Microwave multi-disc cutter body intelligent control method, device, equipment and medium

Abstract

The application relates to the technical field of mining engineering, and discloses a microwave multi-disc cutter rock breaking body intelligent control method, device, equipment and medium. The method collects multi-source sensing data of a microwave multi-disc cutter tunneling operation face in real time, the multi-source sensing data including tunneling machine operation parameters, microwave cracking feedback parameters and lithological characteristic data; the multi-source sensing data is input into a pre-trained body intelligent model, and rock mechanics characteristics and microwave response characteristics of the current operation face are obtained through inversion; based on the rock mechanics characteristics and the microwave response characteristics, optimal working parameter combinations are calculated and output, the optimal working parameter combinations including microwave irradiation parameters and disc cutter tunneling parameters; the optimal working parameter combinations are sent to the microwave multi-disc cutter, so that the microwave multi-disc cutter rock breaking operation is controlled in real time. The scheme is free from manual intervention throughout the whole process, ensures safe operation of the equipment, effectively reduces operation energy consumption, improves tunneling efficiency and working condition adaptability.

Description

Technical Field

[0001] This application relates to the field of mining engineering technology, and in particular to an intelligent control method, device, equipment and medium for microwave multi-disc rock-breaking tool. Background Technology

[0002] Cantilever tunneling machines are widely used rock excavation equipment in tunnels, mines and various underground engineering fields. Microwave fracturing, as a new type of auxiliary rock breaking technology, can effectively weaken the rock mass strength and reduce the mechanical cutting load, and has good engineering application prospects. The combination of microwave and mechanical rock breaking is also gradually becoming an important development direction of underground tunneling technology.

[0003] Currently, when multi-disc cutterhead tunneling equipment equipped with microwave-assisted devices is operating, it mainly relies on manual on-site observation of the cutting face, combined with apparent information such as machine vibration, dust concentration, and microwave-induced fracturing range, to judge the rock strata conditions and manually adjust the equipment's operating parameters based on the operator's experience. This manual control mode has many obvious drawbacks: First, the operational effectiveness is highly dependent on the professional skills and field experience of the operators. Different personnel have varying abilities to judge and respond to cutting loads and rock strata characteristics, directly resulting in unstable tunneling efficiency and low overall equipment utilization. Second, the tunneling face environment is harsh. High dust, high noise, and high-intensity working conditions easily lead to operator fatigue and misjudgment, which not only reduces construction quality but also easily causes excessive wear of the cutter head, equipment overheating, and even damage to core components. Third, the underground working face has complex and variable rock properties, making it impossible for manual operation to match tunneling and microwave working parameters according to rock mechanical properties in real time and accurately. This results in a high proportion of ineffective cutting, increased equipment energy consumption, shortened cutter life, and high overall tunneling costs. Fourth, the coordinated control of the cutter head mechanism and microwave irradiation port is difficult. Manual operation is prone to problems such as component interference and collisions, further reducing operational efficiency and increasing the risk of equipment failure. In addition, although a small number of remote control solutions have emerged, they still rely on manual analysis and parameter tuning and have not yet achieved true autonomous operation of the equipment. At the same time, as underground engineering gradually extends to areas with greater burial depth and more complex geological conditions, the safety risks of on-site manual operation continue to increase.

[0004] At the same time, the limited data collection conditions at the mine site make it difficult for field tests to flexibly change lithological conditions and precisely control various variables as in laboratory tests. It is also impossible to collect full-dimensional working condition data synchronously, resulting in an insufficient number of effective samples that can be used for intelligent model training, which restricts the implementation of intelligent and autonomous control technologies for tunneling equipment. Summary of the Invention

[0005] This application provides a method, device, equipment, and medium for intelligent control of microwave multi-disc cutter rock breaking, aiming to solve the technical problems in related technologies, such as microwave multi-disc cutter joint rock breaking operations relying on manual observation and experience-based parameter adjustment, resulting in inaccurate parameter matching, poor equipment coordination, high operational risks, and insufficient effective on-site data making it difficult to implement intelligent control.

[0006] In a first aspect, embodiments of this application provide a method for intelligent control of a microwave multi-disc cutter rock-breaking tool, the method comprising: Real-time acquisition of multi-source sensing data from the microwave multi-disc cutterhead tunneling face, including tunneling machine operating parameters, microwave fracturing feedback parameters, and lithological characteristic data; The multi-source sensing data is input into a pre-trained embodied intelligent model to invert the rock mechanical properties and microwave response properties of the current working face. Based on the rock mechanical properties and the microwave response properties, the optimal combination of working parameters is calculated and output, which includes microwave irradiation parameters and cutterhead tunneling parameters. The optimal combination of operating parameters is sent to the microwave multi-disc cutter to enable real-time control of the microwave multi-disc cutter's rock-breaking operation.

[0007] In one embodiment, optionally, the training process of the pre-trained embodied intelligence model includes: Rock samples of various specifications were prepared, and indoor tests were conducted to obtain indoor test data. The indoor test data included basic rock characteristic data, microwave fracturing characteristic data, and mechanical rock breaking feedback data. Based on the indoor test data, the numerical simulation model is calibrated, and multi-condition data is generated in batches through multi-factor orthogonal simulation experiments and fused to obtain a training dataset. An initial prediction model was trained based on the training dataset using machine learning algorithms. The initial prediction model included a three-dimensional geological perception model, a microwave fracturing effect prediction model, and a tunneling machine operation efficiency prediction model. The multi-source sensing data is collected at the mine excavation site, and the initial prediction model is iteratively calibrated to obtain the pre-trained embodied intelligent model.

[0008] In one embodiment, optionally, the tunneling machine operating parameters include at least thrust, torque, vibration acceleration, and cutting rate; Microwave-induced fracturing feedback parameters include at least reflected power, standing wave ratio, and temperature field distribution in the irradiated area; The lithological characteristic data include at least rock surface texture characteristics, mineral composition characteristics, and dielectric response characteristics.

[0009] In one embodiment, optionally, the multi-source sensing data is input into a pre-trained embodied intelligent model to invert and obtain the rock mechanical properties and microwave response properties of the current working face, specifically including: Based on the lithological characteristic data, the rock type is identified, and the uniaxial compressive strength, tensile strength, and elastic modulus of the rock are output. Based on the microwave fracturing feedback parameters, the real and imaginary parts of the rock's dielectric constant are inverted, and the microwave penetration depth under the current working conditions is calculated.

[0010] In one embodiment, optionally, calculating and outputting the optimal combination of operating parameters based on the rock mechanical properties and the microwave response characteristics includes: Based on the rock mechanical properties and microwave response characteristics, and under the premise of satisfying the tool safety load constraints and microwave equipment safety constraints, the optimal combination of working parameters is matched and output with the minimum specific energy consumption as the objective function.

