Intelligent adjusting method, system and equipment for cutterhead of tunnel boring machine

Abstract

The invention discloses an intelligent adjusting method, system and equipment for a cutterhead of a tunnel boring machine, and relates to the technical field related to tunnel boring machines, the method comprises the following steps: obtaining characteristic data of a tunneling soil layer, and performing deep detection through a multispectral sensor; carrying out tunneling soil layer derivation to construct a tunneling soil layer feature space; establishing a dynamic safety boundary constraint set which comprises a heading machine hard limit boundary, a geological sudden change boundary, a settlement control boundary and a cutter service life boundary; cutterhead parameter optimization is carried out by taking specific energy minimization, tool wear minimization and settlement risk minimization as multiple targets, and cutterhead execution parameters are output for cutterhead adjustment. The technical problems that in the prior art, dynamic sensing of geological conditions is insufficient, dynamic safety constraint adaptability is poor, multi-target optimization conflicts are difficult to balance, and consequently the tunneling efficiency and stability are low, the service life of a cutter is low and the construction safety is low are solved, and the technical effects of improving the tunneling efficiency, stability and construction safety and prolonging the service life of the cutter are achieved.

Description

Technical Field

[0001] This application relates to the technical field of tunnel boring machines, specifically to intelligent cutterhead adjustment methods, systems, and equipment for tunnel boring machines. Background Technology

[0002] Tunneling projects face increasingly complex geological conditions and higher construction requirements. As the core equipment of modern tunnel construction, the intelligent adjustment of the cutterhead of the tunnel boring machine (TBM) is crucial for improving tunneling efficiency, reducing construction risks, and extending cutter life. However, existing methods still have shortcomings in areas such as deep geological feature perception, dynamic safety boundary modeling, and multi-objective collaborative optimization. For example, different soil layers have significant differences in water content, cohesion, abrasiveness, and lithology, and there may be unexplored weak interlayers or abrupt hard rock zones. Traditional TBM cutterhead adjustment is difficult to adapt to complex and variable geological conditions, directly affecting the cutterhead stress state, tunneling stability, and tunneling efficiency. Furthermore, existing cutterhead control methods struggle to balance multiple objectives such as energy efficiency, cost, and safety.

[0003] Therefore, current technologies suffer from several technical problems, including insufficient dynamic perception of geological conditions, poor adaptability to dynamic safety constraints, and difficulty in balancing multi-objective optimization conflicts, leading to low tunneling efficiency and stability, low tool life, and low construction safety. Summary of the Invention

[0004] This application provides a method, system, and equipment for intelligent adjustment of the cutterhead of a tunnel boring machine, which solves the technical problems in the prior art, such as insufficient dynamic perception of geological conditions, poor adaptability to dynamic safety constraints, and difficulty in balancing multi-objective optimization conflicts, resulting in low tunneling efficiency and stability, cutter life and construction safety. It achieves the technical effect of improving tunneling efficiency, stability, construction safety and extending cutter life.

[0005] This application provides an intelligent cutterhead adjustment method for tunnel boring machines (TBMs). The method includes: acquiring characteristic data of the tunneling soil layer, including water content, cohesion, abrasive index, and lithological distribution, and performing deep detection using a multispectral sensor to obtain the characteristic variation distribution; deducing the tunneling soil layer based on the characteristic data and characteristic variation distribution to construct a tunneling soil layer characteristic space; establishing a dynamic safety boundary constraint set, including the TBM hard limit boundary, geological abrupt change boundary, settlement control boundary, and cutter life boundary; optimizing the cutterhead parameters in the tunneling soil layer characteristic space with multiple objectives of minimizing specific energy, minimizing cutter wear, and minimizing settlement risk, based on the cutterhead adjustment constraints, and using the safety boundary constraint set to trim the optimized solution set, outputting cutterhead execution parameters for cutterhead adjustment.

[0006] In a possible implementation, the intelligent cutterhead adjustment method for a tunnel boring machine acquires characteristic data of the excavated soil layer, including water content, cohesion, abrasion index, and lithological distribution, and performs deep detection using a multispectral sensor to obtain the characteristic variation distribution. This includes: analyzing the soil characteristics of the excavated surface soil layer, including the water content, cohesion, abrasion index, and lithological distribution, to obtain standard soil layer data; performing multispectral signal feature conversion on the standard soil layer data to set an initial standard spectrum; and fitting and comparing the multispectral signals collected from the deep soil layer with the initial standard spectrum to obtain the signal variation gradient of each soil layer, thereby obtaining the characteristic variation distribution.

[0007] In a possible implementation, the intelligent cutterhead adjustment method for a tunnel boring machine deduces the characteristics of the excavated soil layers based on the characteristic data and characteristic variation distribution of the excavated soil layers, and constructs a characteristic space for the excavated soil layers. This includes: establishing a mapping transformation relationship between the initial standard spectrum and the soil characteristics; using the characteristic variation distribution as a mapping input based on the signal variation gradient of each soil layer, and performing soil characteristic mapping output through the mapping transformation relationship to obtain the soil distribution characteristics of each soil layer; and spatially embedding and fusing the soil distribution characteristics of each soil layer according to spatial coordinates to construct the characteristic space for the excavated soil layers.

[0008] In a possible implementation, the intelligent cutterhead adjustment method for a tunnel boring machine establishes a dynamic safety boundary constraint set, including: setting the hard limit boundary of the tunnel boring machine based on its equipment hard limit, rated parameters, and tolerance thresholds; establishing a constraint relationship between the tunneling fragmentation index and the advance speed, and setting the geological abrupt change boundary; determining the settlement control boundary according to the constraint thresholds of the formation loss rate and soil pressure fluctuation; and determining the cutter life boundary based on the cutter distribution structure and cutter performance of the cutterhead, combined with the tunneling constraints of the tunneling soil layer.

[0009] In a possible implementation, the intelligent cutterhead adjustment method for a tunnel boring machine determines the cutter life boundary based on the cutter distribution structure and performance of the cutterhead, combined with the tunneling constraints of the excavated soil layer. This includes: constructing a dynamic model describing the mapping relationship between cutter wear rate and tunneling parameters based on cutter material performance test data and historical tunneling data, where the tunneling parameters include local thrust, cutterhead rotation speed, and soil abrasiveness index; dividing the cutterhead radially into N fan-shaped pressure zones according to the cutter distribution structure, where N is a positive integer ≥ 4; calculating the cutter wear rate of each pressure zone based on the dynamic model and current tunneling parameters, and generating a wear spatial distribution heatmap; analyzing the critical working conditions of cutter performance based on the characteristic parameters of the excavated soil layer, including maximum allowable contact stress, abrasiveness index threshold, and temperature limit; and determining the minimum remaining cutter life value of the entire cutterhead through the shortest board effect, using the mechanical load deviation rate, wear spatial distribution heatmap, and tunneling constraints as inputs, and setting the corresponding tunneling parameter threshold as the cutter life boundary.

