Teaching simulation method and system for heading machine

Abstract

The invention relates to a teaching simulation method and system for a heading machine, and the method comprises the steps: obtaining a model data set for simulating the working condition of the heading machine and a simulation construction data set through a heading machine simulation model and a machine learning algorithm model; judging whether the error of the same type of data in the model data set and the simulation construction data set is smaller than a set threshold value or not; if the data value error is greater than a set threshold value, training the machine learning algorithm model by taking the data in the model data set as a newly added data sample, and obtaining the optimal data value of the type of data again through the trained machine learning algorithm model; and according to the optimal data value, performing data fusion on the model data set and the simulation construction data set to obtain a simulation data set for operating the heading machine. According to the invention, not only can the action of the heading machine be simulated, but also the construction parameters can be simulated and displayed, and meanwhile, the algorithm model is continuously updated and learned through a self-learning mechanism, so that the simulated construction parameters are always in a relatively high precision range.

Description

Technical Field

[0001] The present invention relates to the technical field of tunnel boring machine teaching, and in particular to a tunnel boring machine teaching simulation method and system. Background Art

[0002] Tunnel boring machines are important equipment in the construction process of underground engineering equipment. They have the characteristics of complex structure, complex construction methods and many construction parameters. Before operating a tunnel boring machine, you need to have sufficient understanding of the tunnel boring machine. However, tunnel boring machines are generally expensive, and the geological environment and other working conditions during construction are diverse, so it is impossible to use real machine models to provide teaching guidance.

[0003] At present, most of the training of tunnel boring machines is achieved through teaching simulation systems. However, most of the mainstream teaching simulation systems can only simulate the basic actions of the tunnel boring process. A few teaching systems simplify the geological environment model and the tunnel boring machine model into a model machine, and then build a physical model for simulation teaching, which can simulate some geological environments or working conditions. However, the sensor parameters of the tunnel boring machine cannot be displayed and simulated, because part of the parameter data of each component of the tunnel boring machine during construction can be calculated by the tunnel boring machine model system simulation software, while the other part of the data cannot be calculated by the tunnel boring machine model system simulation software. This part of the data may be from the sensor data of the actual tunnel boring machine or the data coupled with the comprehensive influence of multiple factors such as the geological environment and the human environment, such as the pressure sensor data, speed sensor data, displacement sensor data during the construction process, and some valve switch signals that need to be manually operated during the construction process. If these data are obtained by building hardware equipment (sensors), the expensive cost of each sensor will greatly increase the teaching cost of the tunnel boring machine.

[0004] Therefore, how to restore the data of the construction process as completely as possible at low cost to achieve a better teaching experience has become an urgent problem to be solved. Summary of the invention

[0005] 1. Technical issues to be resolved

[0006] In view of the above-mentioned shortcomings and deficiencies of the prior art, the present invention provides a teaching simulation method and system for a tunnel boring machine, which solves the technical problem of requiring architectural cost to completely restore data during the construction process during the implementation of the tunnel boring machine teaching simulation.

[0007] (II) Technical solution

[0008] In order to achieve the above object, the main technical solutions adopted by the present invention include:

[0009] In a first aspect, an embodiment of the present invention provides a teaching simulation method for a roadheader, which comprises:

[0010] Based on the operation instruction information input by the user, a pre-built tunnel boring machine simulation model is used to perform system simulation to obtain a model data set, and a machine learning algorithm model is used to learn the historical construction data of the tunnel boring machine to obtain a simulated construction data set;

[0011] Calculating the data value error of the same type of data in the model data set and the simulated construction data set, and determining whether the data value error is less than a set threshold;

[0012] If the data value error is not less than the set threshold, the machine learning algorithm model is trained with the data in the model data set as the new data sample, and the historical construction data is learned again through the trained machine learning algorithm model to obtain the optimal data value of this type of data;

[0013] According to the optimal data value, the model data set and the simulated construction data set are subjected to data fusion processing, and the obtained simulated data set of operating the tunnel boring machine is sent to the user display interface for display.

[0014] Optionally, before using a pre-built tunnel boring machine simulation model to perform system simulation based on the operation instruction information input by the user to obtain a model data set, and using a machine learning algorithm model to learn the historical construction data of the tunnel boring machine to obtain a simulated construction data set, the method further includes:

[0015] Build a tunnel boring machine simulation model and geological environment model based on the acquired tunnel boring machine parameter information and historical construction data;

[0016] Using historical construction data as data samples, the initial machine learning algorithm model is trained to obtain a machine learning algorithm model;

[0017] Among them, the machine learning algorithm models include: random forest algorithm model, decision tree algorithm model and Xgboost algorithm model.

