Sand dolomite tunnel surrounding rock category intelligent evaluation method and system

Abstract

The present application belongs to the technical field of tunnel surrounding rock evaluation, and provides a sandified dolomite tunnel surrounding rock category intelligent evaluation method and system, which comprises the following steps: comparing sandification strip parameters of a preceding area and a subsequent area in a tunnel to determine whether a significant increase in sandification strip characteristics occurs; if so, verifying the adaptability of sandification strip characteristic capture in the preceding area and the subsequent area in the initial evaluation unit to determine whether the sandification characteristic quantification classification requirements of the preceding area and the subsequent area can be met simultaneously; if not, screening a general evaluation unit that can take into account the characteristics of the preceding area and the subsequent area; if there is no general evaluation unit, determining a dedicated evaluation unit for the subsequent area, locating the starting point of the significant increase in sandification strip characteristics, and performing scale catastrophe analysis in combination with the dedicated evaluation unit and the initial evaluation unit to determine the switching transition zone of the initial evaluation unit and the dedicated evaluation unit, thereby ensuring the adaptability and continuity of evaluation of different areas.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of tunnel surrounding rock evaluation, in particular to a method and system for intelligently evaluating the category of dolomitic sandstone tunnel surrounding rock. BACKGROUND

[0002] In tunnel engineering safety monitoring, the dynamic change evaluation of sandification strips is crucial for structural stability research and judgment. Current technologies mostly use fixed scale evaluation units to quantify the sandification characteristics of the entire tunnel, which has significant limitations: first, it is difficult to accurately identify the sudden increase of sandification strip characteristics in the pre- and post-sequences, and the rough parameter comparison method often leads to the omission of sudden signals; second, the evaluation unit lacks targeted adaptation verification, and the fixed unit cannot meet the differentiated quantification needs before and after the sudden change of sandification characteristics, which easily leads to feature capture distortion; third, when the fixed unit fails, there is a lack of general unit screening mechanism and exclusive unit construction standard, which leads to the discontinuity of evaluation continuity; fourth, the correlation between sandification parameter gradient change and unit scale mutation is not considered, which cannot accurately locate the evaluation unit switching transition zone, causing imbalance in the connection of evaluation results in different regions. These problems seriously affect the accuracy and continuity of the dynamic evaluation of sandification strips, and urgent technical solutions are needed to break through.

[0003] Therefore, the application provides a method and system for intelligently evaluating the category of dolomitic sandstone tunnel surrounding rock. SUMMARY

[0004] To make up for the deficiencies of the prior art and solve at least one technical problem raised in the background art.

[0005] The technical solution adopted by the application to solve its technical problems is: a method for intelligently evaluating the category of dolomitic sandstone tunnel surrounding rock, comprising the following methods:

[0006] Step S10: Compare the sandification strip parameters of the pre- and post-sequences in the tunnel to determine whether there is a significant increase in the sandification strip characteristics in the post-sequence;

[0007] Step S20: If there is a significant increase in the sandification strip characteristics, adaptively verify the initial evaluation unit for capturing the sandification strip characteristics in the pre- and post-sequences, and determine whether the initial evaluation unit can meet the sandification characteristic quantification classification needs of the pre- and post-sequences;

[0008] Step S30: If the initial unit cannot be adapted, test the adaptability of different scale evaluation units, and select a general evaluation unit that can consider the characteristics of the pre- and post-sequences;

[0009] Step S40: If there is no general evaluation unit, determine the exclusive evaluation unit of the post-sequence based on the smallest feature of the post-sequence according to the feature complete capture principle;

[0010] Step S50: By analyzing the gradient changes of the sandy strip parameters, locate the significant start point of the sudden increase in sandy strip characteristics, perform scale mutation analysis on the dedicated evaluation unit and the initial evaluation unit, and determine the switching transition zone between the initial evaluation unit and the subsequent regional dedicated evaluation unit in combination with the significant start point of the sudden increase.

[0011] Furthermore, the process for determining whether there is a significant and sudden increase in the characteristics of sand-forming bands is as follows:

[0012] By comparing the sand formation parameters of the preceding and following regions respectively, if the comparison results of each sand formation parameter meet the requirements, it indicates that there is a significant and sudden increase in the sand formation characteristics.

[0013] The parameters of the sand formation strips in the preceding and subsequent regions include: average strip spacing, strip density, and connectivity.

[0014] Furthermore, the compatibility verification process is as follows:

[0015] Typical units are selected from the initial evaluation units in the preceding and following regions, respectively;

[0016] For typical cells in the preceding region, compare the actual strip proportion with the strip proportion calculated for the cell, and calculate the strip proportion error. If the strip proportion error is less than or equal to the preset error, the typical cell is marked as a qualified cell; otherwise, the typical cell is marked as an unqualified cell.

[0017] Based on qualified and unqualified cells, the cell adaptation value of the preceding region is obtained;

[0018] For typical cells in the subsequent region, if the output of the typical cell can distinguish the differences in strip morphology and identify connectivity risks, then the typical cell is marked as a qualified cell.

[0019] Based on the qualified cells, the cell adaptation values ​​of the subsequent region are obtained;

[0020] If the unit adaptation value of the preceding region is greater than or equal to the preset threshold and the unit adaptation value of the following region is less than the preset threshold, it means that the initial evaluation unit cannot simultaneously meet the quantitative classification requirements of sandification characteristics in both the preceding and following regions.

[0021] Furthermore, the method for obtaining the unit adaptation value of the preceding region is as follows:

[0022] The unit fitting ratio is obtained by calculating the proportion of qualified units within a typical unit.

[0023] Based on the non-conforming units, the absolute deviation ratio of the strip proportion error corresponding to the non-conforming units and the preset error is calculated to obtain the relative deviation ratio of the error of each non-conforming unit, and then the average value is processed to obtain the unit adaptation deviation value.

