CN113569327B - Method and system for optimizing posture of shield tunneling machine and method and system for training model - Google Patents
Method and system for optimizing posture of shield tunneling machine and method and system for training model Download PDFInfo
- Publication number
- CN113569327B CN113569327B CN202110914150.3A CN202110914150A CN113569327B CN 113569327 B CN113569327 B CN 113569327B CN 202110914150 A CN202110914150 A CN 202110914150A CN 113569327 B CN113569327 B CN 113569327B
- Authority
- CN
- China
- Prior art keywords
- shield
- model
- machine
- stratum
- attitude
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 67
- 238000012549 training Methods 0.000 title claims abstract description 61
- 230000005641 tunneling Effects 0.000 title claims abstract description 61
- 238000010276 construction Methods 0.000 claims abstract description 140
- 238000010801 machine learning Methods 0.000 claims abstract description 72
- 238000012360 testing method Methods 0.000 claims description 27
- 230000015572 biosynthetic process Effects 0.000 claims description 15
- 230000009471 action Effects 0.000 claims description 7
- 230000008569 process Effects 0.000 claims description 7
- 230000002159 abnormal effect Effects 0.000 claims description 4
- 238000007781 pre-processing Methods 0.000 claims description 3
- 238000009826 distribution Methods 0.000 abstract description 6
- 230000011218 segmentation Effects 0.000 abstract description 4
- 230000035772 mutation Effects 0.000 abstract description 3
- 238000013461 design Methods 0.000 description 17
- 238000005755 formation reaction Methods 0.000 description 13
- 230000006870 function Effects 0.000 description 8
- 239000002689 soil Substances 0.000 description 8
- 238000005457 optimization Methods 0.000 description 6
- 230000001276 controlling effect Effects 0.000 description 4
- 238000000605 extraction Methods 0.000 description 4
- 238000013507 mapping Methods 0.000 description 4
- 239000011159 matrix material Substances 0.000 description 4
- 238000010606 normalization Methods 0.000 description 4
- 230000001105 regulatory effect Effects 0.000 description 3
- 239000011435 rock Substances 0.000 description 3
- 230000004913 activation Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000005520 cutting process Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000005553 drilling Methods 0.000 description 2
- 230000008030 elimination Effects 0.000 description 2
- 238000003379 elimination reaction Methods 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 239000002002 slurry Substances 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 239000004568 cement Substances 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000007599 discharging Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000001125 extrusion Methods 0.000 description 1
- 239000007788 liquid Substances 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000000149 penetrating effect Effects 0.000 description 1
- 239000013049 sediment Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/10—Geometric CAD
- G06F30/13—Architectural design, e.g. computer-aided architectural design [CAAD] related to design of buildings, bridges, landscapes, production plants or roads
-
- E—FIXED CONSTRUCTIONS
- E21—EARTH OR ROCK DRILLING; MINING
- E21D—SHAFTS; TUNNELS; GALLERIES; LARGE UNDERGROUND CHAMBERS
- E21D9/00—Tunnels or galleries, with or without linings; Methods or apparatus for making thereof; Layout of tunnels or galleries
- E21D9/06—Making by using a driving shield, i.e. advanced by pushing means bearing against the already placed lining
- E21D9/093—Control of the driving shield, e.g. of the hydraulic advancing cylinders
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/14—Force analysis or force optimisation, e.g. static or dynamic forces
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Geometry (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Software Systems (AREA)
- Computer Hardware Design (AREA)
- Medical Informatics (AREA)
- Mining & Mineral Resources (AREA)
- Mathematical Physics (AREA)
- Environmental & Geological Engineering (AREA)
- Computing Systems (AREA)
- General Life Sciences & Earth Sciences (AREA)
- Geochemistry & Mineralogy (AREA)
- Geology (AREA)
- Architecture (AREA)
- Civil Engineering (AREA)
- Structural Engineering (AREA)
- Computational Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Excavating Of Shafts Or Tunnels (AREA)
Abstract
The invention relates to a method and a system for optimizing the posture of a shield tunneling machine and a method and a system for training a model. The method can classify the stratum type combination according to the shield construction history data by using the clustering method, thereby providing basis for the subsequent segmentation modeling of various machine learning methods. According to the segmentation result, modeling is carried out by using a plurality of machine learning models, and weight distribution is carried out according to the accuracy, so that an integrated learning model suitable for various stratum types and combinations can be obtained. The integrated learning model can allocate higher weight for the machine learning model with higher accuracy aiming at a specified stratum, so that the machine learning model with higher accuracy is different in machine learning model adopted by different stratum types, the phenomenon of inaccurate attitude prediction of the shield tunneling machine caused by stratum mutation is avoided, and the historical data is utilized more fully.
Description
Technical Field
The invention relates to the technical field of tunnel construction by a shield method, in particular to a method and a system for optimizing the posture of a shield machine and a method and a system for training a model.
Background
With the rapid development of Chinese social economy and production, tunnel engineering is greatly emerging, and the construction standard of tunnels is continuously improved. The shield method has high construction speed and small influence on surrounding environment, and is widely applied to the construction of subway tunnels. However, the construction parameters of the shield method are numerous, and if the construction parameters of the shield cannot be reasonably set, the shield method cannot exert the construction advantages and even causes safety accidents. The attitude of the shield machine is an important index to be controlled in the construction process of the shield method, and if the shield construction parameters are set improperly, the shield machine deviates from the design axis of the tunnel, so that the problems of overlarge ground subsidence, difficult segment assembly, incapability of penetrating the tunnel and the like are caused. Reasonably setting the shield construction parameters ensures that the attitude of the shield machine meets the requirements, and has important significance for fully playing the advantages of the shield method, improving the construction efficiency and reducing the construction risk.
Therefore, how to design a method capable of accurately predicting the posture of the shield tunneling machine becomes a technical problem to be solved in the field.
Disclosure of Invention
The invention aims to provide a method and a system for optimizing the posture of a shield machine and a method and a system for training a model. The invention can solve the problem of poor attitude control of the shield machine.
