TBM tunnel face surrounding rock grade excavation-following real-time intelligent sensing method and system

Abstract

The invention provides a TBM tunnel face surrounding rock grade excavation-following real-time intelligent sensing method and system. The method comprises the following steps that rock machine related multi-source information in the TBM tunneling process is collected; building a TBM tunnel face surrounding rock grading standard; based on a TBM tunnel face surrounding rock grading standard, a database used for training rock and machine combination to obtain a grading joint prediction model is constructed; taking the constructed database as a training sample, and constructing a rock-machine composite cascade prediction model based on a ridge regression model and an artificial neural network; and inputting rock machine related multi-source information into the rock machine composite hierarchical joint prediction model, obtaining the composite rock machine score of the current TBM tunnel face, and obtaining the grade of the surrounding rock of the TBM tunnel face according to the grading standard of the surrounding rock of the TBM tunnel face. According to the method, real-time intelligent prediction of the surrounding rock grade of the tunnel face of the TBM is realized, the perception level of the TBM on the tunneling environment is improved, and a solid foundation is laid for intelligent construction of underground engineering.

Description

Technical Field

[0001] The present application relates to the technical field of underground engineering construction, and in particular to a method and system for real-time intelligent perception of surrounding rock grade of a TBM tunnel face during excavation. Background Art

[0002] TBM (Tunnel Boring Machine) tunnel construction method is a highly mechanized tunnel construction method. Compared with drilling and blasting, TBM has many advantages such as less surrounding rock disturbance and high excavation efficiency. It has been widely used in tunnel construction. How to quickly and accurately perceive the surrounding rock grade of the TBM tunnel face is of great significance for optimizing TBM excavation performance.

[0003] At present, it is mainly predicted through advanced geological prediction or TBM operation parameter analysis, but the operation of advanced geological prediction is complicated, time-consuming and labor-intensive. TBM operation parameters are greatly affected by TBM driving behavior and have obvious randomness and discreteness. In addition, the existing surrounding rock classification standard is not proposed for TBM tunnels. It only considers the single factor of surrounding rock geological conditions and ignores the excavation performance of TBM itself. Summary of the invention

[0004] The real-time intelligent perception method of the surrounding rock grade of the TBM tunnel face provided in this application can solve the problem that the advanced geological prediction operation for perceiving the surrounding rock grade of the TBM tunnel face in the prior art is complicated, time-consuming and labor-intensive, and only considers the single factor of the surrounding rock geological conditions while ignoring the technical problem of the TBM's own excavation performance.

[0005] In a first aspect, the present application provides a method for real-time intelligent perception of surrounding rock grade of a TBM tunnel face during excavation, comprising the following steps:

[0006] Collecting and obtaining rock-machine related multi-source information during TBM excavation, wherein the rock-machine related multi-source information includes multi-source information related to the TBM and surrounding rock;

[0007] Establishing the surrounding rock classification standards for TBM tunnel faces;

[0008] Based on the surrounding rock classification standard of TBM tunnel face, a database for training the rock-machine composite score cascade prediction model was constructed;

[0009] Using the constructed database as training samples, a rock-machine composite score cascade prediction model was constructed based on the ridge regression model and artificial neural network.

[0010] The rock-machine related multi-source information collected during the TBM excavation process is input into the constructed rock-machine composite score cascade prediction model to obtain the composite rock-machine score of the current TBM tunnel face. Then, the surrounding rock grade of the TBM tunnel face is obtained according to the constructed TBM tunnel face surrounding rock classification standard.

[0011] In combination with the first aspect, in one implementation, the acquisition of rock machine related multi-source information during TBM excavation specifically includes the following steps:

[0012] Collect rock-machine interaction information during TBM excavation;

[0013] Collect rock debris morphology information during TBM excavation;

[0014] Collect current conduction information during TBM excavation;

[0015] Collect rock-breaking dust information during TBM excavation.

[0016] In combination with the first aspect, in one implementation, establishing the surrounding rock classification standard for the TBM tunnel face specifically includes the following steps:

[0017] Determine surrounding rock score;

[0018] Determine the excavation score;

[0019] The weight coefficients of surrounding rock score and excavation score are determined by using the hierarchical analysis method, and the weighted sum of surrounding rock score and excavation score is calculated to obtain the rock-machine composite score.

[0020] Based on the rock-machine composite score, a classification standard for surrounding rock of TBM tunnel faces is established.

[0021] In combination with the first aspect, in one embodiment, the database for training the rock-machine composite score cascade prediction model based on the TBM tunnel face surrounding rock classification standard is constructed, which specifically includes the following steps:

[0022] Collect rock machine related multi-source information during TBM excavation from excavated sections of multiple TBM projects;

[0023] The multi-source information related to rock and machine collected during TBM excavation is compared with the constructed TBM tunnel face surrounding rock classification standard to obtain the composite rock and machine score;

[0024] Taking the collected rock-machine related multi-source information during TBM excavation as the input part and the corresponding composite rock-machine score as the target output, a database including the input part and the target output for training the rock-machine score cascade prediction model is constructed.

[0025] In combination with the first aspect, in one embodiment, the constructed database is used as a training sample, and a rock-machine composite score cascade prediction model is constructed based on a ridge regression model and an artificial neural network, specifically comprising the following steps:

[0026] Multiple training sets are obtained by sampling the database with replacement, where the size of the training set is the same as the size of the database;

[0027] Ridge regression models were constructed using multiple training sets as the first-level rock-machine composite score prediction model;

[0028] Input the training samples in the database into the first-level rock-machine composite score prediction model to obtain the first-level prediction results;

[0029] The first-level prediction results are combined and used as the input part of the updated training samples while keeping the target output unchanged. An artificial neural network model is constructed based on the updated database, and a second-level rock-machine composite score prediction model is constructed. The output of the second-level rock-machine composite score prediction model is used as the final prediction result.