[0011] In one embodiment, optionally, the method further includes: Real-time monitoring of the tunneling machine thrust, torque, and rock surface temperature field change rate in the multi-source sensing data; calculation of the specific energy consumption change per unit advance distance. When the change in thrust, torque or specific energy consumption exceeds the corresponding preset mutation threshold within a preset time window, it is determined that a sudden change in rock strength has occurred at the current working face. In response to the determination that a sudden change in rock strength has occurred at the current working face, the pre-trained embodied intelligent model is recalled within a preset time, and the updated rock mechanical properties and microwave response properties are retrieved based on the multi-source sensing data at the moment of the sudden change. Based on the updated rock mechanical properties and microwave response characteristics, the optimal combination of working parameters is recalculated and output, and the updated optimal working parameters are sent to the microwave multi-disc cutter to adaptively adjust the rock breaking operation.

[0012] In one embodiment, optionally, the indoor test specifically includes: Block samples of different sizes were prepared, wherein all block samples were taken from the same parent rock and cut in the same direction; The blocky sample was subjected to X-ray diffraction whole-rock analysis, dielectric constant test, closed multimode resonant cavity test, open microwave irradiation test, uniaxial compression and Brazilian splitting test, disc cutter test and microwave-mechanical combined rock breaking test in sequence.

[0013] In one embodiment, optionally, training an initial prediction model based on the training dataset using a machine learning algorithm specifically includes: Multiple candidate prediction models were trained using various machine learning algorithms, including logistic regression, support vector machine, decision tree, k-nearest neighbors, random forest, radial basis function neural network, and XGBoost. Through accuracy, precision, recall, and F1 The performance of the multiple candidate prediction models is evaluated using scores and PR curves to obtain evaluation results. The candidate prediction model with the best evaluation results is selected as the three-dimensional geological sensing model, microwave fracturing effect prediction model, and tunneling machine operation efficiency prediction model. The Stacking ensemble learning method is employed, using the multiple candidate prediction models as primary learners and linear regression or logistic regression models as secondary learners to construct an ensemble prediction model.

[0014] Secondly, embodiments of this application provide an intelligent control device for a microwave multi-disc rock-breaking tool, the device comprising: The acquisition module is used to acquire multi-source sensing data of the microwave multi-disc cutterhead tunneling face in real time. The multi-source sensing data includes tunneling machine operating parameters, microwave fracturing feedback parameters, and lithological characteristic data. The inversion module is used to input the multi-source sensing data into a pre-trained embodied intelligent model to invert the rock mechanical properties and microwave response properties of the current working face. The output module is used to calculate and output the optimal combination of working parameters based on the rock mechanical properties and the microwave response characteristics. The optimal combination of working parameters includes microwave irradiation parameters and cutterhead excavation parameters. The control module is used to send the optimal combination of working parameters to the microwave multi-disc cutter to perform real-time control of the microwave multi-disc cutter's rock breaking operation.

[0015] Thirdly, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the above-described intelligent control method for microwave multi-disc rock-breaking device.

[0016] Fourthly, a computer-readable storage medium is provided, which stores a computer program that, when executed by a processor, implements the steps of the above-described intelligent control method for microwave multi-disc rock-breaking devices.

[0017] In this invention, multi-source sensing data from the microwave multi-disc cutterhead tunneling face is acquired in real time. This multi-source sensing data includes tunneling machine operating parameters, microwave fracturing feedback parameters, and lithological characteristic data. The multi-source sensing data is input into a pre-trained embodied intelligent model to invert and obtain the rock mechanical properties and microwave response characteristics of the current working face. Based on these rock mechanical properties and microwave response characteristics, an optimal combination of working parameters is calculated and output. This optimal combination of working parameters includes microwave irradiation parameters and cutterhead tunneling parameters. The optimal combination of working parameters is then sent to the microwave multi-disc cutterhead for real-time control of its rock-breaking operation. This technical solution firstly acquires multi-source sensing data in real time, including tunneling machine operating parameters, microwave fracturing feedback parameters, and lithological characteristic data. This enables a comprehensive, accurate, and continuous understanding of the on-site operating conditions and rock mass status, solving the problems of incomplete and delayed information from manual observation. Secondly, the multi-source sensing data is input into a pre-trained embodied intelligent model to invert and obtain rock mechanical properties and microwave response characteristics. The model enables intelligent and accurate identification of lithology and rock mass response, with objective and stable identification results unaffected by personnel experience or conditions. Thirdly, based on the inversion results, the optimal combination of microwave irradiation parameters and cutterhead tunneling parameters is calculated and output. This allows for adaptive optimization of parameters based on real-time lithological dynamics, achieving precise matching between parameters and operating conditions and reducing ineffective operations. Finally, the optimal parameters are sent to the execution unit for real-time control, enabling autonomous operation of the equipment throughout the entire process. This effectively improves the shortcomings of traditional manual operation, such as large efficiency fluctuations, easy equipment wear, and poor component coordination. It also reduces on-site operational safety risks, tunneling energy consumption, and maintenance costs, comprehensively enhancing the intelligence level and overall construction efficiency of microwave multi-disc cutterhead joint rock breaking operations. Attached Figure Description

[0018] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 A general flowchart of a microwave multi-disc cutter head intelligent control method according to an embodiment of this application is shown.

[0020] Figure 2 A schematic flowchart of a microwave multi-disc cutter head intelligent control method according to an embodiment of this application is shown.

[0021] Figure 3 A flowchart illustrating the training process of a pre-trained embodied intelligence model according to an embodiment of this application is shown.

[0022] Figure 4A flowchart of a dielectric constant test according to an embodiment of this application is shown.

[0023] Figure 5 A schematic diagram of a multimode resonator test according to an embodiment of this application is shown.

[0024] Figure 6 A schematic diagram of an open microwave test according to an embodiment of this application is shown.

[0025] Figure 7 (a)- Figure 7 (c) A diagram showing the results of a microwave irradiation test according to an embodiment of the present application is shown; (a) a closed microwave irradiation crack diagram, (b) an open microwave irradiation crack feature diagram, and (c) an infrared thermogram.

[0026] Figure 8 A diagram showing the test results of rock mechanical properties according to an embodiment of this application is provided.

[0027] Figure 9 A schematic diagram of a microwave multi-disc knife indoor test according to an embodiment of this application is shown.

[0028] Figure 10 A block diagram of a microwave multi-disc rock-breaking tool intelligent control device according to an embodiment of this application is shown.