In order to achieve the above object, the present invention provides the following solutions:
a training method of a shield tunneling machine attitude prediction model comprises the following steps:
dividing shield construction history data according to stratum category combinations which simultaneously appear on a tunnel face by using a clustering algorithm to obtain n k0 Stratum combination sample sets corresponding to stratum category combinations of the individual categories; the shield construction history data comprise shield construction parameter data, shield machine attitude parameter data and stratum data;
each category is toThe stratum combination sample set is divided into at least one interval sample set according to time continuity; each sample in the interval sample set comprises t corresponding to each other n Shield construction parameter data and t at moment n+1 Attitude parameter data of the shield tunneling machine at moment, wherein t is as follows n+1 The time is the t n The next time of the time;
selecting part of samples in each interval sample set to form a training set, and using t as the n The shield construction parameter data at the moment is taken as an input parameter, and the t is taken as the input parameter n+1 The attitude parameter data of the shield machine at moment is taken as an output parameter, and a machine learning model is trained; forming a test set to test a trained machine learning model by utilizing another part of samples in the interval sample set, and calculating the accuracy of the trained machine learning model;
And determining a final trained machine learning model as a shield tunneling machine attitude prediction model according to the accuracy corresponding to each interval sample set.
Optionally, the machine learning model includes at least 2 algorithm models, the training set trains at least 2 algorithm models, and the test set tests at least 2 trained algorithm models.
Optionally, determining the final trained machine learning model according to the accuracy corresponding to each interval sample set as a shield tunneling machine posture prediction model specifically includes:
calculating the average value of the accuracy of each trained algorithm model corresponding to all the interval sample sets of the same category, and determining the trained algorithm model corresponding to the interval sample set with the highest accuracy as a category target model;
calculating model weights in the same category by using the average value of the accuracy of the various trained algorithm models; and distributing the category target model according to the model weight to obtain a shield tunneling machine attitude prediction model corresponding to the category.
Optionally, the shield construction history data is clustered according to palm by using a clustering algorithm Dividing the simultaneous formation type combinations on the sub-surfaces to obtain n k0 The stratum combination sample set corresponding to the stratum category combination of each category specifically comprises:
acquiring shield construction parameter data, shield machine attitude parameter data and stratum data, and recording the data as shield construction history data; the shield construction parameter data comprises: tool parameters, jack parameters, working condition parameters and grouting parameters; the attitude parameter data of the shield tunneling machine comprises: attitude data of the cutterhead and the shield tail; the formation data includes: the combination of stratum types and stratum characteristic parameters which are simultaneously appeared on the face;
calculating the total number n of stratum category combinations simultaneously appearing on the face s ；
Randomly selecting n in the shield construction history data s The individual samples are used as initial clustering centersSetting the maximum iteration number n max From minimum error variation epsilon min ；
Calculating the distance between each sample and the clustering center, and taking the category with the smallest distance as the category to which the sample belongs;
recalculating a new cluster center for each category;
judging whether an iteration termination condition is reached, if not, taking a new cluster center as the cluster center of the category, and returning to execute 'calculating the distance between each sample and the cluster center'; if yes, determining the category of all the obtained stratum category combinations and a stratum combination sample set corresponding to each category; the category is n k0 And each.
Optionally, selecting part of samples in each interval sample set to form a training set, and using t as the n The shield construction parameter data at the moment is taken as an input parameter, and the t is taken as the input parameter n+1 The moment of time, the attitude parameter data of the shield machine is an output parameter, and the machine learning model is trained, specifically comprising:
establishing a machine learning model for each interval sample set;
training a corresponding machine learning model by utilizing a training set in the interval sample set;
and adjusting the super parameters of the machine learning model by using a grid method in the training process to obtain the trained machine learning model.
Optionally, the machine learning model includes an MLP model, an SVM model and a GBR model, and the step of obtaining the shield tunneling machine posture prediction model specifically includes:
respectively calculating an average value m of accuracy rates of predicting all interval sample sets of the same class by using an MLP model, an SVM model and a GBR model j ，s j And g j ；
According to the average value m j ，s j And g j Calculating model weightsAnd->
And summing the MLP model, the SVM model and the GBR model according to the model weight to obtain a shield tunneling machine attitude prediction model.
Optionally, the acquiring of the stratum combination sample set specifically includes:
Preprocessing the shield construction history data, and removing abnormal values and values of a stopping section to obtain shield construction reference data;
obtaining a posture parameters of the shield machine according to the shield construction reference data, and marking the posture parameters asb shield construction parameters marked as +.>c + d formation parameters,wherein (1)>For the thickness of the combination of the various stratum types on the face, < > for each stratum type>Various physical and mechanical parameters combined for various formation types; i=1, 2 … n;
combining the a shield machine attitude parameters, the b shield construction parameters and the c+d stratum parameters according to time, and simultaneously adding shield machine attitude parameter data at the next moment, wherein the shield machine attitude parameter data at the next moment are recorded asObtaining the stratum combination sample set, wherein the stratum combination sample set is as follows:
the invention also provides a shield tunneling machine attitude optimization method, which comprises the following steps:
inputting the shield construction parameters at the current moment into a shield machine attitude prediction model to obtain shield machine attitude parameter data at the next moment; the shield machine posture prediction model is obtained through the shield machine posture prediction model training method;
judging whether the attitude parameter data of the shield machine at the next moment exceeds a preset deviation limit value or not;
If yes, the shield construction parameters are adjusted to serve as new shield construction parameters at the current moment, and the execution of 'inputting the shield construction parameters at the current moment into a shield machine attitude prediction model' is returned;
and if not, controlling the action of the shield machine according to the attitude parameter data of the shield machine at the next moment, and storing the attitude parameter data of the shield machine at the next moment into the shield construction parameters.
The invention also provides a training system of the attitude prediction model of the shield machine, which comprises the following steps:
the stratum type combination dividing module is used for dividing shield construction history data according to stratum type combinations which simultaneously appear on the face by using a clustering algorithm to obtain n k0 Stratum combination sample sets corresponding to stratum category combinations of the individual categories; the shield construction history data comprise shield construction parameter data, shield machine attitude parameter data and stratum data;
the interval sample set dividing module is used for dividing the stratum combination sample set of each category into at least one interval sample set according to time continuity; each sample in the interval sample set comprises t corresponding to each other n Shield construction parameter data and t at moment n+1 Attitude parameter data of the shield tunneling machine at moment, wherein t is as follows n+1 The time is the t n The next time of the time;
a machine learning model training module for selecting part of samples in each interval sample set to form a training set, and using t as the n The shield construction parameter data at the moment is taken as an input parameter, and the t is taken as the input parameter n+1 The attitude parameter data of the shield machine at moment is taken as an output parameter, and a machine learning model is trained; forming a test set to test a trained machine learning model by utilizing another part of samples in the interval sample set, and calculating the accuracy of the trained machine learning model;
and the shield machine posture prediction model building module is used for determining a final trained machine learning model as a shield machine posture prediction model according to the accuracy rates corresponding to the interval sample sets.
