A method for intelligent determination of a tunnel lining type

Abstract

A kind of intelligent determination method of tunnel lining type, comprising: obtaining the designed tunnel data set;According to the designed tunnel data set, construct lining type sub data set;According to the lining type sub data set, construct GWO-NGBoost lining type prediction model;For the tunnel to be predicted, the section division of surrounding rock grade is determined based on geological survey data;For each section of the tunnel to be predicted, respectively extract features and input into the lining type prediction model of corresponding surrounding rock grade, realize the prediction of lining type.The present application establishes GWO-NGBoost model in each tunnel lining type sub data set, realizes the prediction of lining type, the whole process is standard unified, can be automated and efficiently completed.The present application constructs corresponding lining type prediction model in each tunnel lining type sub data set, finally constructs multiple lining type prediction models, which can effectively improve the precision of lining type prediction.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of tunnel design, and particularly relates to an intelligent determination method of tunnel lining type. BACKGROUND

[0002] At present, the selection of tunnel lining type depends on artificial experience for determination. The main process is: (1) based on geological survey data, the surrounding rock grade is determined, and the tunnel is divided into different surrounding rock grade sections; (2) for a section of the same surrounding rock grade, the lining type is determined according to the geological conditions such as buried depth, bias, and groundwater development; (3) the lining type of each surrounding rock grade section is determined gradually until all sections are determined. Therefore, the existing design method is too vague and highly dependent on the experience of designers, and in addition, the design efficiency is low, and the lining type of each section needs to be determined gradually by artificial, so an intelligent calculation method is needed to improve the determination efficiency and accuracy of the tunnel lining type. SUMMARY

[0003] In view of the above problems, the present application is proposed in order to provide an intelligent determination method of tunnel lining type which overcomes the above problems or at least partially solves the above problems.

[0004] In order to solve the above technical problems, the present application discloses the following technical scheme:

[0005] In a first aspect, the present application discloses an intelligent determination method of tunnel lining type, comprising:

[0006] S100. Obtain a designed tunnel data set;

[0007] S200. According to the designed tunnel data set, a lining type sub-data set is constructed;

[0008] S300. According to the lining type sub-data set, a GWO-NGBoost lining type prediction model is constructed;

[0009] S400. For a tunnel to be predicted, the section division of the surrounding rock grade is determined based on the geological survey data;

[0010] S500. For each section of the tunnel to be predicted, the features are extracted and input into the lining type prediction model corresponding to the surrounding rock grade, to realize the prediction of the lining type.

[0011] Further, in S100, the designed tunnel data set at least includes the longitude and latitude of the tunnel, the length of the tunnel, the starting and ending mileage of the surrounding rock section, the buried depth of multiple mileages in the tunnel, and the corresponding surrounding rock grade and lining type.

[0012] Further, in S200, according to the designed tunnel dataset, a lining type sub-dataset is constructed, and the specific method comprises: feature extraction and conversion are performed on the obtained dataset, and the datasets with the same surrounding rock grade are divided into a sub-dataset, and the sub-dataset is resampled to form a lining type sub-dataset, and finally a plurality of lining type sub-datasets corresponding to a plurality of surrounding rock grades are constructed.

[0013] Further, feature extraction and conversion are performed on the obtained dataset, and the specific method comprises: tunnel longitude and latitude, tunnel length, surrounding rock section length, and buried depth features are extracted from the dataset, and plane coordinate conversion is performed on the tunnel longitude and latitude features, and the geographic coordinates such as longitude and latitude are converted into plane coordinates, and the coordinate conversion formula is as formula (1)-(4):

[0014]

[0015] In the formula, x log , x lat are the longitude and latitude values of the tunnel, x log_sin , x log_cos are the coordinate values after longitude conversion, and x lat_sin , x lat_cos are the coordinate values after latitude conversion.