[0010] In a possible implementation, the intelligent cutterhead adjustment method for a tunnel boring machine optimizes cutterhead parameters within the tunneling soil layer feature space, with multiple objectives including minimizing specific energy, minimizing tool wear, and minimizing settlement risk. This includes: establishing an evaluation function relationship between soil layer characteristics and cutterhead parameters, with the multiple objectives of minimizing specific energy, minimizing tool wear, and minimizing settlement risk; performing a multi-directional search for cutterhead parameters within the cutterhead adjustment constraint range based on the soil layer characteristics in the tunneling soil layer feature space; evaluating the search parameters using the multi-objective evaluation function; and finding a multi-optimized solution set that meets the objective evaluation results as the output.

[0011] In a possible implementation, the intelligent cutterhead adjustment method for a tunnel boring machine further includes: establishing a three-dimensional feature tensor space based on the tunneling axis, wherein the X-axis represents the tunneling mileage, the y-axis represents the radial depth, and the z-axis represents the feature parameter value; mapping the soil feature vectors of each soil layer to the grid nodes of the three-dimensional feature tensor space according to the spatial distribution coordinates; and, based on the three-dimensional feature tensor space, fitting the searched cutterhead parameters to the tunneling mileage and radial depth to construct complete time-series cutterhead parameters and generate the multi-optimization solution set.

[0012] In a possible implementation, the intelligent adjustment method for the cutterhead of a tunnel boring machine maps the soil feature vectors of each soil layer to grid nodes in the three-dimensional feature tensor space according to the spatial distribution coordinates. This includes: establishing the number of grids in the X and Y axes of the three-dimensional feature tensor space, where the number of grids on the X axis corresponds to the tunneling process and the number of grids on the Y axis corresponds to the radial depth; calculating the Euclidean distance from the grid node to the soil layer detection point, calculating the feature value according to the inverse distance weight, and assigning the feature value to the Z-axis feature channel corresponding to the grid node.

[0013] This application also provides an intelligent cutterhead adjustment system for a tunnel boring machine (TBM). The system implements the intelligent cutterhead adjustment method for a TBM as described in this application. The system includes: a deep-layer detection module for acquiring characteristic data of the excavated soil layer, including water content, cohesion, abrasive index, and lithological distribution, and obtaining characteristic variation distributions through a multispectral sensor; an excavated soil layer derivation module for deducing the excavated soil layer based on the characteristic data and characteristic variation distributions, constructing an excavated soil layer characteristic space; a safety boundary constraint configuration module for establishing a dynamic safety boundary constraint set, including the TBM hard limit boundary, geological abrupt change boundary, settlement control boundary, and tool life boundary; and a cutterhead parameter optimization module for optimizing cutterhead parameters within the excavated soil layer characteristic space, with multiple objectives including minimizing specific energy, minimizing tool wear, and minimizing settlement risk, based on cutterhead adjustment constraints. The optimized solution set obtained through the optimization is then pruned using the safety boundary constraint set, and cutterhead execution parameters are output for cutterhead adjustment.

[0014] This application also provides an electronic device, including: a memory for storing executable instructions; and a processor for implementing an intelligent cutterhead adjustment method for a tunnel boring machine when executing the executable instructions stored in the memory.

[0015] This application proposes a method, system, and equipment for intelligent cutterhead adjustment of tunnel boring machines (TBMs). The method acquires characteristic data of the tunneling soil layer and performs deep-layer detection using a multispectral sensor. It then constructs a characteristic space of the tunneling soil layer through layer deduction; establishes a dynamic safety boundary constraint set, including the TBM's hard limit boundary, geological abrupt change boundary, settlement control boundary, and cutter life boundary; and optimizes the cutterhead parameters with multiple objectives: minimizing specific energy, minimizing cutter wear, and minimizing settlement risk. The optimized solution set is then pruned, and the output cutterhead execution parameters are used for cutterhead adjustment. This addresses the technical problems in existing technologies, such as insufficient dynamic perception of geological conditions, poor adaptability of dynamic safety constraints, and difficulty in balancing multi-objective optimization conflicts, leading to low tunneling efficiency and stability, low cutter life, and low construction safety. The proposed method achieves the technical effects of improving tunneling efficiency, stability, construction safety, and extending cutter life. Attached Figure Description

[0016] To more clearly illustrate the technical solutions of the embodiments of this disclosure, the accompanying drawings of the embodiments of this disclosure will be briefly described below. Flowcharts are used in this application to illustrate the operations performed by the system according to the embodiments of this application. It should be understood that the preceding or following operations are not necessarily performed precisely in sequence. Instead, various steps can be processed in reverse order or simultaneously as needed. Furthermore, other operations can be added to these processes, or one or more steps can be removed from these processes.

[0017] Figure 1 This is a schematic flowchart of an intelligent cutterhead adjustment method for a tunnel boring machine provided in an embodiment of this application.

[0018] Figure 2 This is a schematic diagram of the intelligent cutterhead adjustment system for a tunnel boring machine provided in an embodiment of this application.

[0019] Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application.

[0020] Explanation of reference numerals in the attached drawings: Deep exploration module 10, tunneling soil layer deduction module 20, safety boundary constraint configuration module 30, cutterhead parameter optimization module 40, input device 401, processor 402, memory 403, output device 404. Detailed Implementation

[0021] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, specific embodiments of this application are given below.

[0022] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description of this application will be provided in conjunction with the accompanying drawings. The described embodiments should not be considered as limitations on this application. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0023] In the following description, references to "some embodiments" describe a subset of all possible embodiments. However, it is understood that "some embodiments" can be the same or different subsets of all possible embodiments and can be combined with each other without conflict. The terms "first" and "second" are used merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules not explicitly listed or inherent to these processes, methods, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only.

[0024] This application provides an embodiment of a method for intelligent cutterhead adjustment in tunnel boring machines, such as... Figure 1 As shown, the method includes: Step S100: Obtain characteristic data of the tunneling soil layer, including water content, cohesion, abrasive index and lithological distribution, and conduct deep exploration using a multispectral sensor to obtain the characteristic variation distribution.