The invention also provides a shield tunneling machine attitude optimization system, which comprises:
the system comprises a shield machine attitude parameter data acquisition module, a shield machine attitude parameter prediction module and a control module, wherein the shield machine attitude parameter data acquisition module is used for inputting shield construction parameters at the current moment into a shield machine attitude prediction model to obtain shield machine attitude parameter data at the next moment;
the judging module is used for judging whether the attitude parameter data of the shield machine at the next moment exceeds a preset deviation limit value;
If yes, the shield construction parameters are adjusted to serve as new shield construction parameters at the current moment, and the execution of 'inputting the shield construction parameters at the current moment into a shield machine attitude prediction model' is returned;
and if not, controlling the action of the shield machine according to the attitude parameter data of the shield machine at the next moment, and storing the attitude parameter data of the shield machine at the next moment into the shield construction parameters.
According to the specific embodiment provided by the invention, the invention discloses the following technical effects: the invention provides a method and a system for optimizing the posture of a shield machine, a method and a system for training a model, wherein the method and the system can be used for classifying stratum type combinations according to shield construction history data by using a clustering method, so that the method and the system provide a basis for the sectional modeling of various subsequent machine learning methods. According to the segmentation result, modeling is carried out by using a plurality of machine learning models, and weight distribution is carried out according to the accuracy, so that an integrated learning model suitable for various stratum types and combinations can be obtained. The integrated learning model can allocate higher weight for the machine learning model with higher accuracy aiming at a specified stratum, so that the machine learning model with higher accuracy is different in machine learning model adopted by different stratum types, the phenomenon of inaccurate attitude prediction of the shield tunneling machine caused by stratum mutation is avoided, and the historical data is utilized more fully.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are needed in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flowchart of a method for training a model for predicting the attitude of a shield tunneling machine according to embodiment 1 of the present invention;
FIG. 2 is a flow chart of the acquisition of a combined sample set of strata;
FIG. 3 is a flow chart of formation class combination classification;
FIG. 4 is a flow chart of construction of a model for predicting the attitude of the shield tunneling machine;
FIG. 5 is a flowchart of a method for optimizing the posture of a shield tunneling machine according to embodiment 2 of the present invention;
FIG. 6 is a block diagram of a training system for a model for predicting the attitude of a shield tunneling machine according to embodiment 3 of the present invention;
fig. 7 is a structural block diagram of a posture optimization system of a shield tunneling machine provided in embodiment 4 of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The invention aims to develop a shield tunneling machine attitude prediction model training method and a shield tunneling machine attitude optimization method, and the method can allocate weights to different training models according to the test results of the existing machine learning models aiming at different strata, and build an integrated learning model using a plurality of machine learning models, thereby ensuring that the integrated learning model can adapt to various stratum conditions and greatly improving the accuracy and reliability of the integrated learning model.
The definition of the terms of art related to the present invention is as follows:
formation: the aggregate of all layered rock, which may be consolidated rock or unconsolidated sediment, i.e., earth, is a layer or a group of layers of rock (earth) having certain uniform characteristics and properties and being distinctly different from the upper and lower layers.
The shield method comprises the following steps: a tunnel construction method uses a shield machine to excavate stratum and assemble tunnel segments.
Shield machine: a construction machine is composed of a shell, a cutter head, pushing equipment, assembling equipment and other matched equipment, wherein the shell is a cylinder and plays a role in protection, and other equipment is arranged in the shell.
Cutter head: the shield machine is used for cutting stratum, is positioned at the front end of the shield machine, and cuts off soil through rotary extrusion.
Profiling cutter: the cutter extending along the radial direction of the cutter head can expand the cutting range of the cutter head so as to help the shield machine to turn.
Pushing jack: the jack for pushing the shield machine to advance is divided into an upper area, a right area, a lower area and a left area, and the posture of the shield machine can be adjusted by adjusting the pressure values of the four areas.
Hinged jack: the jack for adjusting the shape of the shield is divided into an upper area, a right area, a lower area and a left area, and the shape of the shield machine can be adjusted by adjusting the travel difference of the four areas.
Spiral soil outlet machine: the soil body cut by the cutter disc is conveyed to a machine on the belt conveyor through the rotation of the spiral member, and the speed of the soil discharging is determined by the rotation speed of the spiral construction.
Belt conveyor: and the belt device is used for conveying the soil body from the spiral soil outlet machine to the residue soil vehicle.
Post-wall grouting pipe: and the slurry is transported to a transmission pipeline in a gap between the pipe piece and the stratum, a certain gap exists between the pipe piece and the stratum after the pipe piece is assembled and the pipe piece is separated from the shield tail, the gap is filled with cement slurry and the like to avoid deformation of the stratum, and the grouting pressure and volume can be adjusted to minimize settlement of the stratum and avoid surface elevation.
Shield tail: the rear part of the shield machine is used for protecting the shell of the internal equipment of the shield machine.
Tunnel design axis: the direction line of the extending direction of the tunnel in the design drawing may be either a straight line or a curved line.
Tunnel cross section: the tunnel is designed to excavate a section perpendicular to the tunnel axis.
Construction mileage: the distance traveled by the shield machine from the originating location to the designated location.
Shield construction parameters: various parameters such as cutter head rotating speed and the like which need to be set in the construction process of the shield machine are set, and whether the safety and the quality of the construction of the shield method are reasonably determined by the parameter setting.
Attitude of the shield machine: the position relation of the shield machine axis relative to the tunnel design axis comprises horizontal deviation of a cutter head, vertical deviation of the cutter head, horizontal deviation of a shield tail and vertical deviation of the shield tail.
Horizontal deviation of cutterhead: the distance between the center of the cutter disc and the tunnel design axis at the same mileage is in the horizontal direction and faces the advancing direction of the shield tunneling machine, the center of the cutter disc is positive on the left side of the tunnel design axis, and the center of the cutter disc is negative on the right side of the tunnel design axis.
Vertical deviation of cutterhead: the distance between the center of the cutterhead and the tunnel design axis at the same mileage is in the horizontal direction and faces the advancing direction of the shield tunneling machine, the center of the cutterhead is positive above the tunnel design axis, and the center of the cutterhead is negative below the tunnel design axis.
Horizontal deviation of shield tail: the distance between the center of the shield tail at the same mileage and the tunnel design axis in the horizontal direction faces the advancing direction of the shield machine, and the center of the shield tail is positive on the left side of the tunnel design axis and negative on the right side of the tunnel design axis.