[0016] Further, the sub-dataset is resampled to form a lining type sub-dataset, and the specific method comprises:

[0017] Downsampling is performed on the sub-dataset, and in the sub-dataset, for a certain lining type with too many samples, one sample is randomly selected without repetition each time until a specified number of samples are selected; upsampling is performed on the sub-dataset, and in the sub-dataset, for a certain lining type with too few samples, a specified number of new samples are generated based on the Smote algorithm, and the specific method of the Smote algorithm comprises: a certain lining type with too few samples is randomly selected, the k nearest neighbors of the sample are calculated, 1 sample in the k neighbors is randomly selected, and 1 point between the neighbor sample and the sample point is randomly selected as a newly generated sample, and the steps are repeated until a specified number of samples are generated.

[0018] Further, finally, a plurality of lining type sub-datasets corresponding to a plurality of surrounding rock grades are constructed, and the specific method comprises: the resampling step is repeated for each surrounding rock grade corresponding sub-dataset until all surrounding rock grade corresponding sub-datasets are covered, i.e., all surrounding rock grade corresponding lining type sub-datasets are constructed.

[0019] Further, in S300, according to the lining type sub-dataset, a GWO-NGBoost lining type prediction model is constructed, and the specific method comprises:

[0020] S301. Data set division, in the same lining type sub-data set, the data set is divided into training set and test set, the training set is used to build the model and find the optimal hyperparameter, the test set is used to evaluate the model;

[0021] S302. Finding the optimal hyperparameters of the NGBoost model using the GWO algorithm;

[0022] S303. Building a lining type prediction model, for each lining type sub-data set, based on the obtained optimal hyperparameters, the NGBoost lining type prediction model is established on the entire training set, and finally a plurality of NGBoost lining type prediction models corresponding to a plurality of lining type sub-data sets are obtained.

[0023] Further, in S302, the GWO algorithm is used to find the optimal hyperparameters of the NGBoost model, and the specific method includes:

[0024] S3021. Setting the hyperparameter range, setting the value range of the parameters n_estimators, learning_rate, minibatch_frac, and col_sample in the NGBoost model;

[0025] S3022. Initialize the position of the wolf pack, randomly select a number for each hyperparameter in the value range, and generate a list of M groups of hyperparameters as the initialized wolf pack position;

[0026] S3023. Calculate the fitness value of the wolf individual and save the best three wolves, set the position parameters of each wolf as the hyperparameter values of the NGBoost model, and take the negative value of the average precision based on K-fold cross-validation as the fitness value of the wolf individual, the calculation formula of the precision is shown in formula 5, and the precision represents the proportion of the samples that are actually positive in the samples that are predicted to be positive;

[0027]

[0028] In the formula, Precision is the precision, TP is the sample that is predicted to be positive and actually positive, and FP is the sample that is predicted to be positive and actually negative; after obtaining the fitness value of the wolf individual, take the three wolves with the smallest fitness value as α, β, and δ wolves;

[0029] S3024. Calculate the correlation coefficient, the convergence factor a, the parameter C, and the parameter A three coefficients are calculated according to formulas 6-8:

[0030]

[0031] C i =2×r i (7)

[0032] A i = 2 x a x r i -a (8)

[0033] In the formula, t is the current iteration number, t o is the total iteration number, i takes the value of 1, 2, 3, r i is a random number between 0 and 1;

[0034] S3025. Update the position of the wolf pack, update the position of the wolf pack according to alpha, beta and delta wolves, and the update formula is shown in formula 9-11:

[0035]

[0036] In the formula, X represents the position of the wolf pack, X α , X β , X δ represent the positions of alpha, beta and delta wolves, and X(t+1) represents the updated position of the wolf pack;

[0037] S3206. Obtain the optimal hyperparameter, repeat S3203-S3205 until the iteration number reaches the maximum iteration number, and the position of the alpha wolf is the obtained optimal hyperparameter.

[0038] Further, in S500, for each section of the tunnel to be predicted, features are extracted and input into the corresponding surrounding rock grade lining type prediction model to realize prediction of the lining type, and the specific method includes:

[0039] S501. Feature extraction and conversion, on the surrounding rock grade section, extract the features of buried depth, tunnel longitude and latitude, tunnel length, surrounding rock section length, and perform conversion;

[0040] S502. Obtain the lining type of each section, for each surrounding rock grade section, find the lining type prediction model corresponding to the surrounding rock grade, input the features extracted in S501 into the lining type prediction model, and obtain the lining type of the corresponding section;

[0041] S503. Determine the lining type, detect each adjacent section in S502, if the lining types of adjacent sections are the same, merge them into one section, and the merged section and the corresponding lining type are the final determined lining type.