[0030] In combination with the first aspect, in one implementation, the ridge regression models are constructed using multiple training sets respectively as the first-level rock-machine composite score prediction model step, and the objective function f(w) of the ridge regression model is as shown in the following formula:

[0031] f(w)=||Xw-y||2 2 +α||w||2 2

[0032] Where X refers to the input vector of the training sample; y refers to the target output of the training sample; ||·||2 refers to the Euclidean norm; and w refers to the coefficient to be solved for the ridge regression model.

[0033] In combination with the first aspect, in one implementation, the loss function Loss of the artificial neural network model is as shown in the following formula:

[0034]

[0035] Where weight and bias are the neuron connection weight and bias in the artificial neural network model, respectively.

[0036] In combination with the first aspect, in one implementation, the method further includes the following steps:

[0037] Get the test set;

[0038] Based on the test set, the perceptual performance of the rock-machine composite score prediction model is calculated and the perceptual performance of the model is verified.

[0039] In the second aspect, the present application provides a real-time intelligent perception system for surrounding rock grade of a TBM tunnel face, including:

[0040] Multi-source information acquisition module, used to acquire multi-source information related to rock and machine during TBM excavation;

[0041] The classification standard construction module is used to establish the surrounding rock classification standard for the TBM tunnel face;

[0042] A database construction module, which is in communication connection with the classification standard construction module, and is used to construct a database for training a rock-machine composite score cascade prediction model based on the TBM tunnel face surrounding rock classification standard;

[0043] The prediction model construction module is connected to the database construction module for constructing a rock-machine composite score cascade prediction model based on a ridge regression model and an artificial neural network, using the constructed database as a training sample;

[0044] The tunnel face surrounding rock grade acquisition module is communicated with the multi-source information acquisition module, the prediction model construction module and the classification standard construction module, and is used to input the collected rock-machine related multi-source information during the TBM excavation process into the constructed rock-machine composite score cascade prediction model to obtain the composite rock-machine score of the current TBM tunnel face, and then obtain the TBM tunnel face surrounding rock grade based on the constructed TBM tunnel face surrounding rock classification standard.

[0045] In conjunction with the second aspect, in one implementation, the multi-source information acquisition module includes:

[0046] The rock-machine interaction information acquisition unit is used to collect rock-machine interaction information during TBM excavation;

[0047] The slag morphology information collection unit is used to collect slag morphology information during TBM excavation;

[0048] Current conduction information acquisition unit, used to collect current conduction information during TBM excavation;

[0049] The rock breaking dust information collection unit is used to collect rock breaking dust information during TBM excavation.

[0050] The beneficial effects brought by the technical solution provided by the embodiment of the present application include at least:

[0051] The present application provides a real-time intelligent perception method for the surrounding rock grade of a TBM tunnel face during excavation. By collecting multi-source information related to the rock machine during excavation and constructing a cascade prediction model for the surrounding rock grade suitable for TBM tunnels based on ridge regression and artificial neural networks, real-time intelligent prediction of the surrounding rock grade of the TBM tunnel face can be achieved, which improves the TBM's perception level of the excavation environment and lays a solid foundation for the intelligent construction of underground projects. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 A schematic diagram of the process of a method for real-time intelligent perception of surrounding rock grade of a TBM tunnel face provided in an embodiment of the present application;

[0053] Figure 2 A schematic diagram of the collection process of rock-machine interaction information provided in an embodiment of the present application;

[0054] Figure 3 A schematic diagram of a method flow for collecting slag morphology information provided in an embodiment of the present application;

[0055] Figure 4 A schematic diagram of a method for collecting current conduction information provided in an embodiment of the present application;

[0056] Figure 5 A schematic diagram of a method flow for collecting rock-breaking dust information provided in an embodiment of the present application;

[0057] Figure 6 A schematic diagram of a process for constructing a TBM tunnel surrounding rock classification standard provided in an embodiment of the present application;

[0058] Figure 7 A schematic diagram of a method for building a database provided in an embodiment of the present application;

[0059] Figure 8 A schematic diagram of a method flow chart for constructing a rock-machine composite score cascade prediction model provided in an embodiment of the present application;

[0060] Fig. 9 A schematic flow chart of a method for sensing the surrounding rock grade of a TBM face while digging provided in an embodiment of the present application;

[0061] Fig.10 A schematic diagram of the comparative analysis of the perception performance provided in the embodiment of the present application. DETAILED DESCRIPTION

[0062] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0063] The terms "including" and "having" and any variations thereof in the specification and claims of this application and the above-mentioned drawings are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally includes steps or units that are not listed, or optionally includes other steps or units inherent to these processes, methods, products or devices. The terms "first", "second" and "third" are used to distinguish different objects, etc., and do not represent a sequence, nor do they limit "first", "second" and "third" to different types.

[0064] In the description of the embodiments of the present application, "exemplary", "for example" or "for example" are used to indicate examples, illustrations or descriptions. Any embodiment or design described as "exemplary", "for example" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary", "for example" or "for example" is intended to present related concepts in a specific way.

[0065] In the description of the embodiments of the present application, unless otherwise specified, “ / ” means or. For example, A / B can mean A or B. The “and / or” in the text is merely a description of the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. In addition, in the description of the embodiments of the present application, “multiple” refers to two or more than two.