[0018] Optionally, based on the operation instruction information input by the user, using a pre-built tunnel boring machine simulation model to perform system simulation to obtain a model data set, and using a machine learning algorithm model to learn the historical construction data of the tunnel boring machine to obtain a simulated construction data set includes:

[0019] Obtaining operation instruction information and geological environment information input by the user;

[0020] Based on the geological environment information, the geological environment model is used to simulate the construction conditions of the tunnel boring machine;

[0021] Based on the operation instruction information and the construction conditions of the tunnel boring machine, a simulation model of the tunnel boring machine is used to perform system simulation to obtain a model data set;

[0022] Taking the construction conditions as constraints, the machine learning algorithm model is used to learn the historical construction data of the tunnel boring machine to obtain a simulated construction data set.

[0023] Optionally, the historical construction data of the roadheader is learned using a machine learning algorithm model with the working condition as a constraint, and the simulated construction data set obtained includes:

[0024] Determine data items of the simulation data set according to the operation instruction information;

[0025] Analyze and process the data items, and select the corresponding machine learning algorithm model based on the analysis results to learn the historical construction data;

[0026] Taking the construction conditions as constraints, the historical construction data of the tunnel boring machine is learned in combination with the selected machine learning algorithm model to obtain a simulated construction data set.

[0027] Optionally, calculating the data value error of the same type of data respectively located in the model data set and the simulated construction data set, and judging whether the data value error is less than a set threshold comprises:

[0028] Acquire the same type of data respectively located in the model data set and the simulated construction data set;

[0029] Obtain the data value error of the same type of data according to the error calculation formula, and determine whether the data value error is less than the set threshold;

[0030] If the data value error is less than the set threshold, it is determined that the data of this type is the target data for simulating the action of the tunnel boring machine in both the model data set and the simulated construction data set;

[0031] If the data value error is not less than the set threshold, it is determined that the data of this type does not belong to the target data for simulating the action of the tunnel boring machine;

[0032] The error calculation formula is:

[0033] ε s =|K 1 -K 2 |*(1 / K 1 )

[0034] In the formula, ε s Represents the data value error, K 1 represents the data value of the data in the model dataset, K 2 Represents data values ​​of the same type as those in the simulation construction dataset and the model dataset.

[0035] Optionally, performing data fusion processing on the model data set and the simulated construction data set according to the optimal data value, and sending the obtained simulated data set of operating the tunnel boring machine to the user display interface for display includes:

[0036] Perform data fusion processing on the model data set and the simulated construction data set to obtain an initial simulated data set for operating the tunnel boring machine;

[0037] Based on the optimal data values ​​of the same type of data respectively located in the model data set and the simulated construction data set, the data values ​​of the type of data in the initial simulation data set are replaced with the optimal data values ​​to obtain the initial simulation data set;

[0038] The simulation data set is sent to any one of a display screen, a VR device, and a MR device for display.

[0039] Optionally, it also includes:

[0040] Based on the working mode switching instruction information input by the user, the tunnel boring machine simulation model and the machine learning algorithm model are switched to the corresponding working mode;

[0041] When the working mode is the trainee mode, the simulation data set of the operating roadheader is obtained by controlling the operation of the roadheader simulation model and the machine learning algorithm model;

[0042] When the working mode is the teacher mode, and when the obtained simulation data set does not meet the set data threshold range, the machine learning algorithm model is trained based on the parameter data input by the user as the new sample data.

[0043] In a second aspect, an embodiment of the present invention provides a teaching simulation system for a roadheader, comprising:

[0044] A digital twin system, used to execute the teaching simulation method of a tunnel boring machine as described in any one of claims 1 to 7;

[0045] The tunnel boring machine simulation operating system connected to the digital twin system is used for interactive operations with users to teach the tunnel boring machine.

[0046] Optionally, the digital twin system includes:

[0047] The construction condition simulation module is used to obtain the construction condition of the tunnel boring machine by simulating the geological environment model based on the geological environment information input by the user;

[0048] The tunnel boring machine operation simulation module is used to obtain a model data set by performing system simulation using a tunnel boring machine simulation model based on the operation instruction information and the construction conditions of the tunnel boring machine;

[0049] The simulated construction data acquisition module is used to learn the historical construction data of the tunnel boring machine using the machine learning algorithm model with the construction conditions as constraints to obtain a simulated construction data set;

[0050] The error calculation module is used to calculate the data value error of the same type of data respectively located in the model data set and the simulated construction data set, and determine whether the data value error is less than a set threshold;

[0051] The machine learning algorithm model training module is used to train the machine learning algorithm model with the data in the model data set as the new data sample if the data value error is not less than the set threshold, and to learn the historical construction data again through the trained machine learning algorithm model to obtain the optimal data value of this type of data;

[0052] The data fusion module is used to perform data fusion processing on the model data set and the simulated construction data set according to the optimal data value, and send the obtained simulated data set of operating the tunnel boring machine to the user display interface for display.

