement deformation value. The parameters are then ranked by importance, and parameters are selected based on this ranking. The interpretative and predictive performance of the selected parameters is verified to determine the optimal parameter combination for settlement analysis. The basic boundary conditions for each parameter are determined by combining engineering measured data, equipment performance constraints, and settlement control standards. Using these basic boundary conditions as the data foundation, a dynamic boundary generation system is constructed based on a heatmap correlation analysis model to generate dynamic optimized boundaries for each parameter. This invention achieves high accuracy and adaptability in optimized boundaries.

[0020] Example 1 This invention provides a method for determining the boundary of shield tunneling construction parameters based on a large model. Figure 1 This is a flowchart of the method for determining the boundary of shield tunneling construction parameters based on a large model, as provided in an embodiment of the present invention. Figure 1 As shown, the specific steps include the following: S1. Settlement deformation values, tunneling parameters, and geological parameters are extracted from the shield tunneling construction data according to different extraction rules. Among them, settlement deformation values ​​are extracted using automated extraction rules, tunneling parameters are extracted using shield ring number matching extraction rules, and geological parameters are extracted using composite stratum correction factor quantification extraction rules.

[0021] Shield tunneling construction data comprises all relevant data generated during the shield tunnel construction process, including three main categories: settlement and deformation monitoring logs, shield tunneling operation records, and engineering geological survey data. This data forms the foundation for parameter extraction. Settlement and deformation values ​​are the monitored values ​​of deformation caused to the surrounding strata and surface by the shield machine's excavation. The core data is surface settlement, but it also includes data on deformation of surrounding pipelines and segment convergence, making it a key indicator for assessing construction safety. Excavation parameters are the core operational parameters during shield machine excavation, including soil chamber pressure, grouting pressure, synchronous grouting volume, excavated soil volume, thrust, and advance speed. These directly affect the construction process and the degree of ground disturbance. Geological parameters are the physical and mechanical properties of the strata in the shield construction area, including soil weight, cohesion, and compression modulus.

[0022] Specifically, the settlement deformation value is extracted using automated extraction rules, including: establishing data format parsing rules for shield tunneling deformation monitoring logs; configuring multi-dimensional extraction and retrieval conditions based on monitoring type, spatial location, and construction stage; traversing the target monitoring log file based on the retrieval conditions and parsing rules, automatically extracting settlement monitoring data, and removing invalid data and verifying the rationality of the extracted settlement monitoring data to obtain standardized settlement deformation values.

[0023] The automated extraction rules are standardized operational logic designed to achieve rapid, accurate, and standardized extraction of settlement deformation values, adaptable to shield tunneling deformation monitoring log data. Shield tunneling deformation monitoring logs are daily records of various deformation monitoring data during shield tunneling, typically stored in Excel format, containing multiple worksheets and a large number of monitoring point data points. The data format parsing rules are parsing specifications tailored to the file format and data storage structure of shield tunneling deformation monitoring logs. These rules are used to identify the storage location, type, and relationships of various data types in the logs, forming the foundation for automated extraction.

[0024] By combining the file format and data storage structure of the shield tunneling deformation monitoring logs, a dedicated data format parsing rule is established, enabling the system to identify and interpret various types of settlement monitoring data in the logs. Based on the actual engineering needs of subsequent settlement analysis and parameter optimization, multi-dimensional extraction and retrieval conditions are configured according to monitoring type, spatial location, and construction stage to accurately locate the target data to be extracted. Based on the configured retrieval conditions and established parsing rules, the program comprehensively traverses the target monitoring log files, automatically retrieving and collecting settlement monitoring data that meets the conditions. The initially extracted raw settlement monitoring data is processed, first removing invalid data (meaningless, missing, or erroneous data), then performing a rationality check to verify whether the data conforms to the settlement engineering patterns of shield tunneling construction, ultimately obtaining standardized settlement deformation values ​​that are uniform in format, accurate, and directly usable for subsequent engineering analysis. Through program traversal and automatic extraction, the efficiency of extracting massive amounts of settlement monitoring data is significantly improved.