[0066] In some processes described in the embodiments of the present application, multiple operations or steps that appear in a specific order are included, but it should be understood that these operations or steps may not be executed in the order in which they appear in the embodiments of the present application or in parallel, and the sequence number of the operation is only used to distinguish the different operations, and the sequence number itself does not represent any execution order. In addition, these processes may include more or fewer operations, and these operations or steps may be executed in sequence or in parallel, and these operations or steps may be combined.

[0067] In order to make the objectives, technical solutions and advantages of the present application clearer, the implementation methods of the present application will be further described in detail below with reference to the accompanying drawings.

[0068] First, please refer to Figure 1 The present application provides a method for real-time intelligent perception of surrounding rock grade of a TBM tunnel face during excavation, comprising the following steps:

[0069] Step S1: Collecting and obtaining rock-machine related multi-source information during TBM excavation, wherein the rock-machine related multi-source information includes multi-source information related to the TBM and surrounding rock;

[0070] Step S2: Establishing the surrounding rock classification standard for the TBM tunnel face;

[0071] Step S3: Based on the surrounding rock classification standard of the TBM tunnel face, a database for training the rock-machine composite score cascade prediction model is constructed;

[0072] Step S4: Using the constructed database as a training sample, a rock-machine composite score cascade prediction model is constructed based on a ridge regression model and an artificial neural network;

[0073] Step S5: Input the collected rock-machine related multi-source information during the TBM excavation process into the constructed rock-machine composite score cascade prediction model to obtain the composite rock-machine score of the current TBM tunnel face, and then obtain the TBM tunnel face surrounding rock grade according to the constructed TBM tunnel face surrounding rock grade standard.

[0074] The present application provides a real-time intelligent perception method for the surrounding rock grade of a TBM tunnel face during excavation. By collecting multi-source information related to the rock machine during excavation and constructing a cascade prediction model for the surrounding rock grade suitable for TBM tunnels based on ridge regression and artificial neural networks, real-time intelligent prediction of the surrounding rock grade of the TBM tunnel face can be achieved, which improves the TBM's perception level of the excavation environment and lays a solid foundation for the intelligent construction of underground projects.

[0075] In one embodiment, the step S1: collecting and acquiring rock machine related multi-source information during TBM excavation process specifically includes the following steps:

[0076] Step S1A: Collect rock-machine interaction information during TBM excavation. Specifically, TBM operation parameters contain rich rock-machine interaction information. The sensor system carried by the TBM itself monitors the operation parameters such as cutter head thrust, cutter head torque and cutter head penetration in real time during TBM excavation. Based on the TBM operation parameters collected during excavation monitoring, the field penetration index, torque penetration index and rock breaking ratio are simultaneously extracted as representative features of the rock-machine interaction information. More specifically, Figure 2As shown in the figure, the TBM operating parameters such as cutter head thrust, cutter head torque and cutter head penetration are collected by the sensor system carried by the TBM itself during excavation. The collected TBM operating parameters are then numerically transformed according to equations (1) to (3) to calculate and obtain the representative characteristics of rock-machine interaction information: field penetration index, torque penetration index and rock breaking ratio:

[0077]

[0078] Where FPI, TPI and SE refer to the field penetration index, torque penetration index and rock breaking specific energy, respectively; F, T and PR refer to the cutterhead thrust, cutterhead torque and cutterhead penetration, respectively; N refers to the number of cutters on the cutterhead; R refers to the cutterhead diameter.

[0079] Step S1B: Collecting the slag morphology information during the TBM excavation process to reflect the TBM rock breaking situation; specifically, in order to continuously collect the slag morphology information in real time, such as Figure 3 As shown, a rock slag morphology information acquisition device is installed, for example, a camera is arranged just above the rock slag conveyor belt connected to the slag outlet of the cutter head screw conveyor, and the two-dimensional projection morphology parameters of the rock slag formed on the tunnel face under the rock breaking action of the TBM on the conveyor belt plane are obtained during excavation, that is, the contour perimeter and contour area, and then the two-dimensional projection morphology parameters of the rock slag are synchronously mapped into three-dimensional entity morphology parameters of the rock slag, that is, the three-dimensional volume and three-dimensional sphericity of the rock slag through a geometric transformation method;

[0080] More specifically, step S1B includes the following steps:

[0081] Place a camera directly above the slag conveyor belt connected to the slag outlet of the cutter head screw conveyor, with the camera height 60cm from the conveyor belt plane;

[0082] The two-dimensional projection morphological parameters of the rock slag collected during excavation on the conveyor belt plane are the contour perimeter P. 2D and contour area A 2D ;

[0083] The equivalent diameter D of the two-dimensional projection of the slag is calculated by equations (4) and (5) respectively: 2D Sphericity S 2D :

[0084]

[0085] The slag entity morphological parameters are calculated by equations (6) and (7), namely, the three-dimensional volume V 3D and three-dimensional sphericity S 3D :

[0086] V 3D =3π(D 2D ) 3 / 25 Formula (6)

[0087] S 3D =2.5S 2D -1.525 Formula (7)

[0088] Step S1C: Figure 4 As shown in the figure, the current conduction information during the TBM excavation process is collected. The change of current under isobaric action reflects the change of resistance of the conducting medium, and the size of the resistance is closely related to the rock mass structure and rock properties of the surrounding rock. Specifically, in order to continuously collect the current conduction information of the surrounding rock of the face in real time, an electrode is installed in the cutter grooves on both sides of each TBM disc cutter, and a voltage-stabilizing power supply device and a current detection device are placed on the back of the cutter head. Before the TBM contacts the surrounding rock, the current passing through the disc cutter is first monitored and used as a reference value. Subsequently, the current during the TBM rock breaking process is monitored along the excavation, and the difference between the current and the aforementioned reference value is synchronously calculated as the current conduction characteristic of the surrounding rock of the face.