[0053] Optionally, the tunnel boring machine simulation operating system includes:

[0054] The host computer is used to send operating instructions to the tunnel boring machine;

[0055] The lower computer is connected to the upper computer and the digital twin system respectively, and is used to parse the operation instruction line and send it to the digital twin system, and send the received simulation data set fed back by the digital twin system to the upper computer;

[0056] The virtual reality display device is connected to the digital twin system and is used to display the simulated data set fed back by the digital twin system.

[0057] (III) Beneficial effects

[0058] The beneficial effect of the present invention is that the teaching simulation method of a tunnel boring machine proposed by the present invention uses a tunnel boring machine simulation model to move according to the received operation instructions input by the user to obtain a model data set, and uses a machine learning algorithm model to learn the historical construction data of the tunnel boring machine to obtain a simulated construction data set, and compares and fuses the two sets of data sets to obtain a technical solution of a simulated data set that can fully simulate the operation of the tunnel boring machine. Compared with the prior art, it realizes the simulation of construction data that cannot be simulated, such as sensors, through a machine learning algorithm model, thereby reducing the investment cost of completely restoring the construction data.

[0059] At the same time, in the present invention, the machine learning algorithm model is adaptively learned based on the data in the model data set, and the machine learning algorithm model is continuously corrected and adjusted, thereby improving the accuracy of the simulated tunnel boring machine construction data. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 A schematic diagram of an algorithm implementation flow of a teaching simulation method for a roadheader provided by an embodiment of the present invention;

[0061] Figure 2 A schematic diagram of a teaching simulation system for a tunnel boring machine provided by an embodiment of the present invention;

[0062] Figure 3 A schematic diagram of a digital twin teaching system for a shaft boring machine provided by an embodiment of the present invention;

[0063] Figure 4 A schematic diagram of the downward movement of a shaft boring machine provided by an embodiment of the present invention;

[0064] Figure 5 A schematic diagram of a shaft boring machine cutter head rotating and cutting geology provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0065] In order to better explain the present invention and facilitate understanding, the present invention is described in detail below through specific implementation modes in conjunction with the accompanying drawings.

[0066] refer to Figure 1 As shown, a teaching simulation method and system for a tunnel boring machine proposed in an embodiment of the present invention, the execution subject of the method of this embodiment can be a simulation system using any one of digital twin technology, visualization technology, virtual reality technology and augmented reality technology, the simulation system is electrically connected to the tunnel boring machine operation device and the simulation data display device, and the teaching simulation method includes: first, based on the operation instruction information input by the user, a pre-built tunnel boring machine simulation model is used to perform system simulation to obtain a model data set, and a machine learning algorithm model is used to learn the historical construction data of the tunnel boring machine to obtain a simulated construction data set; secondly, the data value error of the same type of data respectively located in the model data set and the simulated construction data set is calculated, and it is determined whether the data value error is less than a set threshold; then, if the data value error is not less than the set threshold, the machine learning algorithm model is trained with the data in the model data set as the newly added data sample, and the historical construction data is learned again by the trained machine learning algorithm model to obtain the optimal data value of this type of data; finally, according to the optimal data value, the model data set and the simulated construction data set are subjected to data fusion processing, and the obtained simulation data set of operating the tunnel boring machine is sent to the user display interface for display.

[0067] This embodiment uses a technical solution that uses a simulation model of a tunnel boring machine to obtain a model data set by moving according to the operation instructions input by the user, and uses a machine learning algorithm model to learn the historical construction data of the tunnel boring machine to obtain a simulated construction data set, and compares and fuses the two sets of data sets to obtain a simulated data set that can fully simulate the operation of the tunnel boring machine. Compared with the existing technology, it uses a machine learning algorithm model to simulate construction data that cannot be simulated, such as sensors, thereby reducing the investment cost of completely restoring construction data.

[0068] At the same time, in the embodiment, the machine learning algorithm model is adaptively learned based on the data in the model data set, and the machine learning algorithm model is continuously corrected and adjusted, thereby improving the accuracy of the simulated tunnel boring machine construction data.

[0069] In order to better understand the above technical solution, exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided to enable a clearer and more thorough understanding of the present invention and to fully convey the scope of the present invention to those skilled in the art.

[0070] Specifically, this embodiment provides a teaching simulation method for a tunnel boring machine, which includes:

[0071] S100, based on the operation instruction information input by the user, use the pre-built tunnel boring machine simulation model to perform system simulation to obtain a model data set, and use the machine learning algorithm model to learn the historical construction data of the tunnel boring machine to obtain a simulated construction data set.