Vertical deviation of shield tail: the distance between the center of the shield tail and the tunnel design axis at the same mileage is in the horizontal direction, and the advancing direction of the shield machine is faced, wherein the center of the shield tail is positive above the tunnel design axis, and the center of the shield tail is negative below the tunnel design axis.
In order that the above-recited objects, features and advantages of the present invention will become more readily apparent, a more particular description of the invention will be rendered by reference to the appended drawings and appended detailed description.
Example 1:
referring to fig. 1, the invention provides a training method for a posture prediction model of a shield tunneling machine, which comprises the following steps:
s1: dividing shield construction history data according to stratum category combinations which simultaneously appear on a tunnel face by using a clustering algorithm to obtain n k0 Stratum combination sample sets corresponding to stratum category combinations of the individual categories; the shield construction history data comprise shield construction parameter data, shield machine attitude parameter data and stratum data;
S2: combining the formations of each classDividing the sample set into at least one interval sample set according to time continuity; each sample in the interval sample set comprises t corresponding to each other n Shield construction parameter data and t at moment n+1 Attitude parameter data of the shield tunneling machine at moment, wherein t is as follows n+1 The time is the t n The next time of the time;
s3: selecting part of samples in each interval sample set to form a training set, and using t as the n The shield construction parameter data at the moment is taken as an input parameter, and the t is taken as the input parameter n+1 The attitude parameter data of the shield machine at moment is taken as an output parameter, and a machine learning model is trained; forming a test set to test a trained machine learning model by utilizing another part of samples in the interval sample set, and calculating the accuracy of the trained machine learning model;
in the present embodiment, the first p of each interval is train % of samples as training set, post p test % of samples as test set, wherein p train :p test Between 3:1 and 4:1.
S4: and determining a final trained machine learning model as a shield tunneling machine attitude prediction model according to the accuracy corresponding to each interval sample set.
As shown in fig. 2 and 3, in step S1, shield construction history data is divided according to the combination of stratum types simultaneously appearing on the face by using a clustering algorithm to obtain n k0 The stratum combination sample set corresponding to the stratum category combination of each category specifically comprises the following steps:
s11: calling xlwt function packages to obtain shield construction parameter data, shield machine attitude parameter data and stratum data, and recording the data as shield construction history data; the shield construction parameter data and the shield machine attitude parameter data have corresponding data recording systems, such as Robotec systems. The shield construction parameter data comprises: tool parameters, jack parameters, working condition parameters and grouting parameters; the attitude parameter data of the shield tunneling machine comprises: attitude data of the cutterhead and the shield tail; the formation data includes: the combination of stratum types and stratum characteristic parameters which are simultaneously appeared on the face;
specifically, the cutter parameters include: cutter torque, cutter rotational speed and profile modeling cutter travel;
the jack parameters include: the pushing jack is divided into a pushing force and stroke and a hinged jack is divided into a dividing stroke;
the working condition parameters comprise: the rotating speed of the spiral soil outlet machine and the advancing speed of the shield machine;
the grouting parameters include: grouting pressure and grouting amount behind the wall;
the attitude data of the cutterhead and the shield tail comprise: the horizontal deviation of the cutterhead, the vertical deviation of the cutterhead, the horizontal deviation of the shield tail and the vertical deviation of the shield tail;
The stratum characteristic parameters comprise the thickness of each stratum type combination and each physical and mechanical parameter of each stratum type combination; the physical and mechanical parameters comprise cohesion, internal friction angle, horizontal side pressure coefficient, compression modulus, liquid limit and saturation. Wherein the formation type combinations and the corresponding physical and mechanical parameters are obtained from the geological survey report, and the thickness of each formation type combination is obtained from the borehole data according to interpolation.
Obtaining a stratum combination sample set according to the shield construction history data, wherein the obtaining of the stratum combination sample set specifically comprises the following steps:
s111: preprocessing the shield construction history data, and removing abnormal values and values of a stopping section to obtain shield construction reference data;
s112: obtaining a posture parameters of the shield machine according to the shield construction reference data, and marking the posture parameters asb shield construction parameters marked as +.>c + d formation parameters,wherein (1)>For the thickness of the combination of the various stratum types on the face, < > for each stratum type>Various physical and mechanical parameters combined for various formation types; i=1, 2 … n;
s113: combining the a shield machine attitude parameters, the b shield construction parameters and the c+d stratum parameters according to time, and simultaneously adding shield machine attitude parameter data at the next moment, wherein the shield machine attitude parameter data at the next moment are recorded as Obtaining the stratum combination sample set, wherein the stratum combination sample set is as follows:
the pretreatment comprises the following steps: noise reduction and structuring;
the noise reduction comprises abnormal value elimination and shutdown segment elimination;
the outlier rejection specifically includes the following steps:
a1: judging whether the data belongs to the value range of the corresponding parameter, if not, deleting the data;
a2: calculating the mean mu and variance sigma of each parameter;
a3: judging whether the data of each parameter at each moment belongs to a confidence interval [ mu-3 sigma, mu+3 sigma ], if not, deleting the data;
the removing of the shutdown section specifically comprises the following steps:
b1: judging whether the jack stroke is changed or not, if not, eliminating all data in the time period;
the structuring comprises the extraction and mapping of the shield construction parameter data, the extraction and mapping of the stratum data and normalization;
the extraction and mapping of the shield construction parameter data specifically comprises the following steps:
c1: acquiring a monitoring time interval delta t of the attitude parameter data of the shield machine p Each monitoring point corresponds to a time t i Shield tunneling machine posture at this momentA is the number of attitude parameters of the shield tunneling machine, i=1, 2 … n;
C2: selectingCalculating the average value of n shield construction parameters within the time range, wherein the average value of the n shield construction parameters is as follows:
b is the number of shield construction parameters;
and C3: taking the average value of the n shield construction parameters as the delta t p In the time interval t i Obtaining the representative value of the shield construction parameter data at the moment, and obtaining shield construction parameter data consistent with the attitude parameter data of the shield machine;
the extraction and mapping of the stratum data specifically comprises the following steps:
d1: various formation thickness distributions at each drilling location obtained by drilling;
d2: obtaining stratum thickness function h (h) corresponding to construction mileage x one by one through interpolation method 1 ,h 2 ,h 3 ,...,h n )＝f(x)，
D3: calculating the mileage of the shield machine at the moment of recording the posture of each shield machine;
d4: obtaining the thickness of each stratum type combination and each physical and mechanical parameter of each stratum type combination on the face of the shield tunneling machine at the moment through the mileage of the shield tunneling machineWherein c is the number of each physical and mechanical parameter of each stratum type combination, d is the number of the thickness of each stratum type combination on the face, i=1, 2 … n;
the normalization specifically comprises the following steps:
E1: according to the attitude of the shield machine, the shield construction parameters, the thickness on the face of the shield machine, various physical and mechanical parameters of each stratum and the attitude of the shield machine at the next momentObtaining a sample structure, wherein the sample structure is as follows:
wherein i=1, 2 … n;
e2: according to the sample structure, taking the parameter name of the sample structure as a column name, and taking the parameter value of the sample structure as a unit grid value to form a two-dimensional table;
e3: and according to the two-dimensional table, carrying out normalization operation on each column value of the two-dimensional table by using a maximum value-minimum value normalization method to obtain a stratum combination sample set stored in the table.