[0089] More specifically, step S1C comprises the following steps:

[0090] Step S1C1: Install an electrode in the knife grooves on both sides of each hob and place a voltage stabilizer and an ammeter on the back of the cutter disc. Pass the electrode lead back to the back of the cutter disc through the knife groove openings to form a complete current loop with the voltage stabilizer and the ammeter.

[0091] Step S1C2: Measure the current I0 passing through the cutter before the TBM contacts the surrounding rock and use it as a reference value.

[0092] Step S1C3: Monitoring the current I during the rock breaking process of the TBM t And simultaneously calculate the difference ΔI between it and the aforementioned reference value I0 t As the current conduction characteristic of the surrounding rock at the tunnel face, the calculation formula of the difference is as follows:

[0093] ΔI t =I t -I0.

[0094] Step S1D: Collect rock-breaking dust information during TBM excavation; under the penetration and cutting action of the TBM cutter, the surrounding rock of the tunnel face will be destroyed and accompanied by dust generation. The rock-breaking dust information is evaluated and obtained according to the dust particle size generated by the surrounding rock breaking of the tunnel face; specifically, Figure 5 As shown, a rock-breaking dust information collection device is installed, for example, a dust particle size distribution measuring instrument is installed at the slag inlet of the cutter head screw conveyor, the dust particle size distribution generated by the rock breaking of the tunnel face surrounding rock is collected, the dust particle size generated by the rock breaking of the tunnel face surrounding rock is divided into different interval particle size ranges, and the non-uniformity coefficient of the dust mass in different interval particle size ranges is used as the rock-breaking dust characteristics of the tunnel face surrounding rock;

[0095] More specifically, step S1D includes the following steps:

[0096] Step S1D1: a dust particle size distribution measuring instrument is arranged at the slag inlet of the cutter screw conveyor to measure the dust mass with particle sizes >150 μm, 100-150 μm, 50-100 μm and <50 μm respectively;

[0097] Step S1D2: Monitoring the quality of dust of different particle sizes generated during the TBM rock breaking process and And synchronously calculate the difference between it and the monitoring result at the previous moment The mass difference of dust with different particle sizes As the dust information generated by the surrounding rock breaking of the current TBM tunnel face, i is replaced by >150, 100-150, 50-100 and <50 respectively. Specifically, wait.

[0098] Step S1D3: Based on the mass difference of dust with different particle sizes, the non-uniformity coefficient UC of the mass of dust with different particle sizes is calculated by formula (8), and it is used as the rock breaking dust feature of the surrounding rock of the tunnel face:

[0099]

[0100] Where, Mean t refer to and The average value of .

[0101] In one embodiment, if Figure 6 As shown, the step S2: establishing the surrounding rock classification standard of the TBM tunnel face specifically includes the following steps:

[0102] Step S21: determining the surrounding rock score; specifically: determining the surrounding rock score according to the surrounding rock international grade, wherein the surrounding rock international grade is divided into grade I, grade II, grade IV and grade III surrounding rock; more specifically: grade III surrounding rock is most conducive to TBM excavation, and the corresponding surrounding rock score is recorded as 5; grade II and grade IV surrounding rock are second, and the corresponding surrounding rock score is recorded as 3; grade I and grade V surrounding rock are the worst, and the corresponding surrounding rock scores of grade I and grade V surrounding rock are recorded as 1;

[0103] Step S22: Determine the excavation score; specifically, the excavation score is determined by the net excavation rate of the TBM; more specifically: when the net excavation rate of the TBM is greater than 80 mm / min, the corresponding excavation score is recorded as 5; when the net excavation rate of the TBM is between 60 and 80 mm / min, the corresponding excavation score is recorded as 4; when the net excavation rate of the TBM is between 40 and 60 mm / min, the corresponding excavation score is recorded as 3; when the net excavation rate of the TBM is between 20 and 40 mm / min, the corresponding excavation score is recorded as 2; when the net excavation rate of the TBM is less than 20 mm / min, the corresponding excavation score is recorded as 1;

[0104] Step S23: using the analytic hierarchy process to determine the weight coefficients of the surrounding rock score and the excavation score, performing weighted summation of the surrounding rock score and the excavation score, and calculating and obtaining the rock-machine composite score;

[0105] More specifically, step S23 includes the following steps:

[0106] Step S23A: using the analytic hierarchy process to determine the weight coefficients of the surrounding rock score and the excavation score of the Class III surrounding rock, performing weighted summation of the surrounding rock score and the excavation score, and calculating and obtaining the rock-machine composite score to obtain the rock-machine composite score of the Class III surrounding rock for TBM excavation;

[0107] Step S23B: using the analytic hierarchy process to determine the weight coefficients of the surrounding rock scores and the excavation scores of the Class II and Class IV surrounding rocks, respectively, weighted summing the surrounding rock scores and the excavation scores, and calculating the rock-machine composite score to obtain the rock-machine composite score of the Class II or Class IV surrounding rock for TBM excavation; in this step, the surrounding rock scores of Class II and Class IV surrounding rocks are the same, both 3, so the same calculation method of the rock-machine composite score is used;

[0108] Step S23C: Use the hierarchical analysis method to determine the weight coefficients of the surrounding rock scores and excavation scores of Class I and Class V surrounding rock, respectively, perform weighted summation of the surrounding rock scores and excavation scores, and calculate the rock-machine composite score to obtain the rock-machine composite score of Class I or Class V surrounding rock for TBM excavation; in this step, the surrounding rock scores of Class I and Class V surrounding rock are the same, both of which are 1, so the same method for calculating the rock-machine composite score is used.