[0072] Before executing step S100, it is also necessary to build physical simulation models of different types of tunnel boring machines and geological environment models that simulate the construction conditions of tunnel boring machines by inputting parameter data and historical construction data of various types of tunnel boring machines, and to train the selected initial machine learning algorithm model using historical construction data as data samples to obtain a machine learning algorithm model. The specific implementation steps are:

[0073] F100. Build a tunnel boring machine simulation model and geological environment model based on the acquired tunnel boring machine parameter information and historical construction data.

[0074] It is further explained that the parameter information and historical construction data of the tunnel boring machine can be obtained according to the user's self-defined input or obtained from the tunnel boring machine on-site construction database, and the constructed tunnel boring machine simulation model and geological environment model are stored in a pre-set virtual model database, so that when the user selects different types of tunnel boring machines for teaching and training, they can complete the simulation of the tunnel boring machine's actions by calling the corresponding tunnel boring machine simulation model and geological environment model, and then complete the virtual operation of the tunnel boring machine, so that when implementing tunnel boring machine teaching and training for users, there is no need to build a physical model for simulation teaching, which further reduces the training cost of tunnel boring machine teaching.

[0075] F200. Use historical construction data as data samples to train the initial machine learning algorithm model to obtain a machine learning algorithm model.

[0076] Further explanation: historical construction data can be obtained according to user-defined input or from the on-site construction database of the tunnel boring machine. The obtained historical construction data is used as a data sample to train the initial machine learning algorithm model to obtain the machine learning algorithm model, and the machine learning algorithm model is stored in a pre-set algorithm model database, wherein the machine learning algorithm model includes: random forest algorithm model, decision tree algorithm model and Xgboost algorithm model, so that the user can select the corresponding algorithm model according to the proficiency of the algorithm model. After the user selects the corresponding machine learning algorithm model, the system calls the corresponding machine learning algorithm model to learn the target data selected from the historical construction data, and obtains the simulated construction data set corresponding to the operation of the tunnel boring machine.

[0077] In this embodiment, step S100 may include the following sub-steps S110 to S130:

[0078] S110, operation instruction information and geological environment information input by the user.

[0079] S120. Based on the geological environment information, a geological environment model is used to simulate and obtain the construction condition of the tunnel boring machine.

[0080] S130 controls the tunnel boring machine simulation model to perform system simulation in a working environment according to the operation instruction information to obtain a model data set.

[0081] By parsing the operating instructions entered by the user on the operating console or the host computer panel, the type of tunnel boring machine and the action information of the tunnel boring machine under the current operating instructions are obtained, and then the corresponding tunnel boring machine simulation model is called from the virtual model database based on the tunnel boring machine type. According to the action information of the tunnel boring machine, the tunnel boring machine simulation model is controlled to perform system simulation in the geological environment model simulating the working environment of the tunnel boring machine to obtain the model data set.

[0082] S140. Taking the construction conditions as constraints, the historical construction data of the tunnel boring machine is learned using a machine learning algorithm model to obtain a simulated construction data set.

[0083] Since the model data set obtained by system simulation through the tunnel boring machine simulation model lacks real-time detection data during the actual tunnel boring machine excavation process, such as various sensor monitoring data. Therefore, this embodiment uses the working environment as a constraint condition and uses the selected machine learning algorithm model to learn the historical construction data of this type of tunnel boring machine to obtain a simulated construction data set, and the simulated construction data set contains various sensor monitoring data of this type of tunnel boring machine in the determined working environment.

[0084] Furthermore, in step S140, this embodiment can also adaptively adjust the machine learning model to learn the historical construction data. The specific steps are as follows:

[0085] S141. Determine data items of the simulation data set according to the operation instruction information.

[0086] S142, analyzing and processing the data items, and selecting a corresponding machine learning algorithm model based on the analysis results to learn the historical construction data. The available machine learning algorithm models include at least: machine forest algorithm model, decision tree algorithm model and Xgboost algorithm model.

[0087] S143. Taking the construction conditions as constraints, the historical construction data of the tunnel boring machine is learned in combination with the selected machine learning algorithm model to obtain a simulated construction data set.

[0088] For example, when the machine learning algorithm model specified by the user is a decision tree algorithm model, by analyzing each data item in the simulation data set, when there are some data items with high requirements for data accuracy, the decision tree algorithm model is replaced by a random forest algorithm model to learn the historical construction data and improve the accuracy of the data.

[0089] S200, calculating the data value error of the same type of data respectively located in the model data set and the simulated construction data set, and determining whether the data value error is less than a set threshold.

[0090] For example, in this embodiment, step S200 may include the following sub-steps S210 and S230:

[0091] S210: Acquire data of the same type respectively located in the model data set and the simulated construction data set.

[0092] S220, obtaining the data value error of the same type of data according to the error calculation formula, and determining whether the data value error is less than a set threshold.