S12: calculating the total number n of stratum category combinations simultaneously appearing on the face s ；
S13: randomly selecting n in the shield construction history data s The individual samples are used as initial clustering centersSetting the maximum iteration number n max From minimum error variation epsilon min ；
S14: calculating the distance between each sample and the clustering center, and taking the category with the smallest distance as the category to which the sample belongs;
s15: for each categoryn is the total number of samples in the ith class, and a new cluster center of each class is recalculated, wherein the calculation method of the new cluster center is as follows:
S16: judging whether an iteration termination condition is reached, wherein the iteration termination condition is as follows: whether or not the maximum number of iterations n is reached max Or the cluster center change positions of all the categories between two iterations are smaller than the minimum error change epsilon min The method comprises the steps of carrying out a first treatment on the surface of the If not, taking the new cluster center as the cluster center of the category, and returning to execute the step S14; if yes, determining the category of all the obtained stratum category combinations and a stratum combination sample set corresponding to each category; the category is n k0 And each.
As shown in fig. 3, the stratum combination sample set of each category is divided into at least one interval sample set according to time continuity, and n is finally obtained k Intervals.
In step S3, selecting part of samples in each interval sample set to form a training set, and using t as the reference value n The shield construction parameter data at the moment is taken as an input parameter, and the t is taken as the input parameter n+1 The moment of time, the attitude parameter data of the shield machine is an output parameter, and the machine learning model is trained, specifically comprising:
s31: establishing a machine learning model for each interval sample set;
s32: training a corresponding machine learning model by utilizing a training set in the interval sample set;
s33: and adjusting the super parameters of the machine learning model by using a grid method in the training process to obtain the trained machine learning model.
The machine learning model at least comprises 2 algorithm models, the training set respectively trains at least 2 algorithm models, and the testing set respectively tests at least 2 trained algorithm models.
As shown in fig. 4, the machine learning model in this embodiment is 3 algorithm models, namely, an MLP model, an SVM model, and a GBR model. The machine learning algorithm modeling of the present invention is implemented in the python language.
The specific process of MLP model training is as follows:
1) Calling xlwt function package to read stratum combination sample set stored in the table;
2) Calling a PCA algorithm in a sklearn function package, instantiating the stored characteristics, and reducing the dimension of the input parameters of the stratum combination sample set to a value of more than 90; the input parameters include: a shield tunneling machine attitude parameter, b shield construction parameters and c+d stratum parameters;
3) Invoking the MLPRegressor algorithm in sklearn, corresponding to n k Establishing n for each interval k A plurality of MLP models;
4) Training the corresponding MLP model by using the training set of each interval in sequence, calculating the accuracy of the MLP model by using the corresponding test set, selecting the superparameter by using a grid method, and obtaining the MLP model with the highest accuracy and the accuracy on each interval by mainly adjusting the superparameter to be "hiden_layer_sizes", "activation", "solver", "learning_rate" and "displa", and recommending the value to be "hiden_layer_sizes= (50, 100), activation= 'logistic', and software= 'sgd', learning_rate= 'adaptive', alpha=0.01".
The specific flow of SVM model training is as follows:
1) Calling xlwt function package to read stratum combination sample set stored in the table;
2) Calling a PCA algorithm in a sklearn function package, instantiating the stored characteristics, and reducing the dimension of the input parameters of the stratum combination sample set to a value of more than 90; the input parameters include: a shield tunneling machine attitude parameter, b shield construction parameters and c+d stratum parameters;
3) Invoking the SVM algorithm in sklearn, corresponding to n k Establishing n for each interval k A plurality of SVM models;
4) Training the corresponding SVM model by using the training set of each interval in sequence, calculating the accuracy of the SVM model by using the corresponding testing set, selecting the super parameters by using a grid method, and recommending the super parameters with main adjustment of gamma, kernel and C to take the values of gamma= ' auto ', kernel= ' linear ', C=1 ' to obtain the SVM model with highest accuracy and the accuracy of each interval.
The specific flow of the GBR model training is as follows:
1) Calling xlwt function package to read stratum combination sample set stored in the table;
2) Calling a PCA algorithm in a sklearn function package, instantiating the stored characteristics, and reducing the dimension of the input parameters of the stratum combination sample set to a value of more than 90; the input parameters include: a shield tunneling machine attitude parameter, b shield construction parameters and c+d stratum parameters;
3) Invoking the GBR algorithm in sklearn, corresponding to n k Establishing n for each interval k A plurality of GBR models;
4) Training the corresponding GBR model by sequentially using a training set of each interval, calculating the accuracy of the GBR model by using a corresponding test set, selecting the super-parameters by using a grid method, wherein the super-parameters mainly adjusted are 'min_samples_split', 'min_samples_leaf', 'subsamples', 'n_evators', 'loss', 'learning_rate' and 'max_depth', and the recommended values are 'min_samples_split=4, min_samples_leaf=2, subsamples=0.5, n_evators=500, loss=' lad ', learning_rate=0.04 and max_depth=6', so as to obtain the GBR model with the highest accuracy and the accuracy on each interval.
As shown in fig. 4, in step S4, a final trained machine learning model is determined as a shield tunneling machine posture prediction model according to the accuracy corresponding to each interval sample set, and specifically includes the following steps:
s41: calculating the average value of the accuracy of each trained algorithm model corresponding to all the interval sample sets of the same category, and determining the trained algorithm model corresponding to the interval sample set with the highest accuracy as a category target model;
S42: calculating model weights in the same category by using the average value of the accuracy of the various trained algorithm models; and distributing the category target model according to the model weight to obtain a shield tunneling machine attitude prediction model corresponding to the category.