[0109] In one embodiment, in step S23A, step S23B, and step S23C, the weight coefficients of the surrounding rock scores and the excavation scores of each level of surrounding rock are determined by using the hierarchical analysis method, which specifically includes the following steps:

[0110] Step S231: construct a judgment matrix R based on the surrounding rock score s1 and the excavation score s2 of the surrounding rock of this level;

[0111] Step S232: Calculate the eigenvector v corresponding to the maximum eigenvalue of the judgment matrix;

[0112] Step S233: Normalize the feature vector v to obtain the weight coefficient of the surrounding rock score and the weight coefficient of the excavation score.

[0113] In a specific embodiment, step S23A specifically includes the following steps:

[0114] S23A1: The judgment matrix constructed by the surrounding rock score and excavation score of Class III surrounding rock is

[0115]

[0116] Step S23A2: Calculate the eigenvector v=[0.7071, 0.7071] corresponding to the maximum eigenvalue of the judgment matrix;

[0117] Step S23A3: normalizing the feature vector v to obtain weight coefficients of the surrounding rock score and the excavation score of 0.5000 and 0.5000 respectively;

[0118] Step S23A4: Calculate and obtain the rock-machine composite score CS according to the surrounding rock score s1, the excavation score s2, and the weight coefficient w2 of the surrounding rock score w1 and the excavation score.

[0119] In a specific embodiment, step S23B specifically includes the following steps:

[0120] S23B1: The judgment matrix constructed by the II and IV surrounding rock scores and the excavation scores is:

[0121]

[0122] Step S23B2: Calculate the eigenvector v corresponding to the maximum eigenvalue of the judgment matrix = [0.3162, 0.9487]

[0123] Step S23B3: normalizing the feature vector v to obtain weight coefficients of the surrounding rock score and the excavation score of 0.2500 and 0.7500 respectively;

[0124] Step S23B4: Calculate and obtain the rock-machine composite score CS according to the surrounding rock score s1, the excavation score s2, and the weight coefficients w1 and w2 of the surrounding rock score and the excavation score.

[0125] In a specific embodiment, step S23C specifically includes the following steps:

[0126] S23C1: The judgment matrix constructed by the surrounding rock scores and excavation scores of Class I and Class V surrounding rock is

[0127]

[0128] Step S23C2: Calculate the eigenvector vv corresponding to the maximum eigenvalue of the judgment matrix = [0.1961, 0.9806];

[0129] Step S23C3: Normalizing the feature vector v to obtain weight coefficients of the surrounding rock score and the excavation score are 0.1667 and 0.8333 respectively;

[0130] Step S23C4: Calculate and obtain the rock-machine composite score CS according to the surrounding rock score s1, the excavation score s2, and the weight coefficients w1 and w2 of the surrounding rock score and the excavation score.

[0131] Steps S23A1, S23B1, S23C1 determine the elements of the matrix, denoted as r nm , where n is the row, m is the column, and r nm The values ​​are assigned according to the following assignment criteria:

[0132] Table 1 Judgment matrix element r nm The standard for assigning

[0133] <![CDATA[r nm ]]> Assignment criteria 1 Object n is as important as object m 3 Object n and object m are slightly more important 5 Object n and object m are obviously important 1 / 3 Object n and object m are slightly more important 1 / 5 Object n and object m are obviously important

[0134] In Table 1, object n and object m are the surrounding rock score and excavation score, respectively.

[0135] In one embodiment, step S24: based on the rock-machine composite score, a surrounding rock classification standard for the TBM tunnel face is established; specifically, the surrounding rock grade within the rock-machine composite score range of [1, 2) is determined as the first level, the surrounding rock grade within the rock-machine composite score range of [2, 3) is determined as the second level, the surrounding rock grade within the rock-machine composite score range of [3, 4) is determined as the third level, and the surrounding rock grade within the rock-machine composite score range of [4, 5] is determined as the fourth level. As shown in Table 2, the higher the surrounding rock grade, the more conducive it is to TBM excavation:

[0136] Table 2 Comparison table of rock-machine composite score and surrounding rock grade of TBM tunnel face

[0137] Rock Machine Composite Score Surrounding rock grade of TBM tunnel face [1，2) Level 1 [2，3) Level 2 [3，4) Level 3 [4，5] Level 4

[0138] In one embodiment, if Figure 7 As shown, the step S3: based on the surrounding rock classification standard of the TBM tunnel face, constructing a database for training the rock-machine composite score cascade prediction model specifically includes the following steps:

[0139] Step S31: Collecting multi-source rock-machine information during TBM excavation from excavated sections of multiple TBM projects; specifically, the field penetration index, torque penetration index, rock breaking specific energy, three-dimensional volume of rock slag, three-dimensional sphericity of rock slag, current conduction characteristics, and rock breaking dust characteristics collected in the aforementioned steps S1A to S1D, a total of seven dimensions of rock-machine information;

[0140] Step S32: Compare the collected rock-machine related multi-source information during the TBM excavation process with the constructed TBM tunnel face surrounding rock classification standard to obtain a composite rock-machine score;

[0141] Step S33: Using the collected rock-machine related multi-source information during the TBM excavation process as the input part and the corresponding composite rock-machine score as the target output, a database including the input part and the target output for training the rock-machine score cascade prediction model is constructed.