[0029] Among them: 1-multimode resonant cavity, 2-50mm×50mm×100mm cuboid sample, 3-mullite pad, 4-200mm×200mm×200mm cube sample, 5-microwave antenna, 6-waveguide, 7-automatic matching regulator, 8-microwave generator, 100-microwave multi-disc rock-breaking tool intelligent control device, 1001-acquisition module, 1002-inversion module, 1003-output module, 1004-control module. Detailed Implementation

[0030] To better understand the technical solution of this application, the embodiments of this application will be described in detail below with reference to the accompanying drawings.

[0031] It should be understood that the described embodiments are merely some, not all, of the embodiments in this application. All other embodiments obtained by those skilled in the art based on the embodiments in this application without inventive effort are within the scope of protection of this application.

[0032] The terminology used in the embodiments of this application is for the purpose of describing particular embodiments only and is not intended to be limiting of this application. The singular forms “a,” “the,” and “the” used in the embodiments of this application and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise.

[0033] The following detailed description of some embodiments of this application is provided in conjunction with the accompanying drawings. Unless otherwise specified, the following embodiments and features can be combined with each other.

[0034] like Figure 1 As shown, the microwave multi-disc cutter rock-breaking intelligent control method according to an embodiment of this application mainly includes the following four parts: The first part involves collecting lithological data (rock mechanical properties, mineral composition, dielectric constant, etc.), multi-disc cutter data (parameter settings, equipment output mechanical data), microwave fracturing data (parameter settings, irradiation point temperature, reflection power, standing wave ratio), and numerical simulation data (a numerical model was established based on the indoor test model, and data was generated in batches).

[0035] The second part involves constructing a database for training machine learning models using the aforementioned data. Input variables include rock properties, microwave parameters, and tunneling parameters. Output variables include tunneling performance and fracturing effect.

[0036] The third part involves constructing a three-dimensional geological perception model based on a database. This model is used to analyze the lithology of the working face, predict the microwave fracturing effect, and predict the operating efficiency of microwave multi-disc tunneling machines. It also predicts various output parameters during microwave multi-disc tunneling machine operation. The model is iterated using actual mine operation data to form an embodied intelligent system.

[0037] Part Four utilizes an embodied intelligence system to achieve positive prediction of results and reverse deduction of parameters, continuously optimizing the operating parameters of the microwave multi-disc cutterhead tunneling machine.

[0038] in: (1) The rocks need to be processed into various specifications to complete different tests.

[0039] (2) The experimental equipment includes an X-ray diffractometer, a vector network analyzer, a servo single-axis loading system, a multimode resonant cavity 1, an open microwave irradiation device, a disc rock-breaking device, and a microwave-mechanical integrated rock-breaking device. It also includes a remote control, a computer, and other supporting equipment.

[0040] (3) Data acquisition equipment includes infrared thermal imagers, laser scanners, thrust / torque / vibration sensors, three-dimensional geological sensing systems, etc.

[0041] (4) Machine learning models include logistic regression, support vector machine, decision tree, K-nearest neighbor, random forest, BF neural network, XGBoost, and Stacking.

[0042] like Figure 2As shown, the intelligent control method for microwave multi-disc rock-breaking tools includes the following steps: Step S201: Real-time acquisition of multi-source sensing data from the microwave multi-disc cutterhead tunneling face, including tunneling machine operating parameters, microwave fracturing feedback parameters, and lithological characteristic data; In one embodiment, optionally, the tunneling machine operating parameters include at least thrust, torque, vibration acceleration, and cutting rate; Microwave-induced fracturing feedback parameters include at least reflected power, standing wave ratio, and temperature field distribution in the irradiated area; The lithological characteristic data include at least rock surface texture characteristics, mineral composition characteristics, and dielectric response characteristics.

[0043] The operating parameters of a tunneling machine include at least thrust, torque, vibration acceleration, and cutting rate. Thrust is the axial propulsive force exerted by the cutting mechanism of the microwave multi-disc tunneling machine on the rock face, directly determining the depth of the cutter's penetration into the rock mass and the cutting state. Torque is the torsional moment that drives the cutter's rotation, directly reflecting the overall resistance encountered by the cutter in cutting the rock. Vibration acceleration is the acceleration value of the vibration generated by the machine body and cutter mechanism during operation. Combined with the vibration differences under different rock types and cutting conditions, it can help identify the rock type of the working face and abnormal equipment operation. Cutting rate is the tunneling speed, used to characterize the overall tunneling operation efficiency. These parameters are collected in real time by sensors, providing complete equipment operation data support for the embodied intelligent model to invert rock mechanical properties and dynamically optimize tunneling conditions.

[0044] Microwave fracturing feedback parameters include at least reflected power, standing wave ratio (SWR), and temperature field distribution in the irradiated area. Reflected power is the microwave power reflected by the rock mass and equipment cavity after microwave emission. It can be combined with rock mineral composition, dielectric constant, and fracture state to determine the rock mass's absorption level of microwave energy and the microwave fracturing effect. The SWR is calculated from the incident power and reflected power and is used to evaluate the matching status between the microwave antenna 5, waveguide 6, and the rock load. Abnormalities in this parameter not only reduce microwave utilization efficiency but also easily cause damage to components such as the microwave generator 8 and microwave antenna 5. The temperature field distribution in the irradiated area is acquired by an infrared thermal imager, which can reflect the temperature values, high-temperature areas, and temperature gradients of the rock within the microwave irradiation range. It directly reflects the microwave energy transfer law and the degree of rock fracture initiation and propagation, providing comprehensive feedback on the microwave system's operating status and the actual fracturing effect on the rock mass.

[0045] Lithological characteristic data includes at least rock surface texture characteristics, mineral composition characteristics, and dielectric response characteristics. Rock surface texture characteristics are morphological information such as rock surface texture, unevenness, and fracture distribution. This information can be collected non-contactly using laser scanning instruments and vision equipment, and is used to initially determine the integrity of the rock mass and its lithological category. Mineral composition characteristics correspond to the composition and proportion of various minerals in the rock, obtained based on the principle of XRD whole-rock analysis. Mineral composition essentially determines the mechanical properties and microwave absorption capacity of the rock. Dielectric response characteristics are characterized by the real and imaginary parts of the dielectric constant. This characteristic can be combined with corresponding formulas to calculate the penetration depth of microwaves in the rock. This characteristic is closely related to the density of the rock and internal fractures. These three types of data are the core basis for the embodied intelligent model to invert the comprehensive characteristics of the rock and deduce the optimal microwave parameters and tunneling parameters.

[0046] Step S202: Input the multi-source sensing data into the pre-trained embodied intelligent model to invert and obtain the rock mechanical properties and microwave response properties of the current working face; In one embodiment, optionally, step S202 specifically includes: Based on the lithological characteristic data, the rock type is identified, and the uniaxial compressive strength, tensile strength, and elastic modulus of the rock are output. Based on the microwave fracturing feedback parameters, the real and imaginary parts of the rock's dielectric constant are inverted, and the microwave penetration depth under the current working conditions is calculated.