[0093] The error calculation formula is:

[0094] ε s =|K 1 -K 2 |*(1 / K 1 )

[0095] In the formula, ε s Represents the data value error, K 1 represents the data value of the data in the model dataset, K 2 Represents data values ​​of the same type as those in the simulation construction dataset and the model dataset.

[0096] In this embodiment, the threshold value can be set to 0.1, 0.001, etc. according to user needs.

[0097] S230a: If the data value error is less than the set threshold, it is determined that the data of this type is the target data for simulating the action of the tunnel boring machine in both the model data set and the simulated construction data set.

[0098] S220b: If the data value error is not less than the set threshold, it is determined that the data of this type does not belong to the target data for simulating the action of the tunnel boring machine.

[0099] S300. If the data value error is not less than the set threshold, the machine learning algorithm model is trained with the data in the model data set as the new data sample, and the historical construction data is learned again through the trained machine learning algorithm model to obtain the optimal data value of this type of data.

[0100] In this embodiment, when the data value error is not less than the set threshold, the data in the model data set obtained by the tunnel boring machine simulation model is used as a new data sample, combined with the original historical construction data to form a new data sample, and the machine learning algorithm model is retrained until the obtained data value error is less than the set threshold, and the type of data obtained through multiple machine learning algorithm model learning is determined to be the target data for simulating the action of the tunnel boring machine, and its data value is the optimal data value. This embodiment uses an algorithm model with a built-in self-learning mechanism and combines it with a tunnel boring machine simulation model, which can not only simulate the action of the tunnel boring machine during construction, but also simulate and display the complete incalculable construction parameters. At the same time, the algorithm model with a built-in self-learning mechanism is constantly updated and learned, so that the simulated construction parameters are always kept within a relatively high precision range.

[0101] S400: Perform data fusion processing on the model data set and the simulated construction data set according to the optimal data value, and send the obtained simulated data set of operating the tunnel boring machine to the user display interface for display.

[0102] For example, in this embodiment, step S400 may include the following sub-steps S410 and S430:

[0103] S410, performing data fusion processing on the model data set and the simulated construction data set to obtain an initial simulation data set for operating the tunnel boring machine.

[0104] In this embodiment, the initial simulation data set includes different types of data in the model data set and the simulated construction data set, especially the unique data in the simulated construction data set, which is the detection data of various sensors in the process of simulating the operation of the tunnel boring machine, and this data cannot be obtained through the system simulation of the tunnel boring machine simulation model. By simulating the detection data of various sensors of the tunnel boring machine, the tunnel boring machine operation can be simulated more completely, and the effect of the tunnel boring machine teaching simulation can be improved.

[0105] S420. Based on the optimal data values ​​of the same type of data respectively located in the model data set and the simulated construction data set, the data values ​​of the type of data in the initial simulation data set are replaced with the optimal data values ​​to obtain the initial simulation data set.

[0106] In view of the fact that there are two data of the same type in the initial simulation data set, based on the data value error judgment of the same type of data, when the data value of the same type of data is less than the set threshold, the data of this type in the model data set or the simulated construction data set is taken as the target data for simulating the action of the tunnel boring machine. When the data value of the same type of data is not less than the set threshold, the data value of this type is replaced with the optimal data value, and the data of this type that has been numerically replaced is used as the target data for simulating the action of the tunnel boring machine.

[0107] S430: Send the simulation data set to any one of a display screen, a VR device, and an MR device for display.

[0108] For example, when an MR device is selected to display a simulation data set, after the user wears the MR device, the system transmits the simulation data set to the MR device for display, allowing the user to experience the construction process of the tunnel boring machine immersively and improve the teaching effect.

[0109] The teaching simulation method of a roadheader provided in this embodiment also includes:

[0110] S510. Based on the working mode switching instruction information input by the user, the tunnel boring machine simulation model and the machine learning algorithm model are switched to corresponding working modes.

[0111] S520a. When the working mode is the trainee mode, a simulation data set of operating the tunnel boring machine is obtained by controlling the operation of the tunnel boring machine simulation model and the machine learning algorithm model.

[0112] S520b. When the working mode is the teacher mode and the obtained simulation data set does not meet the set data threshold range, the machine learning algorithm model is trained based on the parameter data input by the user as the newly added sample data.

[0113] In this embodiment, when the working mode is switched to the instructor-student mode, first, the user inputs the corresponding operation instructions to run the tunnel boring machine simulation model and the machine learning algorithm model to obtain a simulated data set for operating the tunnel boring machine; then, it automatically determines whether each data in the simulated data set meets the set data threshold range. If not, the data is marked and a warning message of the data error is output; finally, the parameter data input by the user is the newly added sample data, and the machine learning algorithm model is trained to eliminate the erroneous data in the simulated data set. Alternatively, when the working mode is switched to the instructor-student mode, the obtained simulated data set is sent to the user display interface for digital display, so that the user can review the simulated data set, manually modify the obviously unreasonable data or the data that he wants to change, and finally use the modified data as the newly added sample data to train the machine learning algorithm model.