[0025] Preferably, during tunnel excavation, the cutting performance, excavation efficiency, and equipment safety of the cutterhead are highly dependent on real-time perception of geological conditions. This involves dynamically acquiring key characteristic data of the excavated soil layers, including moisture content, cohesion, abrasion index, and lithological distribution. Further, deep exploration is used to construct characteristic variation distributions to achieve accurate geological modeling and intelligent cutterhead adjustment. Specifically, the excavated soil layer characteristic data includes moisture content, cohesion, abrasion index, and lithological distribution. Moisture content refers to the mass percentage of water in the soil layer, reflecting the permeability and stability of the strata. High moisture content may lead to cutterhead slippage and mud cake adhesion, while low moisture content... Moisture content can increase cutting resistance; cohesion refers to the cohesive strength between soil particles, affecting the soil's shear resistance. Strata with high cohesion require greater cutterhead thrust, while loose sand layers are prone to tool idling or settling; abrasiveness index is an indicator of the degree of tool wear caused by soil and rock, usually related to quartz content and particle hardness. Highly abrasive strata such as sandstone and granite will accelerate tool wear, requiring optimization of rotation speed and thrust to extend tool life; lithological distribution refers to the type, thickness, and spatial distribution of different rocks or soil layers on the tunneling section. Abrupt changes in lithology can easily lead to cutterhead vibration or uneven loading, requiring dynamic adjustment of parameters to maintain stable tunneling.

[0026] Preferably, gridded measuring points are arranged in the area ahead of the tunneling. Soil moisture content distribution is inverted through spectral reflectance differences, establishing a quantitative relationship model between moisture content and reflectance intensity with an accuracy of ±1.5%, thereby estimating the soil moisture content. Near-infrared band (1400-1900nm) is used to extract spectral features of soil particle adsorption characteristics. Combined with mechanical testing calibration using borehole core sampling, non-contact cohesion estimation is achieved with an error controlled within 15kPa, thus measuring and obtaining cohesion. Heavy metal ions such as Fe³⁺ and SO₄²⁻ are used in a specific wavelength band of 45 nm. Characteristic absorption peaks in the 0-550nm and 2100-2300nm ranges were identified. Spectral differentiation techniques were used to enhance weak signals, and a database of the correspondence between corrosive substance content and spectral characteristics was constructed. By comparing real-time spectra with database features, the soil corrosion level and abrasive index could be quickly determined, providing a basis for tool material selection and anti-corrosion measures. Based on multispectral imaging and machine learning algorithms, specifically using convolutional neural networks to classify lithology from spectral images with an accuracy rate exceeding 90%, the distribution of rock blocks with a particle size greater than 5cm was identified, ultimately determining the lithological distribution.

[0027] Preferably, a drone equipped with a high-precision multispectral sensor is used to achieve deep detection of the characteristics of the tunneling soil layer through multi-band spectral analysis, including 10 to 20 working bands such as visible light, near infrared, and short-wave infrared. It can simultaneously acquire spatial and spectral information of soil samples and process them to generate three-dimensional geological feature maps. For example, the boundary of aquifer can be identified by infrared band differences to form a water rate gradient, or the distribution of abrasive index can be predicted by hardness inversion model to intuitively show the trend of soil layer change in the tunneling direction.

[0028] Furthermore, step S100 also includes step S110, performing soil characteristic analysis on the excavation surface soil layer, including the water content, cohesion, abrasion index and lithological distribution, obtaining standard soil layer data, and performing multispectral signal feature conversion on the standard soil layer data to set an initial standard spectrum; step S120, fitting and comparing the multispectral signals collected from the deep soil layer with the initial standard spectrum to obtain the signal change gradient of each soil layer and obtain the characteristic change distribution.

[0029] Preferably, soil characteristics analysis is performed on the excavated soil layer in contact with the tunnel boring machine cutterhead, including moisture content, cohesion, abrasion index, and lithological distribution, to establish a standard soil layer database, i.e., to obtain standard soil layer data. Then, the reflectance / absorption spectral signals of the standard soil layer are collected by a multispectral sensor, and these soil parameters are correlated with the reflectance data of the multispectral sensor to set the initial standard spectrum. The multispectral sensor continuously collects the spectral signals of the deep soil layer ahead of the tunneling face, such as within 5 to 30 meters of the tunnel face, and performs nonlinear fitting comparison with the initial standard spectrum. By analyzing the degree of signal shift in different bands, such as the enhanced absorption in the near-infrared band due to increased water content, the signal change gradient of each soil layer is calculated to quantify the spatial variation of geological parameters. Finally, based on the signal gradient data, combined with the inversion model, the spectral differences are mapped to the continuous distribution of geological parameters. For example, if the mid-infrared reflectance drops sharply at a certain depth and the gradient value exceeds the threshold, it is identified as a high-cohesion hard rock layer; if the thermal infrared signal shows a linearly increasing gradient, it is inferred to be a seepage zone where the water content increases with depth, thus ensuring the accurate dynamic optimization of cutterhead parameters.

[0030] Step S200: Based on the tunneling soil layer characteristic data and characteristic change distribution, the tunneling soil layer is deduced to construct the tunneling soil layer characteristic space.

[0031] Step S200 further includes step S210, establishing a mapping transformation relationship between the initial standard spectrum and the soil characteristics; step S220, based on the signal change gradient of each soil layer, using the characteristic change distribution as the mapping input, and performing soil characteristic mapping output through the mapping transformation relationship to obtain the soil distribution characteristics of each soil layer; step S230, spatially embedding and fusing the soil distribution characteristics of each soil layer according to spatial coordinates to construct the tunneling soil layer feature space.

[0032] Preferably, by establishing a spectral-soil characteristic mapping model and combining signal gradient analysis with spatial fusion, multispectral detection data is transformed into a tunneling soil layer characteristic space that can be used for intelligent decision-making. Specifically, based on laboratory or historical data, a quantitative mapping relationship between the initial standard spectrum and soil characteristics is established through neural networks, support vector regression, or physical inversion models. For example, the absorption peak intensity of a specific band is linearly correlated with the water content, while the shape of the spectral reflectance curve can reflect the lithology. The signal change gradient of each soil layer refers to the rate of change of the spectral characteristics of soil layers at different depths with tunneling depth or spatial location when detected by multispectral sensors, reflecting the rate of change of the attenuation, scattering, or reflection intensity of electromagnetic wave signals of different bands when passing through the soil layer.

[0033] Preferably, the signal change gradient and characteristic change distribution of each soil layer are used as the mapping input. The input mapping transformation relationship is used to output the soil feature mapping, and the corresponding soil parameters are output layer by layer. For example, if the spectral gradient at a certain depth shows a sudden increase in near-infrared absorption rate and an increase in gradient value, the water content at that point is calculated by the water content mapping model. If the thermal infrared gradient changes abruptly, the lithological interface between sandstone and granite is identified by combining the lithology classifier, thereby obtaining the soil distribution characteristics of each soil layer, including the water content and cohesion of each spatial point. Each spatial point includes multiple parameters such as coordinates, water content, cohesion, and lithology code, thereby improving the accuracy of soil inversion.