[0047] like Figure 3 As shown, in one embodiment, optionally, the training process of the pre-trained embodied intelligence model includes: Step S301: Prepare rock samples of various specifications, conduct indoor tests to obtain indoor test data, wherein the indoor test data includes basic rock characteristic data, microwave fracturing characteristic data and mechanical rock breaking feedback data; In one embodiment, optionally, the indoor test specifically includes: Block samples of different sizes were prepared, wherein all block samples were taken from the same parent rock and cut in the same direction; Specifically, the required powdered sample preparation method involves crushing different types of rocks into particles no larger than 1 cm, then grinding them in a ball mill at 600 r / min for 30 minutes. Powder samples with a particle size less than 0.045 mm are separated using a mesh sieve and placed in a drying oven at 110°C for 48 hours. The processing standards for the 200 mm × 200 mm × 200 mm cube sample, the 50 mm diameter and 25 mm height cylindrical sample, and the 600 mm × 600 mm × 500 mm cuboid sample are as follows: allowable error for side length ±0.2 mm, allowable error for cross-sectional unevenness ±0.02 mm, allowable error for perpendicularity of adjacent faces ±0.25°, and allowable error for non-parallelism of opposite end faces ±0.02 mm. To reduce the influence of anisotropy and dispersion, all samples are taken from the same source material and cut in the same direction. For 600mm×600mm×500mm cuboid specimens, opposite faces must be parallel to each other and connected faces must be perpendicular to each other, as observed by the naked eye. After each specimen is prepared, it is photographed from six directions: top, bottom, left, right, front, and back, and a three-dimensional model of the specimen's appearance is obtained using a laser scanner.

[0048] The blocky sample was subjected to X-ray diffraction whole-rock analysis, dielectric constant test, closed multimode resonant cavity test, open microwave irradiation test, uniaxial compression and Brazilian splitting test, disc cutter test and microwave-mechanical combined rock breaking test in sequence.

[0049] The procedure for conducting whole-rock X-ray diffraction analysis is as follows: Before the experiment, turn on the X-ray generator and preheat for at least 30 minutes. Then, use a standard sample (such as silica powder) for angle calibration to ensure a 2θ error < 0.01°. Next, uniformly fill the sample holder groove with the sample, smoothing the surface to avoid absorption effects caused by uneven thickness. Adjust the sample position using a laser positioner to focus the X-ray beam on the sample center (deviation must be < 0.1 mm) and perform scanning. Monitor the diffraction pattern and use software to perform image search to obtain the rock sample composition.

[0050] For the process of obtaining dielectric constant data, please refer to [link / reference]. Figure 4 First, a sample pressing machine was used to prepare a ring-shaped cylinder with an inner diameter of 3.05 mm, an outer diameter of 6.97 mm, and a height of 10 mm. An appropriate amount of dry powder sample was taken and evenly filled into the sample pressing mold. After cleaning the outer shell of the sample pressing mold, it was placed in the center of the sample pressing platform, and the screw was brought into contact with the sample pressing mold. Pressure was applied to the sample pressing mold by raising and lowering the pressure bar at a constant speed. When the set pressure value was reached, the pressure was maintained for 60 seconds and then released. The sample pressing mold was then removed, producing a synthetic sample with an outer ring diameter of 6.97 mm, an inner ring diameter of 3.05 mm, and a thickness not exceeding 10 mm. Three sets of samples were prepared for each lithology to reduce error. Finally, the test fixture containing the synthetic sample was placed on a dielectric property testing device for testing.

[0051] Before testing the dielectric properties of the samples, the dielectric property testing equipment is calibrated. After calibration, the test fixture is placed in the testing equipment, and the combined dielectric constant of air and paraffin is compared with the standard values ​​(the actual standard value of the combined dielectric constant of air is 1.00, and the standard value of the fictitious component is 0.00; the actual standard value of the combined dielectric constant of paraffin is 2.00–2.50, and the standard value of the fictitious component is 0.00) to verify the calibration effect. If the average relative error between the measured combined dielectric constant of air and paraffin and their standard values ​​is within 10%, the testing requirements are met. Then, the test fixture containing the sample is placed in the testing equipment for dielectric constant testing. The dielectric constant can be used to characterize the penetration depth of microwave-irradiated rocks, and the calculation formula is:

[0052] In the formula, The microwave penetration depth is (cm). Let be the real part of the dielectric constant. This is the imaginary part of the dielectric constant. The wavelength is the microwave wavelength (cm).

[0053] When conducting microwave irradiation experiments, refer to the closed irradiation experiment. Figure 5 1 is a multimode resonant cavity, a closed microwave heating device; 2 is a rock sample; 3 is a mullite pad, a heat-insulating material that does not easily absorb microwaves, serving as a pad. In the closed irradiation test, microwaves reflect back and forth within the metal cavity, creating resonance and heating the sample from multiple directions. For the open irradiation test, see [link to relevant documentation]. Figure 6 The test setup consists of: 4 (rock sample), 5 (microwave antenna for transmitting microwaves), 6 (waveguide for transmitting microwaves), 7 (automatic matching regulator to prevent microwave reflection from damaging the equipment), and 8 (microwave generator for producing microwaves). In the open irradiation test, microwaves emitted by the microwave radiator propagate in all directions through the air. Some enter the rock, and some are reflected by the rock. The microwaves attenuate as the propagation distance increases. For the closed microwave irradiation test, a 50mm×50mm×100mm cuboid sample 2 was used, with irradiation power and irradiation time as variables. For the open microwave irradiation test, a 200mm×200mm×200mm cube sample 4 was used, with irradiation power, irradiation time, and irradiation distance as variables. After irradiation, an infrared thermal imager was used to acquire infrared images, which were then analyzed to determine the highest and average temperatures within the irradiation range. A camera was used to photograph the samples to obtain their fracture characteristics. The results of the microwave irradiation test are as follows: Figure 7 (a)- Figure 7 As shown in (c), where Figure 7 (a) is a diagram showing the fracture characteristics of a closed-circuit microwave irradiation. Figure 7(b) is a fracture characteristic diagram of open-air irradiation. Figure 7 (c) is an infrared thermal image.

[0054] When using a press to test the mechanical properties of rocks, the tests include: conducting uniaxial compression tests on 50mm×50mm×100mm rectangular specimens (with and without microwave irradiation) to obtain parameters such as uniaxial compressive strength and Young's modulus; and conducting Brazilian splitting tests on cylindrical specimens with a diameter of 50mm and a height of 25mm. The stress-strain curves are output from the press's computer as follows: Figure 8 As shown in the figure, the stress-strain curves of the sample under different microwave irradiation parameters are displayed.

