TBM operation parameter intelligent decision-making method based on stratum information dynamic identification

Abstract

The invention discloses a TBM operation parameter intelligent decision-making method based on stratum information dynamic identification, and the method comprises the steps: obtaining rock-machine parameters in a TBM tunneling process, carrying out the preprocessing of parameter information, constructing an original database, building three tunneling working conditions, and carrying out the deep learning of a structure based on a table. A rock mass parameter multi-output prediction model and a tunneling parameter multi-output prediction model are established, tunneling efficiency, construction cost, equipment safety and engineering experience are comprehensively considered, a multi-target optimization algorithm is further constructed, rock mass parameters are predicted through the rock mass parameter multi-output prediction model, tunneling working conditions are identified, and tunneling efficiency is improved. And the tunneling parameter multi-output prediction model is used as a fitting function, the TBM tunneling state is optimized according to actual tunneling working condition requirements, and an operation parameter decision value is given. According to the method, not only are the tunneling rate and the cutter consumption considered, but also the equipment safety, the engineering experience and the like are considered, and the given operation parameter decision value can effectively improve the construction efficiency, reduce the cutter cost and ensure the construction safety.

Description

Technical Field

[0001] The invention belongs to the technical field of tunnel construction and relates to an intelligent decision method for TBM operation parameters based on dynamic identification of stratum information. Background Art

[0002] Tunnel boring machine (TBM) is an important construction method for underground infrastructure construction, and has been widely used in projects such as water conservancy transportation systems, urban subway tunnels, and energy infrastructure. During the construction of underground projects, the geological environment is often highly complex and uncertain. When the operating parameters do not match the geological environment well, it will lead to low excavation efficiency, and even cause problems such as machine jamming, abnormal tool wear, and increased construction costs.

[0003] At present, construction personnel mainly judge the geological conditions of the excavation section and the excavation status of the TBM based on experience, and then repeatedly adjust the operating parameters to make the TBM reach a relatively stable working state. This is time-consuming and requires a lot of experience, and it is difficult to achieve optimal excavation under complex geological conditions. Therefore, it is of great significance to establish a method that can automatically provide corresponding TBM operating parameter decision values ​​for different geological conditions to ensure safe and efficient TBM excavation.

[0004] At present, most of the decision-making methods for operating parameters are based on the premise of geological information. In actual engineering, geological information needs to be obtained through coring experiments, which consumes a lot of time and energy and is difficult to guide construction in a timely and effective manner. In addition, most of the existing decision-making methods only consider efficiency and cost, while in actual engineering, it is also necessary to pay attention to equipment safety under different working conditions and respect the engineering experience of construction personnel. Summary of the invention

[0005] The purpose of the present invention is to provide an intelligent decision-making method for TBM operating parameters based on dynamic identification of stratum information, which solves the problem that the existing TBM operating parameter decision-making method is time-consuming and difficult to guide construction in a timely and effective manner.

[0006] The technical solution adopted by the present invention is a TBM operation parameter intelligent decision-making method based on dynamic identification of formation information, comprising the following steps: Step 1: Obtain the operating parameters, passive parameters and rock mass parameters of the TBM excavation process, pre-process the parameter information to build an original database, divide the original database into a training set and a test set at a ratio of 8:2, and establish three excavation conditions: soft rock formation, balanced formation and hard rock formation; Step 2: Based on the tabular deep learning architecture TabNet, a multi-output prediction model for rock mass parameters and a multi-output prediction model for tunneling parameters are constructed using the training set, which are used for formation information identification and characterization mapping relationship, respectively, and the performance of the multi-output prediction model for rock mass parameters and the multi-output prediction model for tunneling parameters are verified using the test set; Step 3: Construct a multi-objective optimization algorithm to take into account excavation efficiency, construction cost, equipment safety and engineering experience. Use the rock mass parameter multi-output prediction model to predict rock mass parameters and then identify excavation conditions. Use the excavation parameter multi-output prediction model as the fitting function to optimize the TBM excavation state and give the operating parameter decision value according to the actual excavation conditions.

[0007] The operating parameters include the cutterhead speed, cutterhead torque and cutterhead thrust. The passive parameters include the support force of the support shoe, the left pressure of the top shield cylinder, the right pressure of the top shield cylinder, the left pressure of the side shield cylinder and the right pressure of the side shield cylinder. The rock mass parameters include the uniaxial compressive strength of the rock mass and the rock mass quality indicators.

[0008] The parameter information is preprocessed to build a database, including using the isolation forest algorithm to identify outliers in the parameter information, then using linear interpolation to replace outliers, and using the Min-Max method to normalize the data. The calculation formula is as follows;

[0009] in, x , x * is the characteristic value before and after normalization, min ( x ), max ( x ) are the minimum and maximum values ​​in the feature, respectively.

[0010] Three excavation conditions are established according to the uniaxial compressive strength of the rock mass, namely, soft rock formations are those not exceeding 30MPa, balanced formations are between 30MPa and 90MPa, and hard rock formations are those exceeding 90MPa.

[0011] In step 2, based on the tabular deep learning architecture TabNet, a multi-output prediction model for rock mass parameters is constructed with operating parameters and passive parameters as input and rock mass uniaxial compressive strength and rock mass quality indicators as output. Operating parameters and passive parameters are directly generated during the TBM tunneling process. Using them as input to the multi-output prediction model for rock mass parameters, the excavation stratum information of the tunnel ring can be identified.

[0012] In step 2, based on the tabular deep learning architecture TabNet, a multi-output prediction model for excavation parameters is constructed with operating parameters and rock parameters as input and excavation rate and tool consumption per linear meter as output. During the excavation process, the newly collected parameter information is dynamically added to the database, and the multi-output prediction model for rock parameters and the multi-output prediction model for excavation parameters are continuously trained to enhance the model performance.

[0013] In step 3, by establishing the torque safety factor K T and thrust safety factor K TF To achieve equipment safety considerations, namely, to limit excessive torque and excessive thrust in the decision-making process;

[0014] in, Actual torque is the actual cutter head torque value of the TBM equipment, TBM maximum torque The maximum cutter head torque allowed for TBM equipment;

[0015] in, Actual thust force is the actual cutterhead thrust value of the TBM equipment, TBM maximum current force The maximum cutterhead thrust allowed for TBM equipment.