[0142] In one embodiment, if Figure 8 As shown, the step S4: using the constructed database as a training sample, constructing a rock-machine composite score cascade prediction model based on a ridge regression model and an artificial neural network, specifically includes the following steps:

[0143] Step S41: obtaining multiple training sets by sampling the database with replacement, wherein the size of the training set is the same as the size of the database, that is, the sample size is the same;

[0144] Step S42: using multiple training sets to construct ridge regression models respectively as the first-level rock-machine composite score prediction model;

[0145] Step S43: input the training samples in the database into the first-level rock-machine composite score prediction model to obtain the first-level prediction results;

[0146] Step S44: Combine the first-level prediction results and use them as the input part of the updated training samples while keeping the target output unchanged. Based on the updated database, use the artificial neural network algorithm to construct a second-level rock-machine composite score prediction model, and use the output of the second-level rock-machine composite score prediction model as the final prediction result, that is, use the second-level rock-machine composite score prediction model as the constructed rock-machine composite score prediction model.

[0147] In a specific embodiment, the step S4: using the constructed database as a training sample, constructing a rock-machine composite score cascade prediction model based on a ridge regression model and an artificial neural network, specifically includes the following steps:

[0148] The constructed database was sampled with replacement to obtain ten sub-training sets, where the size of the sub-training set was kept the same as the database size;

[0149] Based on each training set, a ridge regression model is constructed as the first-level rock-machine composite score prediction model, where the objective function f(w) of the ridge regression model is shown as follows:

[0150] f(w)=||Xw-y||2 2 +α||w||2 2

[0151] In the formula, X refers to the input vector of the training sample; y refers to the target output of the training sample; ‖·‖2 refers to the Euclidean norm; w refers to the coefficient to be solved for the ridge regression model, which is solved by minimizing f(w) using the least squares method. The least squares method minimizes f(w) as shown in the following formula:

[0152]

[0153] Input the training samples in the database into each first-level rock-machine composite score prediction model to obtain multiple first-level prediction results, and use the set of multiple first-level prediction results as the training samples of the new input part, that is, in, Refers to the prediction result of the i-th ridge regression model, while the target output y remains unchanged, and the updated database is constructed based on the new input part and the new target output;

[0154] An artificial neural network model is constructed based on the updated database as the second-level rock-machine composite score prediction model and its output is used as the final result. The neuron connection weight weight and bias bias in the artificial neural network model are solved by minimizing Loss(weight, bias) using the error back propagation method, as shown in the following formula:

[0155]

[0156] In one embodiment, if Fig. 9 As shown, step S5 includes the following steps:

[0157] Input the rock-machine interaction information, rock slag morphology information, current conduction information and rock-breaking dust information collected as the excavation progresses;

[0158] The fusion input is fed into the rock-machine composite score cascade prediction model to obtain the rock-machine composite score of the current tunnel face;

[0159] The surrounding rock grade of the current tunnel face is obtained according to the established TBM tunnel surrounding rock grade standard.

[0160] In a specific embodiment, the method for real-time intelligent perception of surrounding rock grade of a TBM tunnel face provided by the present application further includes the following steps:

[0161] Step S6: verifying the prediction performance, i.e., the perception accuracy, of the constructed rock-machine composite score cascade prediction model to obtain the model performance verification result; specifically, the following steps are included:

[0162] Get the test set;

[0163] Based on the test set, the prediction accuracy and F1-score of the rock-machine composite score cascade prediction model are calculated, the model prediction performance, i.e., the perception accuracy, is verified, and the model performance verification results are obtained;

[0164] In a specific embodiment, step S6 includes the following steps:

[0165] In order to verify the perception effect of the real-time intelligent perception method of surrounding rock grade of TBM tunnel face in this application, a total of 120,576 samples were collected from a deep buried TBM water diversion tunnel project in my country as a test set, including 26,261 first-level surrounding rock samples, 32,765 second-level surrounding rock samples, 32,657 third-level surrounding rock samples, and 28,883 fourth-level surrounding rock samples, as shown in Table 3:

[0166] Table 3 Confusion matrix of the prediction results of the real-time intelligent perception method for the surrounding rock grade of the TBM tunnel face in this application for these 120,576 samples

[0167]

[0168] In order to quantitatively evaluate the perception accuracy of this application, two indicators, accuracy and F1-score, are introduced. The former is a global indicator, which is used to measure the prediction performance of this application on the entire test set, calculated by formula 10; the latter is a local indicator, which is used to measure the prediction performance of this application on each type of surrounding rock in the test set, calculated by formula 11. The evaluation results show that the prediction accuracy of this application is 97.88%, and the F1-scores for the first-level surrounding rock, second-level surrounding rock, third-level surrounding rock and fourth-level surrounding rock are 0.9774, 0.9796, 0.9794 and 0.9786 respectively.

[0169]

[0170] In the formula, K refers to the number of sample categories; n ij Refers to the number of samples whose true category is i and predicted category is j.

[0171] The perception performance of this application using rock machine related multi-source information and using only TBM operating parameters as input information is compared and analyzed. Table 4 shows the confusion matrix of the perception model prediction results for these 120576 samples when only TBM operating parameters are used as partial input information, where the prediction accuracy is 61.81%, and the F1-scores for the first-level surrounding rock, second-level surrounding rock, third-level surrounding rock, and fourth-level surrounding rock are 0.5651, 0.6461, 0.6473, and 0.6012, respectively.