[0055] When conducting rock mechanical cutting tests using mechanical cutting equipment, and when carrying out microwave-mechanical combined rock breaking tests, please refer to the experimental schematic diagram. Figure 9 The test used a master-slave servo disc cutter (also known as a multi-disc cutter). The main motor drives the main cutter disc to rotate as a whole, so that the five small disc cutters operate alternately and produce a continuous impact effect on the rock. At the same time, when the blade of the servo small disc cutter cuts into the rock, under the action of cutting resistance, the bearing structure inside the small disc cutter allows it to passively rotate around its own central axis.

[0056] The experiment was conducted on a 600mm × 600mm × 500mm rectangular specimen. During the cutting test, a cantilever crane was used to hoist the sample onto the loading platform. The height of the loading platform was adjusted to align the specimen with the center of the cutter head. The loading platform was then horizontally moved to the position corresponding to the selected cutting depth. The rotating platform was adjusted to rotate the disc cutter to the selected cutting angle. Finally, the back plate was installed, and the sample was compressed by loading the upper and left plates. After loading, the disc cutter was moved to the left side of the sample, and the cutting parameters (push speed, rotation speed, and cutting angle, i.e., the angle between the cutter plane and the free surface of the rock) were set, and the cutting test was conducted. An infrared thermal imager was placed on the left side of the disc cutter to detect its temperature rise. A vibration sensor was placed at the bottom of the loading platform to record vibration data during the cutting process. A data acquisition system was used to record information related to equipment performance (thrust, torque, etc.) during rock cutting. After cutting, the surfaces of the cut rock and rock chips were scanned and imaged to record the surface roughness and chip size. When conducting microwave-mechanical combined rock-breaking experiments, microwave irradiation methods include three types: fixed-point irradiation, moving-point irradiation, and irradiation with the microwave port and the rock-breaking disc moving together. The microwave irradiation operation method is the same as the aforementioned open-type microwave experiment.

[0057] A numerical model is established based on the results of indoor experiments. The input parameters of the numerical model are adjusted, and data is generated in batches to build a database.

[0058] Step S302: Based on the indoor test data, calibrate the numerical simulation model, generate multi-condition data in batches through multi-factor orthogonal simulation experiments, and fuse them to obtain a training dataset; After completing the aforementioned indoor and simulation experiments, all data from various sources were consolidated into a raw data table. Samples with obvious anomalies or incomplete records were removed, such as missing values ​​due to sensor malfunctions or noise points exceeding reasonable limits. A small number of missing values ​​were handled using interpolation methods (e.g., linear interpolation) or deletion methods to ensure a complete correspondence between the input and output variables for each sample. Categorical variables (e.g., rock and mineral types) were then uniquely coded or labeled. All continuous input variables were then normalized, using min-max scaling to map the values ​​to the [0,1] interval, or Z-score standardization. Normalization parameters (maximum, minimum, mean, and standard deviation) were saved for future use on new data. The processed data was organized into a table, with each row representing an independent experimental sample (or simulation sample) and each column corresponding to an input or output variable.

[0059] The database can be in CSV, Excel, or HDF5 format for easy reading by machine learning platforms such as Python and MATLAB. A unique identifier (e.g., experiment number) should be added to each sample, and the data source label (indoor experiment, numerical simulation) should be retained for subsequent analysis of distribution differences across different data sources. The database should be randomly divided into training and test sets in an 8:2 or 7:3 ratio. The training set is used for model training and parameter tuning, while the test set is used for final model performance evaluation.

[0060] Step S303: Using machine learning algorithms, an initial prediction model is trained based on the training dataset. The initial prediction model includes a three-dimensional geological perception model, a microwave fracturing effect prediction model, and a tunneling machine operation efficiency prediction model. In one embodiment, optionally, step S303 specifically includes: Multiple candidate prediction models were trained using various machine learning algorithms, including logistic regression, support vector machine, decision tree, k-nearest neighbors, random forest, radial basis function neural network, and XGBoost. The performance of the multiple candidate prediction models was evaluated using accuracy, precision, recall, F1-score, and PR curves to obtain the evaluation results. The candidate prediction model with the best evaluation results is selected as the three-dimensional geological sensing model, microwave fracturing effect prediction model, and tunneling machine operation efficiency prediction model. The Stacking ensemble learning method is employed, using the multiple candidate prediction models as primary learners and linear regression or logistic regression models as secondary learners to construct an ensemble prediction model.

[0061] When building the model, different machine learning models are used for prediction, and then the performance of each model is evaluated. The best-performing model is selected as the prediction model. The principles of each method are as follows: Logistic Regression: Despite its name, it is a classic linear classification model. It maps the predicted values ​​of linear regression to a range of 0 and 1 using the sigmoid function, thus outputting the probability that a sample belongs to a particular class. Due to its simple form, fast computation, and strong interpretability, it is often used as the baseline model for handling binary classification problems in industry.

[0062] Support Vector Machine (SVM): Its core idea is to find an optimal hyperplane in the feature space that maximizes the geometric margin between samples of different classes and that hyperplane. For linearly inseparable data, SVM uses the kernel trick to map low-dimensional features to a high-dimensional space to make them linearly separable, exhibiting strong generalization ability on small to medium-sized datasets.

[0063] Decision Tree: A nonparametric supervised learning method for decision-making based on a tree structure. It starts from the root node and partitions the data according to a metric (such as information gain or Gini coefficient) to form internal and leaf nodes. The model is intuitive and easy to understand, fully conforming to human IF-THEN thinking logic, and requires no strict pre-assumptions about the data.

[0064] K-Nearest Neighbors (KNN): An instance-based "lazy learning" algorithm. It does not have an explicit model training process. When a new sample is input, the algorithm finds the K nearest neighbors in the training set and determines the label of the new sample through majority voting (classification) or average (regression). Its core elements lie in the choice of distance metric and the value of K.

[0065] Random Forest: An ensemble learning algorithm based on the Bagging strategy. It constructs multiple independent decision trees through bootstrapping resampling and randomly selects a subset of features when training the nodes of each tree. The final prediction result is determined by voting or averaging of all decision trees. This method effectively reduces the risk of overfitting and has good robustness to outliers and noise.

[0066] RF Neural Network (Radial Function Network): A three-layer feedforward neural network consisting of an input layer, hidden layers, and an output layer. Its unique feature is that the hidden layer nodes use radial basis functions (such as Gaussian functions) as activation functions, locally mapping the input space to a higher dimension. It possesses global approximation capabilities and has a relatively fast training speed.