In the same category, model weights are calculated by means of average values of accuracy of the various trained algorithm models, so that training results can be more accurate, objective absolute factors can be eliminated, and a foundation is laid for obtaining a shield tunneling machine posture prediction model subsequently.
In step S42, the step of obtaining the attitude prediction model of the shield tunneling machine specifically includes:
s421: respectively calculating an average value m of accuracy rates of predicting all interval sample sets of the same class by using an MLP model, an SVM model and a GBR model j ，s j And g j ；
S422: according to the average value m j ，s j And g j Calculating model weightsAnd->
S423: and summing the MLP model, the SVM model and the GBR model according to the model weights to form a result weight distribution matrix of the three models, wherein the row names of the matrix are stratum type combinations, the column names of the matrix are model algorithms, and a shield tunneling machine attitude prediction model is obtained according to the weight distribution matrix.
So far, the shield posture prediction model based on the integrated learning is already built, and it is noted that the embodiment only introduces three algorithms of the MLP model, the SVM model and the GBR model, and the machine learning method has a plurality of methods, and can be added into other various methods for the integrated learning.
Example 2:
referring to fig. 5, the invention provides a method for optimizing the posture of a shield tunneling machine, which comprises the following steps:
1) Inputting the shield construction parameters at the current moment into a shield machine attitude prediction model to obtain shield machine attitude parameter data at the next moment; the shield machine posture prediction model is obtained by the shield machine posture prediction model training method described in embodiment 1, and is not described in detail herein;
2) Judging whether the attitude parameter data of the shield machine at the next moment exceeds a preset deviation limit value or not;
if yes, the shield construction parameters are adjusted to serve as new shield construction parameters at the current moment, and the execution of 'inputting the shield construction parameters at the current moment into a shield machine attitude prediction model' is returned;
and if not, controlling the action of the shield machine according to the attitude parameter data of the shield machine at the next moment, and storing the attitude parameter data of the shield machine at the next moment into the shield construction parameters.
When in practical application, firstly, setting the attitude deviation limit value of the shield machine, wherein the specification is regulated to be +/-50 mm, and the specification can be generally selected to be +/-30 mm; secondly, selecting construction parameters which are frequently regulated by a driver in the shield construction process as parameters to be regulated, wherein the parameters comprise: pushing jack force and stroke, cutter disc rotating speed and torque, hinging jack stroke, grouting amount and pushing speed. Thirdly, taking the construction parameter at the current moment as an initial value of construction parameter optimization, inputting an integrated learning prediction model, and obtaining a predicted value of the shield machine posture at the next moment; and finally, judging whether the predicted value of the posture of the shield machine at the next moment exceeds a deviation limit value, if not, taking the construction parameter as the construction parameter value at the current moment to control the action of the shield machine, and if so, sequentially adjusting the construction parameters frequently adjusted by a shield driver, and judging once by inputting the model every time the construction parameters are adjusted until the posture of the shield machine at the next moment meets the requirement or the construction parameters exceed the value range.
Example 3:
the invention provides a training system for a posture prediction model of a shield tunneling machine, which comprises the following components:
stratum type combination dividing module 1 for simultaneously outputting shield construction history data according to the face by using a clustering algorithm Dividing the existing stratum types to obtain n k0 Stratum combination sample sets corresponding to stratum category combinations of the individual categories; the shield construction history data comprise shield construction parameter data, shield machine attitude parameter data and stratum data;
the interval sample set dividing module 2 is used for dividing the stratum combination sample set of each category into at least one interval sample set according to time continuity; each sample in the interval sample set comprises t corresponding to each other n Shield construction parameter data and t at moment n+1 Attitude parameter data of the shield tunneling machine at moment, wherein t is as follows n+1 The time is the t n The next time of the time;
a machine learning model training module 3 for selecting part of the samples in each interval sample set to form a training set, and using t as the n The shield construction parameter data at the moment is taken as an input parameter, and the t is taken as the input parameter n+1 The attitude parameter data of the shield machine at moment is taken as an output parameter, and a machine learning model is trained; forming a test set to test a trained machine learning model by utilizing another part of samples in the interval sample set, and calculating the accuracy of the trained machine learning model;
and the shield machine posture prediction model building module 4 is used for determining a final trained machine learning model as a shield machine posture prediction model according to the accuracy corresponding to each interval sample set.
Example 4:
the invention provides a shield tunneling machine attitude optimization system, which comprises:
the shield machine attitude parameter data acquisition module 5 is used for inputting the shield construction parameters at the current moment into the shield machine attitude prediction model to obtain shield machine attitude parameter data at the next moment;
the judging module 6 is used for judging whether the attitude parameter data of the shield machine at the next moment exceeds a preset deviation limit value;
if yes, the shield construction parameters are adjusted to serve as new shield construction parameters at the current moment, and the execution of 'inputting the shield construction parameters at the current moment into a shield machine attitude prediction model' is returned;
and if not, controlling the action of the shield machine according to the attitude parameter data of the shield machine at the next moment, and storing the attitude parameter data of the shield machine at the next moment into the shield construction parameters.
In summary, the method can classify the stratum type combination according to the shield construction history data by using the clustering method, thereby providing a basis for the segment modeling of various subsequent machine learning methods. According to the segmentation result, modeling is carried out by using a plurality of machine learning models, and weight distribution is carried out according to the accuracy, so that an integrated learning model suitable for various stratum types and combinations can be obtained. The integrated learning model can allocate higher weight for the machine learning model with higher accuracy aiming at a specified stratum, so that the machine learning model with higher accuracy is different in machine learning model adopted by different stratum types, the phenomenon of inaccurate attitude prediction of the shield tunneling machine caused by stratum mutation is avoided, and the historical data is utilized more fully.
The principles and embodiments of the present invention have been described herein with reference to specific examples, the description of which is intended only to assist in understanding the methods of the present invention and the core ideas thereof; also, it is within the scope of the present invention to be modified by those of ordinary skill in the art in light of the present teachings. In view of the foregoing, this description should not be construed as limiting the invention.