[0025] When extracting tunneling parameters using the shield ring number matching extraction rule, the shield tunneling records are used as the original data source. These records contain all tunneling parameters for each shield ring, including soil chamber pressure, grouting pressure, and synchronous grouting volume. Based on the layout pattern of surface settlement monitoring points, for example, the spatial distribution characteristic of one settlement monitoring point every 10 shield rings, the shield ring number corresponding to each settlement monitoring point is determined. Finally, using the corresponding shield ring number as the matching basis, the tunneling parameters at the corresponding shield ring of each monitoring point are accurately selected from the tunneling records. This completes the spatially accurate matching and extraction of tunneling parameters and settlement monitoring points, forming a tunneling parameter dataset that can be analyzed in correspondence with settlement deformation values.

[0026] When extracting geological parameters using the composite stratum correction factor quantitative extraction rule, the engineering geological survey data is first used as the data foundation. Combined with the composite stratum characteristics of the shield tunneling area, the construction strata are divided into two categories: the overlying strata and the strata being traversed. Layer thickness factor, layer depth factor, and the correction factor formed by their product are defined to comprehensively consider the influence of spatial factors such as the thickness of each soil layer and the burial depth of the foundation slab on tunnel construction. For various soil physical and mechanical parameters such as cohesion, internal friction angle, compression modulus, and SPT blow count, the parameters of each individual soil layer in the overlying and traversing strata are weighted and corrected based on the correction factor. Finally, the comprehensive geological parameters of the overlying and traversing strata are obtained by summing or weighted averaging, respectively, achieving the scientific quantitative extraction of geological parameters and forming a geological parameter dataset that closely reflects the actual engineering situation.

[0027] By developing specific extraction rules for different parameters, the problems of misalignment and error in manual extraction are avoided. At the same time, the shortcomings of traditional geological parameter extraction that does not take into account the characteristics of composite strata are solved, ensuring the accuracy of parameters and their suitability for engineering.

[0028] S2. Perform pairwise rank correlation analysis on the extracted settlement deformation values, tunneling parameters and geological parameters to obtain the correlation coefficient matrix between each parameter, and determine the degree of correlation between each parameter and the settlement deformation value based on the correlation coefficient matrix.

[0029] The total parameters refer to all settlement deformation values, tunneling parameters, and geological parameters obtained after extraction using specific rules. It is the complete data object for correlation analysis, covering all core parameters related to shield tunneling settlement analysis. Pairwise rank correlation analysis analyzes the correlation between any two parameters by ranking them according to their rank. It effectively handles discrete data, outliers, and monotonically nonlinear correlations, without requiring the data to satisfy a normal distribution. The correlation coefficient matrix is ​​a matrix formed after pairwise rank correlation analysis of the total parameters, with correlation coefficients as its elements. Each value in the matrix represents the strength and direction of the correlation between the corresponding two sets of parameters.

[0030] Specifically, pairwise rank correlation analysis was performed on the extracted settlement deformation values, tunneling parameters, and geological parameters to obtain the correlation coefficient matrix between each parameter. Based on the correlation coefficient matrix, the degree of correlation between each parameter and the settlement deformation value was determined. This included: using Spearman correlation analysis to sort the parameters by value and assign ranks; calculating the Spearman correlation coefficient between each pair of parameters and integrating them to form the correlation coefficient matrix; and accurately determining the degree of correlation between each parameter and the settlement deformation value based on the correlation coefficient matrix.

[0031] Spearman correlation analysis is a nonparametric statistical correlation analysis method. Its core focus is not on the magnitude of the original variable values, but rather on analyzing the degree and direction of monotonic correlation between two variables by ranking the variable data. It does not require the data to meet preconditions such as normal distribution or homogeneity of variance, making it suitable for correlation analysis of discrete, non-normal, and outlier data. The ranking assigns a sequential number to each parameter value after arranging them in ascending order. The Spearman correlation coefficient is a statistical value calculated from the parameter ranks, ranging from -1 to 1. The absolute value represents the strength of the correlation between parameters, and positive or negative values ​​represent the direction of the correlation; it is a quantitative indicator of the degree of correlation. The correlation coefficient matrix is ​​a matrix formed by organizing the Spearman correlation coefficients between all parameters pairwise according to their row and column correspondences. The matrix elements visually present the quantitative results of the correlation between any two sets of parameters. The degree of correlation between each parameter and the settlement deformation value is categorized as strong, weak, and moderate.