[0067] XGBoost (Extreme Gradient Boosting): An efficient gradient boosting decision tree implementation based on the boosting strategy. Building upon traditional GBDT, it performs a second-order Taylor expansion of the objective function and introduces L1 and L2 regularization terms to control model complexity. It boasts high prediction accuracy due to its support for parallel computation, built-in missing value handling, and support for parallel computation.

[0068] Stacking: A hierarchical ensemble learning framework. It first uses multiple different primary learners to predict the original data, then uses the outputs of these primary learners as new features to feed into secondary learners (usually simple linear models) for final training and prediction. By combining the advantages of heterogeneous models, it can often further improve prediction performance.

[0069] Because rocks exhibit significant dispersion and data distribution is uneven, multiple metrics are used to evaluate model performance to ensure universality and prevent overfitting. These include accuracy, precision, recall, F1-score, and PR curve. Considering the operational requirements of tunneling machines, recall (ensuring no high-strength rocks are missed) should be prioritized, followed by accuracy.

[0070] Step S304: Collect the multi-source sensing data at the mine excavation site, iteratively calibrate the initial prediction model, and obtain the pre-trained embodied intelligent model.

[0071] After the model is built, a microwave-mechanical integrated continuous mining tunneling machine is used to go to the mine site for tunneling. The model is then iteratively optimized using on-site data to form an embodied intelligent system.

[0072] Step S203: Based on the rock mechanical properties and the microwave response properties, calculate and output the optimal combination of working parameters, which includes microwave irradiation parameters and cutterhead excavation parameters; In one embodiment, optionally, calculating and outputting the optimal combination of operating parameters based on the rock mechanical properties and the microwave response characteristics includes: Based on the rock mechanical properties and microwave response characteristics, and under the premise of satisfying the tool safety load constraints and microwave equipment safety constraints, the optimal combination of working parameters is matched and output with the minimum specific energy consumption as the objective function.

[0073] This step uses the rock mechanics properties and microwave response characteristics obtained from model inversion as input conditions to optimize parameters. Tool safety load constraints mean that the operating parameters of the cutterhead, such as thrust, torque, and cutting force, are limited within the rated load range of the equipment and the allowable force range of the cutterhead, preventing overload, breakage, and excessive wear of the cutterhead. Microwave equipment safety constraints mean that parameters such as microwave irradiation power, standing wave ratio, and reflected power are controlled within the rated operating range of devices such as the microwave generator 8, waveguide 6, and microwave antenna 5, preventing microwave equipment failure due to load mismatch or abnormal power. Specific energy consumption is the total energy consumed per unit excavation depth or per unit of broken rock mass during rock breaking operations, and is a core indicator for measuring operational economy and energy efficiency. This step uses safe equipment operation as a prerequisite boundary condition, and uses minimum specific energy consumption as the optimization objective to solve for the parameter combination, ultimately outputting a complete set of optimal operating parameters including microwave irradiation parameters and cutterhead excavation parameters. In this way, while ensuring the long-term safe and stable operation of the cutting disc and microwave equipment, the working parameters are optimized globally, minimizing the overall energy consumption of rock breaking operations, and taking into account both equipment lifespan and construction economy. This solves the problem that traditional manual parameter adjustment cannot balance safety and energy efficiency.

[0074] Step S204: The optimal combination of working parameters is sent to the microwave multi-disc cutter to perform real-time control of the microwave multi-disc cutter's rock breaking operation.

[0075] This step pertains to the instruction execution phase. The system takes the optimal combination of working parameters calculated in the previous phase and sends it through the control link to the microwave transmitting unit and the cutterhead tunneling execution unit of the microwave multi-disc cutterhead system. This directly changes the operating conditions of the equipment, enabling the tunneling equipment to perform a combined microwave fracturing and mechanical cutting operation according to the optimized parameters, thus completing closed-loop real-time control. This achieves seamless integration from intelligent decision-making to on-site execution, completing fully automated control, completely replacing manual on-site parameter adjustment, reducing human intervention, and improving the timeliness and accuracy of parameter execution.

[0076] In one embodiment, optionally, the method further includes: Real-time monitoring of the tunneling machine thrust, torque, and rock surface temperature field change rate in the multi-source sensing data; calculation of the specific energy consumption change per unit advance distance. This step continuously collects two types of mechanical parameters of the tunneling machine: thrust and torque, using on-site sensors and infrared detection equipment. Simultaneously, it collects the rock surface temperature field and calculates its rate of change. Combined with equipment energy consumption data and tunneling progress, it calculates the specific energy consumption per unit advance distance in real time and tracks its dynamic trend. Thrust and torque directly reflect the rock cutting resistance, the rate of change of the rock surface temperature field characterizes the fluctuation of the microwave fracturing effect, and the change in specific energy consumption comprehensively reflects the stability of the overall operating conditions. These three types of monitoring data complement each other, achieving comprehensive dynamic perception of the working face condition.

[0077] When the change in thrust, torque or specific energy consumption exceeds the corresponding preset mutation threshold within a preset time window, it is determined that a sudden change in rock strength has occurred at the current working face. In this step, the system pre-sets corresponding abrupt change thresholds and statistical time windows for thrust, torque, and specific energy consumption based on different lithologies and operating conditions. If any monitored data exceeds the corresponding threshold within the specified time range, it is determined that a sudden change in rock strength has occurred at the tunneling face, such as encountering hard rock layers, fractured rock layers, or interbedded rock masses. This judgment logic is based on quantitative determination using measured data, which differs from traditional subjective manual judgment.

[0078] In response to the determination that a sudden change in rock strength has occurred at the current working face, the pre-trained embodied intelligent model is recalled within a preset time, and the updated rock mechanical properties and microwave response properties are retrieved based on the multi-source sensing data at the moment of the sudden change. Once a sudden change in rock strength is detected, the system will proactively restart the embodied intelligent model within a set response time, extract the complete set of multi-source sensing data at the moment of the change as input, recalculate the latest rock mechanical properties and microwave response properties of the current rock mass, and complete the dynamic update of rock mass characteristic data to adapt to the changed geological conditions.

[0079] Based on the updated rock mechanical properties and microwave response characteristics, the optimal combination of working parameters is recalculated and output, and the updated optimal working parameters are sent to the microwave multi-disc cutter to adaptively adjust the rock breaking operation.

[0080] Based on the updated rock mass characteristics, the system recalculates the optimal combination of working parameters, with equipment safety as a constraint and minimum specific energy consumption as the objective. The new parameters are immediately sent to the actuators, simultaneously adjusting various operating parameters for microwave irradiation and cutterhead excavation, thus achieving adaptive adjustment of the rock-breaking operation. In this way, based on the embodied intelligent system and combined with the tunneling machine's sensor data, an intelligent closed loop of "real-time monitoring → characteristic identification → parameter optimization" is realized. The system's functions include forward prediction of rock-breaking effects and backward deduction of equipment parameters, enabling continuous automatic parameter optimization.