[0034] Preferably, the soil distribution characteristics of each soil layer are associated with the mileage station, radial depth, and circumferential angle of the tunneling machine coordinate system to form a feature point cloud with spatial labels. Then, the discrete points are fused into a continuous tunneling soil layer feature space through three-dimensional interpolation.

[0035] Step S300: Establish a dynamic safety boundary constraint set, including the tunneling machine hard limit boundary, geological mutation boundary, settlement control boundary, and tool life boundary.

[0036] Step S300 further includes step S310, setting the hard limit boundary of the tunneling machine based on the rated parameters and tolerance threshold according to the equipment hard limit of the tunneling machine; step S320, establishing the constraint relationship between the tunneling fragmentation index and the advance speed, and setting the geological mutation boundary; step S330, determining the settlement control boundary according to the constraint threshold of the formation loss rate and the soil pressure fluctuation; and step S340, determining the tool life boundary based on the tool distribution structure and tool performance of the cutterhead, combined with the tunneling constraints of the tunneling soil layer.

[0037] Preferably, through multi-dimensional dynamic constraint analysis, a dynamic safety boundary constraint set is established, including the tunnel boring machine's hard limit boundary, geological change boundary, settlement control boundary, and cutter life boundary, to ensure the safe and efficient operation of the tunnel boring machine under complex geological conditions. Specifically, the tunnel boring machine's hard limit boundary is an absolute safety threshold based on the equipment's mechanical performance to prevent overload or structural damage. According to the tunnel boring machine's design parameters, such as maximum torque, rated thrust, and motor power, a rated working range is set. Then, engineering experience and safety specifications are superimposed to set tolerance thresholds. For example, 90% of the rated torque is the warning threshold, and 105% is the hard limit shutdown threshold. This is used to monitor the load of the cutterhead drive system in real time. If the parameters approach the hard limit, speed reduction or shutdown protection is immediately triggered.

[0038] Preferably, the geological abrupt change boundary is a dynamic constraint for geological abrupt changes such as alternating soft and hard strata and fractured zones. Geological stability is quantified through tunneling fracturing indices such as rock mass fracturing degree and cutting energy abrupt change rate. A negative correlation constraint is established between the tunneling fracturing degree index and the tunnel boring machine speed. For example, when the tunneling fracturing degree index is greater than 0.8, the tunnel boring machine speed needs to be reduced to below 50% to avoid cutterhead jamming, cutter impact breakage, or tunnel face collapse due to geological abrupt changes. The settlement control boundary is a dynamic threshold to ensure that surface settlement does not exceed the allowable value. It sets dual-indicator constraints based on stratum loss rate and soil chamber pressure fluctuations. The bottom layer loss rate is the ratio of excavation volume to settlement trough volume. Tunneling parameters are dynamically adjusted in conjunction with urban environmental requirements. By adjusting the thrust and grouting pressure, the excavation face balance is maintained to prevent surface collapse.

[0039] Preferably, the tool life boundary is a replacement or maintenance threshold set according to the tool wear prediction model. Based on the tool distribution structure and soil abrasiveness, the dynamic wear rate is calculated. The tool distribution structure may include the cutter spacing, cutting edge material, etc. Then, dual thresholds of wear amount and tunneling mileage are set, such as 15% cutting edge height loss and mandatory inspection every 500 meters, to ensure optimized cutterhead speed / thrust combination, avoid unplanned downtime caused by excessive tool wear, thereby achieving accurate risk warning, extending tool life, and ensuring the continuity and stability of tunnel excavation.

[0040] Furthermore, step S340 also includes step S341, which constructs a dynamic model describing the mapping relationship between tool wear rate and tunneling parameters based on tool material performance test data and historical tunneling data, wherein the tunneling parameters include local propulsion force, cutterhead rotation speed, and soil abrasiveness index; step S342, which divides the cutterhead radially into N sector-shaped pressure regions according to the tool distribution structure of the cutterhead, where N is a positive integer ≥ 4; step S343, which calculates the tool wear rate of each pressure region based on the dynamic model and the current tunneling parameters, and generates a wear spatial distribution heat map; step S344, which analyzes the critical working conditions of tool performance based on the characteristic parameters of the tunneling soil layer, including the maximum allowable contact stress, abrasiveness index threshold, and temperature limit; and step S345, which uses the mechanical load deviation rate, wear spatial distribution heat map, and tunneling constraints as inputs, determines the minimum remaining tool life value of the entire cutterhead through the shortest board effect, and sets the corresponding tunneling parameter threshold as the tool life boundary.

[0041] Preferably, the wear rate curves of laboratory tool material performance tests under different stresses and temperatures are obtained, such as the compressive strength and heat resistance of cemented carbide, as well as historical tunneling data, such as tool wear records under different rock strata. Then, a dynamic model mapping the tool wear rate to tunneling parameters is established based on nonlinear regression or neural networks to quantify the impact of tunneling parameters on tool life. Among these, tunneling parameters include local thrust, cutterhead rotation speed, and soil abrasiveness index. Local thrust is the axial thrust borne by a single tool or local area on the cutterhead. Insufficient thrust will lead to reduced cutting efficiency, while excessive thrust will accelerate tool wear or even breakage. Cutterhead rotation speed refers to the angular velocity of the cutterhead rotation. Too high a rotation speed will increase centrifugal force, leading to increased tool temperature, while too low a rotation speed will reduce tunneling efficiency and may cause cutterhead vibration.

[0042] Preferably, the tangential and radial forces borne by the cutters at different radii are unevenly distributed. Based on the cutter installation position (e.g., center cutter, main cutter, side cutter) and the cutter distribution structure (e.g., radial / circumferential arrangement of hobbing cutters and scrapers), the cutterhead is radially divided into N sector regions, where N is a positive integer ≥ 4, typically 8~12 regions. Each region contains several cutters, thereby achieving spatially refined monitoring of cutter wear and avoiding the global average from masking local high-risk areas. Based on the dynamic model, the current tunneling parameters and soil abrasiveness are input in real time to calculate the instantaneous wear rate of each sector region. An interpolation algorithm is used to generate a wear spatial distribution heatmap to visually display high-wear and low-wear areas. The high-wear area is a red warning zone, where the side cutters wear faster due to high linear velocity; the low-wear area is a blue safety zone, where the center cutter wears more slowly due to low rotational speed.