[0024] The difference between the element adaptation ratio and the element adaptation deviation is calculated to obtain the element adaptation value of the preceding region.

[0025] Furthermore, the method for obtaining the unit adaptation value of the subsequent region is as follows:

[0026] The unit fitting ratio is obtained by calculating the proportion of qualified units within a typical unit.

[0027] For the sanded strips within the qualified unit, the dimensional differentiation error ratio of each sanded strip is calculated and averaged to obtain the dimensional differentiation value; wherein, the dimensional differentiation error ratio of the sanded strip is obtained by summing the width error ratio and the thickness error ratio of the sanded strip;

[0028] For the sanded strips within the qualified unit, the proportion of the connected length error of each sanded strip is calculated and averaged to obtain the connectivity identification value. The size differentiation value and the connectivity identification value are summed to obtain the unit adaptation deviation value.

[0029] The difference between the element adaptation ratio and the element adaptation deviation is calculated to obtain the element adaptation value of the subsequent region.

[0030] Furthermore, the process of determining the dedicated evaluation unit for the subsequent region is as follows:

[0031] Obtain the minimum strip spacing in the subsequent region, and set up a dedicated evaluation unit based on the minimum strip spacing in the subsequent region, wherein the dedicated evaluation unit is ≤ 1 / 2 of the minimum strip spacing.

[0032] Furthermore, the process of determining the starting point of the significant increase in the characteristics of the sandy bands is as follows:

[0033] Parameters of sand formation strips were collected along the tunnel axis, resulting in multiple sets of sand formation strip parameters. Each set of sand formation strip parameters includes the average strip spacing, strip density, and connectivity.

[0034] The location point corresponding to the parameter group of the sandy strip that first meets the threshold for the sudden increase in subsequent regions is marked as the significant starting point of the sudden increase in sandy strip characteristics.

[0035] Furthermore, the process of determining the transition zone between the initial evaluation unit and the subsequent region-specific evaluation unit is as follows:

[0036] By using correlation analysis, the correlation dimension parameters are marked in the distribution parameters of the sandy strips output by the different evaluation units before and after the switch. The proportion of the correlation dimension parameters in the distribution parameters of the sandy strips is counted to obtain the scale correlation rate of the evaluation units before and after the switch.

[0037] Based on the scale of the evaluation units before and after the switchover, and the scale correlation rate, a scale correlation data set of the evaluation units before and after the switchover is constructed. Based on the scale of the initial evaluation unit and the exclusive evaluation units of the subsequent region, a matching scale correlation data set is found, thereby determining the scale correlation rate between the initial evaluation unit and the exclusive evaluation units of the subsequent region.

[0038] If the scale correlation rate is less than the preset scale correlation rate, it means that the scale correlation rate is not qualified. Then, the transition zone length corresponding to different scale correlation rates is obtained.

[0039] Different scale correlation rates are integrated into a scale correlation sequence, and different transition band lengths are integrated into a transition sequence;

[0040] By performing correlation analysis on scale-related sequences and transition sequences, a scale-related-transition model is constructed. The scale-related rate compensation value is obtained by calculating the absolute deviation between the scale-related rate and the preset scale-related rate, and then input into the scale-related-transition model to obtain the target transition zone length.

[0041] The transition zone is set at the starting point of a significant increase in the characteristics of sandy strips, and the switching transition zone between the initial evaluation unit and the subsequent regional dedicated evaluation unit is set in combination with the target transition zone length.

[0042] Furthermore, the process of constructing the scale-related-transition model is as follows:

[0043] Calculate the Pearson correlation coefficient between the scale-related sequence and the transition sequence. If the Pearson correlation coefficient is within the preset range, it indicates that there is a linear effect between the scale correlation rate and the transition band length. Then, use the least squares method to fit the transition band length under different scale correlation rates to obtain the scale-related-transition model.

[0044] If the Pearson correlation coefficient is not within the preset range, it indicates that there is a non-linear effect between the scale correlation rate and the transition band length. In this case, the scale correlation rate and the transition band length are integrated into a training data set, and the LSTM model is trained through multiple training data sets to obtain the scale correlation-transition model.

[0045] A smart assessment system for the surrounding rock type of sandy dolomite tunnels includes the following modules:

[0046] Sandification strip sudden change detection module: By comparing the sandification strip parameters of the preceding and subsequent regions in the tunnel, it can determine whether there is a significant sudden increase in the sandification strip characteristics in the subsequent region;

[0047] Evaluation unit adaptation verification module: If there is a significant increase in sand formation characteristics, the initial evaluation unit is adapted to capture sand formation characteristics in the preceding and subsequent regions to determine whether the initial evaluation unit can simultaneously meet the quantitative classification requirements of sand formation characteristics in the preceding and subsequent regions.

[0048] General unit screening and analysis module: If the initial units cannot be adapted simultaneously, test the adaptability of evaluation units at different scales, and screen general evaluation units that can take into account the characteristics of preceding and following regions.

[0049] Dedicated evaluation unit construction module: If there is no general evaluation unit, the dedicated evaluation unit for the subsequent region is determined based on the minimum feature of the subsequent region and the principle of feature complete capture.

[0050] The transition zone analysis module analyzes the gradient changes of sandy strip parameters to locate the starting point of a significant increase in sandy strip characteristics. It performs scale mutation analysis on the dedicated evaluation unit and the initial evaluation unit, and determines the transition zone between the initial evaluation unit and the subsequent regional dedicated evaluation unit by combining the starting point of the significant increase.