[0032] Using the extracted and standardized settlement deformation values, tunneling parameters, and geological parameters as the analysis objects, Spearman correlation analysis was employed. The original values ​​of each type of parameter were sorted in ascending order, and a corresponding rank was assigned to each sorted value, completing the transformation of all parameters from original values ​​to ranks. Based on the transformed rank sequences of various parameters, pairwise calculations were performed on all parameters to obtain the Spearman correlation coefficient between each pair of parameters. All calculated correlation coefficients were then systematically integrated according to the parameter pairing relationships to form a correlation coefficient matrix that fully presents the pairwise correlation results between all parameters. Using the generated correlation coefficient matrix as the core basis, the Spearman correlation coefficients corresponding to the settlement deformation values ​​for each type of tunneling parameter and geological parameter were extracted. Based on the absolute value and positive / negative characteristics of the coefficients, the strength and direction of the correlation between each parameter and the settlement deformation value were accurately determined, identifying the core parameters that have a significant impact on the settlement deformation value.

[0033] By employing Spearman correlation analysis to rank-normalize the parameters, the problems of discrete, non-normally distributed, and outlier original parameters in shield tunneling are avoided. This approach accurately uncovers the true monotonic correlations between parameters, resulting in results that better reflect engineering realities compared to traditional analysis methods. By calculating Spearman correlation coefficients for each pair of all parameters and integrating them into a correlation coefficient matrix, a systematic and quantitative representation of the correlations between all parameters is achieved. This comprehensively elucidates the inherent correlation patterns between tunneling parameters, geological parameters, settlement deformation values, and each parameter, avoiding the biases inherent in localized analyses.

[0034] S3. Based on the correlation between each parameter and the settlement deformation value, the parameters are ranked by importance, and the parameters are screened according to the importance ranking. The interpretation and prediction performance of settlement deformation is verified based on the screened parameters, and the optimal parameter combination suitable for settlement analysis is determined.

[0035] Interpretive performance refers to the ability of the selected parameters to interpret, fit, and predict the variation patterns of settlement deformation. The optimal parameter combination refers to the set of parameters that, after importance ranking, screening, and performance verification, most accurately reflects the settlement deformation patterns and is most suitable for settlement analysis.

[0036] Specifically, based on the degree of correlation between each parameter and the settlement deformation value, the parameters are ranked in order of importance, and the parameters are screened according to the order of importance. This includes: taking the degree of correlation between each parameter and the settlement deformation value as the core basis, and combining the correlation coefficient matrix to reflect the correlation pattern between parameters, the parameters are ranked in order of importance gradient; low-importance redundant parameters are gradually eliminated in order of importance from low to high.

[0037] Based on the correlation between each parameter and the settlement deformation value, and combined with the correlation coefficient matrix reflecting the inter-parameter correlation patterns, all tunneling parameters and geological parameters are ranked by importance. Parameters with low to high importance are gradually eliminated, eliminating those with little impact on settlement deformation, weak correlation, or redundant information, thus completing parameter simplification and optimization. By combining correlation degree and inter-parameter correlation patterns for importance ranking, the determination of parameter importance becomes more objective and comprehensive, avoiding the one-sidedness of relying solely on a single correlation. Gradually eliminating redundant parameters from low to high importance effectively reduces the number of parameters while retaining core influencing parameters, lowering the complexity and computational load of subsequent analysis. It also eliminates redundant information interference, providing a more concise and reliable parameter foundation for subsequent settlement prediction performance verification and determination of the optimal parameter combination.

[0038] The interpretation and prediction performance of settlement deformation is verified based on the screened parameters, and the optimal parameter combination suitable for settlement analysis is determined. This includes: verifying the interpretation and prediction performance of the remaining parameters after each screening for settlement deformation; stopping the screening when the interpretation and prediction performance of the screened parameters reaches the set conditions, and confirming that the current parameter combination is the optimal parameter combination suitable for settlement analysis.

[0039] Based on the remaining parameters after importance-based sorting, the settlement deformation interpretation and prediction performance of each parameter combination retained after each round of sorting is verified sequentially, focusing on the explanatory power of the parameter combination for settlement deformation variation patterns and the accuracy of settlement value prediction. Simultaneously, performance judgment conditions are preset, continuously advancing the verification and sorting process. When the interpretation and prediction performance of the remaining parameter combination after a round of sorting meets the set conditions, subsequent sorting stops, confirming the current parameter combination as the optimal parameter combination for settlement analysis. By verifying the interpretation and prediction performance of parameters after each round of sorting, the actual suitability of parameter combinations for settlement analysis can be accurately controlled.