[0043] Preferably, based on the characteristic parameters of the tunneling soil layer, the critical working conditions of the cutting tool performance are analyzed, including the maximum allowable contact stress, the abrasion index threshold, and the temperature limit. Specifically, the maximum allowable contact stress is set based on the material compressive strength test, setting the maximum stress limit per unit area of ​​the cutting edge, such as 2000 MPa. Exceeding this limit may lead to chipping. The abrasion index threshold is based on historical data statistics. When the abrasion index > 5, it indicates a highly abrasive formation, and the rotation speed is forcibly reduced by 20% to extend the tool's life. The temperature limit is a critical value used to ensure whether the cutting tool needs cooling. The cutting temperature of the cutting tool is monitored by infrared, and the cooling system is triggered when it exceeds 300°C to prevent the predicted results from deviating from the actual working conditions.

[0044] Preferably, the minimum remaining tool life of the entire cutterhead is determined by using the mechanical load deviation rate, wear spatial distribution heat map, and tunneling constraints as inputs, through the shortest board effect. The mechanical load deviation rate refers to the deviation between the actual load and the design rated value in each sector of the cutterhead's pressure zone, used to quantify the uniformity of load distribution. The wear spatial distribution heat map visually displays the real-time wear rate or cumulative wear in different areas of the cutterhead. Tunneling constraints refer to the engineering boundary conditions that must be met during tunnel construction, including rigid limitations such as mechanical performance, safety regulations, and schedule requirements. By traversing the wear data of all tools and combining it with their material property curves, the remaining life of each tool is calculated, i.e., the difference between the cutting edge height scrap threshold and the current measured value. The ratio of the wear rate to the remaining life is used to screen out the minimum remaining life and its corresponding cutting tool, such as the No. 3 side tool with 48 hours remaining. Then, based on the dynamic model of the cutting tool, the limit working parameters that enable its life to reach are solved in reverse. If the minimum remaining life under the current parameter combination is 48 hours, the tunneling parameter thresholds that maintain a safety margin of 72 hours or more for the minimum remaining life are obtained through model inversion, namely, the maximum allowable thrust of 1800kN and the maximum speed of 2.8rpm. Then, the tunneling parameter thresholds are compared with the tunneling constraints, and the more stringent value is taken as the final boundary. Finally, the corresponding tunneling parameter thresholds are set as the tool life boundary to ensure that the overall tool life is accurately extended and the safety and stability of tunnel excavation are guaranteed.

[0045] In step S400, within the characteristic space of the tunneling soil layer, with multiple objectives of minimizing specific energy, minimizing tool wear, and minimizing settlement risk, the cutterhead parameters are optimized based on the cutterhead adjustment constraints. The optimized solution set obtained by optimization is then trimmed using the safety boundary constraint set, and the cutterhead execution parameters are output for cutterhead adjustment.

[0046] Step S400 further includes step S410, which establishes an evaluation function relationship between soil layer characteristics and cutterhead parameters with the minimization of specific energy, the minimum of cutter wear, and the minimum of settlement risk as multiple objectives; step S420, which performs a multi-directional search of cutterhead parameters within the cutterhead adjustment constraint range based on the soil layer characteristics in the tunneling soil layer characteristic space, evaluates the search parameters using the multi-objective evaluation function, and finds a multi-optimized solution set that meets the objective evaluation results as the output.

[0047] Preferably, by establishing a quantitative mapping relationship between soil layer characteristics, cutterhead parameters, and construction objectives, and under the premise of meeting safety constraints, the system automatically searches for the optimal combination of tunneling parameters that comprehensively optimizes energy consumption, tool wear, and settlement risk. This enables intelligent and adaptive adjustment of the tunnel boring machine cutterhead. Specifically, energy consumption is the energy consumed per unit volume of tunneling. The energy minimization objective refers to establishing the physical relationship between soil hardness, moisture content, and other characteristics and energy consumption, focusing on reducing energy consumption. The tool wear minimization objective refers to mapping soil abrasiveness, propulsion force, and rotational speed to a wear rate index based on a wear dynamics model, aiming to minimize the wear rate. The settlement risk minimization objective refers to constructing a settlement risk scoring model using parameters such as ground loss rate and soil chamber pressure stability to suppress the risk of exceeding limits. Furthermore, a weighted comprehensive... A functional relationship between soil layer characteristics and cutterhead parameters is established. Then, in a multi-dimensional parameter space including cutterhead rotation speed, zone thrust, and torque, the cutterhead adjustment constraint range is defined, including red lines such as not exceeding the equipment's rated power and settlement threshold, and prioritizing the satisfaction of the cutter life boundary. Next, based on the soil layer characteristics in the tunneling soil layer characteristic space, a multi-directional search for cutterhead parameters is performed within the cutterhead adjustment constraint range. Specifically, hundreds of sets of randomly generated rotation speed and thrust parameter combinations are initialized for parameter evaluation. The evaluation score of each set of parameters is calculated using a multi-objective evaluation function, retaining high-quality solutions that are close to low specific energy, low wear, and low risk. Then, a new generation of parameter combinations is generated through simulation mutation and crossover, and the search parameter evaluation continues. Finally, a multi-optimized solution set that meets the target evaluation results is output.

[0048] Furthermore, step S420 also includes step S421, establishing a three-dimensional feature tensor space based on the tunneling axis, wherein the X-axis represents the tunneling mileage, the y-axis represents the radial depth, and the z-axis represents the feature parameter value; step S422, mapping the soil feature vectors of each soil layer to the grid nodes of the three-dimensional feature tensor space according to the spatial distribution coordinates; step S423, based on the three-dimensional feature tensor space, fitting the searched cutterhead parameters to the tunneling mileage and radial depth to construct complete time-series cutterhead parameters and generate the multi-optimization solution set.