[0016] In step 3, by establishing the engineering experience coefficient K e Take engineering experience into account, i.e. limit the adjustment range of operating parameters in the decision-making process;

[0017] in, T original is the initial value of the cutter head thrust, T optimized is the decision value of the cutterhead thrust, TF original is the initial value of the cutter head torque, TF optimized is the decision value of the cutter head torque, RPM original is the initial value of the cutter head speed, RPM optimized is the decision value of the cutter head speed.

[0018] In step 3, the Black Kite algorithm is improved into a multi-objective optimization algorithm through the weighted Chebyshev scalarization strategy, and then the initial population generation strategy of the Black Kite algorithm is replaced by the Tent chaotic mapping strategy, so as to construct a multi-objective Black Kite optimization algorithm. The operating parameters and passive parameters directly generated in the process of TBM tunneling are used as the input of the multi-output prediction model of rock mass parameters to obtain the actual rock mass parameters. The excavation working conditions are identified according to the uniaxial compressive strength of the rock mass in the actual rock mass parameters, and then the optimization scheme is determined. In the optimization process, the target value and weight are set according to the engineering requirements, and the rock parameters are kept unchanged. The multi-output prediction model of excavation parameters is used as the fitting function, and the operating parameters and actual rock mass parameters directly generated in the process of TBM tunneling are used as input to obtain the excavation rate and the tool consumption per linear meter. Then, the weights and target values ​​of the excavation rate and the tool consumption per linear meter are set in combination with the excavation plan and the construction budget. The excavation efficiency, construction cost, equipment safety and engineering experience are comprehensively considered, and different optimization schemes are formulated for different working conditions, and finally the corresponding decision values ​​of the cutter head speed, cutter head torque and cutter head thrust are given.

[0019] The parameter information collected by the model update module is processed using the isolation forest algorithm to establish a database, build and continuously update the model performance, and the excavation stratum information identification module uses the operating parameters and passive parameters of the current excavation section as input. The rock parameters of the current excavation section are predicted through a tabular deep learning model to identify the excavation working conditions. The operating parameter optimization decision module uses a multi-objective black kite optimization algorithm and a multi-objective prediction model of excavation parameters as a mapping relationship. According to the actual working conditions, the decision values ​​of cutter head speed, cutter head torque and cutter head thrust are given.

[0020] The beneficial effects of the present invention are as follows: (1) By constructing a multi-output prediction model for rock mass parameters, the parameters obtained in real time by TBM were used to accurately predict the uniaxial compressive strength and rock mass quality indicators, thus avoiding the coring and testing process and effectively improving construction efficiency. (2) The Black Kite Algorithm (BKA) is improved by using the weighted Chebyshev scalarization strategy and the Tent chaos mapping strategy, and a multi-objective Black Kite Optimization Algorithm (MOBK) is constructed. Compared with the multi-objective particle swarm optimization algorithm and the fast elite multi-objective genetic algorithm, the MOBK algorithm has better optimization performance and is more adaptable to optimization problems under multi-constraints. (3) The tabular deep learning architecture is applied to the TBM field, which achieves multi-target prediction more accurately than existing machine learning methods such as random forest, support vector regression, and feedforward neural network; (4) The present invention not only takes into account the two key performance indicators of TBM, namely, excavation rate and tool consumption, but also takes into account equipment safety and engineering experience. It achieves the purpose of improving efficiency and reducing costs by using different optimization schemes under different working conditions, and gives corresponding operating parameter decision values, which has guiding significance for tunnel construction. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 It is a flow chart of the intelligent decision-making method for TBM operation parameters based on dynamic identification of formation information of the present invention; Figure 2 is the prediction result of the uniaxial compressive strength of the rock mass by the rock mass parameter multi-output prediction model in Example 6 of the present invention; Figure 3 The prediction results of the rock mass quality index by the rock mass parameter multi-output prediction model in Example 6 of the present invention; Figure 4 The prediction result of the tunneling rate by the tunneling parameter multi-output prediction model in Example 6 of the present invention; Figure 5 The prediction result of the tool consumption per linear meter by the tunneling parameter multi-output prediction model in Example 6 of the present invention; Figure 6 The optimization result of the excavation rate in the soft rock formation in Example 6 of the present invention; Figure 7 The optimization result of the tool consumption per linear meter of the soft rock formation in Example 6 of the present invention; Figure 8 is the decision result of the cutter head rotation speed of the soft rock formation in Example 6 of the present invention; Fig. 9 is the decision result of the cutter head torque of the soft rock formation in Example 6 of the present invention; Fig.10 is the decision result of the cutterhead thrust in the soft rock formation in Example 6 of the present invention; Fig.11 The optimized result of the balanced stratum excavation rate in Example 6 of the present invention; Fig.12 This is the optimization result of the cutter consumption per linear meter of the balanced formation in Example 6 of the present invention; Fig.13 is the decision result of the rotation speed of the cutterhead in the balanced formation in Example 6 of the present invention; Fig.14 is the decision result of the balanced formation cutter head torque in Example 6 of the present invention; Fig.15 The decision result of the balanced formation cutterhead thrust in Example 6 of the present invention; Fig.16 The optimization result of the excavation rate of the hard rock formation in Example 6 of the present invention; Fig.17 The optimization result of the tool consumption per linear meter of the hard rock formation in Example 6 of the present invention; Fig.18 is the decision result of the cutter head rotation speed of the hard rock formation in Example 6 of the present invention; Fig.19 is the decision result of the cutter head torque of the hard rock formation in Example 6 of the present invention; Fig. 20 This is the decision result of the cutter head thrust in the hard rock formation in Example 6 of the present invention. DETAILED DESCRIPTION

[0022] The present invention is described in detail below with reference to the accompanying drawings and specific embodiments.