[0040] S4. Based on the optimal parameter combination, combined with engineering measured data, equipment performance constraints and settlement control standards, determine the basic boundary conditions of each parameter.

[0041] Specifically, this includes: eliminating abnormal data from the optimal parameter combination by combining engineering measured data, defining the physical feasible boundary of each parameter; delineating the operational feasible boundary of each parameter based on equipment performance; clarifying the safety constraint boundary of each parameter by referring to settlement control standards; and integrating the physical feasible boundary, operational feasible boundary, and safety constraint boundary to form the basic boundary conditions of each parameter.

[0042] Actual measured data from the project refers to various data monitored and recorded on-site during tunnel boring machine (TBM) construction, serving as the practical basis for defining parameter boundaries. Equipment performance constraints, encompassing the TBM's mechanical performance, operational limits, and operating procedures, define the operable range of each tunneling parameter and are the core basis for delineating parameter operating boundaries. Settlement control standards, clearly defined in industry specifications and engineering design documents, specify the allowable settlement range and control requirements for TBM construction, and are the core basis for clarifying parameter safety constraint boundaries.

[0043] Based on the optimal parameter combination as the core foundation, and combined with actual engineering measurement data, outliers in the optimal parameter combination are identified and eliminated. The physical feasible boundaries of each parameter are defined according to the overall data patterns. Based on the tunnel boring machine's mechanical performance, operational limits, and operating procedures, the operational feasible boundaries of each parameter in the optimal parameter combination are delineated to ensure that parameter values ​​meet equipment operating requirements. Referring to industry standards and settlement control standards specified in engineering design documents, the safety constraint boundaries that ensure settlement safety for each parameter in the optimal parameter combination are defined. The physical feasible boundaries, operational feasible boundaries, and safety constraint boundaries are integrated to form basic boundary conditions, providing a benchmark for subsequent dynamic parameter optimization.

[0044] By integrating the optimal parameter combination with engineering measured data, equipment performance, and settlement control standards—three core bases—physical, operational, and safety boundaries were defined and integrated after eliminating abnormal data. This ensured that the basic boundary conditions were consistent with the actual on-site construction and the equipment's operating limits, while also meeting the requirements for settlement safety management. It effectively avoided the one-sidedness and theoretical problems of setting a single boundary, laying a solid data and rule foundation for the subsequent generation of dynamic optimization boundaries, and guaranteeing the safety, rationality, and engineering practicality of shield tunneling construction parameter optimization.

[0045] S5. Based on the basic boundary conditions of each parameter, and according to the heat map correlation analysis model, a dynamic boundary generation system is constructed. The optimal parameter combination, the basic boundary conditions of each parameter, and real-time construction geological data are input into the dynamic boundary generation system to generate the dynamic optimized boundary of each parameter.

[0046] The heatmap correlation analysis model is a parameter correlation analysis model based on heatmap visualization technology. It can intuitively present the correlation strength between parameters and between parameters and real-time geological data, and accurately capture the dynamic change patterns of data.

[0047] The heatmap correlation analysis model is a correlation coefficient matrix heatmap model constructed based on Spearman's correlation analysis method. It performs pairwise rank correlation analysis of all parameters using Spearman's correlation analysis to obtain the correlation coefficient matrix between each parameter. Using this correlation coefficient matrix as data support, heatmap visualization technology transforms the correlation coefficients in the matrix into an intuitive heatmap, forming a heatmap correlation analysis model that combines quantitative analysis and visualization, adapting to the needs of dynamic correlation analysis of shield tunneling construction parameters. The optimal parameter combination, the basic boundary conditions of each parameter, and real-time construction geological data collected during shield tunneling are synchronously input into the dynamic boundary generation system. The system uses the heatmap correlation analysis model to analyze the correlation between the optimal parameter combination and real-time geological data, and dynamically calculates and adjusts the data in conjunction with the basic boundary conditions to generate dynamically optimized boundaries for each parameter that change dynamically with construction conditions and geological conditions.