[0049] Preferably, a three-dimensional feature tensor space is established with the X-axis as the tunneling mileage, the y-axis as the radial depth, and the z-axis as the feature parameter value, based on the tunneling axis. The X-axis is based on the tunnel axis and is divided into grids per meter / ring; the y-axis extends from the excavation surface to the depth of the strata, reflecting the range of advance detection; the z-axis is a normalized multi-dimensional geological feature, such as water content of 0% to 100% and abrasion index of 1 to 10. Then, the soil layer feature vectors obtained from multispectral detection are inserted into the corresponding grid nodes in the three-dimensional feature tensor space according to their spatial distribution coordinates, and Kriging interpolation is used to fill in the undetected areas to form a continuous three-dimensional tensor field. Next, based on the three-dimensional feature tensor space, the searched cutterhead parameters are fitted with the tunneling mileage and radial depth, that is, the spatial dimension is added according to the cutter rotation speed and thrust, including radial adaptation and mileage prediction. Radial adaptation refers to matching the strata at different depths on the y-axis according to the radial position of the cutterhead partition, and mileage prediction refers to preloading the strata features 50m ahead along the X-axis in advance to pre-generate parameter plans. Then, dynamic fitting is performed, that is, extracting the features of the directly excavated area and loading the features of the area to be excavated during the tunneling process, and generating the corresponding complete time-series cutterhead parameters. Finally, in the three-dimensional tensor space, the local optimal parameters are solved in parallel with the feature values ​​of each grid node as input, and multiple optimized solution sets are output, which may include parameters matching different radial depths for each mileage segment, thereby ensuring the stability and reliability of the tunnel boring machine operation.

[0050] Furthermore, step S422 also includes step S4221, establishing the number of grids in the X and Y axes of the three-dimensional feature tensor space, where the number of grids on the X axis corresponds to the tunneling process and the number of grids on the Y axis corresponds to the radial depth; step S4222, calculating the Euclidean distance from the grid node to the soil layer detection point, calculating the feature value according to the inverse distance weight, and assigning the feature value to the Z-axis feature channel corresponding to the grid node.

[0051] Preferably, by discretizing the tunneling environment into a structured grid and accurately interpolating the geological features of discrete detection points to the full-space grid nodes based on spatial distance weights, a quantifiable three-dimensional digital twin geological model is constructed, providing high-precision spatial data for cutterhead parameter optimization. Specifically, in the X-axis tunneling mileage direction, divided according to construction ring number or mileage station, the number of grids in the X-axis direction in the three-dimensional feature tensor space is the ratio of the total tunneling length to the grid spacing. For example, if the total tunnel length is 1000 meters and the grid spacing is 0.5 meters, then the number of grids in the X-axis = 2000. In the Y-axis radial depth direction, divided from the excavation face to the depth of the stratum, the number of grids in the Y-axis direction in the three-dimensional feature tensor space is the ratio of the detection depth to the layer thickness. For example, if the detection depth is 30 meters and each meter is divided into 1 layer, the number of grids in the Y-axis is 30. Key areas can be densified, such as dividing 0-5 meters into 10 layers, to reflect the differences in the cutting strata of the cutterhead at different positions. Then, for each grid node, the Euclidean distance to the soil detection point is calculated, and an eigenvalue is calculated according to the inverse distance weighting, that is, the reciprocal of the Euclidean distance is used as the eigenvalue of the node. The closer the grid node is to the detection point, the greater the influence. Finally, the eigenvalue is assigned to the corresponding Z-axis feature channel of the grid node. Each grid node can contain multiple Z-axis feature channels. By adjusting the grid density to balance calculation efficiency and control accuracy, the operational stability and reliability of the tunnel boring machine are ensured.

[0052] Preferably, the optimized solution set obtained through optimization is pruned using a set of safety boundary constraints. Specifically, tunneling time-series simulation is performed based on the three-dimensional feature tensor space and multiple optimized solution sets to analyze ground loss, tunneling fracturing, and cutter and equipment wear during the tunneling process. This includes ground loss simulation, rock fracturing analysis, and wear accumulation calculation. Ground loss simulation calculates the over-excavation rate based on the soil chamber pressure and advance speed to predict the development of surface settlement troughs. For example, a thrust of 2000 kN and a rotation speed of 3 rpm in a sand layer indicates a predicted settlement of 8 mm, close to the control boundary. Rock fracturing analysis assesses the rock mass fracturing degree by inverting the rock mass fracturing degree through cutting energy. Collapse risks, such as when the fragmentation index is >0.7, are marked as potential spalling zones. The cumulative wear calculation is based on the dynamic model to iteratively calculate the tool wear depth, for example, the cumulative wear of the side tool in the granite layer is 0.3mm / ring. Then, the simulation results are checked ring by ring using the hard limit boundary of the tunnel boring machine, the geological mutation boundary, the settlement control boundary and the tool life boundary in the safety boundary constraint set. The cutterhead parameters that cause any boundary to exceed the limit are eliminated and the final cutterhead execution parameters are output for the cutterhead adjustment of the tunnel boring machine, thereby improving the tunnel excavation efficiency, stability, construction safety and extending the tool life and reducing tool consumption.

[0053] In the above text, refer to Figure 1 A method for intelligent adjustment of the cutterhead of a tunnel boring machine according to an embodiment of the present invention is described in detail. Next, reference will be made to... Figure 2A cutterhead intelligent adjustment system for a tunnel boring machine is described according to an embodiment of the present invention.

[0054] The intelligent cutterhead adjustment system for tunnel boring machines according to embodiments of the present invention addresses the technical problems in existing technologies, such as insufficient dynamic perception of geological conditions, poor adaptability to dynamic safety constraints, and difficulty in balancing multi-objective optimization conflicts, leading to low tunneling efficiency and stability, cutter life, and construction safety. It achieves the technical effects of improving tunneling efficiency, stability, construction safety, and extending cutter life. Figure 2 As shown, the intelligent cutterhead adjustment system for tunnel boring machines includes: a deep detection module 10, a soil layer deduction module 20, a safety boundary constraint configuration module 30, and a cutterhead parameter optimization module 40.

[0055] The deep exploration module 10 is used to acquire characteristic data of the tunneling soil layer, including water content, cohesion, abrasive index, and lithological distribution, and to obtain the characteristic variation distribution through deep exploration using a multispectral sensor. The tunneling soil layer derivation module 20 is used to deduce the tunneling soil layer based on the tunneling soil layer characteristic data and characteristic variation distribution, and to construct the tunneling soil layer characteristic space. The safety boundary constraint configuration module 30 is used to establish a dynamic safety boundary constraint set, including the tunneling machine hard limit boundary, geological abrupt change boundary, settlement control boundary, and cutter life boundary. The cutterhead parameter optimization module 40 is used to optimize the cutterhead parameters in the tunneling soil layer characteristic space with multiple objectives, including minimizing specific energy, minimizing cutter wear, and minimizing settlement risk, based on the cutterhead adjustment constraints, and to prune the optimized solution set obtained by optimization using the safety boundary constraint set, and output the cutterhead execution parameters for cutterhead adjustment.

[0056] The specific configuration of the deep detection module 10 will be described in detail below. The deep detection module 10 further includes: performing soil characteristic analysis on the excavation surface soil layer, including the water content, cohesion, abrasion index, and lithological distribution, obtaining standard soil layer data, and performing multispectral signal feature conversion on the standard soil layer data to set an initial standard spectrum; fitting and comparing the multispectral signals collected from the deep soil layer with the initial standard spectrum to obtain the signal change gradient of each soil layer and obtain the characteristic change distribution.