[0042] In a second aspect, the embodiments of the present application disclose an electronic device, comprising:

[0043] One or more processors;

[0044] Memory for storing one or more programs;

[0045] When the one or more programs are executed by the one or more processors, the one or more processors implement the intelligent determination method of the tunnel lining type.

[0046] The above technical solution provided by the embodiment of the present application has at least the following beneficial effects:

[0047] The present application discloses an intelligent determination method of tunnel lining type, comprising: acquiring a designed tunnel data set; constructing a lining type sub-data set according to the designed tunnel data set; constructing a GWO-NGBoost lining type prediction model according to the lining type sub-data set; determining the section division of surrounding rock grade based on geological survey data for the tunnel to be predicted; for each section of the tunnel to be predicted, respectively extracting features and inputting them into the lining type prediction model corresponding to the surrounding rock grade to realize the prediction of the lining type. The present application realizes the prediction of the lining type by establishing a GWO-NGBoost model in each tunnel lining type sub-data set, and the whole process is standardized and can be automatically and efficiently completed. The present application constructs a corresponding lining type prediction model in each tunnel lining type sub-data set, and finally constructs multiple lining type prediction models, which can effectively improve the prediction accuracy of the lining type.

[0048] The technical solutions of the present application will be further described in detail below with the aid of the accompanying drawings and embodiments. BRIEF DESCRIPTION OF DRAWINGS

[0049] The accompanying drawings are used to provide a further understanding of the present application, and constitute a part of the specification, together with the embodiments of the present application, to explain the present application, and do not constitute a limitation on the present application. In the drawings:

[0050] Figure 1 For the flowchart of the intelligent determination method of tunnel lining type in the embodiment 1 of the present application;

[0051] Figure 2 For the flowchart of constructing the GWO-NGBoost lining type model in the embodiment 1 of the present application;

[0052] Figure 3 For the lining type prediction process chart in the embodiment 1 of the present application;

[0053] Figure 4 For the structural schematic diagram of an electronic device in the embodiment 2 of the present application. DETAILED DESCRIPTION

[0054] Exemplary embodiments of the present disclosure will be described in greater detail below with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it is understood that the present disclosure can be embodied in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the present disclosure to those skilled in the art.

[0055] To solve the problems in the prior art, the embodiment of the present application provides a tunnel lining type intelligent determination method.

[0056] Embodiment 1

[0057] The embodiment of the present application discloses a tunnel lining type intelligent determination method, which comprises Figure 1 , comprising:

[0058] S100. Obtain a designed tunnel data set; wherein the designed tunnel data set at least includes tunnel longitude and latitude, tunnel length, surrounding rock section start and end mileage, buried depth of multiple mileages in the tunnel, and corresponding surrounding rock grade and lining type.

[0059] Specifically, the specific data of the designed tunnel data set is shown in Table 1, and a tunnel to be designed includes tunnel longitude and latitude, tunnel length, surrounding rock section start and end mileage, buried depth of multiple mileages in the tunnel, and corresponding surrounding rock grade, which is specifically shown in Table 2.

[0060] Table 1: Data set of a designed tunnel

[0061]

[0062] Table 2: Data set of a tunnel to be designed

[0063]

[0064] S200. Construct a lining type sub-data set according to the designed tunnel data set; in S200 of the embodiment, a lining type sub-data set is constructed according to the designed tunnel data set, and the specific method comprises:

[0065] S201. Feature extraction and conversion, extract tunnel longitude and latitude, tunnel length, surrounding rock section length, and buried depth features from the data set, and perform plane coordinate conversion on the tunnel longitude and latitude features, convert the geographic coordinates of longitude and latitude into plane coordinates, and the coordinate conversion formula is as formula (1)-(4):

[0066]

[0067] In the formula, x log , x latlongitude and latitude values of the tunnel, respectively, x log_sin , x log_cos longitude converted coordinate value, x lat_sin , x lat_cos latitude converted coordinate value.

[0068] S202. Splitting sub-datasets. The data with the same surrounding rock grade in the data set after feature extraction is divided into a same sub-dataset, so as to divide the data set into multiple sub-datasets. Each sub-dataset corresponds to a surrounding rock grade.

[0069] S203. Resampling. In a same sub-dataset, there are multiple lining types, and the corresponding number of lining types is not the same. Sample imbalance will lead to a subsequent model tending to predict a sample with a large number of a certain lining type, and the prediction performance of a sample with a small number of a certain lining type is poor. Therefore, resampling is needed for the sub-datasets. The specific steps are as follows:

[0070] S2031. Downsampling the sub-dataset. In the sub-dataset, for a sample with too many samples of a certain lining type, one sample is randomly selected without repetition each time until a specified number of samples is selected.

[0071] S2032. Upsampling the sub-dataset. In the sub-dataset, for a sample with too few samples of a certain lining type, a specified number of new samples are generated based on the Smote algorithm. The specific method of the Smote algorithm includes: randomly selecting a sample with too few samples of a certain lining type, calculating the k nearest neighbors of the sample, randomly selecting 1 sample from the k neighbors, and randomly selecting 1 point between the neighbor sample and the sample point as a newly generated sample. Repeat the steps until a specified number of samples are generated.

[0072] S204. Constructing a lining type sub-dataset. Repeat S203 for each sub-dataset corresponding to a surrounding rock grade until all sub-datasets corresponding to surrounding rock grades are covered, i.e., all lining type sub-datasets corresponding to surrounding rock grades are constructed.

[0073] For example, according to the data in Table 2, the feature extraction and conversion results of the designed tunnel are shown in Table 3. The data with the same surrounding rock grade in the case after feature extraction is divided into a sub-dataset, so that sub-dataset one is: {sample 1, sample 2}, and the corresponding surrounding rock grade is V level. Sub-dataset two is: {sample 3, sample 5, sample 7, sample 9, sample 11, sample 13}, and the corresponding surrounding rock grade is IV level. Sub-dataset three is: {sample 4, sample 6, sample 8, sample 10, sample 12, sample 14}, and the corresponding surrounding rock grade is III level. Sub-dataset four is: {sample 15}, and the corresponding surrounding rock grade is II level.

[0074] Table 3 Tunnel feature extraction and conversion results

[0075]

[0076] In the collected all tunnel data set, the surrounding rock grade V corresponding to the sub data set I contains 3075 Vb type lining samples, 290 Va type lining samples, 72 open tunnel lining samples, and the resampling is performed, that is, the corresponding lining type sub data set I is obtained, containing 200 Vb type lining samples, 200 Va type lining samples and 200 open tunnel lining samples.

[0077] The surrounding rock grade IV corresponding to the sub data set II contains 1711 IVa type lining samples and 1717 IVb type lining samples. Since the number of IVa and IVb type lining type samples is not much different, resampling is not required, and the sub data set II is the corresponding lining type sub data set II, containing 1711 IVa type lining samples and 1717 IVb type lining samples.

[0078] The surrounding rock grade III corresponding to the sub data set III contains 1418 IIIa type lining samples and 831 IIIb type lining samples. Resampling is performed, that is, the corresponding lining type sub data set III is obtained, containing 831 IIIa type lining samples and 831 IIIb type lining samples.

[0079] The surrounding rock grade II corresponding to the sub data set IV contains 558 IIa type lining samples and 45 IIb type lining samples. Resampling is performed, that is, the corresponding lining type sub data set IV is obtained, containing 150 IIIa type lining samples and 150 IIIb type lining samples.

[0080] S300. According to the lining type sub data set, a GWO-NGBoost lining type prediction model is constructed; in S300 of the embodiment, a GWO-NGBoost lining type prediction model is constructed according to the lining type sub data set, as shown in Figure 2 , the specific method comprising:

[0081] S301. Data set division, in the same lining type sub data set, the data set is divided into a training set and a test set, training set: test set = 80:20. The training set is used to construct the model and find the optimal hyperparameter, and the test set is used to evaluate the model;

[0082] S302. The optimal hyperparameters of the NGBoost model are found by using the GWO algorithm; in S302, the optimal hyperparameters of the NGBoost model are found by using the GWO algorithm, and the specific method comprises:

[0083] S3021. Set the hyperparameter range, set the value range of the parameters n_estimators, learning_rate, minibatch_frac, col_sample in the NGBoost model;

[0084] S3022. Initialize the position of the wolf pack, randomly select a number in the value range for each hyperparameter, generate a list of M groups of hyperparameters as the initialized wolf pack position;

[0085] S3023. Calculate the fitness value of the wolf pack and save the best three wolves, set the position parameters of each wolf as the hyperparameters of the NGBoost model, and take the negative value of the average precision based on K-fold cross-validation as the fitness value of the wolf individual, as shown in formula 5, and the precision represents the proportion of samples that are actually positive in the samples predicted to be positive by the model.

[0086]

[0087] In the formula, Precision is the precision, TP is the sample predicted to be positive and actually positive, and FP is the sample predicted to be positive and actually negative; after obtaining the fitness value of the wolf individual, take the three wolves with the smallest fitness as α, β, and δ wolves;

[0088] S3024. Calculate the correlation coefficient, the convergence factor a, the parameter C, and the parameter A are calculated according to formulas 6-8:

[0089]

[0090] C i =2×r i (7)

[0091] A i =2×a×r i -a (8)

[0092] In the formula, t is the current iteration number, t o is the total iteration number, i takes the value of 1, 2, 3, and r i is a random number between 0 and 1;

[0093] S3025. Update the position of the wolf pack, update the position of the wolf pack according to α, β, and δ wolves, and the update formula is shown in formulas 9-11:

[0094]

[0095] In the formula, X represents the position of the wolf pack, X α , X β , and X δX(t) represents the position of the alpha, beta, and delta wolves, and X(t+1) represents the updated position of the wolf group;

[0096] S3206. Obtain the optimal hyperparameters, repeat S3203-S3205 until the number of iterations reaches the maximum number of iterations, and the position of the alpha wolf is the optimal hyperparameters obtained.

[0097] S303. Construct a lining type prediction model. For each lining type sub-data set, establish an NGBoost lining type prediction model based on the obtained optimal hyperparameters on the entire training set, and finally obtain multiple NGBoost lining type prediction models corresponding to multiple lining type sub-data sets.

[0098] According to the extracted features and conversion results in Table 3, the optimal hyperparameters are found in the training set of each lining type sub-data set based on the GWO-NGBoost model, and the corresponding prediction model is established. The prediction accuracy evaluation of the model on the training set and the test set is shown in Table 4. It can be seen that the GWO-NGBoost prediction model established on each lining type sub-data set has an accuracy of more than 80% on the training set and the test set, which indicates that the established prediction model has good performance and can accurately realize the prediction of the lining type.

[0099] Table 4 GWO-NGBoost model prediction results table

[0100]

[0101] S400. For the tunnel to be predicted, determine the section division of the surrounding rock grade based on the geological survey data. The division of the tunnel surrounding rock grade based on the geological survey data of the tunnel to be designed is shown in Table 2.

[0102] S500. For each section of the tunnel to be predicted, respectively extract features and input into the lining type prediction model corresponding to the surrounding rock grade to realize the prediction of the lining type.

[0103] In S500 of the present embodiment, for each section of the tunnel to be predicted, respectively extract features and input into the lining type prediction model corresponding to the surrounding rock grade to realize the prediction of the lining type, as shown in Figure 3 , the specific method includes:

[0104] S501. Extract features and convert. On the surrounding rock grade section, extract the features of buried depth, tunnel longitude and latitude, tunnel length, and surrounding rock section length, and convert them;

[0105] S502. Obtain the lining type of each section. For each surrounding rock grade section, find the lining type prediction model corresponding to the surrounding rock grade, input the features extracted in S501 into the lining type prediction model, and obtain the lining type of the corresponding section.

[0106] S503. Determine the lining type, detect each adjacent section in S502, if the lining types of adjacent sections are the same, merge them into one section, and the merged section and the corresponding lining type are the final determined lining type.

[0107] For the tunnel to be designed in Table 2, the corresponding tunnel length, surrounding rock section length, buried depth, and features converted from latitude and longitude are extracted, and the extracted features are input into the lining type prediction model corresponding to the surrounding rock grade, so that the lining type prediction result of each mileage can be obtained, as shown in Table 5. Since adjacent surrounding rock grade sections do not have the same lining type prediction result, the merging of the sections is not required, and the prediction result is the final determined lining type.

[0108] Table 5 Characteristic values and prediction results of uniform mileage of a tunnel to be designed

[0109]

[0110] The embodiment discloses an intelligent determination method of a tunnel lining type, including: acquiring a designed tunnel data set; constructing a lining type sub-data set according to the designed tunnel data set; constructing a GWO-NGBoost lining type prediction model according to the lining type sub-data set; for a tunnel to be predicted, determining the division of a surrounding rock grade section based on geological survey data; for each section of the tunnel to be predicted, respectively extracting features and inputting the features into a lining type prediction model corresponding to the surrounding rock grade, so as to realize the prediction of the lining type. The application realizes the prediction of the lining type by establishing a GWO-NGBoost model in each tunnel lining type sub-data set, and the whole process is standardized and can be automatically and efficiently completed. The application constructs a corresponding lining type prediction model in each tunnel lining type sub-data set, and finally constructs multiple lining type prediction models, which can effectively improve the prediction accuracy of the lining type.

[0111] Embodiment 2

[0112] Based on the same inventive concept, the embodiment of the disclosure also provides an electronic device. FIG. 4 is a structural schematic diagram of an electronic device according to an embodiment of the disclosure. As shown in the figure, the electronic device according to the embodiment of the disclosure includes one or more processors 101, a memory 102, and one or more I / O interfaces 103. The memory 102 stores one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors implement the optimization method in any of the above embodiments. Figure 4 The one or more I / O interfaces 103 are connected between the processor and the memory, and are configured to realize the information interaction between the processor and the memory.

[0113] The processor 101 is a device with data processing capability, including but not limited to a central processing unit (CPU) and the like; the memory 102 is a device with data storage capability, including but not limited to a random access memory (RAM, more specifically SDRAM, DDR, etc.), a read-only memory (ROM), an electrically erasable programmable read-only memory (EEPROM), a flash memory (FLASH); the I / O interface (read-write interface) 103 is connected between the processor 101 and the memory 102, and can realize information interaction between the processor 101 and the memory 102, including but not limited to a data bus (Bus) and the like.

[0114] In some embodiments, the processor 101, the memory 102 and the I / O interface 103 are connected with each other through the bus 104, and further connected with other components of the computing device.

[0115] In some embodiments, the one or more processors 101 include a field programmable gate array.

[0116] According to embodiments of the present disclosure, a computer readable medium is also provided. The computer readable medium has a computer program stored thereon, wherein the program, when executed by a processor, implements the steps in any of the optimization methods in the above embodiments.

[0117] It should be understood that the specific order or hierarchy of steps in the processes disclosed is an example. Based upon design preferences, it should be understood that the specific order or hierarchy of steps in the processes can be re-arranged while remaining within the scope of the present disclosure. The accompanying method claims present elements of the various steps in a sample order, and are not meant to be limited to the specific order or hierarchy presented.

[0118] In the above detailed description, various features are grouped together in single embodiments for the purpose of streamlining the disclosure. Such disclosed methods should not be interpreted as reflecting an intention that the claimed embodiments require more features than are explicitly recited in each claim. On the contrary, as indicated above, the inventiveness lies in less, the claimed subject matter than in the disclosed single embodiments. Thus, the following claims, which are hereby expressly incorporated by reference, are hereby expressly incorporated by reference, are hereby expressly incorporated into this detailed description, in which each claim independently represents an additional preferred embodiment of the present invention.