[0081] like Figure 10 As shown, in a second aspect, embodiments of this application provide a microwave multi-disc rock-breaking tool intelligent control device 100, the device comprising: The acquisition module 1001 is used to acquire multi-source sensing data of the microwave multi-disc cutterhead tunneling face in real time. The multi-source sensing data includes tunneling machine operating parameters, microwave fracturing feedback parameters, and lithological characteristic data. The inversion module 1002 is used to input the multi-source sensing data into a pre-trained embodied intelligent model to invert the rock mechanical properties and microwave response properties of the current working face. Output module 1003 is used to calculate and output the optimal combination of working parameters based on the rock mechanical properties and the microwave response characteristics. The optimal combination of working parameters includes microwave irradiation parameters and cutterhead excavation parameters. The control module 1004 is used to send the optimal combination of working parameters to the microwave multi-disc cutter to perform real-time control of the microwave multi-disc cutter rock breaking operation.

[0082] In one embodiment, optionally, the training process of the pre-trained embodied intelligence model includes: Rock samples of various specifications were prepared, and indoor tests were conducted to obtain indoor test data. The indoor test data included basic rock characteristic data, microwave fracturing characteristic data, and mechanical rock breaking feedback data. Based on the indoor test data, the numerical simulation model is calibrated, and multi-condition data is generated in batches through multi-factor orthogonal simulation experiments and fused to obtain a training dataset. An initial prediction model was trained based on the training dataset using machine learning algorithms. The initial prediction model included a three-dimensional geological perception model, a microwave fracturing effect prediction model, and a tunneling machine operation efficiency prediction model. The multi-source sensing data is collected at the mine excavation site, and the initial prediction model is iteratively calibrated to obtain the pre-trained embodied intelligent model.

[0083] In one embodiment, optionally, the tunneling machine operating parameters include at least thrust, torque, vibration acceleration, and cutting rate; Microwave-induced fracturing feedback parameters include at least reflected power, standing wave ratio, and temperature field distribution in the irradiated area; The lithological characteristic data include at least rock surface texture characteristics, mineral composition characteristics, and dielectric response characteristics.

[0084] In one embodiment, optionally, the inversion module specifically includes: The output unit is used to identify rock types based on the lithological characteristic data and output the uniaxial compressive strength, tensile strength and elastic modulus of the rock. The calculation unit is used to invert the real and imaginary parts of the dielectric constant of the rock based on the microwave fracturing feedback parameters, and to calculate the microwave penetration depth under the current working conditions.

[0085] In one embodiment, optionally, the output module is used for: Based on the rock mechanical properties and microwave response characteristics, and under the premise of satisfying the tool safety load constraints and microwave equipment safety constraints, the optimal combination of working parameters is matched and output with the minimum specific energy consumption as the objective function.

[0086] In one embodiment, optionally, the apparatus further includes: The monitoring module is used to monitor the tunneling machine thrust, torque and rock surface temperature field change rate in the multi-source sensing data in real time, and calculate the change in specific energy consumption per unit advance distance; The determination module is used to determine that a rock strength mutation has occurred at the current working face when the change in thrust, torque or specific energy consumption exceeds the corresponding preset mutation threshold within a preset time window. The update module is used to respond to the determination that a sudden change in rock strength has occurred at the current working face, and to re-call the pre-trained embodied intelligent model within a preset time to invert the updated rock mechanical properties and microwave response properties based on the multi-source sensing data at the moment of the sudden change. The adjustment module is used to recalculate and output the optimal combination of working parameters based on the updated rock mechanical properties and microwave response characteristics, and send the updated optimal working parameters to the microwave multi-disc cutter to adaptively adjust the rock breaking operation.

[0087] In one embodiment, optionally, the indoor test specifically includes: Block samples of different sizes were prepared, wherein all block samples were taken from the same parent rock and cut in the same direction; The blocky sample was subjected to X-ray diffraction whole-rock analysis, dielectric constant test, closed multimode resonant cavity test, open microwave irradiation test, uniaxial compression and Brazilian splitting test, disc cutter test and microwave-mechanical combined rock breaking test in sequence.

[0088] In one embodiment, optionally, training an initial prediction model based on the training dataset using a machine learning algorithm specifically includes: Multiple candidate prediction models were trained using various machine learning algorithms, including logistic regression, support vector machine, decision tree, k-nearest neighbors, random forest, radial basis function neural network, and XGBoost. Through accuracy, precision, recall, and F1 The performance of the multiple candidate prediction models is evaluated using scores and PR curves to obtain evaluation results. The candidate prediction model with the best evaluation results is selected as the three-dimensional geological sensing model, microwave fracturing effect prediction model, and tunneling machine operation efficiency prediction model. The Stacking ensemble learning method is employed, using the multiple candidate prediction models as primary learners and linear regression or logistic regression models as secondary learners to construct an ensemble prediction model.

[0089] Thirdly, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the above-described intelligent control method for microwave multi-disc rock-breaking device.

[0090] Fourthly, a computer-readable storage medium is provided, which stores a computer program that, when executed by a processor, implements the steps of the above-described intelligent control method for microwave multi-disc rock-breaking devices.

[0091] Specific limitations regarding the intelligent control device for microwave multi-disc rock-breaking cutter bodies can be found in the limitations of the intelligent control method for microwave multi-disc rock-breaking cutter bodies described above, and will not be repeated here. Each module in the aforementioned intelligent control device for microwave multi-disc rock-breaking cutter bodies can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the corresponding operations of each module.

[0092] The software tools or components not belonging to our company that appear in the embodiments of this application are merely examples and do not represent actual use.

[0093] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method described in any one of the first aspect embodiments.

[0094] It should be noted that the functions or steps that can be implemented by the computer-readable storage medium or electronic device described above can be referred to the relevant descriptions in the foregoing method embodiments. To avoid repetition, they will not be described one by one here.

[0095] It should be understood that the term "and / or" used in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Additionally, the character " / " in this article generally indicates that the preceding and following related objects have an "or" relationship.

[0096] It should be understood that although the terms "first," "second," etc., may be used to describe the setting units in the embodiments of this application, these setting units should not be limited to these terms. These terms are only used to distinguish the setting units from each other. For example, without departing from the scope of the embodiments of this application, the first setting unit may also be referred to as the second setting unit, and similarly, the second setting unit may also be referred to as the first setting unit.