[0057] The specific configuration of the tunneling soil layer derivation module 20 will be described in detail below. The tunneling soil layer derivation module 20 further includes: establishing a mapping transformation relationship between the initial standard spectrum and the soil characteristics; based on the signal change gradient of each soil layer, using the characteristic change distribution as a mapping input, and performing soil characteristic mapping output through the mapping transformation relationship to obtain the soil distribution characteristics of each soil layer; and spatially embedding and fusing the soil distribution characteristics of each soil layer according to spatial coordinates to construct the tunneling soil layer feature space.

[0058] The specific configuration of the safety boundary constraint configuration module 30 will be described in detail below. The safety boundary constraint configuration module 30 further includes: setting the hard limit boundary of the tunneling machine based on its equipment hard limits, rated parameters, and tolerance thresholds; establishing a constraint relationship between the tunneling fragmentation index and the advance speed, and setting the geological abrupt change boundary; determining the settlement control boundary according to the constraint thresholds of the formation loss rate and soil pressure fluctuation; and determining the cutter life boundary based on the cutter distribution structure and cutter performance of the cutterhead, combined with the tunneling constraints of the tunneling soil layer.

[0059] The specific configuration of the safety boundary constraint configuration module 30 will be described in detail below. The safety boundary constraint configuration module 30 further includes: constructing a dynamic model describing the mapping relationship between tool wear rate and tunneling parameters based on tool material performance test data and historical tunneling data, wherein the tunneling parameters include local thrust, cutterhead rotation speed, and soil abrasiveness index; dividing the cutterhead radially into N sector-shaped pressure regions according to the tool distribution structure of the cutterhead, where N is a positive integer ≥ 4; calculating the tool wear rate of each pressure region based on the dynamic model and current tunneling parameters, and generating a wear spatial distribution heat map; analyzing the critical working conditions of tool performance based on the characteristic parameters of the tunneling soil layer, including maximum allowable contact stress, abrasiveness index threshold, and temperature limit; using mechanical load deviation rate, wear spatial distribution heat map, and tunneling constraints as inputs, determining the minimum remaining tool life value of the entire cutterhead through the shortest board effect, and setting the corresponding tunneling parameter threshold as the tool life boundary.

[0060] The specific configuration of the cutterhead parameter optimization module 40 will be described in detail below. The cutterhead parameter optimization module 40 further includes: establishing an evaluation function relationship between soil layer characteristics and cutterhead parameters with multiple objectives, including minimizing specific energy, minimizing tool wear, and minimizing settlement risk; performing multi-directional search of cutterhead parameters within the cutterhead adjustment constraint range based on the soil layer characteristics in the tunneling soil layer characteristic space; evaluating the search parameters using the multi-objective evaluation function; and finding a multi-optimized solution set that meets the objective evaluation results as the output.

[0061] The following will describe the specific configuration of the cutterhead parameter optimization module 40 in detail. The cutterhead parameter optimization module 40 further includes: establishing a three-dimensional feature tensor space based on the tunneling axis, where the X-axis represents the tunneling mileage, the y-axis represents the radial depth, and the z-axis represents the feature parameter values; mapping the soil feature vectors of each soil layer to the grid nodes of the three-dimensional feature tensor space according to their spatial distribution coordinates; and fitting the searched cutterhead parameters to the tunneling mileage and radial depth based on the three-dimensional feature tensor space to construct complete time-series cutterhead parameters and generate the multi-optimization solution set.

[0062] The specific configuration of the cutterhead parameter optimization module 40 will be described in detail below. The cutterhead parameter optimization module 40 further includes: establishing the number of grids in the X and Y axes of the three-dimensional feature tensor space, where the number of grids on the X axis corresponds to the corresponding tunneling process, and the number of grids on the Y axis corresponds to the corresponding radial depth; calculating the Euclidean distance from the grid node to the soil layer detection point, calculating feature values ​​according to inverse distance weights, and assigning the feature values ​​to the Z-axis feature channel corresponding to the grid node.

[0063] The intelligent cutterhead adjustment system for tunnel boring machines provided in this embodiment of the invention can execute the intelligent cutterhead adjustment method for tunnel boring machines provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the method.

[0064] Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention, showing a block diagram of an exemplary electronic device suitable for implementing the embodiments of the present invention. Figure 3 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of the embodiments of the present invention. This electronic device is in the form of a general-purpose computing device, and its components may include, but are not limited to, an input device 401, a processor 402, a memory 403, and an output device 404. The processor 402 may be one or more; the memory 403 may include a computer-readable medium and at least one program product having a set (at least one) of program modules configured to perform the functions of the embodiments of this application.

[0065] The memory 403 shown in this embodiment of the invention can be any combination of one or more computer-readable media. The computer-readable storage medium can be, but is not limited to, infrared, semiconductor systems, devices or components, or any combination thereof, used to store software programs, computer-executable programs and modules, such as the program instructions / modules corresponding to the intelligent cutterhead adjustment method for tunnel boring machines in this embodiment of the invention. The processor 402 executes various functional applications and data processing of the computer device by running the software programs, instructions and modules stored in the memory 403, thereby realizing the above-mentioned intelligent cutterhead adjustment method for tunnel boring machines.

[0066] Although this application makes various references to certain modules in the system according to the embodiments of this application, any number of different modules can be used and run on user terminals and / or servers. The various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of each functional unit are only for easy distinction between each other and are not used to limit the scope of protection of this invention.

[0067] The specific embodiments described above do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand that various modifications, combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for intelligent adjustment of the cutterhead in a tunnel boring machine, characterized in that, include: Acquire characteristic data of the tunneling soil layer, including water content, cohesion, abrasion index and lithological distribution, and use multispectral sensors to conduct deep exploration to obtain the characteristic variation distribution; Based on the characteristic data and characteristic variation distribution of the tunneled soil layers, the tunneled soil layers are deduced, and a characteristic space of the tunneled soil layers is constructed. Establish a dynamic safety boundary constraint set, including the tunneling machine hard limit boundary, geological change boundary, settlement control boundary, and tool life boundary; In the tunneling soil layer characteristic space, with multiple objectives of minimizing specific energy, minimizing tool wear, and minimizing settlement risk, the cutterhead parameters are optimized based on the cutterhead adjustment constraints. The optimized solution set obtained by optimization is then trimmed using the safety boundary constraint set, and the output cutterhead execution parameters are used for cutterhead adjustment.