[0119] Those of skill would further appreciate that the various illustrative logical blocks, modules, circuits, and algorithm steps described in connection with the embodiments disclosed herein can be implemented as electronic hardware, computer software, or combinations of both. To clearly illustrate this interchangeability of hardware and software, various illustrative components, blocks, modules, circuits, and steps have been described above generally in terms of their functionality. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the overall system. Skilled artisans can implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present disclosure.

[0120] The steps of a method or algorithm described in connection with the embodiments disclosed herein can be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. A software module can reside in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art. An exemplary storage medium is coupled to the processor such that the processor can read information from, and write information to, the storage medium. In the alternative, the storage medium can be integral to the processor. The processor and the storage medium can reside in an ASIC. The ASIC can reside in a user terminal. In the alternative, the processor and the storage medium can reside as discrete components in a user terminal.

[0121] For a software implementation, the techniques described herein can be implemented with modules (e.g., procedures, functions, and so on) that perform the functions described herein. The software codes can be stored in memory units and executed by processors. The memory unit can be implemented within the processor or external to the processor, in which case it can be communicatively coupled to the processor via various means as is known in the art.

[0122] The above description includes one or more examples of the embodiments. Of course, not all possible combinations of components or method steps described above can be claimed as embodiments. One of ordinary skill in the art can recognize that modifications and variations of the described embodiments are possible and are within the scope of the present disclosure. It is therefore intended that the embodiments described herein be considered in all respects as only illustrative and not restrictive. Specifically, the description of the embodiments should be considered to be illustrative and not exhaustive, and should be considered in the light of the claims. Further, the use of the term "comprise" in the specification is to be construed in the same way as the term "comprise" as defined in the claims. In addition, the use of any term "or" in the specification is to be construed as "non-exclusive or".

Claims

1. A method for intelligently determining tunnel lining type, characterized in that, include: S100. Obtain the dataset of the designed tunnel; S200. Based on the designed tunnel dataset, construct a sub-dataset of lining types; S300. Construct a GWO-NGBoost lining type prediction model based on the lining type subset dataset; In S300, the specific method for constructing the GWO-NGBoost lining type prediction model based on the lining type subset dataset includes: S301. Dataset partitioning: In the same lining type sub-dataset, the dataset is divided into a training set and a test set. The training set is used to build the model and find the optimal hyperparameters, and the test set is used to evaluate the model. S302. Use the GWO algorithm to find the optimal hyperparameters of the NGBoost model; S303. Construct a lining type prediction model. For each lining type subset, build an NGBoost lining type prediction model on the entire training set based on the obtained optimal hyperparameters, and finally obtain multiple NGBoost lining type prediction models corresponding to multiple lining type subsets. S400. For tunnels to be predicted, the segment division of the surrounding rock grade is determined based on geological survey data; S500. For each section of the tunnel to be predicted, features are extracted and input into the lining type prediction model corresponding to the surrounding rock grade to achieve prediction of the lining type.

2. The intelligent method for determining tunnel lining type as described in claim 1, characterized in that, In S100, the designed tunnel dataset includes at least the tunnel's latitude and longitude, tunnel length, starting and ending points of the surrounding rock section, burial depth of multiple mileages in the tunnel, and the corresponding surrounding rock grade and lining type.

3. The intelligent method for determining tunnel lining type as described in claim 1, characterized in that, In S200, based on the designed tunnel dataset, a lining type sub-dataset is constructed. The specific method includes: extracting and transforming features from the acquired dataset, dividing data with the same surrounding rock grade into a sub-dataset, resampling the sub-dataset to form a lining type sub-dataset, and finally constructing multiple lining type sub-datasets corresponding to multiple surrounding rock grades.

4. The intelligent method for determining tunnel lining type as described in claim 3, characterized in that, The acquired dataset is subjected to feature extraction and transformation. The specific methods include: extracting the tunnel latitude and longitude, tunnel length, surrounding rock section length, and burial depth features from the dataset, and performing planar coordinate transformation on the tunnel latitude and longitude features to convert geographical coordinates such as latitude and longitude into planar coordinates. The coordinate transformation formulas are as shown in equations (1) to (4): ； In the formula, x log , x lat These are the longitude and latitude values ​​of the tunnel, respectively. x log_sin , x log_cos These are the coordinate values ​​after longitude transformation. x lat_sin , x lat_cos These are the coordinate values ​​after latitude conversion.