[0051] The beneficial effects of this invention are as follows: By comparing the sand formation parameters of the preceding and subsequent regions within the tunnel, a significant surge in sand formation characteristics in the subsequent region can be accurately identified. When a significant surge exists, the initial evaluation unit can be validated for adaptability to determine whether it can simultaneously meet the quantitative classification requirements of sand formation characteristics in both preceding and subsequent regions. If not, a universal evaluation unit that can accommodate both can be selected. If no universal unit exists, a dedicated evaluation unit can be constructed based on the principle of capturing the minimum features and complete features of the subsequent region. Simultaneously, by analyzing the gradient changes in sand formation parameters, the starting point of the significant surge can be located. Combined with scale mutation analysis, the transition zone between the initial and dedicated evaluation units can be determined. Ultimately, a comprehensive and accurate capture and quantitative evaluation of changes in sand formation characteristics within the tunnel can be achieved, ensuring the adaptability and continuity of evaluations in different regions. Attached Figure Description

[0052] The invention will now be further described with reference to the accompanying drawings.

[0053] Figure 1 This is a flowchart illustrating the steps of an intelligent assessment method for the surrounding rock type of a sandy dolomite tunnel according to an embodiment of the present invention.

[0054] Figure 2 This is a logical schematic diagram of an intelligent assessment method for the surrounding rock category of sandy dolomite tunnels according to an embodiment of the present invention;

[0055] Figure 3 This is a flowchart of an intelligent evaluation system for the surrounding rock type of sandy dolomite tunnels, as described in an embodiment of the present invention. Detailed Implementation

[0056] To make the technical means, creative features, objectives and effects of this invention easier to understand, the invention will be further described below in conjunction with specific embodiments.

[0057] Example 1: Please refer to Figures 1-2 As shown in the embodiment of the present invention, an intelligent assessment method for the surrounding rock category of sandy dolomite tunnels includes the following steps:

[0058] Step S10: By comparing the sand formation parameters of the preceding and subsequent regions within the tunnel, determine whether there is a significant increase in the sand formation characteristics in the subsequent region;

[0059] In step S10, the preceding and subsequent regions are adjacent evaluation sections defined based on the spatial progression sequence of tunnel construction or exploration and the sequence of geological condition changes. Taking the tunnel axis as the reference, according to the direction of construction or exploration, the section evaluated first is the preceding region, and the adjacent section evaluated immediately afterward is the subsequent region. Adjacent surrounding rock sections belonging to the same tunnel project are usually bounded by a certain mileage point (such as K0+150), where K0 is the way to represent the mileage station. For example, the preceding region is (K0+100~K0+150), and the subsequent region is (K0+150~K0+200).

[0060] In step S10, the sand formation strip parameters of the preceding and subsequent regions respectively include: average strip spacing, strip density, and connectivity.

[0061] Among them, the average strip spacing represents the distance between the center lines of two adjacent sandy strips; the strip density represents the number of sandy strips per unit area; and the connectivity rate represents the proportion of strips connected to fissures / solution pores.

[0062] In step S10, the process of determining whether there is a significant and sudden increase in the characteristics of sand-forming stripes is as follows:

[0063] By comparing the sand formation strip parameters of the preceding and following regions respectively, if the comparison results of each sand formation strip parameter meet the requirements, it indicates that there is a significant sudden increase in the sand formation strip characteristics. Conversely, if any comparison result of the sand formation strip parameters does not meet the requirements, it indicates that there is no significant sudden increase in the sand formation strip characteristics.

[0064] For example, see Table 1 below;

[0065] Table 1: Criteria for determining a significant and sudden increase in the sand formation parameters and characteristics of the preceding and subsequent regions;

[0066]

[0067] For example, assuming the preceding region is segment A and the following region is segment B, the scan data of segment A shows: D=15cm, N=3 stripes / m², C=5% (1 stripe is connected to the fracture); the scan data of segment B shows: D=4cm, N=10 stripes / m², C=40% (4 stripes are connected to the fracture).

[0068] If the thresholds are: D=4cm<5cm, N=10>9, C=40%>30%, then the characteristic is judged to have increased significantly and sharply.

[0069] Step S20: If there is a significant increase in sand formation characteristics, perform an adaptation verification of the initial evaluation unit for capturing sand formation characteristics in the preceding and subsequent regions, and determine whether the initial evaluation unit can simultaneously meet the quantitative classification requirements of sand formation characteristics in the preceding and subsequent regions.

[0070] In step S20, the initial evaluation unit is a pre-set minimum analysis unit (or basic data acquisition and calculation unit) for the precise quantification and classification of the sandification characteristics of the surrounding rock. Its core function is to "discretize" and "standardize" the complex and continuous geological body of the surrounding rock of the tunnel so that the system can efficiently identify the distribution pattern, characteristic parameters and regional differences of the sandification strips.

[0071] In step S20, if there is a significant sudden increase in the sand-like stripe characteristics, the adaptation verification process for capturing the sand-like stripe characteristics in the preceding and subsequent regions of the initial evaluation unit is as follows:

[0072] Typical units are selected from the initial evaluation units in the preceding and following regions, respectively;

[0073] For a typical cell in the preceding region, compare the actual strip proportion with the strip proportion calculated for the cell, and calculate the strip proportion error;

[0074] Wherein, the strip proportion error = |strip proportion calculated by the unit - actual strip proportion| / actual strip proportion × 100%;

[0075] For example, the actual strip area = 600cm², the unit area = 2500cm², the actual percentage = 600 / 2500×100% = 24%, the calculated strip percentage for the unit is 24% (consistent with the actual strip percentage), and the error = |24%-24%| / 24%×100% = 0 / 24%×100% = 0%;

[0076] If the strip proportion error is less than or equal to the preset error, the typical unit is marked as a qualified unit;

[0077] If the strip proportion error is greater than the preset error, the typical unit will be marked as a non-conforming unit.