[0048] By inputting real-time construction geological data, the dynamic optimization boundary can accurately match the real-time fluctuations in geological conditions during construction, avoiding the drawbacks of fixed boundaries that cannot adapt to changes in working conditions, and improving the flexibility and adaptability of the parameter boundaries. Simultaneously, relying on the visualization and precise correlation capabilities of the heat map correlation analysis model, the accuracy of the dynamic optimization boundary is ensured, conforming to both equipment operating limits and settlement control requirements, and adapting to real-time construction conditions on site. This provides a reliable basis for the dynamic control of tunnel boring machine (TBM) construction parameters, improving the engineering practicality and accuracy of parameter optimization.

[0049] Example 2 This invention also provides a system for determining the boundary of shield tunneling construction parameters based on a large model. Figure 2 This is a schematic diagram of the structure of the shield tunneling construction parameter optimization boundary determination system based on a large model provided in an embodiment of the present invention, as shown below. Figure 2 As shown, the system includes: The parameter extraction module is used to extract settlement deformation values, tunneling parameters, and geological parameters from shield tunneling construction data according to different extraction rules. Among them, the settlement deformation values ​​are extracted using automated extraction rules, the tunneling parameters are extracted using shield ring number matching extraction rules, and the geological parameters are extracted using composite stratum correction factor quantification extraction rules. The correlation analysis module is used to perform pairwise rank correlation analysis on the extracted and sorted settlement deformation values, tunneling parameters and geological parameters to obtain the correlation coefficient matrix between each parameter, and to determine the degree of correlation between each parameter and the settlement deformation value based on the correlation coefficient matrix. The filtering module is used to rank the parameters by importance based on the correlation coefficient matrix and the degree of correlation between each parameter and the settlement deformation value, filter the parameters according to the importance ranking, verify the interpretation and prediction performance of settlement deformation based on the filtered parameters, and determine the optimal parameter combination suitable for settlement analysis. The basic boundary determination module is used to determine the basic boundary conditions of each parameter based on the optimal parameter combination, combined with engineering measured data, equipment performance constraints and settlement control standards. The dynamic boundary generation module is used to construct a dynamic boundary generation system based on the basic boundary conditions of each parameter and combined with the heat map correlation analysis model. The optimal parameter combination, the basic boundary conditions of each parameter, and real-time construction geological data are input into the dynamic boundary generation system to generate dynamic optimized boundaries for each parameter.

[0050] The shield tunneling parameter optimization boundary determination system based on a large model provided in this embodiment is used to execute the shield tunneling parameter optimization boundary determination method based on a large model in any of the above embodiments, and has the beneficial effects of any of the above embodiments, which will not be repeated here.

[0051] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the technical solutions of the embodiments of the present invention.

Claims

1. A method for determining a boundary of shield tunneling parameter optimization based on a large model, characterized in that, include: S1. Settlement deformation values, tunneling parameters, and geological parameters are extracted from the shield tunneling construction data according to different extraction rules. Among them, settlement deformation values ​​are extracted using automated extraction rules, tunneling parameters are extracted using shield ring number matching extraction rules, and geological parameters are extracted using composite stratum correction factor quantification extraction rules. S2. Perform pairwise rank correlation analysis on the extracted settlement deformation values, tunneling parameters and geological parameters to obtain the correlation coefficient matrix between each parameter, and determine the degree of correlation between each parameter and the settlement deformation value based on the correlation coefficient matrix. S3. Based on the correlation between each parameter and the settlement deformation value, the importance of each parameter is ranked, the parameters are screened according to the importance ranking, and the interpretation and prediction performance of settlement deformation is verified based on the screened parameters to determine the optimal parameter combination suitable for settlement analysis. S4. Based on the optimal parameter combination, combined with engineering measured data, equipment performance constraints and settlement control standards, determine the basic boundary conditions of each parameter; S5. Based on the basic boundary conditions of each parameter, and according to the heat map correlation analysis model, a dynamic boundary generation system is constructed. The optimal parameter combination, the basic boundary conditions of each parameter, and real-time construction geological data are input into the dynamic boundary generation system to generate the dynamic optimized boundary of each parameter.

2. The large model-based tunneling parameter optimization boundary determination method of claim 1, wherein, In step S1, the settlement deformation value is extracted using automated extraction rules, including: Establish data format parsing rules for shield tunneling deformation monitoring logs; Configure multi-dimensional extraction and retrieval conditions based on monitoring type, spatial location, and construction stage; Based on the search criteria and parsing rules, the target monitoring log file is traversed to automatically extract settlement monitoring data. Invalid data is removed and the reasonableness of the extracted settlement monitoring data is verified to obtain standardized settlement deformation values.