[0114] In a specific embodiment, the user chooses to conduct simulation teaching training for a shaft boring machine. First, the corresponding shaft boring machine simulation model and geological environment model are selected from the database to build a shaft boring machine twin teaching system, such as Figure 3 Then, after the user issues a downward excavation instruction, the shaft boring machine simulation model simulates the shaft boring machine's cutter head and the geology to complete the downward excavation action, as shown in Figure 1. Figure 4 Then, the twin teaching system of the shaft boring machine takes the construction conditions as input and combines the two telescopic cylinders ( Figure 5 The stroke, pressure and other parameters of the telescopic oil cylinder (the blue part in the middle) are simulated and calculated to obtain the model data set; at the same time, based on the machine learning model, the stroke, pressure and other parameters of the two telescopic oil cylinders under the same construction conditions are obtained from the historical data set as the simulated construction data set. Finally, by comparing the stroke data and pressure data of the two telescopic oil cylinders in the model data set and the simulated construction data set, and when the error between the two does not exceed one percent, the data calculated by the system simulation model is taken as the final display data of the two telescopic oil cylinders. Similarly, the other moving parts of the well boring machine can also be evaluated in the same way.

[0115] On the other hand, this embodiment also provides a teaching simulation system for a roadheader, which includes:

[0116] The digital twin system is used to execute the teaching simulation method of the tunnel boring machine described above.

[0117] The tunnel boring machine simulation operating system connected to the digital twin system is used for interactive operations with users to teach the tunnel boring machine.

[0118] Furthermore, the digital twin system includes:

[0119] The construction condition simulation module is used to obtain the construction condition of the tunnel boring machine by simulating the geological environment model based on the geological environment information input by the user.

[0120] The tunnel boring machine operation simulation module is used to obtain a model data set by performing system simulation using a tunnel boring machine simulation model based on the operation instruction information and the construction conditions of the tunnel boring machine.

[0121] The simulated construction data acquisition module is used to learn the historical construction data of the tunnel boring machine using the machine learning algorithm model with the construction conditions as constraints to obtain a simulated construction data set.

[0122] The error calculation module is used to calculate the data value error of the same type of data respectively located in the model data set and the simulated construction data set, and determine whether the data value error is less than a set threshold.

[0123] The machine learning algorithm model training module is used to train the machine learning algorithm model with the data in the model data set as the new data sample if the data value error is not less than the set threshold, and to learn the historical construction data again through the trained machine learning algorithm model to obtain the optimal data value of this type of data.

[0124] The data fusion module is used to perform data fusion processing on the model data set and the simulated construction data set according to the optimal data value, and send the obtained simulated data set of operating the tunnel boring machine to the user display interface for display.

[0125] Furthermore, the tunnel boring machine simulation operating system includes:

[0126] The host computer is used to send operating instructions to the tunnel boring machine.

[0127] The lower computer is connected to the upper computer and the digital twin system respectively, and is used to parse the operation instruction line and send it to the digital twin system, and send the simulation data set fed back by the digital twin system to the upper computer.

[0128] The virtual reality display device is connected to the digital twin system and is used to display the simulated data set fed back by the digital twin system.

[0129] Further explanation of the tunnel boring machine simulation operating system, the upper computer may include an operation console and an upper computer host. After the user inputs the tunnel boring machine working condition that the user wants to simulate, the operation console will form an operation command signal and send it to the lower computer. After the user selects the corresponding tunnel boring machine working mode through the operation screen on the upper computer host, the operation command information will be formed and sent to the lower computer. The lower computer builds a PLC system, and the PLC adopts a mature S7 series PLC system, which is used to parse the operation command line and send it to the digital twin system, and send the received simulation data set fed back by the digital twin system to the upper computer. Virtual reality display equipment includes MR equipment and VR equipment, which is used to display the simulation data set fed back by the digital twin system, so that users can experience the construction process of the tunnel boring machine in person. Among them, the communication between the operation console and the lower computer is established through communication protocols such as Can bus and serial port; the communication between the upper computer host and the lower computer, the communication between the upper computer host and the operation console, and the communication between the lower computer and the digital twin system are realized through the TCP / IP protocol, realizing the forward and reverse transmission of data.

[0130] In summary, the embodiment of the present invention provides a teaching simulation method and system for a tunnel boring machine, and the teaching simulation method includes: first, based on the operation instructions of the tunnel boring machine input by the user, the tunnel boring machine simulation model and the machine learning algorithm model are controlled to simulate the working conditions of the tunnel boring machine to obtain a model data set and a simulated construction data set. Then, the data respectively located in the model data set and the simulated construction data set are compared and analyzed. Then, for the data that cannot be obtained by the system simulation of the tunnel boring machine simulation model, the data simulated by the machine learning algorithm model is directly taken and retained; for the same type of data, the data values ​​of the same type of data are compared. If the data value is within the allowable error range, it means that the simulated data is valid. If the data value is not within the allowable error range, the data obtained by the system simulation of the tunnel boring machine simulation model is put into the machine learning algorithm model as sample data for re-learning to obtain the optimal data value of the type of data. Finally, based on the optimal data value, the model data set and the simulated construction data set are subjected to data fusion processing to obtain a simulated data set for operating the tunnel boring machine, and the simulated data set is sent to the user display interface for display.

[0131] The technical solution of the present invention proposes a teaching simulation method and system for a tunnel boring machine. By building a tunnel boring machine simulation model and a machine learning algorithm model to simulate the working conditions of the tunnel boring machine, and combining with a self-learning mechanism built into the machine learning algorithm model, it can not only simulate the actions of the tunnel boring machine during the construction process, but also simulate and display the complete incalculable construction parameters. At the same time, the machine learning algorithm model is continuously updated and learned through the self-learning mechanism, so that the simulated construction parameters are always maintained within a relatively high precision range.

[0132] Since the system / device described in the above embodiments of the present invention is a system / device used to implement the method of the above embodiments of the present invention, a person skilled in the art can understand the specific structure and deformation of the system / device based on the method described in the above embodiments of the present invention, and thus will not be described in detail here. All systems / devices used in the method of the above embodiments of the present invention belong to the scope of protection of the present invention.

[0133] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0134] The present invention is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems) and computer program products according to the embodiments of the present invention. It should be understood that each process and / or box in the flowchart and / or block diagram and the combination of the processes and / or boxes in the flowchart and / or block diagram can be implemented by computer program instructions.

[0135] It should be noted that in the claims, any reference numerals placed between brackets shall not be construed as limiting the claims. The word "comprising" does not exclude the presence of components or steps not listed in the claims. The word "a" or "an" preceding a component does not exclude the presence of a plurality of such components. The invention may be implemented by means of hardware comprising several different components and by means of a suitably programmed computer. In the claims enumerating several means, several of these means may be embodied by the same hardware. The use of the words first, second, third, etc., is for convenience of expression only and does not indicate any order. These words may be understood as part of the component name.

[0136] In addition, it should be noted that, in the description of this specification, the description of the terms "one embodiment", "some embodiments", "embodiment", "example", "specific example" or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, unless they are contradictory.

[0137] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments after knowing the basic creative concept. Therefore, the claims should be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present invention.

[0138] Obviously, those skilled in the art can make various modifications and variations to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention should also include these modifications and variations.

Claims

1. A teaching simulation method for a tunnel boring machine, characterized in that: include: Based on the operation instruction information input by the user, a pre-built tunnel boring machine simulation model is used to perform system simulation to obtain a model data set, and a machine learning algorithm model is used to learn the historical construction data of the tunnel boring machine to obtain a simulated construction data set; Calculating the data value error of the same type of data in the model data set and the simulated construction data set, and determining whether the data value error is less than a set threshold; If the data value error is not less than the set threshold, the machine learning algorithm model is trained with the data in the model data set as the new data sample, and the historical construction data is learned again through the trained machine learning algorithm model to obtain the optimal data value of this type of data; According to the optimal data value, the model data set and the simulated construction data set are subjected to data fusion processing, and the obtained simulated data set of operating the tunnel boring machine is sent to the user display interface for display.

2. The method according to claim 1, characterized in that Before the system simulation is performed based on the operation instruction information input by the user using the pre-built tunnel boring machine simulation model to obtain the model data set, and the historical construction data of the tunnel boring machine is learned using the machine learning algorithm model to obtain the simulated construction data set, the method also includes: Build a tunnel boring machine simulation model and geological environment model based on the acquired tunnel boring machine parameter information and historical construction data; Using historical construction data as data samples, the initial machine learning algorithm model is trained to obtain a machine learning algorithm model; Among them, the machine learning algorithm models include: random forest algorithm model, decision tree algorithm model and Xgboost algorithm model.

3. The method according to claim 2, characterized in that Based on the operation instruction information input by the user, the system simulation is performed using the pre-built virtual tunnel boring machine simulation model to obtain the model data set, and the historical construction data of the tunnel boring machine is learned using the machine learning algorithm model to obtain the simulated construction data set including: Obtaining operation instruction information and geological environment information input by the user; Based on the geological environment information, the geological environment model is used to simulate the construction conditions of the tunnel boring machine; Based on the operation instruction information and the construction conditions of the tunnel boring machine, a simulation model of the tunnel boring machine is used to perform system simulation to obtain a model data set; Taking the construction conditions as constraints, the machine learning algorithm model is used to learn the historical construction data of the tunnel boring machine to obtain a simulated construction data set.

4. The method according to claim 3, characterized in that Taking the construction conditions as constraints, the machine learning algorithm model is used to learn the historical construction data of the roadheader to obtain the simulated construction data set including: Determine data items of the simulation data set according to the operation instruction information; Analyze and process the data items, and select the corresponding machine learning algorithm model based on the analysis results to learn the historical construction data; Taking the construction conditions as constraints, the historical construction data of the tunnel boring machine is learned in combination with the selected machine learning algorithm model to obtain a simulated construction data set.

5. The method according to claim 1, characterized in that Calculating the data value error of the same type of data in the model data set and the simulated construction data set, and judging whether the data value error is less than a set threshold, includes: Acquire the same type of data respectively located in the model data set and the simulated construction data set; Obtain the data value error of the same type of data according to the error calculation formula, and determine whether the data value error is less than the set threshold; If the data value error is less than the set threshold, it is determined that the data of this type is the target data for simulating the action of the tunnel boring machine in both the model data set and the simulated construction data set; If the data value error is not less than the set threshold, it is determined that the data of this type does not belong to the target data for simulating the action of the tunnel boring machine; The error calculation formula is: e s =|K1-K2|*(1 / K1) In the formula, ε s represents the data value error, K1 represents the data value of the data in the model data set, and K2 represents the data value of the same type of data in the simulation construction data set and the model data set.

6. The method according to claim 1, characterized in that According to the optimal data value, the model data set and the simulated construction data set are subjected to data fusion processing, and the obtained simulated data set of operating the tunnel boring machine is sent to the user display interface for display, including: Perform data fusion processing on the model data set and the simulated construction data set to obtain an initial simulated data set for operating the tunnel boring machine; Based on the optimal data values ​​of the same type of data respectively located in the model data set and the simulated construction data set, the data values ​​of the type of data in the initial simulation data set are replaced with the optimal data values ​​to obtain the initial simulation data set; The simulation data set is sent to any one of a display screen, a VR device, and a MR device for display.

7. The method according to claim 1, characterized in that Also includes: Based on the working mode switching instruction information input by the user, the tunnel boring machine simulation model and the machine learning algorithm model are switched to the corresponding working mode; When the working mode is the trainee mode, the simulation data set of the operating roadheader is obtained by controlling the operation of the roadheader simulation model and the machine learning algorithm model; When the working mode is the teacher mode, and when the obtained simulation data set does not meet the set data threshold range, the machine learning algorithm model is trained based on the parameter data input by the user as the new sample data.

8. A teaching simulation system for a tunnel boring machine, characterized in that: include: A digital twin system, used to execute the teaching simulation method of a tunnel boring machine as described in any one of claims 1 to 7; The tunnel boring machine simulation operating system connected to the digital twin system is used for interactive operations with users to teach the tunnel boring machine.

9. The system according to claim 8, characterized in that The digital twin system includes: The construction condition simulation module is used to obtain the construction condition of the tunnel boring machine by simulating the geological environment model based on the geological environment information input by the user; The tunnel boring machine operation simulation module is used to obtain a model data set by performing system simulation using a tunnel boring machine simulation model based on the operation instruction information and the construction conditions of the tunnel boring machine; The simulated construction data acquisition module is used to learn the historical construction data of the tunnel boring machine using the machine learning algorithm model with the construction conditions as constraints to obtain a simulated construction data set; The error calculation module is used to calculate the data value error of the same type of data respectively located in the model data set and the simulated construction data set, and determine whether the data value error is less than a set threshold; The machine learning algorithm model training module is used to train the machine learning algorithm model with the data in the model data set as the new data sample if the data value error is not less than the set threshold, and to learn the historical construction data again through the trained machine learning algorithm model to obtain the optimal data value of this type of data; The data fusion module is used to perform data fusion processing on the model data set and the simulated construction data set according to the optimal data value, and send the obtained simulated data set of operating the tunnel boring machine to the user display interface for display.

10. The system according to claim 8, characterized in that The tunnel boring machine simulation operating system includes: The host computer is used to send operating instructions to the tunnel boring machine; The lower computer is connected to the upper computer and the digital twin system respectively, and is used to parse the operation instruction line and send it to the digital twin system, and send the received simulation data set fed back by the digital twin system to the upper computer; The virtual reality display device is connected to the digital twin system and is used to display the simulated data set fed back by the digital twin system.

Images