Claims (8)
1. The method for training the attitude prediction model of the shield tunneling machine is characterized by comprising the following steps of:
dividing shield construction history data according to stratum category combinations which simultaneously appear on a tunnel face by using a clustering algorithm to obtain n k0 Stratum combination sample sets corresponding to stratum category combinations of the individual categories; the shield construction history data comprise shield construction parameter data, shield machine attitude parameter data and stratum data;
dividing the stratigraphic combination sample set of each category into at least one in time continuityA set of interval samples; each sample in the interval sample set comprises t corresponding to each other n Shield construction parameter data and t at moment n+1 Attitude parameter data of the shield tunneling machine at moment, wherein t is as follows n+1 The time is the t n The next time of the time;
selecting part of samples in each interval sample set to form a training set, and using t as the n The shield construction parameter data at the moment is taken as an input parameter, and the t is taken as the input parameter n+1 The attitude parameter data of the shield machine at moment is taken as an output parameter, and a machine learning model is trained; forming a test set to test a trained machine learning model by utilizing another part of samples in the interval sample set, and calculating the accuracy of the trained machine learning model;
determining a final trained machine learning model as a shield tunneling machine attitude prediction model according to the accuracy rate corresponding to each interval sample set;
the machine learning model at least comprises 2 algorithm models, the training set respectively trains at least 2 algorithm models, and the test set respectively tests at least 2 trained algorithm models;
determining a final trained machine learning model as a shield tunneling machine attitude prediction model according to the accuracy corresponding to each interval sample set, wherein the method specifically comprises the following steps of:
calculating the average value of the accuracy of each trained algorithm model corresponding to all the interval sample sets of the same category, and determining the trained algorithm model corresponding to the interval sample set with the highest accuracy as a category target model;
Calculating model weights in the same category by using the average value of the accuracy of the various trained algorithm models; and distributing the category target model according to the model weight to obtain a shield tunneling machine attitude prediction model corresponding to the category.
2. The method for training a model for predicting the posture of a shield tunneling machine according to claim 1, wherein the history data of shield construction is obtainedDividing according to stratum category combinations which simultaneously appear on the face by using a clustering algorithm to obtain n k0 The stratum combination sample set corresponding to the stratum category combination of each category specifically comprises:
acquiring shield construction parameter data, shield machine attitude parameter data and stratum data, and recording the data as shield construction history data; the shield construction parameter data comprises: tool parameters, jack parameters, working condition parameters and grouting parameters; the attitude parameter data of the shield tunneling machine comprises: attitude data of the cutterhead and the shield tail; the formation data includes: the combination of stratum types and stratum characteristic parameters which are simultaneously appeared on the face;
calculating the total number n of stratum category combinations simultaneously appearing on the face s ；
Randomly selecting n in the shield construction history data s The individual samples are used as initial clustering centers Setting the maximum iteration number n max From minimum error variation epsilon min ；
Calculating the distance between each sample and the clustering center, and taking the category with the smallest distance as the category to which the sample belongs;
recalculating a new cluster center for each category;
judging whether an iteration termination condition is reached, if not, taking a new cluster center as the cluster center of the category, and returning to execute 'calculating the distance between each sample and the cluster center'; if yes, determining the category of all the obtained stratum category combinations and a stratum combination sample set corresponding to each category; the category is n k0 And each.
3. The method for training a model for predicting the attitude of a shield tunneling machine according to claim 1, wherein said selecting a portion of samples in each of said interval sample sets forms a training set, and said t n The shield construction parameter data at the moment is taken as an input parameter, and the t is taken as the input parameter n+1 The attitude parameter data of the shield machine at moment is taken as an output parameter, and the machine learning model is adoptedThe model is used for training and specifically comprises the following steps:
establishing a machine learning model for each interval sample set;
training a corresponding machine learning model by utilizing a training set in the interval sample set;
and adjusting the super parameters of the machine learning model by using a grid method in the training process to obtain the trained machine learning model.
4. The method for training a posture prediction model of a shield machine according to claim 1, wherein the machine learning model includes an MLP model, an SVM model and a GBR model, and the step of obtaining the posture prediction model of the shield machine specifically includes:
respectively calculating an average value m of accuracy rates of predicting all interval sample sets of the same class by using an MLP model, an SVM model and a GBR model j ，s j And g j ；
According to the average value m j ，s j And g j Calculating model weightsAnd->
And summing the MLP model, the SVM model and the GBR model according to the model weight to obtain a shield tunneling machine attitude prediction model.
5. The method for training the attitude prediction model of the shield tunneling machine according to claim 1, wherein the obtaining of the stratum combination sample set specifically comprises:
preprocessing the shield construction history data, and removing abnormal values and values of a stopping section to obtain shield construction reference data;
obtaining a posture parameters of the shield machine according to the shield construction reference data, and marking the posture parameters asb shield construction parameters marked as +.>c+d stratum parameters, < >>Wherein (1)>For the thickness of the combination of the various stratum types on the face, < > for each stratum type>Various physical and mechanical parameters combined for various formation types; i=1, 2 … n;
Combining the a shield machine attitude parameters, the b shield construction parameters and the c+d stratum parameters according to time, and simultaneously adding shield machine attitude parameter data at the next moment, wherein the shield machine attitude parameter data at the next moment are recorded asObtaining the stratum combination sample set, wherein the stratum combination sample set is as follows:
6. the method for optimizing the posture of the shield tunneling machine is characterized by comprising the following steps of:
inputting the shield construction parameters at the current moment into a shield machine attitude prediction model to obtain shield machine attitude parameter data at the next moment; the shield machine posture prediction model is obtained by the shield machine posture prediction model training method according to claim 1;
judging whether the attitude parameter data of the shield machine at the next moment exceeds a preset deviation limit value or not;
if yes, the shield construction parameters are adjusted to serve as new shield construction parameters at the current moment, and the execution of 'inputting the shield construction parameters at the current moment into a shield machine attitude prediction model' is returned;
and if not, controlling the action of the shield machine according to the attitude parameter data of the shield machine at the next moment, and storing the attitude parameter data of the shield machine at the next moment into the shield construction parameters.