5. The intelligent method for determining tunnel lining type as described in claim 3, characterized in that, Resampling of a subset of data to form a lining-type subset can be achieved through methods including: The subset of data is downsampled. In the subset, for samples with an excessive number of lining types, one sample is randomly selected each time without repetition until a specified number of samples are selected. The subset of data is upsampled. In the subset, for samples with an insufficient number of lining types, a new specified number of samples are generated based on the Smote algorithm. The specific method of the Smote algorithm includes: randomly selecting a sample with an insufficient number of lining types, calculating the k nearest neighbors of the sample, randomly selecting one sample from the k neighbors, and randomly selecting one point between the neighbor sample and the sample point as the newly generated sample. The above steps are repeated until a specified number of samples are generated.

6. The intelligent method for determining tunnel lining type as described in claim 3, characterized in that, Finally, multiple lining type subsets corresponding to multiple surrounding rock grades are constructed. The specific method includes: repeating the resampling step for each sub-dataset corresponding to the surrounding rock grade until all sub-datasets corresponding to the surrounding rock grades are covered, thus constructing a sub-dataset of lining types corresponding to all surrounding rock grades.

7. The intelligent method for determining tunnel lining type as described in claim 1, characterized in that, In S302, the GWO algorithm is used to find the optimal hyperparameters of the NGBoost model. The specific methods include: S3021. Set the hyperparameter range, set the value range of the parameters n_estimators, learning_rate, minibatch_frac, and col_sample in the NGBoost model; S3022. Initialize the wolf pack position. For each hyperparameter, randomly select a number within the range of values ​​to generate a list of M hyperparameters, which is the initial wolf pack position. S3023. Calculate the fitness values ​​of individual wolves and save the three best wolves. Set the position parameters of each wolf to the hyperparameter values ​​of the NGBoost model, and use the negative value of the average precision based on K-fold cross-validation as the fitness value of that individual wolf. The formula for calculating the precision is shown in 5. The precision represents the proportion of samples that are actually positive among those predicted as positive by the model. ； In the formula, Precision is the accuracy rate, TP represents the samples predicted as positive and actually positive, and FP represents the samples predicted as positive and actually negative. After obtaining the fitness values ​​of individual wolves, the three wolves with the lowest fitness are selected and saved as... , , Wolf; S3024. Calculate the correlation coefficient. The convergence factor a, parameter C, and parameter A are calculated according to equations 6-8: ； In the formula, t This represents the current iteration number. t o This represents the total number of iterations. i The values ​​are 1, 2, and 3. r i A random number between 0 and 1; S3025. Update wolf pack location, based on... , , The wolf pack's location is updated using formulas shown in equations 9-11: ； In the formula, X represents the location of the wolf pack. , , represent , , The wolf's position, X(t+1) represents the updated wolf pack position; S3206. Obtain the optimal hyperparameters, and repeat S3203-S3205 until the maximum number of iterations is reached. The wolf's position is the optimal hyperparameter obtained.

8. The intelligent method for determining tunnel lining type as described in claim 1, characterized in that, In S500, for each section of the tunnel to be predicted, features are extracted and input into the lining type prediction model corresponding to the surrounding rock grade to predict the lining type. Specific methods include: S501. Feature Extraction and Transformation: Extract and transform features such as burial depth, tunnel latitude and longitude, tunnel length, and surrounding rock section length in sections with different surrounding rock grades. S502. Obtain the lining type of each section. For each section with a surrounding rock grade, find the lining type prediction model corresponding to the surrounding rock grade. Input the features extracted in S501 into the lining type prediction model to obtain the lining type of the corresponding section. S503. Determine the lining type. Test each adjacent section in S502. If the lining types of adjacent sections are the same, merge them into one section. The merged section and the corresponding lining type are the final determined lining type.

9. An electronic device, characterized in that, include: One or more processors; Memory, used to store one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the intelligent determination method for any tunnel lining type in claims 1-8.

Images