[0078] For typical cells in the subsequent region, determine whether they are qualified cells by the output of the typical cells;

[0079] If the output of a typical cell can distinguish the differences in strip morphology and identify connectivity risks, then the typical cell is marked as a qualified cell.

[0080] If the output of a typical cell cannot distinguish the differences in strip morphology or identify connectivity risks, the typical cell will be marked as a non-compliant cell.

[0081] For example, a typical cell contains 10 thin strips of 3cm each (total area = 10 × 3 × 30 = 900cm², accounting for 36%) and 2 thick strips of 15cm each (total area = 2 × 15 × 50 = 1500cm², accounting for 60%). Cell calculation: It can only output the proportions of 36% and 60%, and cannot distinguish between the dispersed thin strips and the continuous thick strips, which indicates that the shape recognition has failed, and the typical cell output cannot distinguish the differences in strip shape.

[0082] For example, in a typical cell, one 10cm strip is connected to two cracks (high risk); cell calculation: only the strip percentage is counted = (10×40) / 2500 = 16%, and the connectivity feature is not marked, which means that the typical cell output cannot identify the connectivity risk;

[0083] For the preceding region;

[0084] The unit fitting ratio is obtained by calculating the proportion of qualified units within a typical unit.

[0085] Based on the non-conforming units, the absolute deviation ratio of the strip proportion error corresponding to the non-conforming units and the preset error is calculated to obtain the relative deviation ratio of the error of the non-conforming units. The relative deviation ratios of the errors of all non-conforming units are averaged to obtain the unit adaptation deviation value.

[0086] The difference between the element adaptation ratio and the element adaptation deviation value is calculated to obtain the element adaptation value of the preceding region.

[0087] For subsequent regions;

[0088] The unit fitting ratio is obtained by calculating the proportion of qualified units within a typical unit.

[0089] For the sanded strips within the qualified unit, the dimensional differentiation error ratio of each sanded strip is calculated. The dimensional differentiation error ratio of the sanded strip is obtained by summing the width error ratio and the thickness error ratio of the sanded strip.

[0090] The width error ratio of the sanded strip = |width of the sanded strip output by the qualified unit - actual width of the sanded strip| / actual width of the sanded strip;

[0091] The thickness error ratio of the sanded strip = |thickness of the sanded strip output by the qualified unit - actual thickness of the sanded strip| / actual thickness of the sanded strip;

[0092] The dimensional differentiation error ratios of all sanded strips are averaged to obtain the dimensional differentiation values;

[0093] For each sanded strip within a qualified unit, the percentage error of the connected length of the sanded strip is calculated. The percentage error of the connected length is calculated as follows: |Percentage of the connected length of the sanded strip output by the qualified unit - Percentage of the actual connected length of the sanded strip| / Percentage of the actual connected length of the sanded strip.

[0094] The connectivity identification value is obtained by averaging the error ratio of the connectivity length of all sandy strips.

[0095] The element fitting deviation value is obtained by summing the size differentiation value and the connectivity identification value.

[0096] The difference between the element adaptation ratio and the element adaptation deviation value is calculated to obtain the element adaptation value of the subsequent region.

[0097] In some embodiments, the cell adaptation value is compared with the cell adaptation threshold;

[0098] If the unit adaptation values ​​of both the preceding and following regions are greater than or equal to the preset threshold, it means that the initial evaluation unit can simultaneously meet the quantitative classification requirements of sandification characteristics in both the preceding and following regions.

[0099] If the unit adaptation value of the preceding region is greater than or equal to the preset threshold and the unit adaptation value of the following region is less than the preset threshold, it means that the initial evaluation unit cannot simultaneously meet the quantitative classification requirements of sandification characteristics in both the preceding and following regions.

[0100] It should be noted that, due to the significant increase in sandy strip characteristics in the subsequent region, the initial evaluation unit usually satisfies the preceding region. Therefore, the unit fit values ​​of both the preceding and subsequent regions are not analyzed here because they are all less than the preset threshold.

[0101] In step S20, the method for selecting typical units in the initial evaluation units of the preceding and following regions is as follows:

[0102] For the preceding region, the selection criteria for typical units are as follows:

[0103] Standard 1: The spacing, thickness, and distribution of sandy strips within a unit must be consistent with the overall characteristics of the preceding area;

[0104] Standard 2: The unit boundary does not cut the main body of the sanded strip;

[0105] Standard 3: No local anomalies within the unit (such as isolated dissolution pores or traces of construction disturbance);

[0106] For example, a 50cm×50cm unit (consistent with the initial assessment unit size of the preceding area) is selected, located in the middle of the preceding area (e.g., the right side of the K0+120 tunnel face). The unit contains a 15cm thick and 40cm long sandy strip (with a 30° angle to the tunnel axis), and the strip spacing (with adjacent strips) is 18cm (in the range of 10-20cm). The strips have no fracture intersections and a connectivity rate of 0%. The rest of the area within the unit is intact dolomite (without sandy characteristics). This unit can be used as a typical unit of the preceding area.

[0107] For subsequent regions, the selection criteria for typical units are as follows:

[0108] Standard 1: The unit must contain a dense interweaving pattern of strips (such as more than 3 intersecting strips) and at least one connection point between the strips and the crack;

[0109] Standard 2: The strip spacing, density, and connectivity must meet the criteria for a significant increase in the characteristics of the subsequent region (e.g., spacing ≤ 5cm, density ≥ 8 strips / m²).

[0110] Standard 3: Ensure that the element can fully contain the local features of a single strip (such as strip thickness and details of intersection with cracks);

[0111] For example, a 10cm×10cm unit (a candidate scale adapted to the dense characteristics of the subsequent region) is selected and located in the front part of the subsequent region (such as the left side of the K0+160 face). The unit contains three sandy strips with a thickness of 3-5cm (parallel to each other, spaced 4cm apart). One strip intersects with a 2mm wide fissure (forming a connection point), with a connectivity rate of 33% (1 / 3 of the strips are connected). The strip density is converted to 10 strips / m² (meeting the standard of ≥8 strips / m²). There are no intact rock blocks in the unit, and the whole is in a state of "interwoven sandy strips". This unit can be used as a typical unit of the subsequent region.

[0112] Step S30: If the initial unit cannot be adapted simultaneously, test the adaptability of evaluation units at different scales and screen for a general evaluation unit that can take into account the characteristics of preceding and following regions.

[0113] In step S30, if the initial units cannot be adapted simultaneously, the process of testing the adaptability of units at different scales is as follows:

[0114] Among them, the evaluation units of different scales are designed based on the minimum scale of the strip. For example, the minimum scale of the strip is (10cm thick, 3cm spacing), and three sets of evaluation units are designed: 25cm×25cm, 10cm×10cm, and 5cm×5cm.

[0115] Determine whether different evaluation units can simultaneously meet the quantitative classification requirements of sandification characteristics in both preceding and subsequent regions. If they do, select a general evaluation unit that can take into account the characteristics of both preceding and subsequent regions.

[0116] If the condition cannot be met, proceed to step S40;

[0117] It should be noted that the determination of whether the quantitative classification requirements of sandification characteristics of the preceding and subsequent regions can be met simultaneously has been explained in step S20 above, and will not be explained here again.

[0118] Step S40: If there is no universal evaluation unit, then based on the minimum features of the subsequent region, determine the specific evaluation unit of the subsequent region according to the principle of feature complete capture;

[0119] In step S40, if there is no universal evaluation unit, the process of determining the specific evaluation unit for the subsequent region based on the minimum features of the subsequent region and according to the principle of feature complete capture is as follows:

[0120] Obtain the minimum strip spacing in the subsequent region, and set up a dedicated evaluation unit based on the minimum strip spacing in the subsequent region;

[0121] Among them, the dedicated evaluation unit is ≤ 1 / 2 of the minimum strip spacing. For example, if the minimum strip spacing in the subsequent region is 3cm, then the dedicated evaluation unit is 1.5cm (to ensure that single strips and interlaced strips can be distinguished).

[0122] Step S50: By analyzing the gradient changes in the parameters of the sandy strips, locate the starting point of a significant and abrupt increase in the characteristics of the sandy strips.

[0123] Scale mutation analysis was performed on the dedicated evaluation unit and the initial evaluation unit, and the switching transition zone between the initial evaluation unit and the subsequent regional dedicated evaluation unit was determined by combining the significant increase start point.

[0124] In step S50, the process of determining the starting point of a significant increase in the characteristics of the sandy bands is as follows:

[0125] The parameters of the sand formation strips were collected along the tunnel axis, and multiple sets of sand formation strip parameters were obtained. Each set of sand formation strip parameters includes the average strip spacing, strip density and connectivity. Each set of sand formation strip parameters was compared with the subsequent region sudden increase threshold.

[0126] The location point corresponding to the parameter group of the sandy strip that first meets the threshold of subsequent region increase is marked as the significant start point of the sandy strip characteristics.

[0127] Among them, the sandy strip parameter set that satisfies the subsequent region abrupt increase threshold means that the average strip spacing, strip density and connectivity of the sandy strip parameter set all satisfy the subsequent region abrupt increase threshold.

[0128] In step S50, the process of performing scale mutation analysis on the dedicated evaluation unit and the initial evaluation unit, and determining the switching transition zone between the initial evaluation unit and the subsequent region's dedicated evaluation unit based on the significant abrupt increase starting point, is as follows:

[0129] Acquire historical data during the intelligent assessment of the surrounding rock category of sandy dolomite tunnels. The historical data includes: distribution parameters of sandy stripes output by assessment units before different switches and distribution parameters of sandy stripes output by assessment units after different switches during multiple historical assessments of the surrounding rock category of sandy dolomite tunnels. The distribution parameters include, but are not limited to, the length, density, connectivity, and distribution uniformity of the sandy stripes (the number of distribution parameters is N).

[0130] Correlation analysis was performed on the same sandy strip distribution parameters output by the evaluation unit before and after the switch;

[0131] If the same sand formation strip distribution parameters output by the evaluation unit before and after the switch can be directly converted or logically correspond, then the sand formation strip distribution parameters are marked as related dimension parameters.

[0132] If the same sandy strip distribution parameters output by the evaluation unit before and after the switch cannot be directly converted or logically corresponded, then the sandy strip distribution parameters are marked as non-correlated dimension parameters.

[0133] For example, the logical correspondence is represented as follows:

[0134] Suppose we analyze the sand formation bands in the same region across two evaluation units at different scales:

[0135] Large-scale units (such as 1:10000 maps, with "area blocks" as the evaluation unit): The output parameter is "average width level of sandification strips", which is divided into 3 levels - narrow (<2m), medium (2-5m), and wide (>5m);

[0136] Small-scale units (such as 1:1000 maps, with "plots" as the evaluation unit): The output parameter is the "actual average width of the sandy strips", which is directly expressed by specific values ​​(such as 1.8m, 3.5m, 6.2m).

[0137] The parameters of the two units have a clear logical correspondence:

[0138] The small-scale "1.8m" corresponds to the large-scale "narrow";

[0139] The small-scale "3.5m" corresponds to the large-scale "medium";

[0140] The small-scale "6.2m" corresponds to the large-scale "width";

[0141] The "average width grade of sandy strips" (large scale) and the "actual average width of sandy strips" (small scale) are related dimension parameters. Because they describe the same attribute (strip width) and can be converted or corresponded across scales through clear logical rules (numerical range corresponds to grade), they can be directly correlated without additional transition processing (for example, using specific values ​​at the small scale to verify the accuracy of the large-scale grade division).

[0142] For example, a direct conversion can be expressed as: the output of the large-scale unit is "total length of the sandy strip (km)";

[0143] The small-scale unit outputs "total length of sand-formed strips (m)";

[0144] The only difference between the two is the unit conversion (1km=1000m), which means they are directly convertible related dimension parameters.

[0145] The proportion of statistical correlation dimension parameters in the distribution parameters of sand formation strips is used to obtain the evaluation unit before and after the switch.

[0146] Scale correlation rate (in %)

[0147] The scales of the evaluation units before and after the switchover are obtained respectively, and the scale correlation data sets of the evaluation units before and after the switchover are constructed by combining the scale correlation rate.

[0148] For example, if the evaluation unit size before the switch is (50cm×50cm) and after the switch it is (20cm×20cm), and the scale correlation rate is 60%, then the scale-correlated data group is [50cm-20cm-60%];

[0149] The scale-related data sets of different evaluation units before and after the switch are aggregated to obtain a scale-related database;

[0150] Based on the scale of the initial evaluation unit and the dedicated evaluation units of the subsequent region, a matching scale association data set is found in the scale association database, and the scale association rate between the initial evaluation unit and the dedicated evaluation units of the subsequent region is determined based on the scale association data set.

[0151] For example, finding a matching scale-related data set means that the scale of the initial evaluation unit and the dedicated evaluation unit of the subsequent region corresponds to the scale of the evaluation units before and after the switch contained in the scale-related data set. For example, the scale of the initial evaluation unit and the dedicated evaluation unit of the subsequent region is 50cm and 20cm, which is the same as the scale of the scale-related data set [50cm-20cm-60%].

[0152] In some embodiments, the scale correlation rate is compared with a preset scale correlation rate;

[0153] If the scale correlation rate is greater than or equal to the preset scale correlation rate, it means that the scale correlation rate is qualified and no processing is required.

[0154] If the scale correlation rate is less than the preset scale correlation rate, it means that the scale correlation rate is not qualified. In this case, the historical switching data of the intelligent assessment of the surrounding rock category of the sandy dolomite tunnel is obtained. The historical switching data includes the length of the transition zone corresponding to different scale correlation rates.

[0155] Among them, the transition band length corresponding to different scale correlation rates indicates that the scale correlation rate is adapted to the corresponding transition band length.

[0156] It should be noted that when switching from one scale of evaluation unit (e.g., large scale) to another scale (e.g., small scale), the data may become "incompatible" due to differences in observation dimensionality and precision, resulting in information loss. The role of the transition zone is to compensate for excessive information loss, i.e., insufficient scale correlation rate. The transition zone is the connecting area between two evaluation units. Here, data is not collected using only a single scale, but simultaneously using multi-scale data collection (e.g., recording both macroscopic statistics at the large scale and microscopic details at the small scale). This allows for cross-validation and fusion of the output data from the two scale evaluation units, thereby reducing information loss and compensating for the scale correlation rate.

[0157] The correlation rates at different scales are integrated into a scale correlation sequence, and the transition band lengths corresponding to the correlation rates at different scales are integrated into a transition sequence.

[0158] Calculate the Pearson correlation coefficient between the scale-related sequence and the transition sequence. If the Pearson correlation coefficient is within the preset range, it indicates that there is a linear effect between the scale correlation rate and the transition band length. Then, use the least squares method to fit the transition band length under different scale correlation rates to obtain the scale-related-transition model.

[0159] If the Pearson correlation coefficient is not within the preset range, it indicates that there is a non-linear effect between the scale correlation rate and the transition band length. In this case, the scale correlation rate and the transition band length are integrated into a training data set, and the LSTM model is trained through multiple training data sets to obtain the scale correlation-transition model.

[0160] Calculate the absolute deviation between the scale correlation rate and the preset scale correlation rate to obtain the scale correlation rate compensation value;

[0161] The scale correlation rate compensation value is input into the scale correlation-transition model to obtain the target transition band length;

[0162] The transition zone is set at the starting point of a significant increase in the characteristics of sandy strips, and the switching transition zone between the initial evaluation unit and the subsequent regional-specific evaluation unit is set in combination with the target transition zone length.

[0163] For example, assuming the starting point of the significant increase is K0+120 and the target transition zone length is 20, then the starting point of the transition zone is K0+110 and the ending point is K0+130.

[0164] The technical solution of this invention is as follows: By comparing the sand formation parameters of the preceding and subsequent regions within the tunnel, a significant sudden increase in the sand formation characteristics of the subsequent region is accurately identified. When a significant sudden increase exists, the adaptability of the initial evaluation unit can be verified to determine whether it can simultaneously meet the quantitative classification requirements of the sand formation characteristics of the preceding and subsequent regions. If not, a universal evaluation unit that can take both into account is selected. If there is no universal unit, a dedicated evaluation unit is constructed based on the principle of capturing the minimum feature and complete feature of the subsequent region. At the same time, by analyzing the gradient change of the sand formation parameters, the starting point of the significant sudden increase is located. Combined with scale mutation analysis, the transition zone between the initial and dedicated evaluation units is determined. Ultimately, a comprehensive and accurate capture and quantitative evaluation of the changes in the sand formation characteristics within the tunnel is achieved, ensuring the adaptability and continuity of the evaluation in different regions.

[0165] Example 2: Please refer to Figure 3 As shown in the embodiment of the present invention, an intelligent assessment system for the surrounding rock type of sandy dolomite tunnels includes the following modules:

[0166] Sandification strip sudden change detection module: By comparing the sandification strip parameters of the preceding and subsequent regions in the tunnel, it can determine whether there is a significant sudden increase in the sandification strip characteristics in the subsequent region;

[0167] Evaluation unit adaptation verification module: If there is a significant increase in sand formation characteristics, the initial evaluation unit is adapted to capture sand formation characteristics in the preceding and subsequent regions to determine whether the initial evaluation unit can simultaneously meet the quantitative classification requirements of sand formation characteristics in the preceding and subsequent regions.

[0168] General unit screening and analysis module: If the initial units cannot be adapted simultaneously, test the adaptability of evaluation units at different scales, and screen general evaluation units that can take into account the characteristics of preceding and following regions.

[0169] Dedicated evaluation unit construction module: If there is no general evaluation unit, the dedicated evaluation unit for the subsequent region is determined based on the minimum feature of the subsequent region and the principle of feature complete capture.

[0170] The transition zone analysis module analyzes the gradient changes of sandy strip parameters to locate the starting point of a significant increase in sandy strip characteristics. It performs scale mutation analysis on the dedicated evaluation unit and the initial evaluation unit, and determines the transition zone between the initial evaluation unit and the subsequent regional dedicated evaluation unit by combining the starting point of the significant increase.

[0171] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for intelligently assessing the surrounding rock type of sandy dolomite tunnels, characterized in that: Including the following methods: Step S10: By comparing the sand formation parameters of the preceding and subsequent regions within the tunnel, determine whether there is a significant increase in the sand formation characteristics in the subsequent region; The process for determining whether there is a significant and sudden increase in the characteristics of sand-forming bands is as follows: By comparing the sand formation parameters of the preceding and following regions respectively, if the comparison results of each sand formation parameter meet the requirements, it indicates that there is a significant and sudden increase in the sand formation characteristics. The sand-forming strip parameters of the preceding and subsequent regions respectively include: average strip spacing, strip density, and connectivity. Step S20: If there is a significant increase in sand formation characteristics, perform an adaptation verification of the initial evaluation unit for capturing sand formation characteristics in the preceding and subsequent regions, and determine whether the initial evaluation unit can simultaneously meet the quantitative classification requirements of sand formation characteristics in the preceding and subsequent regions. The compatibility verification process is as follows: Typical units are selected from the initial evaluation units in the preceding and following regions, respectively; For typical cells in the preceding region, compare the actual strip proportion with the strip proportion calculated for the cell, and calculate the strip proportion error. If the strip proportion error is less than or equal to the preset error, the typical cell is marked as a qualified cell; otherwise, the typical cell is marked as an unqualified cell. Based on qualified and unqualified cells, the cell adaptation value of the preceding region is obtained; For typical cells in the subsequent region, if the output of the typical cell can distinguish the differences in strip morphology and identify connectivity risks, then the typical cell is marked as a qualified cell. Based on the qualified cells, the cell adaptation values ​​of the subsequent region are obtained; If the unit adaptation value of the preceding region is greater than or equal to the preset threshold and the unit adaptation value of the following region is less than the preset threshold, it means that the initial evaluation unit cannot simultaneously meet the quantitative classification requirements of sandification characteristics in both the preceding and following regions. Step S30: If the initial unit cannot be adapted simultaneously, test the adaptability of evaluation units at different scales and screen for a general evaluation unit that can take into account the characteristics of preceding and following regions. Step S40: If there is no universal evaluation unit, then based on the minimum features of the subsequent region, determine the specific evaluation unit of the subsequent region according to the principle of feature complete capture; Step S50: By analyzing the gradient changes of the sandy strip parameters, locate the significant start point of the sudden increase in sandy strip characteristics, perform scale mutation analysis on the dedicated evaluation unit and the initial evaluation unit, and determine the switching transition zone between the initial evaluation unit and the subsequent regional dedicated evaluation unit in combination with the significant start point of the sudden increase.

2. The intelligent assessment method for the surrounding rock type of sandy dolomite tunnels according to claim 1, characterized in that: The method for obtaining the unit adaptation value of the preceding region is as follows: The unit fitting ratio is obtained by calculating the proportion of qualified units within a typical unit. Based on the non-conforming units, the absolute deviation ratio of the strip proportion error corresponding to the non-conforming units and the preset error is calculated to obtain the relative deviation ratio of the error of each non-conforming unit, and then the average value is processed to obtain the unit adaptation deviation value. The difference between the element adaptation ratio and the element adaptation deviation is calculated to obtain the element adaptation value of the preceding region.

3. The intelligent assessment method for the surrounding rock category of sandy dolomite tunnels according to claim 1, characterized in that: The method for obtaining the unit adaptation value of the subsequent region is as follows: The unit fitting ratio is obtained by calculating the proportion of qualified units within a typical unit. For the sanded strips within the qualified unit, the dimensional differentiation error ratio of each sanded strip is calculated and averaged to obtain the dimensional differentiation value; wherein, the dimensional differentiation error ratio of the sanded strip is obtained by summing the width error ratio and the thickness error ratio of the sanded strip; For the sanded strips within the qualified unit, the proportion of the connected length error of each sanded strip is calculated and averaged to obtain the connectivity identification value. The size distinction value and the connectivity identification value are summed to obtain the unit adaptation deviation value. The difference between the element adaptation ratio and the element adaptation deviation is calculated to obtain the element adaptation value of the subsequent region.

4. The intelligent assessment method for the surrounding rock category of sandy dolomite tunnels according to claim 3, characterized in that: The process of determining the dedicated evaluation unit for the subsequent region is as follows: Obtain the minimum strip spacing in the subsequent region, and set up a dedicated evaluation unit based on the minimum strip spacing in the subsequent region, wherein the dedicated evaluation unit is ≤ 1 / 2 of the minimum strip spacing.

5. The intelligent assessment method for the surrounding rock category of sandy dolomite tunnels according to claim 4, characterized in that: The process of identifying the starting point of a significant and sudden increase in the characteristics of the sandy strips is as follows: Parameters of sand formation strips were collected along the tunnel axis, resulting in multiple sets of sand formation strip parameters. Each set of sand formation strip parameters includes the average strip spacing, strip density, and connectivity. The location point corresponding to the parameter group of the sandy strip that first meets the threshold for the sudden increase in subsequent regions is marked as the significant starting point of the sudden increase in sandy strip characteristics.

6. The intelligent assessment method for the surrounding rock category of sandy dolomite tunnels according to claim 5, characterized in that: The process of determining the transition zone between the initial evaluation unit and the subsequent region-specific evaluation unit is as follows: By using correlation analysis, the correlation dimension parameters are marked in the distribution parameters of the sandy strips output by the different evaluation units before and after the switch. The proportion of the correlation dimension parameters in the distribution parameters of the sandy strips is counted to obtain the scale correlation rate of the evaluation units before and after the switch. Based on the scale of the evaluation units before and after the switchover, and the scale correlation rate, a scale correlation data set of the evaluation units before and after the switchover is constructed. Based on the scale of the initial evaluation unit and the exclusive evaluation units of the subsequent region, a matching scale correlation data set is found, thereby determining the scale correlation rate between the initial evaluation unit and the exclusive evaluation units of the subsequent region. If the scale correlation rate is less than the preset scale correlation rate, it means that the scale correlation rate is not qualified. Then, the transition zone length corresponding to different scale correlation rates is obtained. Different scale correlation rates are integrated into a scale correlation sequence, and different transition band lengths are integrated into a transition sequence; By performing correlation analysis on scale-related sequences and transition sequences, a scale-related-transition model is constructed. The scale-related rate compensation value is obtained by calculating the absolute deviation between the scale-related rate and the preset scale-related rate, and then input into the scale-related-transition model to obtain the target transition zone length. The transition zone is set at the starting point of a significant increase in the characteristics of sandy strips, and the switching transition zone between the initial evaluation unit and the subsequent regional dedicated evaluation unit is set in combination with the target transition zone length.

7. The intelligent assessment method for the surrounding rock category of sandy dolomite tunnels according to claim 6, characterized in that: The process of constructing the scale correlation-transition model is as follows: Calculate the Pearson correlation coefficient between the scale-related sequence and the transition sequence. If the Pearson correlation coefficient is within the preset range, it indicates that there is a linear effect between the scale correlation rate and the transition band length. Then, use the least squares method to fit the transition band length under different scale correlation rates to obtain the scale-related-transition model. If the Pearson correlation coefficient is not within the preset range, it indicates that there is a non-linear effect between the scale correlation rate and the transition band length. In this case, the scale correlation rate and the transition band length are integrated into a training data set, and the LSTM model is trained through multiple training data sets to obtain the scale correlation-transition model.

8. An intelligent assessment system for the surrounding rock type of sandy dolomite tunnels, characterized in that: Includes the following modules: Sandification strip sudden change detection module: By comparing the sandification strip parameters of the preceding and subsequent regions in the tunnel, it can determine whether there is a significant sudden increase in the sandification strip characteristics in the subsequent region; The process for determining whether there is a significant and sudden increase in the characteristics of sand-forming bands is as follows: By comparing the sand formation parameters of the preceding and following regions respectively, if the comparison results of each sand formation parameter meet the requirements, it indicates that there is a significant and sudden increase in the sand formation characteristics. The sand-forming strip parameters of the preceding and subsequent regions respectively include: average strip spacing, strip density, and connectivity. Evaluation unit adaptation verification module: If there is a significant increase in sand formation characteristics, the initial evaluation unit is adapted to capture sand formation characteristics in the preceding and subsequent regions to determine whether the initial evaluation unit can simultaneously meet the quantitative classification requirements of sand formation characteristics in the preceding and subsequent regions. The compatibility verification process is as follows: Typical units are selected from the initial evaluation units in the preceding and following regions, respectively; For typical cells in the preceding region, compare the actual strip proportion with the strip proportion calculated for the cell, and calculate the strip proportion error. If the strip proportion error is less than or equal to the preset error, the typical cell is marked as a qualified cell; otherwise, the typical cell is marked as an unqualified cell. Based on qualified and unqualified cells, the cell adaptation value of the preceding region is obtained; For typical cells in the subsequent region, if the output of the typical cell can distinguish the differences in strip morphology and identify connectivity risks, then the typical cell is marked as a qualified cell. Based on the qualified cells, the cell adaptation values ​​of the subsequent region are obtained; If the unit adaptation value of the preceding region is greater than or equal to the preset threshold and the unit adaptation value of the following region is less than the preset threshold, it means that the initial evaluation unit cannot simultaneously meet the quantitative classification requirements of sandification characteristics in both the preceding and following regions. General unit screening and analysis module: If the initial units cannot be adapted simultaneously, test the adaptability of evaluation units at different scales, and screen general evaluation units that can take into account the characteristics of preceding and following regions. Dedicated evaluation unit construction module: If there is no general evaluation unit, the dedicated evaluation unit for the subsequent region is determined based on the minimum feature of the subsequent region and the principle of feature complete capture. The transition zone analysis module analyzes the gradient changes of sandy strip parameters to locate the starting point of a significant increase in sandy strip characteristics. It performs scale mutation analysis on the dedicated evaluation unit and the initial evaluation unit, and determines the transition zone between the initial evaluation unit and the subsequent regional dedicated evaluation unit by combining the starting point of the significant increase.

Images