2. The intelligent cutterhead adjustment method for tunnel boring machines according to claim 1, characterized in that, Acquire characteristic data of the tunneling soil layers, including water content, cohesion, abrasion index, and lithological distribution, and conduct deep exploration using multispectral sensors to obtain characteristic variation distributions, including: Soil characteristics analysis is performed on the surface soil layer of the tunnel, including water content, cohesion, abrasion index and lithology distribution, to obtain standard soil layer data, and multispectral signal feature conversion is performed on the standard soil layer data to set the initial standard spectrum; The multispectral signals collected from deep soil layers are fitted and compared with the initial standard spectrum to obtain the signal change gradient of each soil layer and the characteristic change distribution.

3. The intelligent cutterhead adjustment method for tunnel boring machines according to claim 2, characterized in that, Based on the aforementioned tunneling soil layer characteristic data and characteristic variation distribution, the tunneling soil layer is deduced, and a tunneling soil layer characteristic space is constructed, including: Establish the mapping and transformation relationship between the initial standard spectrum and the soil characteristics; Based on the signal change gradient of each soil layer, the characteristic change distribution is used as the mapping input, and the soil feature mapping output is performed through the mapping transformation relationship to obtain the soil distribution characteristics of each soil layer. The soil distribution characteristics of each soil layer are spatially embedded and fused according to spatial coordinates to construct the feature space of the tunneling soil layer.

4. The intelligent cutterhead adjustment method for tunnel boring machines according to claim 1, characterized in that, Establish a dynamic safety boundary constraint set, including: Based on the hard limit of the tunneling machine, and using the rated parameters and tolerance threshold, the hard limit boundary of the tunneling machine is set. Establish a constraint relationship between the tunneling fragmentation index and the advance speed, and set the geological abrupt change boundary; Settlement control boundaries are determined based on the constraint thresholds of formation loss rate and soil pressure fluctuation. Based on the cutterhead's cutter distribution structure and cutter performance, and combined with the tunneling constraints of the excavated soil layer, the cutter life boundary is determined.

5. The intelligent cutterhead adjustment method for tunnel boring machines according to claim 4, characterized in that, Based on the cutterhead's cutter distribution structure and cutter performance, combined with the tunneling constraints of the excavated soil layer, the cutter life boundary is determined, including: Based on test data of tool material performance and historical tunneling data, a dynamic model describing the mapping relationship between tool wear rate and tunneling parameters is constructed. The tunneling parameters include local propulsion force, cutterhead rotation speed and soil abrasiveness index. According to the tool distribution structure of the cutter head, the cutter head is radially divided into N sector-shaped pressure regions, where N is a positive integer ≥ 4; Based on the aforementioned dynamic model, the tool wear rate in each pressure zone is calculated in conjunction with the current tunneling parameters, and a thermal map of the wear spatial distribution is generated. Based on the characteristic parameters of the tunneling soil layer, the critical working conditions of the cutting tool performance are analyzed, including the maximum allowable contact stress, the abrasive index threshold, and the temperature limit. Using mechanical load deviation rate, wear spatial distribution heat map and tunneling constraints as inputs, the minimum remaining tool life value of the entire cutterhead is determined by the shortest board effect, and the corresponding tunneling parameter threshold is set as the tool life boundary.

6. The intelligent cutterhead adjustment method for tunnel boring machines according to claim 3, characterized in that, In the characteristic space of the tunneling soil layer, with multiple objectives of minimizing specific energy, minimizing cutter wear, and minimizing settlement risk, the cutterhead parameters are optimized based on cutterhead adjustment constraints, including: With minimizing specific energy, minimizing tool wear, and minimizing settlement risk as multiple objectives, an evaluation function relationship between soil layer characteristics and cutterhead parameters is established; Based on the soil layer characteristics in the tunneling soil layer characteristic space, the cutterhead parameters are searched in multiple directions within the cutterhead adjustment constraint range. The search parameters are evaluated using a multi-objective evaluation function, and a multi-optimized solution set that meets the objective evaluation results is found as the output.

7. The intelligent cutterhead adjustment method for a tunnel boring machine according to claim 6, characterized in that, Also includes: A three-dimensional feature tensor space is established with the tunneling axis as the reference, where the X-axis represents the tunneling mileage, the y-axis represents the radial depth, and the z-axis represents the feature parameter values; The soil feature vectors of each soil layer are mapped to the grid nodes of the three-dimensional feature tensor space according to the spatial distribution coordinates. Based on the three-dimensional feature tensor space, the searched cutterhead parameters are fitted with tunneling mileage and radial depth to construct complete time-series cutterhead parameters and generate the multi-optimization solution set.

8. The intelligent cutterhead adjustment method for a tunnel boring machine according to claim 7, characterized in that, Mapping the soil feature vectors of each soil layer to the grid nodes of the three-dimensional feature tensor space according to their spatial distribution coordinates includes: Establish the number of grids in the X and Y axes of the three-dimensional feature tensor space, where the number of grids along the X axis corresponds to the tunneling process and the number of grids along the Y axis corresponds to the radial depth; Calculate the Euclidean distance from the grid node to the soil layer detection point, calculate the feature value according to the inverse distance weight, and assign the feature value to the Z-axis feature channel corresponding to the grid node.

9. An intelligent cutterhead adjustment system for tunnel boring machines, characterized in that, The system is used to implement the intelligent cutterhead adjustment method for a tunnel boring machine according to any one of claims 1 to 8, and the system includes: The deep exploration module is used to acquire characteristic data of the tunneling soil layer, including water content, cohesion, abrasion index and lithological distribution, and to obtain the characteristic variation distribution through multispectral sensors for deep exploration. The excavation soil layer deduction module is used to deduce the excavation soil layer based on the excavation soil layer characteristic data and characteristic change distribution, and to construct the excavation soil layer characteristic space; The safety boundary constraint configuration module is used to establish a dynamic safety boundary constraint set, including the tunneling machine hard limit boundary, geological change boundary, settlement control boundary, and tool life boundary; The cutterhead parameter optimization module is used to optimize the cutterhead parameters in the characteristic space of the tunneling soil layer with multiple objectives, including minimizing specific energy, minimizing tool wear, and minimizing settlement risk, based on the cutterhead adjustment constraints. The module also uses the safety boundary constraint set to trim the optimized solution set and outputs the cutterhead execution parameters for cutterhead adjustment.

10. An electronic device, characterized in that, The electronic device includes: Memory, used to store executable instructions; The processor, when executing executable instructions stored in the memory, implements the intelligent cutterhead adjustment method for a tunnel boring machine as described in any one of claims 1-8.

Images