[0172] Table 4 Confusion matrix of the prediction results of the perception model for these 120,576 samples when only TBM operating parameters are used as partial input information

[0173]

[0174] In contrast, when using the rock-machine related multi-source information of the present application, the prediction accuracy of the perception model increased by 36.07%, and the F1-score of the primary surrounding rock, secondary surrounding rock, tertiary surrounding rock and quaternary surrounding rock increased by 0.4123, 0.3335, 0.3322 and 0.3775 respectively. Fig.10 shown.

[0175] In the second aspect, the present application provides a real-time intelligent perception system for surrounding rock grade of a TBM tunnel face, including:

[0176] Multi-source information acquisition module, used to acquire multi-source information related to rock and machine during TBM excavation;

[0177] The classification standard construction module is used to establish the surrounding rock classification standard for the TBM tunnel face;

[0178] A database construction module, which is in communication connection with the classification standard construction module, and is used to construct a database for training a rock-machine composite score cascade prediction model based on the TBM tunnel face surrounding rock classification standard;

[0179] The prediction model construction module is connected to the database construction module for constructing a rock-machine composite score cascade prediction model based on a ridge regression model and an artificial neural network, using the constructed database as a training sample;

[0180] The tunnel face surrounding rock grade acquisition module is communicated with the multi-source information acquisition module, the prediction model construction module and the classification standard construction module, and is used to input the collected rock-machine related multi-source information during the TBM excavation process into the constructed rock-machine composite score cascade prediction model to obtain the composite rock-machine score of the current TBM tunnel face, and then obtain the TBM tunnel face surrounding rock grade based on the constructed TBM tunnel face surrounding rock classification standard.

[0181] In conjunction with the second aspect, in one implementation, the multi-source information acquisition module includes:

[0182] The rock-machine interaction information acquisition unit is used to collect rock-machine interaction information during TBM excavation;

[0183] The slag morphology information collection unit is used to collect slag morphology information during TBM excavation;

[0184] Current conduction information acquisition unit, used to collect current conduction information during TBM excavation;

[0185] The rock-breaking dust information collection unit is used to collect rock-breaking dust information during TBM excavation. The functions of each module in the above-mentioned TBM tunnel face surrounding rock grade real-time intelligent perception system correspond to the steps in the above-mentioned TBM tunnel face surrounding rock grade real-time intelligent perception method embodiment, and their functions and implementation processes are not repeated here one by one.

[0186] On the third aspect, an embodiment of the present application provides a real-time intelligent perception device for the surrounding rock grade of a TBM tunnel face as it is excavated. The real-time intelligent perception device for the surrounding rock grade of a TBM tunnel face as it is excavated can be a personal computer (PC), a laptop computer, a server, or other device with data processing capabilities.

[0187] The communication interface includes input / output (I / O) interface, physical interface and logical interface, etc., which are used to realize the interconnection of devices inside the TBM tunnel face surrounding rock grade real-time intelligent perception device, and the interface used to realize the interconnection of the TBM tunnel face surrounding rock grade real-time intelligent perception device with other devices (such as other computing devices or user devices). The physical interface can be an Ethernet interface, a fiber optic interface, an ATM interface, etc.; the user device can be a display (Display), a keyboard (Keyboard), etc.

[0188] The memory can be various types of storage media, such as random access memory (RAM), read-only memory (ROM), non-volatile RAM (NVRAM), flash memory, optical storage, hard disk, programmable ROM (PROM), erasable PROM (EPROM), electrically erasable PROM (EEPROM), etc.

[0189] The processor may be a general-purpose processor, which may call the real-time intelligent perception program of the surrounding rock grade of the TBM tunnel face stored in the memory, and execute the real-time intelligent perception method of the surrounding rock grade of the TBM tunnel face provided in the embodiment of the present application. For example, the general-purpose processor may be a central processing unit (CPU). The method executed when the real-time intelligent perception program of the surrounding rock grade of the TBM tunnel face is called may refer to the various embodiments of the real-time intelligent perception method of the surrounding rock grade of the TBM tunnel face in the present application, which will not be repeated here.

[0190] In a fourth aspect, an embodiment of the present application also provides a readable storage medium.

[0191] The readable storage medium of the present application stores a real-time intelligent perception program for the surrounding rock grade of the TBM tunnel face as it is excavated. When the real-time intelligent perception program for the surrounding rock grade of the TBM tunnel face as it is excavated is executed by the processor, the steps of the real-time intelligent perception method for the surrounding rock grade of the TBM tunnel face as it is excavated as described above are implemented.

[0192] Among them, the method implemented when the real-time intelligent perception program of the surrounding rock grade of the TBM tunnel face is executed can refer to the various embodiments of the real-time intelligent perception method of the surrounding rock grade of the TBM tunnel face in this application, and will not be repeated here.

[0193] It should be noted that the serial numbers of the above-mentioned embodiments of the present application are only for description and do not represent the advantages or disadvantages of the embodiments.

[0194] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, disk, CD) as described above, and includes a number of instructions for a terminal device to execute the methods described in each embodiment of the present application.

[0195] The above are only preferred embodiments of the present application, and are not intended to limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A real-time intelligent perception method for surrounding rock grade of a TBM tunnel face during excavation, characterized in that: The following steps are involved: Collecting and obtaining rock-machine related multi-source information during TBM excavation, wherein the rock-machine related multi-source information includes multi-source information related to the TBM and surrounding rock; Establishing the surrounding rock classification standards for TBM tunnel faces; Based on the surrounding rock classification standard of TBM tunnel face, a database for training the rock-machine composite score cascade prediction model was constructed; Using the constructed database as training samples, a rock-machine composite score cascade prediction model was constructed based on the ridge regression model and artificial neural network. The rock-machine related multi-source information collected during the TBM excavation process is input into the constructed rock-machine composite score cascade prediction model to obtain the composite rock-machine score of the current TBM tunnel face. Then, the surrounding rock grade of the TBM tunnel face is obtained according to the constructed TBM tunnel face surrounding rock classification standard.

2. The method for real-time intelligent perception of surrounding rock grade of a TBM tunnel face as described in claim 1 is characterized in that: The method of collecting and acquiring rock machine related multi-source information during TBM excavation specifically includes the following steps: Collect rock-machine interaction information during TBM excavation; Collect rock debris morphology information during TBM excavation; Collect current conduction information during TBM excavation; Collect rock-breaking dust information during TBM excavation.

3. The method for real-time intelligent perception of surrounding rock grade of a TBM tunnel face as described in claim 1 is characterized in that: The establishment of the surrounding rock classification standard for the TBM tunnel face specifically includes the following steps: Determine surrounding rock score; Determine the excavation score; The weight coefficients of surrounding rock score and excavation score are determined by using the hierarchical analysis method, and the weighted sum of surrounding rock score and excavation score is calculated to obtain the rock-machine composite score. Based on the rock-machine composite score, a classification standard for surrounding rock of TBM tunnel faces is established.

4. The method for real-time intelligent perception of surrounding rock grade of a TBM tunnel face as described in claim 1 is characterized in that: The method of constructing a database for training a rock-machine composite score cascade prediction model based on the surrounding rock classification standard of the TBM tunnel face specifically includes the following steps: Collect rock machine related multi-source information during TBM excavation from excavated sections of multiple TBM projects; The multi-source information related to rock and machine collected during TBM excavation is compared with the constructed TBM tunnel face surrounding rock classification standard to obtain the composite rock and machine score; Taking the collected rock-machine related multi-source information during TBM excavation as the input part and the corresponding composite rock-machine score as the target output, a database including the input part and the target output for training the rock-machine score cascade prediction model is constructed.

5. The method for real-time intelligent perception of surrounding rock grade of a TBM tunnel face as described in claim 1 is characterized in that: The method uses the constructed database as a training sample and constructs a rock-machine composite score cascade prediction model based on a ridge regression model and an artificial neural network, specifically comprising the following steps: Multiple training sets are obtained by sampling the database with replacement, where the size of the training set is the same as the size of the database; Ridge regression models were constructed using multiple training sets as the first-level rock-machine composite score prediction model; Input the training samples in the database into the first-level rock-machine composite score prediction model to obtain the first-level prediction results; The first-level prediction results are combined and used as the input part of the updated training samples while keeping the target output unchanged. An artificial neural network model is constructed based on the updated database, and a second-level rock-machine composite score prediction model is constructed. The output of the second-level rock-machine composite score prediction model is used as the final prediction result.

6. The method for real-time intelligent perception of surrounding rock grade of a TBM tunnel face as described in claim 5 is characterized in that: The ridge regression model is constructed by using multiple training sets respectively as the objective function f(w) of the ridge regression model in the first-level rock-machine composite score prediction model step, as shown in the following formula: f(w)=‖Xw-y‖2 2 +α‖w‖2 2 Where X refers to the input vector of the training sample; y refers to the target output of the training sample; ‖·‖2 refers to the Euclidean norm; and w refers to the coefficient to be solved for the ridge regression model.

7. The method for real-time intelligent perception of surrounding rock grade of a TBM tunnel face as described in claim 5 is characterized in that: The loss function Loss of the artificial neural network model is shown in the following formula: Where weight and bias are the neuron connection weight and bias in the artificial neural network model, respectively.

8. The method for real-time intelligent perception of surrounding rock grade of a TBM tunnel face as described in claim 1 is characterized in that: The following steps are also included: Get the test set; Based on the test set, the perceptual performance of the rock-machine composite score prediction model is calculated and the perceptual performance of the model is verified.

9. A real-time intelligent perception system for surrounding rock grade of TBM tunnel face, characterized in that: include: Multi-source information acquisition module, used to acquire multi-source information related to rock and machine during TBM excavation; The classification standard construction module is used to establish the surrounding rock classification standard for the TBM tunnel face; A database construction module, which is in communication connection with the classification standard construction module, and is used to construct a database for training a rock-machine composite score cascade prediction model based on the TBM tunnel face surrounding rock classification standard; The prediction model construction module is connected to the database construction module for constructing a rock-machine composite score cascade prediction model based on a ridge regression model and an artificial neural network, using the constructed database as a training sample; The tunnel face surrounding rock grade acquisition module is communicated with the multi-source information acquisition module, the prediction model construction module and the classification standard construction module, and is used to input the collected rock-machine related multi-source information during the TBM excavation process into the constructed rock-machine composite score cascade prediction model to obtain the composite rock-machine score of the current TBM tunnel face, and then obtain the TBM tunnel face surrounding rock grade based on the constructed TBM tunnel face surrounding rock classification standard.

10. The real-time intelligent perception system for surrounding rock grade of TBM tunnel face as described in claim 9, characterized in that: The multi-source information acquisition module includes: The rock-machine interaction information acquisition unit is used to collect rock-machine interaction information during TBM excavation; The slag morphology information collection unit is used to collect slag morphology information during TBM excavation; Current conduction information acquisition unit, used to collect current conduction information during TBM excavation; The rock breaking dust information collection unit is used to collect rock breaking dust information during TBM excavation.

Images