[0023] Example 1 Intelligent decision-making method for TBM operation parameters based on dynamic identification of formation information, see Figure 1 , including the following steps: Step 1, obtaining the operating parameters, passive parameters and rock mass parameters of the TBM excavation process, the operating parameters include the cutter head speed, cutter head torque and cutter head thrust, the passive parameters include the support force of the support shoe, the left pressure of the top shield cylinder, the right pressure of the top shield cylinder, the left pressure of the side shield cylinder and the right pressure of the side shield cylinder, and the rock mass parameters include the uniaxial compressive strength of the rock mass and the rock mass quality index; Preprocessing the acquired parameter information to construct an original database, including using an isolation forest algorithm to quickly and accurately process outliers in the parameter information, and establishing three excavation conditions according to the uniaxial compressive strength of the rock mass, namely, soft rock formations not exceeding 30MPa, balanced formations between 30MPa and 90MPa, and hard rock formations exceeding 90MPa; The original database is divided into training set and test set in a ratio of 8:2; Step 2: Based on the tabular deep learning architecture TabNet, a multi-output prediction model for rock mass parameters and a multi-output prediction model for tunneling parameters are constructed using the training set, which are used for formation information identification and characterization mapping relationship, respectively, and the performance of the multi-output prediction model for rock mass parameters and the multi-output prediction model for tunneling parameters are verified using the test set; Step 3: Based on the considerations of tunneling efficiency, construction cost, equipment safety and engineering experience, a multi-objective optimization algorithm is constructed. The rock mass parameters are predicted through the multi-output prediction model of rock mass parameters to identify the working conditions. The tunneling parameter multi-output prediction model is used as the fitting function. According to the actual working conditions, the TBM tunneling state is optimized and the operating parameter decision values ​​are given to achieve the highest tunneling rate while minimizing the tool consumption per linear meter.

[0024] Example 2 Step 1, obtaining the operating parameters, passive parameters and rock mass parameters of the TBM excavation process, the operating parameters include the cutter head speed, cutter head torque and cutter head thrust, the passive parameters include the support force of the support shoe, the left pressure of the top shield cylinder, the right pressure of the top shield cylinder, the left pressure of the side shield cylinder and the right pressure of the side shield cylinder, and the rock mass parameters include the uniaxial compressive strength of the rock mass and the rock mass quality index; The acquired parameter information is preprocessed to construct the original database, including using the isolation forest algorithm to quickly identify outliers in the parameter information, then using linear interpolation to replace outliers, and using the Min-Max method to normalize the data. The calculation formula is as follows;

[0025] in, x , x * is the characteristic value before and after normalization, min ( x ), max ( x ) are the minimum and maximum values ​​in the feature, respectively.

[0026] The original database is divided into training set and test set in a ratio of 8:2; Three excavation conditions are established according to the uniaxial compressive strength of the rock mass, namely, soft rock formations with a compressive strength not exceeding 30MPa, balanced formations with a compressive strength between 30MPa and 90MPa, and hard rock formations with a compressive strength exceeding 90MPa. Step 2: Based on the tabular deep learning architecture TabNet, a multi-output prediction model for rock mass parameters and a multi-output prediction model for tunneling parameters are constructed using the data in the training set, which are used for formation information identification and characterization mapping relationship, respectively; Based on the tabular deep learning architecture TabNet, a multi-output prediction model for rock mass parameters is constructed with operating parameters and passive parameters as input and rock mass uniaxial compressive strength and rock mass quality indicators as output. In the process of TBM excavation of the tunnel ring, operating parameters and passive parameters will be directly generated. Taking them as the input of the multi-output prediction model for rock mass parameters, the excavation stratum information of the tunnel ring can be identified.

[0027] Based on the tabular deep learning architecture TabNet, a multi-output prediction model for tunneling parameters is constructed with operating parameters and rock mass parameters as inputs and tunneling rate and tool consumption per linear meter as outputs. Use the test set to verify the performance of the rock mass parameter multi-output prediction model and the tunneling parameter multi-output prediction model; During the excavation process, the newly collected parameters are dynamically added to the database, and the rock mass parameter multi-output prediction model and the excavation parameter multi-output prediction model are continuously trained to enhance the model performance.

[0028] Step 3: Based on the considerations of tunneling efficiency, construction cost, equipment safety and engineering experience, a multi-objective optimization algorithm is constructed. The rock mass parameters are predicted through the multi-output prediction model of rock mass parameters to identify the working conditions. The tunneling parameter multi-output prediction model is used as the fitting function. According to the actual working conditions, the TBM tunneling state is optimized and the operating parameter decision values ​​are given to achieve the highest tunneling rate while minimizing the tool consumption per linear meter.

[0029] Example 3 Step 1, obtaining the operating parameters, passive parameters and rock mass parameters of the TBM excavation process, the operating parameters include the cutter head speed, cutter head torque and cutter head thrust, the passive parameters include the support force of the support shoe, the left pressure of the top shield cylinder, the right pressure of the top shield cylinder, the left pressure of the side shield cylinder and the right pressure of the side shield cylinder, and the rock mass parameters include the uniaxial compressive strength of the rock mass and the rock mass quality index; Preprocessing the acquired parameter information to construct the original database includes using the isolation forest algorithm to quickly identify outliers in the parameter information, then using linear interpolation to replace outliers, and using the Min-Max method to normalize the data; The original database is divided into training set and test set in a ratio of 8:2; Three excavation conditions are established according to the uniaxial compressive strength of the rock mass, namely, soft rock formations with a compressive strength not exceeding 30MPa, balanced formations with a compressive strength between 30MPa and 90MPa, and hard rock formations with a compressive strength exceeding 90MPa. Step 2: Based on the tabular deep learning architecture TabNet, a multi-output prediction model for rock mass parameters and a multi-output prediction model for tunneling parameters are constructed using the training set, which are used for formation information identification and characterization mapping relationship, respectively; Based on the tabular deep learning architecture TabNet, a multi-output prediction model for rock mass parameters is constructed with operating parameters and passive parameters as input and rock mass uniaxial compressive strength and rock mass quality indicators as output. In the process of TBM excavation of the tunnel ring, operating parameters and passive parameters will be directly generated. Taking them as the input of the multi-output prediction model for rock mass parameters, the excavation stratum information of the tunnel ring can be identified.