[0097] Depending on the context, the word "if" as used here can be interpreted as "when," "when," "in response to determination," or "in response to detection." Similarly, depending on the context, the phrase "if determination" or "if detection (of the stated condition or event)" can be interpreted as "when determination," "in response to determination," "when detection (of the stated condition or event)," or "in response to detection (of the stated condition or event)."

[0098] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0099] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in a combination of hardware and software functional units.

[0100] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0101] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A method for intelligent control of a microwave multi-disc rock-breaking tool, characterized in that, The method includes: Real-time acquisition of multi-source sensing data from the microwave multi-disc cutterhead tunneling face, including tunneling machine operating parameters, microwave fracturing feedback parameters, and lithological characteristic data; The multi-source sensing data is input into a pre-trained embodied intelligent model to invert the rock mechanical properties and microwave response properties of the current working face. Based on the rock mechanical properties and the microwave response properties, the optimal combination of working parameters is calculated and output, which includes microwave irradiation parameters and cutterhead tunneling parameters. The optimal combination of working parameters is sent to the microwave multi-disc cutter to control the rock breaking operation of the microwave multi-disc cutter in real time. The multi-source sensing data is input into a pre-trained embodied intelligent model to invert and obtain the rock mechanical properties and microwave response properties of the current working face, specifically including: Based on the lithological characteristic data, the rock type is identified, and the uniaxial compressive strength, tensile strength, and elastic modulus of the rock are output. Based on the microwave fracturing feedback parameters, the real and imaginary parts of the rock's dielectric constant are inverted, and the microwave penetration depth under the current working conditions is calculated. Real-time monitoring of the tunneling machine thrust, torque, and rock surface temperature field change rate in the multi-source sensing data; calculation of the specific energy consumption change per unit advance distance. When the change in thrust, torque or specific energy consumption exceeds the corresponding preset mutation threshold within a preset time window, it is determined that a sudden change in rock strength has occurred at the current working face. In response to the determination that a sudden change in rock strength has occurred at the current working face, the pre-trained embodied intelligent model is recalled within a preset time, and the updated rock mechanical properties and microwave response properties are retrieved based on the multi-source sensing data at the moment of the sudden change. Based on the updated rock mechanical properties and microwave response characteristics, the optimal combination of working parameters is recalculated and output, and the updated optimal working parameters are sent to the microwave multi-disc cutter to adaptively adjust the rock breaking operation.

2. The method according to claim 1, characterized in that, The training process of the pre-trained embodied intelligence model includes: Rock samples of various specifications were prepared, and indoor tests were conducted to obtain indoor test data. The indoor test data included basic rock characteristic data, microwave fracturing characteristic data, and mechanical rock breaking feedback data. Based on the indoor test data, the numerical simulation model is calibrated, and multi-condition data is generated in batches through multi-factor orthogonal simulation experiments and fused to obtain a training dataset. An initial prediction model was trained based on the training dataset using machine learning algorithms. The initial prediction model included a three-dimensional geological perception model, a microwave fracturing effect prediction model, and a tunneling machine operation efficiency prediction model. The multi-source sensing data is collected at the mine excavation site, and the initial prediction model is iteratively calibrated to obtain the pre-trained embodied intelligent model.

3. The method according to claim 1, characterized in that, The operating parameters of the tunneling machine include at least thrust, torque, vibration acceleration, and cutting rate; Microwave-induced fracturing feedback parameters include at least reflected power, standing wave ratio, and temperature field distribution in the irradiated area; The lithological characteristic data include at least rock surface texture characteristics, mineral composition characteristics, and dielectric response characteristics.

4. The method according to claim 1, characterized in that, The calculation and output of the optimal combination of operating parameters based on the rock mechanical properties and the microwave response characteristics includes: Based on the rock mechanical properties and microwave response characteristics, and under the premise of satisfying the tool safety load constraints and microwave equipment safety constraints, the optimal combination of working parameters is matched and output with the minimum specific energy consumption as the objective function.

5. The method according to claim 2, characterized in that, The indoor tests specifically include: Block samples of different sizes were prepared, wherein all block samples were taken from the same parent rock and cut in the same direction; The blocky sample was subjected to X-ray diffraction whole-rock analysis, dielectric constant test, closed multimode resonant cavity test, open microwave irradiation test, uniaxial compression and Brazilian splitting test, disc cutter test and microwave-mechanical combined rock breaking test in sequence.

6. A microwave multi-disc cutter rock-breaking tool intelligent control device, used in the microwave multi-disc cutter rock-breaking tool intelligent control method according to any one of claims 1 to 5, characterized in that, The device includes: The acquisition module is used to acquire multi-source sensing data of the microwave multi-disc cutterhead tunneling face in real time. The multi-source sensing data includes tunneling machine operating parameters, microwave fracturing feedback parameters, and lithological characteristic data. The inversion module is used to input the multi-source sensing data into a pre-trained embodied intelligent model to invert the rock mechanical properties and microwave response properties of the current working face. The output module is used to calculate and output the optimal combination of working parameters based on the rock mechanical properties and the microwave response characteristics. The optimal combination of working parameters includes microwave irradiation parameters and cutterhead excavation parameters. The control module is used to send the optimal combination of working parameters to the microwave multi-disc cutter to control the rock breaking operation of the microwave multi-disc cutter in real time. The inversion module specifically includes: The output unit is used to identify rock types based on the lithological characteristic data and output the uniaxial compressive strength, tensile strength and elastic modulus of the rock. The calculation unit is used to invert the real and imaginary parts of the dielectric constant of the rock based on the microwave fracturing feedback parameters, and to calculate the microwave penetration depth under the current working conditions. The monitoring module is used to monitor the tunneling machine thrust, torque and rock surface temperature field change rate in the multi-source sensing data in real time, and calculate the change in specific energy consumption per unit advance distance; The determination module is used to determine that a rock strength mutation has occurred at the current working face when the change in thrust, torque or specific energy consumption exceeds the corresponding preset mutation threshold within a preset time window. The update module is used to respond to the determination that a sudden change in rock strength has occurred at the current working face, and to re-call the pre-trained embodied intelligent model within a preset time to invert the updated rock mechanical properties and microwave response properties based on the multi-source sensing data at the moment of the sudden change. The adjustment module is used to recalculate and output the optimal combination of working parameters based on the updated rock mechanical properties and microwave response characteristics, and send the updated optimal working parameters to the microwave multi-disc cutter to adaptively adjust the rock breaking operation.

7. A computer device, characterized in that, include: At least one processor; And, a memory communicatively connected to the at least one processor; The memory stores instructions that can be executed by the at least one processor; The instructions are configured to perform the method described in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, The device stores computer-executable instructions for performing the method as described in any one of claims 1 to 5.

Images