3. The large model-based tunneling parameter optimization boundary determination method of claim 1, wherein, In step S2, pairwise rank correlation analysis is performed on the extracted settlement deformation values, tunneling parameters, and geological parameters to obtain the correlation coefficient matrix between each parameter. Based on the correlation coefficient matrix, the degree of correlation between each parameter and the settlement deformation value is determined, including: Spearman correlation analysis was used to sort the parameters by their numerical values ​​and assign them ranks. Calculate the Spearman correlation coefficient between each pair of parameters and integrate them to form a correlation coefficient matrix; The correlation coefficient matrix is ​​used to accurately determine the degree of correlation between each parameter and the settlement deformation value.

4. The large model-based tunneling parameter optimization boundary determination method of claim 3, wherein, The degree of correlation between each parameter and the settlement deformation value is classified into three types: strong correlation, weak correlation, and moderate correlation.

5. The method of claim 1, wherein, In step S3, based on the correlation between each parameter and the settlement deformation value, the parameters are ranked by importance, and the parameters are then filtered according to the importance ranking, including: Based on the degree of correlation between each parameter and the settlement deformation value, and combined with the correlation coefficient matrix reflecting the correlation pattern between parameters, the parameters are ranked by importance gradient. Redundant parameters with low importance are gradually eliminated in order of increasing importance.

6. The large model-based tunneling parameter optimization boundary determination method of claim 5, wherein, Based on the verified performance of settlement deformation interpretation after parameter screening, the optimal parameter combination suitable for settlement analysis was determined, including: The interpretative and predictive performance of the remaining parameters after each screening on settlement deformation was verified successively. When the interpretation and prediction performance of the filtered parameters meets the set conditions, the filtering is stopped, and the current parameter combination is confirmed to be the optimal parameter combination for settlement analysis.

7. The large model-based tunneling parameter optimization boundary determination method of claim 1, wherein, In step S4, based on the optimal parameter combination, combined with engineering measured data, equipment performance constraints, and settlement control standards, the basic boundary conditions for each parameter are determined, including: By combining measured engineering data to eliminate outliers in the optimal parameter combination, the physical feasible boundaries of each parameter are defined. Based on the equipment performance, define the operational feasibility boundaries for each parameter; Refer to settlement control standards to clarify the safety constraint boundaries of each parameter; Integrate the physical feasible boundary, operational feasible boundary, and safety constraint boundary to form the basic boundary conditions for each parameter.

8. The method for determining the boundary of shield tunneling construction parameters based on a large model according to claim 1, characterized in that, In S5, the heatmap correlation analysis model is a correlation coefficient matrix heatmap model constructed based on Spearman correlation analysis.

9. A system for determining the boundary of shield tunneling construction parameters based on a large model, characterized in that, The system is used to execute the shield tunneling construction parameter optimization boundary determination method based on a large model as described in any one of claims 1-8, and the system comprises: The parameter extraction module is used to extract settlement deformation values, tunneling parameters, and geological parameters from shield tunneling construction data according to different extraction rules. Among them, the settlement deformation values ​​are extracted using automated extraction rules, the tunneling parameters are extracted using shield ring number matching extraction rules, and the geological parameters are extracted using composite stratum correction factor quantification extraction rules. The correlation analysis module is used to perform pairwise rank correlation analysis on the extracted and sorted settlement deformation values, tunneling parameters and geological parameters to obtain the correlation coefficient matrix between each parameter, and to determine the degree of correlation between each parameter and the settlement deformation value based on the correlation coefficient matrix. The filtering module is used to rank the parameters by importance based on the correlation coefficient matrix and the degree of correlation between each parameter and the settlement deformation value, filter the parameters according to the importance ranking, verify the interpretation and prediction performance of settlement deformation based on the filtered parameters, and determine the optimal parameter combination suitable for settlement analysis. The basic boundary determination module is used to determine the basic boundary conditions of each parameter based on the optimal parameter combination, combined with engineering measured data, equipment performance constraints and settlement control standards. The dynamic boundary generation module is used to construct a dynamic boundary generation system based on the basic boundary conditions of each parameter and combined with the heat map correlation analysis model. The optimal parameter combination, the basic boundary conditions of each parameter, and real-time construction geological data are input into the dynamic boundary generation system to generate dynamic optimized boundaries for each parameter.

Images