[0030] Based on the tabular deep learning architecture TabNet, a multi-output prediction model for tunneling parameters is constructed with operating parameters and rock mass parameters as inputs and tunneling rate and tool consumption per linear meter as outputs. Use the test set to verify the performance of the rock mass parameter multi-output prediction model and the tunneling parameter multi-output prediction model; During the excavation process, the newly collected parameters are dynamically added to the database, and the rock mass parameter multi-output prediction model and the excavation parameter multi-output prediction model are continuously trained to enhance the model performance.

[0031] Step 3: Based on the considerations of tunneling efficiency, construction cost, equipment safety and engineering experience, a multi-objective optimization algorithm is constructed. The rock mass parameters are predicted through the multi-output prediction model of rock mass parameters to identify the working conditions. The tunneling parameter multi-output prediction model is used as the fitting function to optimize the TBM tunneling state according to the actual working conditions and give the decision values ​​of the operating parameters.

[0032] By establishing the torque safety factor K T and thrust safety factor K TF To achieve equipment safety considerations, namely, to limit excessive torque and excessive thrust in the decision-making process;

[0033] in, Actual torque is the actual cutter head torque value of the TBM equipment, TBM maximum torque The maximum cutter head torque allowed for TBM equipment;

[0034] in, Actual thust force is the actual cutterhead thrust value of the TBM equipment, TBM maximum current force The maximum cutterhead thrust allowed for TBM equipment.

[0035] By establishing the engineering experience coefficient K e Take engineering experience into account, i.e. limit the adjustment range of operating parameters in the decision-making process;

[0036] in, T original is the initial value of the cutter head thrust, T optimized is the decision value of the cutterhead thrust, TF original is the initial value of the cutter head torque, TF optimized is the decision value of the cutter head torque, RPM original is the initial value of the cutter head speed, RPM optimized is the decision value of the cutter head speed.

[0037] Example 4 Step 1, obtain the operating parameters, passive parameters and rock parameters of the TBM excavation process. The operating parameters include the cutter head speed, cutter head torque and cutter head thrust. The passive parameters include the support force of the support shoe, the left pressure of the top shield cylinder, the right pressure of the top shield cylinder, the left pressure of the side shield cylinder and the right pressure of the side shield cylinder. The rock parameters include the uniaxial compressive strength of the rock and the rock quality index.

[0038] The acquired parameter information is preprocessed to construct the original database, including using the isolation forest algorithm to identify outliers in the parameter information, then using linear interpolation to replace the outliers, and using the Min-Max method to normalize the data. The calculation formula is as follows;

[0039] in, x , x * is the characteristic value before and after normalization, min ( x ), max ( x ) are the minimum and maximum values ​​in the feature, respectively.

[0040] The original database is divided into training set and test set in a ratio of 8:2; Three excavation conditions are established according to the uniaxial compressive strength of the rock mass, namely, soft rock formations with a compressive strength not exceeding 30MPa, balanced formations with a compressive strength between 30MPa and 90MPa, and hard rock formations with a compressive strength exceeding 90MPa. Step 2: Based on the tabular deep learning architecture TabNet, a multi-output prediction model for rock mass parameters and a multi-output prediction model for tunneling parameters are constructed using the training set, which are used for formation information identification and characterization mapping relationship, respectively, and the performance of the multi-output prediction model for rock mass parameters and the multi-output prediction model for tunneling parameters are verified using the test set; Based on the tabular deep learning architecture TabNet, a multi-output prediction model for rock mass parameters is constructed with operating parameters and passive parameters as input and rock mass uniaxial compressive strength and rock mass quality indicators as output. In the process of TBM excavation of the tunnel ring, operating parameters and passive parameters will be directly generated. Taking them as the input of the multi-output prediction model for rock mass parameters, the excavation stratum information of the tunnel ring can be identified.

[0041] Based on the tabular deep learning architecture TabNet, a multi-output prediction model for tunneling parameters is constructed with operating parameters and rock mass parameters as inputs and tunneling rate and tool consumption per linear meter as outputs. During the excavation process, newly collected parameters are dynamically added to the database to continuously enhance the model performance.

[0042] Step 3: Based on the considerations of tunneling efficiency, construction cost, equipment safety and engineering experience, a multi-objective optimization algorithm is constructed. The rock mass parameters are predicted through the multi-output prediction model of rock mass parameters to identify the working conditions. The tunneling parameter multi-output prediction model is used as the fitting function to optimize the TBM tunneling state according to the actual working conditions and give the decision values ​​of the operating parameters.

[0043] By establishing the torque safety factor K T and thrust safety factor K TFTo achieve equipment safety considerations, namely, to limit excessive torque and excessive thrust in the decision-making process;

[0044] in, Actual torque is the actual cutter head torque value of the TBM equipment, TBM maximum torque The maximum cutter head torque allowed for TBM equipment;

[0045] in, Actual thust force is the actual cutterhead thrust value of the TBM equipment, TBM maximum current force The maximum cutterhead thrust allowed for TBM equipment.

[0046] By establishing the engineering experience coefficient K e Take engineering experience into account, i.e. limit the adjustment range of operating parameters in the decision-making process;

[0047] in, T original is the initial value of the cutter head thrust, T optimized is the decision value of the cutterhead thrust, TF original is the initial value of the cutter head torque, TF optimized is the decision value of the cutter head torque, RPM original is the initial value of the cutter head speed, RPM optimized is the decision value of the cutter head speed.

[0048] The Black Kite algorithm is improved into a multi-objective optimization algorithm through the weighted Chebyshev scalarization strategy, and then the initial population generation strategy of the Black Kite algorithm is replaced by the Tent chaotic mapping strategy to construct a multi-objective Black Kite optimization algorithm. The operating parameters and passive parameters directly generated during the TBM tunneling process are used as the input of the multi-output prediction model of rock mass parameters to obtain the actual rock mass parameters. The excavation working conditions are identified according to the uniaxial compressive strength of the rock mass in the actual rock mass parameters, and then the optimization scheme is determined. In the optimization process, the target value and weight are set according to the engineering requirements, and the rock parameters are kept unchanged. The multi-output prediction model of excavation parameters is used as the fitting function, and the operating parameters and actual rock mass parameters directly generated during the TBM tunneling process are used as input to obtain the excavation rate and the tool consumption per linear meter. Then, the weights and target values ​​of the excavation rate and the tool consumption per linear meter are set in combination with the excavation plan and the construction budget. Taking into account the excavation efficiency, construction cost, equipment safety and engineering experience, different optimization schemes are formulated for different working conditions, and finally the corresponding decision values ​​of the cutterhead speed, cutterhead torque and cutterhead thrust are given.

[0049] Example 5 Step 1, obtaining the operating parameters, passive parameters and rock mass parameters of the TBM excavation process, the operating parameters include the cutter head speed, cutter head torque and cutter head thrust, the passive parameters include the support force of the support shoe, the left pressure of the top shield cylinder, the right pressure of the top shield cylinder, the left pressure of the side shield cylinder and the right pressure of the side shield cylinder, and the rock mass parameters include the uniaxial compressive strength of the rock mass and the rock mass quality index; The acquired parameter information is preprocessed to construct the original database, including using the isolation forest algorithm to quickly identify outliers in the parameter information, then using linear interpolation to replace outliers, and using the Min-Max method to normalize the data. The calculation formula is as follows;

[0050] in, x , x * is the characteristic value before and after normalization, min ( x ), max ( x ) are the minimum and maximum values ​​in the feature, respectively.

[0051] The original database is divided into training set and test set in a ratio of 8:2; Three excavation conditions are established according to the uniaxial compressive strength of the rock mass, namely, soft rock formations with a compressive strength not exceeding 30MPa, balanced formations between 30MPa and 90MPa, and hard rock formations with a compressive strength exceeding 90MPa. Step 2: Based on the tabular deep learning architecture TabNet, a multi-output prediction model for rock mass parameters and a multi-output prediction model for tunneling parameters are constructed using the data in the training set, which are used for formation information identification and characterization mapping relationship, respectively; Based on the tabular deep learning architecture TabNet, a multi-output prediction model for rock mass parameters is constructed with operating parameters and passive parameters as input and rock mass uniaxial compressive strength and rock mass quality indicators as output. In the process of TBM excavation of the tunnel ring, operating parameters and passive parameters will be directly generated. Taking them as the input of the multi-output prediction model for rock mass parameters, the excavation stratum information of the tunnel ring can be identified.

[0052] Based on the tabular deep learning architecture TabNet, with operating parameters and rock parameters as input and excavation rate and tool consumption per linear meter as output, a multi-output prediction model for excavation parameters is constructed. During the excavation process, the newly collected parameters are dynamically added to the database to continuously enhance the model performance.

[0053] Use the test set to verify the performance of the rock mass parameter multi-output prediction model and the tunneling parameter multi-output prediction model; During the excavation process, the newly collected parameter information is dynamically added to the database, and the rock mass parameter multi-output prediction model and the excavation parameter multi-output prediction model are continuously trained to enhance the model performance; Step 3: Based on the consideration of excavation efficiency, construction cost, equipment safety and engineering experience, a multi-objective optimization algorithm is constructed to achieve the minimum adjustment range of operating parameters, improve excavation efficiency and reduce construction costs while ensuring equipment safety. The specific process is as follows: Step 3.1, improve the Black Kite algorithm into a multi-objective optimization algorithm through the weighted Chebyshev scalarization strategy, and then use the Tent chaotic mapping strategy to replace the initial population generation strategy of the Black Kite algorithm, so as to construct a multi-objective Black Kite optimization algorithm; Step 3.2, predicting rock mass parameters through a rock mass parameter multi-output prediction model and then identifying the working condition, that is, using the operating parameters and passive parameters directly generated during the TBM tunneling process as the input of the rock mass parameter multi-output prediction model, obtaining the actual rock mass parameters, and identifying the tunneling working condition based on the rock mass uniaxial compressive strength in the actual rock mass parameters; Step 3.3, during the optimization process, the rock parameters are kept unchanged, the multi-output prediction model of excavation parameters is used as the fitting function, and the operating parameters and actual rock mass parameters directly generated during the TBM tunneling process are used as input. The excavation state of the TBM is optimized by adjusting the decision variables, and the value of the final decision variable is given, that is, by comparing the actual excavation rate and tool consumption per linear meter output by the multi-output prediction model of excavation parameters with the target excavation rate and tool consumption per linear meter required by the project, combined with considerations of excavation efficiency, construction cost, equipment safety and engineering experience, different optimization schemes are formulated for different working conditions, so as to achieve the minimum adjustment range of operating parameters, improve excavation efficiency and reduce construction costs while ensuring equipment safety, and finally give the corresponding decision values ​​of cutter head speed, cutter head torque and cutter head thrust.

[0054] By establishing the torque safety factor K T and thrust safety factor K TF To achieve equipment safety considerations, namely, to limit excessive torque and excessive thrust in the decision-making process;

[0055] in, Actual torque is the actual cutter head torque value of the TBM equipment, TBM maximum torque The maximum cutter head torque allowed for TBM equipment;

[0056] in, Actual thust force is the actual cutterhead thrust value of the TBM equipment, TBM maximum current force The maximum cutterhead thrust allowed for TBM equipment.

[0057] By establishing the engineering experience coefficient K e Take engineering experience into account, i.e. limit the adjustment range of operating parameters in the decision-making process;

[0058] in, T original is the initial value of the cutter head thrust, T optimized is the decision value of the cutterhead thrust, TF original is the initial value of the cutter head torque, TF optimized is the decision value of the cutter head torque, RPM original is the initial value of the cutter head speed, RPM optimized is the decision value of the cutter head speed.

[0059] Develop corresponding optimization plans for different working conditions: (1) For soft rock formations, the TBM torque is likely to approach the maximum value allowed by the equipment. It is necessary to limit the increase of TBM torque, while also ensuring that the adjustment of operating parameters is not too large. The optimization solution at this time is to increase K T , reduce K e , improve the excavation rate and reduce tool consumption. T and K e The target values ​​are 1 and 0 respectively. The target value of the excavation rate is set according to the excavation plan, and the target value of the tool consumption per linear meter is set according to the construction budget. The weights are generally the same, that is, 1 / 4. If a certain target needs to be focused on during construction, the weight of the target is increased.

[0060] (2) For balanced formations, the thrust and torque of the TBM vary in similar magnitude and are moderate in size, and the equipment as a whole is safe, so K is not considered. T and K TF , but the adjustment of operating parameters should not be too large. The optimization solution at this time is to reduce K e , improve the excavation rate and reduce tool consumption. e The target value is 0. The target value of the excavation rate is set according to the excavation plan, and the target value of the tool consumption per linear meter is set according to the construction budget. The weights are generally the same, that is, 1 / 3. If a certain target needs to be focused on during construction, the weight of the target is increased.

[0061] (3) For hard rock formations, the TBM thrust is likely to approach the maximum value allowed by the equipment. It is necessary to limit the increase in TBM thrust while also ensuring that the adjustment of operating parameters is not too large. The optimization solution at this time is to increase K TF , reduce K e , improve the excavation rate and reduce tool consumption. TF and K e The target values ​​are 1 and 0 respectively. The target value of the excavation rate is set according to the excavation plan, and the target value of the tool consumption per linear meter is set according to the construction budget. The weights are generally the same, that is, 1 / 4. If a certain target needs to be focused on during construction, the weight of the target is increased.

[0062] The present invention collects and obtains parameter information through a model updating module, processes the parameter information using an isolation forest algorithm, establishes a database, builds and continuously updates the performance of the model, uses an excavation stratum information identification module to take the operating parameters and passive parameters of the current excavation section as input, and predicts the rock mass parameters of the current excavation section through a table deep learning model to achieve identification of the excavation working condition, uses an operating parameter optimization decision module using a multi-objective black kite optimization algorithm, and uses an excavation parameter multi-objective prediction model as a mapping relationship, and gives decision values ​​for the cutter head speed, cutter head torque, and cutter head thrust according to the actual working condition's requirements for the excavation rate and tool consumption.

[0063] Example 6 Step 1: Obtain the operating parameters, passive parameters and rock mass parameters recorded during the TBM excavation process, and pre-process these parameter information to build an original database. The specific steps are as follows: Step 1.1, dividing the parameters recorded during the TBM excavation process into each individual excavation ring, the operating parameters include the cutter head speed, the cutter head torque and the cutter head thrust, the passive parameters include the support force of the support shoe, the left pressure of the top shield cylinder, the right pressure of the top shield cylinder, the left pressure of the side shield cylinder and the right pressure of the side shield cylinder, and the rock mass parameters include the uniaxial compressive strength of the rock mass and the rock mass quality index; Step 1.2, preprocess the obtained parameter information to build the original database, use the isolation forest algorithm to quickly identify outliers in the parameter information, then use linear interpolation to replace the outliers, and use the Min-Max method to normalize the data. The calculation formula is as follows;

[0064] in, x , x * is the characteristic value before and after normalization, min ( x ), max ( x ) are the minimum and maximum values ​​in the feature, respectively.

[0065] Step 1.3, divide the original database into training set and test set in a ratio of 8:2; Step 1.4, three excavation conditions are established according to the uniaxial compressive strength of the rock mass, i.e., the stratum with a compressive strength not exceeding 30 MPa is a soft rock stratum, the stratum with a compressive strength between 30 MPa and 90 MPa is a balanced stratum, and the stratum with a compressive strength exceeding 90 MPa is a hard rock stratum; Step 2: Based on the tabular deep learning architecture TabNet, the training set data is used to construct a multi-output prediction model for rock mass parameters and a multi-output prediction model for tunneling parameters, which are used for stratum information identification and characterization mapping relationship, respectively, as follows: Step 2.1, based on the tabular deep learning architecture TabNet, with operating parameters and passive parameters as input, and rock mass uniaxial compressive strength and rock mass quality indicators as output, a rock mass parameter multi-output prediction model is constructed. In the process of TBM tunneling the tunnel ring, operating parameters and passive parameters will be directly generated. Using them as the input of the rock mass parameter multi-output prediction model, the excavation stratum information of the tunnel ring, that is, the excavation conditions, can be identified.

[0066] Step 2.2, based on the tabular deep learning architecture TabNet, with operating parameters and rock mass parameters as inputs and excavation rate and tool consumption per linear meter as outputs, a multi-output prediction model for excavation parameters is constructed; Step 2.3, using the test set to verify the performance of the rock mass parameter multi-output prediction model and the tunneling parameter multi-output prediction model. In this embodiment, the determination coefficient is used. R 2 , Root Mean Square Error RMSE and mean absolute percentage error MAPE To verify the performance of the model, the calculation formula is as follows:

[0067]

[0068]

[0069] in, , and represent the actual value, predicted value and average value respectively. Represents the total number of samples.

[0070] R 2 The closer the value is to 1, RMSE and MAPE The closer the value is to zero, the closer the model prediction is to the actual value of the dataset, that is, the more accurate the model prediction is.

[0071] Draw a scatter plot to visualize the performance of the two multi-output prediction models on the test set, see Figure 2-3 , the rock mass parameter multi-output prediction model is effective in predicting the uniaxial compressive strength of rock mass R 2 , RMSE, MAPE They are 0.9115, 4.7358, and 0.0496 respectively; R 2 , RMSE, MAPE They are 0.8986, 5.6703, and 0.0923 respectively. Figure 4-5 , the tunneling parameter multi-output prediction model is effective in predicting tunneling rate R 2 , RMSE, MAPE They are 0.9598, 2.0560, and 0.0333 respectively; in predicting the tool consumption per linear meter R 2 , RMSE, MAPE They are 0.9371, 0.0485, and 0.1180 respectively. The test results show that the performance of these two models meets the actual engineering needs.

[0072] Step 2.4, during the excavation process, dynamically add the newly collected parameter information to the database, and continuously train the rock mass parameter multi-output prediction model and the excavation parameter multi-output prediction model to enhance the model performance; Step 3: Based on the considerations of tunneling efficiency, construction cost, equipment safety and engineering experience, a multi-objective optimization algorithm is constructed. The rock mass parameters are predicted by the multi-output prediction model of rock mass parameters to identify the working conditions. The tunneling parameter multi-output prediction model is used as the fitting function to optimize the TBM tunneling state and give the decision value of the operating parameters according to the actual working conditions. The specific steps are as follows: Step 3.1, the Black Kite algorithm is improved into a multi-objective optimization algorithm by using a weighted Chebyshev scalarization strategy, and then the initial population generation strategy of the Black Kite algorithm is replaced by a Tent chaotic mapping strategy, thereby constructing a multi-objective Black Kite optimization algorithm. In this embodiment, the initial population size is defined as 100, and the maximum number of iterations is set to 50.

[0073] Step 3.2, using the cutter head speed, cutter head torque and cutter head thrust as decision variables, using the excavation rate and tool consumption per linear meter as target variables, using the operating parameters and passive parameters directly generated during the TBM tunneling process as inputs to the rock mass parameter multi-output prediction model, obtaining the actual rock mass parameters, identifying the excavation conditions according to the uniaxial compressive strength of the rock mass in the actual rock mass parameters, and then determining the optimization scheme. In the optimization process, the target values ​​and weights of the excavation rate and tool consumption per linear meter are set in combination with the engineering requirements (excavation plan and construction budget), keeping the rock parameters unchanged, using the excavation parameter multi-output prediction model as the fitting function, using the operating parameters and actual rock mass parameters directly generated during the TBM tunneling process as inputs, optimizing the excavation state of the TBM by adjusting the decision variables, and giving the value of the final decision variable. In this embodiment, the same weight is set, and the optimal value in the database is taken as the target value.

[0074] Step 3.3, comprehensively consider the excavation efficiency, construction cost, equipment safety and engineering experience, formulate different optimization plans for different working conditions, and achieve the minimum adjustment range of operating parameters. While ensuring equipment safety, improve the excavation efficiency and reduce the construction cost, and finally give the corresponding decision values ​​of cutter head speed, cutter head torque and cutter head thrust.

[0075] By establishing the torque safety factor KT and thrust safety factor K TF To achieve equipment safety considerations, namely, to limit excessive torque and excessive thrust in the decision-making process;

[0076] in, Actual torque is the actual cutter head torque value of the TBM equipment, TBM maximum torque The maximum cutter head torque allowed for TBM equipment;

[0077] in, Actual thust force is the actual cutterhead thrust value of the TBM equipment, TBM maximum current force The maximum cutterhead thrust allowed for TBM equipment.

[0078] By establishing the engineering experience coefficient K e Take engineering experience into account, i.e. limit the adjustment range of operating parameters in the decision-making process;

[0079] in, T original is the initial value of the cutter head thrust, T optimized is the decision value of the cutterhead thrust, TF original is the initial value of the cutter head torque, TF optimized is the decision value of the cutter head torque, RPM original is the initial value of the cutter head speed, RPM optimized is the decision value of the cutter head speed.

[0080] Develop corresponding optimization plans for different working conditions: (1) For soft rock formations, the TBM torque is likely to approach the maximum value allowed by the equipment. It is necessary to limit the increase of TBM torque, while also ensuring that the adjustment of operating parameters is not too large. The optimization solution at this time is to increase K T , reduce K e , improve the excavation rate and reduce tool consumption. T and K e The target values ​​are 1 and 0 respectively. The target value of the excavation rate is set according to the excavation plan, and the target value of the tool consumption per linear meter is set according to the construction budget. The weights are generally the same, that is, 1 / 4. If a certain target needs to be focused on during construction, the weight of the target is increased.

[0081] (2) For balanced formations, the thrust and torque of the TBM vary in similar magnitude and are moderate, and the equipment as a whole is safe, so K is not considered. T and K TF , but the adjustment of operating parameters should not be too large. The optimization solution at this time is to reduce K e , improve the excavation rate and reduce tool consumption. e The target value is 0. The target value of the excavation rate is set according to the excavation plan, and the target value of the tool consumption per linear meter is set according to the construction budget. The weights are generally the same, that is, 1 / 3. If a certain target needs to be focused on during construction, the weight of the target is increased.

[0082] (3) For hard rock formations, the TBM thrust is likely to approach the maximum value allowed by the equipment. It is necessary to limit the increase in TBM thrust while also ensuring that the adjustment of operating parameters is not too large. The optimization solution at this time is to increase K TF , reduce K e , improve the excavation rate and reduce tool consumption. TF and K e The target values ​​are 1 and 0 respectively. The target value of the excavation rate is set according to the excavation plan, and the target value of the tool consumption per linear meter is set according to the construction budget. The weights are generally the same, that is, 1 / 4. If a certain target needs to be focused on during construction, the weight of the target is increased.

[0083] The final optimization results of tunneling parameters and operation parameter decision results of the three working conditions are as follows: Figure 6-20 As shown in the figure, taking the soft rock formation working condition as an example, the average excavation rate of the rock formation increased from 48.89mm / min to 56.19mm / min, and the average tool consumption decreased from 0.30cutter / m to 0.21cutter / m. The excavation rate and tool consumption were improved by 15.37% and 28.58%, respectively. In addition, the corresponding decision values ​​of cutter head speed, cutter head torque and cutter head thrust are given. It can be seen that the soft rock formation working condition focuses on limiting the increase of torque, and the average torque is reduced from 2384.37 kN·m to 2106.78 kN·m. The adjustment range is small, with an average adjustment of 25.56%. This shows that this method can optimize the excavation state of TBM with a suitable adjustment scheme, give the decision values ​​of the operating parameters, and contribute to the efficient construction of TBM.

[0084] During actual TBM excavation, incremental learning is used to improve the performance of the optimization method. That is, new data is continuously collected during the TBM excavation process and added to the database. The two multi-output prediction models are retrained and the optimization framework is automatically improved to achieve automatic updating of the entire method.

Claims

1. An intelligent decision-making method for TBM operation parameters based on dynamic identification of formation information, characterized in that: The following steps are involved: Step 1: Obtain the operating parameters, passive parameters and rock mass parameters of the TBM excavation process, pre-process the parameter information to build an original database, divide the original database into a training set and a test set at a ratio of 8:2, and establish three excavation conditions: soft rock formation, balanced formation and hard rock formation; Step 2: Based on the tabular deep learning architecture TabNet, a multi-output prediction model for rock mass parameters and a multi-output prediction model for tunneling parameters are constructed using the training set, which are used for formation information identification and characterization mapping relationship, respectively, and the performance of the multi-output prediction model for rock mass parameters and the multi-output prediction model for tunneling parameters are verified using the test set; Step 3: Based on the considerations of tunneling efficiency, construction cost, equipment safety and engineering experience, a multi-objective optimization algorithm is constructed. The rock mass parameters are predicted through the multi-output prediction model of rock mass parameters, and the tunneling conditions are identified. The multi-output prediction model of tunneling parameters is used as the fitting function. According to the actual tunneling conditions, the TBM tunneling state is optimized and the decision values ​​of the operating parameters are given.

2. The intelligent decision-making method for TBM operation parameters based on dynamic identification of formation information according to claim 1 is characterized in that: The operating parameters include the cutter head rotation speed, cutter head torque and cutter head thrust; the passive parameters include the support force of the support shoe, the left pressure of the top shield cylinder, the right pressure of the top shield cylinder, the left pressure of the side shield cylinder and the right pressure of the side shield cylinder; and the rock mass parameters include the uniaxial compressive strength of the rock mass and the rock mass quality index.

3. The intelligent decision-making method for TBM operation parameters based on dynamic identification of formation information according to claim 2 is characterized in that: The parameter information is preprocessed to construct a database, including using an isolation forest algorithm to identify outliers in the parameter information, then using linear interpolation to replace the outliers, and using the Min-Max method to normalize the data. The calculation formula is as follows; in, x , x * is the characteristic value before and after normalization, min ( x ), max ( x ) are the minimum and maximum values ​​in the feature, respectively.

4. The intelligent decision-making method for TBM operation parameters based on dynamic identification of formation information according to claim 3 is characterized in that: The three excavation conditions are established according to the uniaxial compressive strength of the rock mass, i.e., a soft rock formation is defined as not exceeding 30 MPa, a balanced formation is defined as between 30 MPa and 90 MPa, and a hard rock formation is defined as a formation above 90 MPa.

5. The intelligent decision-making method for TBM operation parameters based on dynamic identification of formation information according to claim 4 is characterized in that: In step 2, based on the tabular deep learning architecture TabNet, a multi-output prediction model for rock mass parameters is constructed with operating parameters and passive parameters as inputs and rock mass uniaxial compressive strength and rock mass quality indicators as outputs. In the process of TBM excavation of the tunnel ring, operating parameters and passive parameters are directly generated. Using them as inputs of the multi-output prediction model for rock mass parameters, the excavation stratum information of the tunnel ring can be identified.

6. The intelligent decision-making method for TBM operation parameters based on dynamic identification of formation information according to claim 5 is characterized in that: In the step 2, based on the tabular deep learning architecture TabNet, a multi-output prediction model for excavation parameters is constructed with operating parameters and rock mass parameters as inputs and excavation rate and tool consumption per linear meter as outputs. During the excavation process, the newly collected parameter information is dynamically added to the database, and the multi-output prediction model for rock mass parameters and the multi-output prediction model for excavation parameters are continuously trained to enhance the model performance.

7. The intelligent decision-making method for TBM operation parameters based on dynamic identification of formation information according to claim 6 is characterized in that: In step 3, by establishing the torque safety factor K T and thrust safety factor K TF To achieve equipment safety considerations, namely, to limit excessive torque and excessive thrust in the decision-making process; in, Actual torque is the actual cutter head torque value of the TBM equipment, TBM maximum torque The maximum cutter head torque allowed for TBM equipment; in, Actual thust force is the actual cutterhead thrust value of the TBM equipment, TBM maximum current force The maximum cutterhead thrust allowed for TBM equipment.

8. The intelligent decision-making method for TBM operation parameters based on dynamic identification of formation information according to claim 7 is characterized in that: In step 3, by establishing the engineering experience coefficient K e Take engineering experience into account, i.e. limit the adjustment range of operating parameters in the decision-making process; in, T original is the initial value of the cutter head thrust, T optimized is the decision value of the cutterhead thrust, TF original is the initial value of the cutter head torque, TF optimized is the decision value of the cutter head torque, RPM original is the initial value of the cutter head speed, RPM optimized is the decision value of the cutter head speed.

9. The intelligent decision-making method for TBM operation parameters based on dynamic identification of formation information according to claim 8 is characterized in that: In the step 3, the Black Kite algorithm is improved into a multi-objective optimization algorithm by using a weighted Chebyshev scalarization strategy, and then the initial population generation strategy of the Black Kite algorithm is replaced by the Tent chaotic mapping strategy, so as to construct a multi-objective Black Kite optimization algorithm, and the operating parameters and passive parameters directly generated in the TBM tunneling process are used as the input of the rock mass parameter multi-output prediction model to obtain the actual rock mass parameters, and the tunneling working conditions are identified according to the uniaxial compressive strength of the rock mass in the actual rock mass parameters, and then the optimization scheme is determined. In the optimization process, the target value and weight are set according to the engineering requirements, and the rock parameters are kept unchanged. The tunneling parameter multi-output prediction model is used as the fitting function, and the operating parameters and actual rock mass parameters directly generated in the TBM tunneling process are used as input to obtain the tunneling rate and the tool consumption per linear meter. Then, the weights and target values ​​of the tunneling rate and the tool consumption per linear meter are set in combination with the excavation plan and the construction budget, and the tunneling efficiency, construction cost, equipment safety and engineering experience are comprehensively considered, and different optimization schemes are formulated for different working conditions, and finally the corresponding decision values ​​of the cutter head speed, cutter head torque and cutter head thrust are given.

10. The intelligent decision-making method for TBM operation parameters based on dynamic identification of formation information according to claim 1 or 9, characterized in that: The parameter information collected by the model update module is processed using the isolation forest algorithm to establish a database, build and continuously update the model performance, and the excavation stratum information identification module uses the operating parameters and passive parameters of the current excavation section as input. The rock parameters of the current excavation section are predicted through a tabular deep learning model to identify the excavation working conditions. The operating parameter optimization decision module uses a multi-objective black kite optimization algorithm and a multi-objective prediction model of excavation parameters as a mapping relationship. According to the actual working conditions, the decision values ​​of cutter head speed, cutter head torque and cutter head thrust are given.

Images