7. The utility model provides a shield constructs quick-witted gesture prediction model training system which characterized in that includes:
the stratum type combination dividing module is used for dividing shield construction history data according to stratum type combinations which simultaneously appear on the face by using a clustering algorithm to obtain n k0 Stratum combination sample sets corresponding to stratum category combinations of the individual categories; the shield construction history data comprise shield construction parameter data, shield machine attitude parameter data and stratum data;
the interval sample set dividing module is used for dividing the stratum combination sample set of each category into at least one interval sample set according to time continuity; each sample in the interval sample set comprises t corresponding to each other n Shield construction parameter data and t at moment n+1 Attitude parameter data of the shield tunneling machine at moment, wherein t is as follows n+1 The time is the t n The next time of the time;
a machine learning model training module for selecting part of samples in each interval sample set to form a training set, and using t as the n The shield construction parameter data at the moment is taken as an input parameter, and the t is taken as the input parameter n+1 The attitude parameter data of the shield machine at moment is taken as an output parameter, and a machine learning model is trained; forming a test set to test a trained machine learning model by utilizing another part of samples in the interval sample set, and calculating the accuracy of the trained machine learning model;
The shield machine attitude prediction model building module is used for determining a final trained machine learning model as a shield machine attitude prediction model according to the accuracy rates corresponding to the interval sample sets;
the machine learning model at least comprises 2 algorithm models, the training set respectively trains at least 2 algorithm models, and the test set respectively tests at least 2 trained algorithm models;
determining a final trained machine learning model as a shield tunneling machine attitude prediction model according to the accuracy corresponding to each interval sample set, wherein the method specifically comprises the following steps of:
calculating the average value of the accuracy of each trained algorithm model corresponding to all the interval sample sets of the same category, and determining the trained algorithm model corresponding to the interval sample set with the highest accuracy as a category target model;
calculating model weights in the same category by using the average value of the accuracy of the various trained algorithm models; and distributing the category target model according to the model weight to obtain a shield tunneling machine attitude prediction model corresponding to the category.
8. The utility model provides a shield constructs quick-witted gesture optimizing system which characterized in that includes:
The system comprises a shield machine attitude parameter data acquisition module, a shield machine attitude parameter prediction module and a control module, wherein the shield machine attitude parameter data acquisition module is used for inputting shield construction parameters at the current moment into a shield machine attitude prediction model to obtain shield machine attitude parameter data at the next moment; the shield machine posture prediction model is obtained by the shield machine posture prediction model training method according to claim 1;
the judging module is used for judging whether the attitude parameter data of the shield machine at the next moment exceeds a preset deviation limit value;
if yes, the shield construction parameters are adjusted to serve as new shield construction parameters at the current moment, and the execution of 'inputting the shield construction parameters at the current moment into a shield machine attitude prediction model' is returned;
and if not, controlling the action of the shield machine according to the attitude parameter data of the shield machine at the next moment, and storing the attitude parameter data of the shield machine at the next moment into the shield construction parameters.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110914150.3A CN113569327B (en) | 2021-08-10 | 2021-08-10 | Method and system for optimizing posture of shield tunneling machine and method and system for training model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110914150.3A CN113569327B (en) | 2021-08-10 | 2021-08-10 | Method and system for optimizing posture of shield tunneling machine and method and system for training model |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113569327A CN113569327A (en) | 2021-10-29 |
| CN113569327B true CN113569327B (en) | 2024-02-09 |
Family
ID=78171113
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110914150.3A Active CN113569327B (en) | 2021-08-10 | 2021-08-10 | Method and system for optimizing posture of shield tunneling machine and method and system for training model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113569327B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2019192172A1 (en) * | 2018-04-04 | 2019-10-10 | 歌尔股份有限公司 | Attitude prediction method and apparatus, and electronic device |
| CN110533065A (en) * | 2019-07-18 | 2019-12-03 | 西安电子科技大学 | Based on the shield attitude prediction technique from coding characteristic and deep learning regression model |
| CN112879024A (en) * | 2021-01-23 | 2021-06-01 | 西安建筑科技大学 | Dynamic prediction method, system and equipment for shield attitude |
-
2021
- 2021-08-10 CN CN202110914150.3A patent/CN113569327B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2019192172A1 (en) * | 2018-04-04 | 2019-10-10 | 歌尔股份有限公司 | Attitude prediction method and apparatus, and electronic device |
| CN110533065A (en) * | 2019-07-18 | 2019-12-03 | 西安电子科技大学 | Based on the shield attitude prediction technique from coding characteristic and deep learning regression model |
| CN112879024A (en) * | 2021-01-23 | 2021-06-01 | 西安建筑科技大学 | Dynamic prediction method, system and equipment for shield attitude |
Non-Patent Citations (2)
| Title |
|---|
| 基于BP神经网络的盾构机姿态与轨迹控制研究;丁海英;;机械设计与制造工程(12);全文 * |
| 基于机器学习的地铁隧道施工扰动控制研究;王鹏;;现代隧道技术(S2);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113569327A (en) | 2021-10-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| RU2594372C1 (en) | Propagating update of fracturing plane | |
| WO2019042483A2 (en) | Tbm boring rock state real-time sensing system and method | |
| CN110617074B (en) | Incidence relation method for ground settlement and tunneling parameters in shield construction | |
| CN108868805A (en) | Shield method for correcting error based on statistical analysis in conjunction with XGboost | |
| Yan et al. | Identification of geological characteristics from construction parameters during shield tunnelling | |
| CN112364422B (en) | MIC-LSTM-based dynamic prediction method for shield construction earth surface deformation | |
| CN110630274B (en) | Shield tunnel composite stratum section implementation system and implementation method | |
| CN116522692B (en) | Underground engineering surrounding rock structural feature in-situ detection and classification method | |
| CN113298220A (en) | Neural network optimization-based shield tunneling machine tunneling speed prediction method | |
| CN112901183A (en) | Method for determining geological features of shield construction process based on machine learning | |
| CN113569327B (en) | Method and system for optimizing posture of shield tunneling machine and method and system for training model | |
| CN110359841A (en) | The track of horizontal well determines method, the Drilling Control method and apparatus of horizontal well | |
| CN116432855A (en) | Tunnel collapse condition prediction method based on tunneling data | |
| CN115898442A (en) | Method and system for determining grade of surrounding rock in tunneling based on shield construction parameters | |
| CN115239108B (en) | Weak broken surrounding rock sensing method based on TBM real-time broken rock data | |
| CN115640759A (en) | Drill jamming early warning method and system based on machine learning | |
| CN115640996A (en) | Evaluation method for water inrush disaster of tunnel in water-rich complex stratum | |
| CN104462792B (en) | Log data lithological stratum numerical value reduction method | |
| Unterlass et al. | Identifying rock loads on TBM shields during standstills (non-advance-periods) | |
| US11216700B2 (en) | Automated material classification by structural features | |
| CN115324557A (en) | Method for predicting deformation risk degree of fracturing-induced casing based on multi-factor analysis | |
| CA2943189C (en) | Bin constraints for generating a histogram of microseismic data | |
| CN114075979A (en) | Real-time identification system and method for shield tunneling environment change | |
| CN107060747B (en) | Early warning method and system for crack development zone during drilling | |
| CN110728438B (en) | Stock ground reserve analysis and evaluation method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |