CN117473689A - Intelligent decision-making method and device for self-adaptive rock lithology and grade TBM tunneling - Google Patents
Intelligent decision-making method and device for self-adaptive rock lithology and grade TBM tunneling Download PDFInfo
- Publication number
- CN117473689A CN117473689A CN202210849366.0A CN202210849366A CN117473689A CN 117473689 A CN117473689 A CN 117473689A CN 202210849366 A CN202210849366 A CN 202210849366A CN 117473689 A CN117473689 A CN 117473689A
- Authority
- CN
- China
- Prior art keywords
- lithology
- data
- tbm tunneling
- surrounding rock
- model
- 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.)
- Pending
Links
- 230000005641 tunneling Effects 0.000 title claims abstract description 347
- 239000011435 rock Substances 0.000 title claims abstract description 279
- 238000000034 method Methods 0.000 title claims abstract description 81
- 238000007781 pre-processing Methods 0.000 claims abstract description 20
- 238000012549 training Methods 0.000 claims description 147
- 238000012545 processing Methods 0.000 claims description 31
- 238000000605 extraction Methods 0.000 claims description 19
- 238000004590 computer program Methods 0.000 claims description 17
- 230000009467 reduction Effects 0.000 claims description 14
- 238000003860 storage Methods 0.000 claims description 10
- 208000022971 Tuberculous meningitis Diseases 0.000 description 291
- 208000001223 meningeal tuberculosis Diseases 0.000 description 291
- 230000003044 adaptive effect Effects 0.000 description 27
- 230000008569 process Effects 0.000 description 19
- 230000006870 function Effects 0.000 description 9
- 238000010586 diagram Methods 0.000 description 8
- 238000010276 construction Methods 0.000 description 7
- 238000011217 control strategy Methods 0.000 description 7
- 230000002159 abnormal effect Effects 0.000 description 6
- 230000008859 change Effects 0.000 description 6
- 230000000694 effects Effects 0.000 description 6
- 238000004891 communication Methods 0.000 description 5
- 238000013135 deep learning Methods 0.000 description 5
- 239000011159 matrix material Substances 0.000 description 5
- 238000001514 detection method Methods 0.000 description 4
- 230000002194 synthesizing effect Effects 0.000 description 4
- 230000004913 activation Effects 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 3
- 230000015572 biosynthetic process Effects 0.000 description 3
- 238000004422 calculation algorithm Methods 0.000 description 3
- 238000003062 neural network model Methods 0.000 description 3
- 238000012706 support-vector machine Methods 0.000 description 3
- 238000003786 synthesis reaction Methods 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000013527 convolutional neural network Methods 0.000 description 2
- 238000012937 correction Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000005553 drilling Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000004927 fusion Effects 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000035515 penetration Effects 0.000 description 2
- ORILYTVJVMAKLC-UHFFFAOYSA-N Adamantane Natural products C1C(C2)CC3CC1CC2C3 ORILYTVJVMAKLC-UHFFFAOYSA-N 0.000 description 1
- 238000012935 Averaging Methods 0.000 description 1
- 238000010220 Pearson correlation analysis Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000001174 ascending effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 238000013075 data extraction Methods 0.000 description 1
- 238000013079 data visualisation Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000008034 disappearance Effects 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 238000003064 k means clustering Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003204 osmotic effect Effects 0.000 description 1
- 230000001568 sexual effect Effects 0.000 description 1
- 238000010008 shearing Methods 0.000 description 1
- 239000002893 slag Substances 0.000 description 1
- 230000000087 stabilizing effect Effects 0.000 description 1
- 238000011425 standardization method Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Classifications
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0637—Strategic management or analysis, e.g. setting a goal or target of an organisation; Planning actions based on goals; Analysis or evaluation of effectiveness of goals
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/02—Agriculture; Fishing; Forestry; Mining
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- Tourism & Hospitality (AREA)
- Educational Administration (AREA)
- Marketing (AREA)
- Entrepreneurship & Innovation (AREA)
- General Business, Economics & Management (AREA)
- Game Theory and Decision Science (AREA)
- Agronomy & Crop Science (AREA)
- General Engineering & Computer Science (AREA)
- Geometry (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Evolutionary Computation (AREA)
- Computer Hardware Design (AREA)
- Life Sciences & Earth Sciences (AREA)
- Development Economics (AREA)
- Animal Husbandry (AREA)
- Marine Sciences & Fisheries (AREA)
- Mining & Mineral Resources (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides an intelligent decision-making method and device for self-adaptive rock lithology and grade TBM tunneling, wherein the method comprises the following steps: acquiring TBM tunneling related data; preprocessing TBM tunneling related data to obtain TBM tunneling characteristic data; acquiring surrounding rock grades corresponding to the TBM tunneling related data; obtaining a decision model according to the surrounding rock grade and the intelligent decision model library; and predicting TBM tunneling parameters according to the TBM tunneling characteristic data and the decision model. The device is used for executing the method. The intelligent decision method and the intelligent decision device for self-adaptive rock mass lithology and grade TBM tunneling can improve the accuracy of TBM tunneling parameter prediction.
Description
Technical Field
The invention relates to the technical field of artificial intelligence, in particular to a self-adaptive rock lithology and grade TBM tunneling intelligent decision method and device.
Background
The rock tunnel boring machine (Tunnel Boring Machine, TBM for short) is a machine for tunnel construction, and is widely applied to industries such as water conservancy, hydropower, urban construction, traffic and the like by crushing rock through strong thrust and shearing force of the machine.
In the tunneling process of the TBM, a driver needs to adjust equipment tunneling parameters in real time according to geological conditions of the rock mass. However, the traditional control strategy mainly carries out artificial judgment according to the driving experience of a TBM driver and the fluctuation condition of tunneling parameters, but the mode is easily influenced by the artificial subjectivity and the singleness of a decision rule, so that the parameter adjustment and the actual value generate larger deviation, the decision error of the TBM control strategy is caused, and the safety and the reliability of a man-machine are seriously threatened.
Disclosure of Invention
Aiming at the problems in the prior art, the embodiment of the invention provides an intelligent decision method and device for self-adaptive rock lithology and grade TBM tunneling, which can at least partially solve the problems in the prior art.
In a first aspect, the invention provides an intelligent decision-making method for self-adaptive rock lithology and grade TBM tunneling, which comprises the following steps:
acquiring TBM tunneling related data;
preprocessing TBM tunneling related data to obtain TBM tunneling characteristic data;
acquiring surrounding rock grades corresponding to the TBM tunneling related data;
obtaining a decision model corresponding to the TBM tunneling related data according to the surrounding rock grade corresponding to the TBM tunneling related data and an intelligent decision model library;
Predicting TBM tunneling parameters according to the TBM tunneling characteristic data and a decision model corresponding to the TBM tunneling related data; the decision model corresponding to the TBM tunneling related data is obtained through pre-training.
Further, training the decision model corresponding to the TBM tunneling related data comprises:
acquiring TBM tunneling historical data corresponding to various surrounding rock grades;
extracting key parameters of TBM tunneling historical data corresponding to each surrounding rock grade, and obtaining TBM tunneling training data corresponding to each surrounding rock grade;
and training to obtain a decision model corresponding to the TBM tunneling related data according to TBM tunneling training data corresponding to each surrounding rock grade and the initial model.
Further, the acquiring the surrounding rock grade corresponding to the TBM tunneling related data includes:
acquiring TBM tunneling data based on the rock-machine data;
according to TBM tunneling data and lithology prediction models, lithology corresponding to TBM tunneling related data is obtained; wherein the lithology prediction model is obtained by pre-training;
acquiring surrounding rock grades corresponding to TBM tunneling related data according to TBM tunneling data and a surrounding rock grade prediction model of lithology corresponding to TBM tunneling related data; the surrounding rock grade prediction model is obtained through pre-training.
Further, the step of training the lithology prediction model comprises:
acquiring rock-machine historical data and corresponding lithology;
according to the rock-machine history data and the corresponding lithology, lithology prediction training data are obtained;
respectively training to obtain a plurality of lithology prediction sub-models according to the lithology prediction training data and a plurality of original models;
obtaining a prediction error of each lithology predictor model according to the evidence set and each lithology predictor model; the evidence set is obtained based on random extraction of lithology prediction training data;
determining a weight coefficient of each lithology predictor model according to the prediction error of each lithology predictor model;
and obtaining the lithology prediction model according to each lithology prediction sub-model and the corresponding weight coefficient.
Further, the step of training the lithology prediction model comprises:
acquiring rock-machine historical data and corresponding lithology;
according to the rock-machine history data and the corresponding lithology, lithology prediction training data are obtained;
respectively training to obtain a plurality of lithology prediction sub-models according to the lithology prediction training data and a plurality of original models;
according to the evidence set and each lithology predictor model, obtaining a basic probability value of each lithology predictor model; the evidence set is obtained based on random extraction of lithology prediction training data;
Obtaining a weight coefficient of each lithology predictor model based on the basic probability value of the credibility factor for each lithology predictor model; wherein the confidence factor is obtained based on the evidence set;
and obtaining the lithology prediction model according to each lithology prediction sub-model and the corresponding weight coefficient.
Further, the step of training the surrounding rock grade prediction model includes:
acquiring rock-machine historical data of the same lithology and corresponding surrounding rock grades;
according to the rock-machine history data of the same lithology and the corresponding surrounding rock grade, obtaining surrounding rock grade prediction training data;
respectively training to obtain a plurality of surrounding rock grade predictor models according to the surrounding rock prediction training data and a plurality of original models;
according to the evidence set and each surrounding rock predictor model, obtaining a basic probability value of each surrounding rock grade predictor model; the evidence set is obtained based on random extraction of surrounding rock prediction training data;
based on the credibility factors and the basic probability value of each surrounding rock grade predictor model, obtaining the weight coefficient of each surrounding rock grade predictor model; wherein the confidence factor is obtained based on the evidence set;
And obtaining the surrounding rock grade prediction model according to each surrounding rock grade prediction sub-model and the corresponding weight coefficient.
Further, the step of training the surrounding rock grade prediction model includes:
acquiring rock-machine historical data of the same lithology and corresponding surrounding rock grades;
according to the rock-machine history data of the same lithology and the corresponding surrounding rock grade, obtaining surrounding rock grade prediction training data;
respectively training to obtain a plurality of surrounding rock grade predictor models according to the surrounding rock prediction training data and a plurality of original models;
according to the evidence set and each surrounding rock predictor model, obtaining a prediction error of each surrounding rock grade predictor model; the evidence set is obtained based on random extraction of lithology prediction training data;
determining a weight coefficient of each surrounding rock grade predictor model according to the prediction error of each surrounding rock grade predictor model;
and obtaining the surrounding rock grade prediction model according to each surrounding rock grade prediction sub-model and the corresponding weight coefficient.
Further, the preprocessing the TBM tunneling related data to obtain TBM tunneling characteristic data includes:
and carrying out outlier processing, noise reduction processing and standardization processing on the TBM tunneling related data.
In a second aspect, the invention provides an intelligent decision device for adaptive rock mass lithology and grade TBM tunneling, comprising:
the first acquisition unit is used for acquiring TBM tunneling related data;
the pretreatment unit is used for carrying out pretreatment on TBM tunneling related data to obtain TBM tunneling characteristic data;
the second acquisition unit is used for acquiring surrounding rock grades corresponding to the TBM tunneling related data;
the obtaining unit is used for obtaining a decision model corresponding to the TBM tunneling related data according to the surrounding rock grade corresponding to the TBM tunneling related data and an intelligent decision model library;
the prediction unit is used for predicting TBM tunneling parameters according to the TBM tunneling characteristic data and a decision model corresponding to the TBM tunneling related data; the decision model corresponding to the TBM tunneling related data is obtained through pre-training.
In a third aspect, the present invention provides an electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the steps of the adaptive rock mass lithology and grade TBM tunneling intelligent decision method of any of the embodiments described above when the program is executed.
In a fourth aspect, the present invention provides a computer readable storage medium having stored thereon a computer program which when executed by a processor performs the steps of the adaptive rock mass lithology and grade TBM tunneling intelligent decision method of any of the embodiments described above.
According to the intelligent decision method and device for self-adaptive rock mass lithology and grade TBM tunneling, TBM tunneling related data are obtained, preprocessing is carried out on the TBM tunneling related data to obtain TBM tunneling characteristic data, surrounding rock grades corresponding to the TBM tunneling related data are obtained, decision models corresponding to the TBM tunneling related data are obtained according to the surrounding rock grades corresponding to the TBM tunneling related data and an intelligent decision model library, TBM tunneling parameters are predicted according to the TBM tunneling characteristic data and the decision models corresponding to the TBM tunneling related data, and accuracy of TBM tunneling parameter prediction can be improved.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art. In the drawings:
fig. 1 is a flow chart of an intelligent decision-making method for adaptive rock mass lithology and grade TBM tunneling provided by a first embodiment of the present invention.
Fig. 2 is a flow chart of an intelligent decision-making method for adaptive rock mass lithology and grade TBM tunneling according to a second embodiment of the present invention.
Fig. 3 is a flow chart of an intelligent decision-making method for adaptive rock mass lithology and grade TBM tunneling according to a third embodiment of the present invention.
Fig. 4 is a flow chart of an intelligent decision-making method for adaptive rock mass lithology and grade TBM tunneling according to a fourth embodiment of the present invention.
Fig. 5 is a flow chart of an intelligent decision-making method for adaptive rock mass lithology and grade TBM tunneling according to a fifth embodiment of the present invention.
Fig. 6 is a flow chart of an intelligent decision-making method for adaptive rock mass lithology and grade TBM tunneling according to a sixth embodiment of the present invention.
Fig. 7 is a flow chart of an intelligent decision making method for adaptive rock mass lithology and grade TBM tunneling according to a seventh embodiment of the present invention.
Fig. 8 is a schematic structural diagram of an intelligent decision making device for adaptive rock mass lithology and grade TBM tunneling according to an eighth embodiment of the present invention.
Fig. 9 is a schematic physical structure of an electronic device according to a ninth embodiment of the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the embodiments of the present invention will be described in further detail with reference to the accompanying drawings. The exemplary embodiments of the present invention and their descriptions herein are for the purpose of explaining the present invention, but are not to be construed as limiting the invention. It should be noted that, in the case of no conflict, the embodiments and features in the embodiments may be arbitrarily combined with each other.
In order to facilitate understanding of the technical solutions provided in the present application, the following description will first explain relevant content of the technical solutions of the present application.
Considering that the tunneling targets concerned by TBM construction under different geological conditions have differences, the tunneling operation state of the TBM needs to be continuously adjusted along with the change of the geological conditions, and the tunneling parameter control strategy needs to be adaptively adjusted along with the change of the geological conditions. Therefore, the embodiment of the invention provides an intelligent decision method for self-adaptive rock lithology and grade TBM tunneling, which predicts TBM tunneling parameters based on collected TBM tunneling related data, so that the TBM tunneling operation state continuously adjusts tunneling control strategies along with the change of geological conditions.
In addition, according to the intelligent decision method for self-adaptive rock mass lithology and grade TBM tunneling, the empty pushing section, the ascending section and the stabilizing section in the TBM tunneling process do not need to be considered, and the TBM tunneling parameters are predicted for complete TBM tunneling circulation, so that the data operand can be reduced, and the prediction accuracy is improved.
Fig. 1 is a flow chart of an intelligent decision method for adaptive rock mass lithology and grade TBM tunneling provided by a first embodiment of the present invention, as shown in fig. 1, and the intelligent decision method for adaptive rock mass lithology and grade TBM tunneling provided by the embodiment of the present invention includes:
S101, acquiring TBM tunneling related data;
specifically, the data acquisition system in the on-site TBM monitoring platform can acquire TBM tunneling related data, the TBM tunneling related data can include tunnel surrounding rock state information, tunnel geological forecast information, fine exploration information, accurate lining information, real-time tunneling effect information and the like, and the TBM tunneling related data are set according to actual needs, and the embodiment of the invention is not limited.
S102, preprocessing TBM tunneling related data to obtain TBM tunneling characteristic data;
specifically, after the TBM tunneling related data is obtained, preprocessing the TBM tunneling related data to obtain TBM tunneling characteristic data. The TBM tunneling characteristic data is used for predicting TBM tunneling parameters. The preprocessing may include outlier processing, noise reduction processing, normalization processing, and the like, and is set according to actual needs, which is not limited by the embodiment of the present invention.
S103, acquiring surrounding rock grades corresponding to the TBM tunneling related data;
specifically, the surrounding rock grade corresponding to the TBM tunneling related data can be obtained from a pre-geological exploration result. In order to obtain the surrounding rock grade more accurately, the surrounding rock grade corresponding to the TBM tunneling related data can be predicted based on the rock-machine data. The specific process of predicting the surrounding rock grade from the rock-machine data is described in detail below.
S104, obtaining a decision model corresponding to the TBM tunneling related data according to the surrounding rock grade corresponding to the TBM tunneling related data and an intelligent decision model library;
specifically, inquiring in an intelligent decision model library according to surrounding rock grades corresponding to the TBM tunneling related data, and taking a decision model corresponding to the surrounding rock grades corresponding to the TBM tunneling related data as a decision model corresponding to the TBM tunneling related data. The intelligent decision model library is preset and comprises decision models corresponding to all surrounding rock grades. Through different surrounding rock grades, the corresponding decision model is selected, so that the requirement of using working condition diversity under complex geological conditions in the TBM tunneling process can be met. The setting of the intelligent decision model library is beneficial to realizing the selection of proper decision models according to different surrounding rock grades.
S105, predicting TBM tunneling parameters according to the TBM tunneling characteristic data and a decision model corresponding to the TBM tunneling related data; the decision model corresponding to the TBM tunneling related data is obtained through pre-training.
Specifically, the TBM tunneling characteristic data is input into a decision model corresponding to the TBM tunneling related data, and TBM tunneling parameters are output after the decision model corresponding to the TBM tunneling related data is processed. The decision model corresponding to the TBM tunneling related data is obtained through pre-training.
According to the intelligent decision method for self-adaptive rock mass lithology and grade TBM tunneling, TBM tunneling related data is obtained, preprocessing is carried out on the TBM tunneling related data to obtain TBM tunneling characteristic data, surrounding rock grades corresponding to the TBM tunneling related data are obtained, decision models corresponding to the TBM tunneling related data are obtained according to the surrounding rock grades corresponding to the TBM tunneling related data and an intelligent decision model library, TBM tunneling parameters are predicted according to the TBM tunneling characteristic data and the decision models corresponding to the TBM tunneling related data, and accuracy of TBM tunneling parameter prediction can be improved. In addition, different decision models can be selected for TBM tunneling parameter prediction according to surrounding rock grades of different geological conditions, and the reliability of TBM tunneling parameter prediction is improved. The more accurate and reliable the TBM tunneling parameter prediction, the more accurate and reliable the TBM control strategy obtained.
Fig. 2 is a flow chart of an intelligent decision method for adaptive rock mass lithology and grade TBM tunneling according to a second embodiment of the present invention, as shown in fig. 2, further, on the basis of the foregoing embodiments, training a decision model corresponding to TBM tunneling related data includes:
s201, TBM tunneling historical data corresponding to various surrounding rock grades are obtained;
Specifically, data from past TBM development processes may be collected from which TBM development history data is obtained. And collecting surrounding rock grades corresponding to TBM tunneling historical data, and acquiring TBM tunneling historical data corresponding to various surrounding rock grades. The TBM tunneling history data can comprise TBM tunneling parameter information, tunnel surrounding rock state information, tunnel geological forecast information, fine exploration information, accurate lining information, real-time tunneling effect information and the like, and are set according to actual needs, and the embodiment of the invention is not limited. The surrounding rock grade may be obtained based on a geological advance forecast.
The TBM tunneling parameter information can comprise cutter torque, total propelling force, cutter rotating speed, propelling speed, penetration and cutter power. The tunnel surrounding rock state information may include uniaxial saturation compressive strength of the rock, number of joints per unit rock mass volume, and surrounding rock grade. The tunnel geological forecast information is obtained by the advanced geological detection system, and the advanced geological detection system can adopt the advanced geological detection system which takes the horizontal overlength drilling technology as the core and has the characteristics of long detection distance and capability of effectively detecting bad geological bodies such as faults, broken zones and the like existing in front of the face. The fine exploration information comprises information such as ground stress, rock strength, geological structure, osmotic pressure distribution and the like, and is obtained by exploration of fine geological conditions through a tunnel geological exploration technology of horizontal ultra-long drilling holes. The accurate lining information comprises intelligent assessment information of local collapse risk, intelligent collapse support decision information and cleaning information of slag accumulated at the bottom of the tunnel, and is provided by an accurate lining intelligent operation system. The real-time tunneling effect information comprises tunneling direction planning data in a design drawing, actual measurement data of the tunneling direction of a tunneling machine tunneling guide system, single-cutter thrust, cutter torque, cutter rotating speed and propulsion speed of main parameters of the tunneling machine, and real-time tunneling effect index values calculated based on the data.
The real-time tunneling effect index value A (x, y, z) is calculated by the following formula:
wherein,representing the average error value, +.>Averaging the first error average value and the second error average value, wherein the first error average value is obtained by acquiring N average values of N relative errors measured by the tunneling direction planning data and the tunneling guiding system in N current periods according to a preset acquisition frequency, and N is a positive integer greater than 1; and acquiring the set tunneling parameter information and the real-time tunneling parameter information of the tunneling machine in N groups of current periods according to the preset acquisition frequency, calculating the relative error of each parameter in each group of tunneling parameter information, and further calculating a second error average value of each parameter.
TBM operation state information refers to TBM operation state information related to tunneling operation and comprising TBM tunneling parameters such as cutter torque, total propulsion, cutter rotating speed, propulsion speed, penetration and cutter power, and specifically comprises working condition information and state information. Wherein, the working condition information such as voltage, current and pressure; status information such as the position of the shoe supporting hydraulic cylinder, the rear supporting state, the motor and hydraulic motor state, the hydraulic system state of each hydraulic power station, the jumbolter and the conveyor, various shutdown states caused by faults, various shutdown states caused by tool checking/replacement and the like.
S202, extracting key parameters of TBM tunneling historical data corresponding to each surrounding rock grade, and obtaining TBM tunneling training data corresponding to each surrounding rock grade;
specifically, key parameters are extracted from TBM tunneling history data corresponding to each surrounding rock grade, the key parameters are used as TBM tunneling training data corresponding to each surrounding rock grade, and the TBM tunneling training data corresponding to each surrounding rock grade comprises TBM tunneling parameter information. The dimension reduction of TBM tunneling historical data is realized through the extraction of key parameters, the dimension reduction of the data can reduce errors caused by redundant information, the precision of identification or other applications is improved, and the training cost is reduced; searching for intrinsic structural features in the data through a dimension reduction algorithm; and data visualization is facilitated.
For example, the key parameter extraction is implemented by a one-dimensional convolutional neural network or a pearson correlation analysis method.
And S203, training to obtain a decision model corresponding to TBM tunneling related data according to TBM tunneling training data corresponding to each surrounding rock grade and the initial model.
Specifically, training is performed on the initial model according to TBM tunneling training data corresponding to each surrounding rock grade, and a decision model corresponding to each surrounding rock grade can be obtained through training. The surrounding rock grade corresponding to the TBM tunneling related data determines which decision model corresponding to the surrounding rock grade is adopted as the decision model corresponding to the TBM tunneling related data.
The initial model includes, but is not limited to, RNN, LSTM, GRU, bi-RNN, bi-LSTM, bi-GRU, PSO-RNN, PSO-LSTM, PSO-GRU, PSO-Bi-RNN, PSO-Bi-LSTM, PSO-Bi-GRU, bi-LSTM+EMB-ATT, bi-GRU-ATT, etc., and is selected according to the actual situation, and the embodiments of the present invention are not limited. For example, model training may be performed on each of the initial models to obtain a plurality of candidate decision models, and a model with the most accurate prediction may be selected as a final decision model.
To facilitate model training, different deep learning frameworks may be employed, and multiple deep learning frameworks may be switched through a framework switching interface, including but not limited to Keras and TensorFlow deep learning frameworks.
Further, the super-parameter adjustment of the model is used for adjusting super-parameters, wherein the super-parameters comprise learning rate, batch size, optimizer, iteration times, activation function and the like. The learning rate control adjusts the weight of the network to meet the gradient loss, and the lower the value is, the slower the value is along the gradient; the batch size is the number of samples sent to the model by each training of the neural network, in the convolutional neural network, a large batch can generally lead the network to converge more quickly, but due to the limitation of memory resources, the excessive batch can cause the memory to be insufficient or the program kernel to crash, so the batch size value range is generally set as [16,32,64,128]; the optimizer is realized by adopting a common Adam optimizer; the iteration times are the times of inputting the whole training set into the neural network for training; the activation function is selected from the relu activation function which is most used in deep learning, so that the method is simple and gradient disappearance can be avoided.
Under complex geological conditions, a proper decision model can be selected according to different surrounding rock grades, and the advantages of various control strategies are exerted, so that the control performance of the system can meet the requirement of using working condition diversity under the complex geological conditions in the TBM tunneling process.
Fig. 3 is a flow chart of an intelligent decision method for adaptive rock lithology and grade TBM tunneling according to a third embodiment of the present invention, as shown in fig. 3, further, based on the foregoing embodiments, obtaining a surrounding rock grade corresponding to the TBM tunneling related data includes:
s301, acquiring TBM tunneling data based on rock-machine data;
specifically, a TBM tunneling parameter matrix can be extracted from the rock-machine data and used as TBM tunneling data.
For example, rock-machine data extraction methods: taking the working data of a day and the total propulsion of the TBM as an example. Firstly, original data are read, and each piece of data is judged according to the total propulsion; recording the first piece of data with non-zero total propulsion force, and recording as P1; reading the next piece of data according to the row, searching the second piece of data with zero total propulsion force, and marking the second piece of data as P2; judging whether the number of lines between P1 and P2 is more than 500s and less than 5000s; if the data is not within the range of 500 s-5000 s, discarding the segment of data; if not, outputting the P1 and P2 data to the designated files, wherein each file has 300 cycles at most until all the data are extracted.
According to the method, effective tunneling cycle segment data in TBM original data are extracted, but due to the influence of geological conditions of a construction segment, the data stored in the construction process of the TBM have extremely individual abnormal phenomena, the model precision is influenced, and the TBM needs to be removed. Taking the propulsion speed as an example, the 3 times standard deviation method is adopted for correcting the extreme individual outliers in the data samples. And calculating the average value and standard deviation of the propulsion speed in the whole digging progress, and if the data value is out of the range of the standard deviation of the average value 3, taking the data value as an abnormal point, and correcting by adopting a Lagrange interpolation method.
S302, obtaining lithology corresponding to TBM tunneling related data according to TBM tunneling data and lithology prediction models; wherein the lithology prediction model is obtained by pre-training;
specifically, the TBM tunneling data are input into a lithology prediction model, and lithology corresponding to the TBM tunneling related data can be output through processing of the lithology prediction model, wherein the TBM tunneling data are used for determining lithology of a tunneling section rock body corresponding to the TBM tunneling related data. Wherein the lithology prediction model is obtained by pre-training.
S303, obtaining surrounding rock grades corresponding to TBM tunneling data according to the TBM tunneling data and a surrounding rock grade prediction model of lithology corresponding to the TBM tunneling related data; the surrounding rock grade prediction model is obtained through pre-training;
Specifically, the TBM tunneling data is input into a surrounding rock grade prediction model of lithology corresponding to the TBM tunneling related data, and the surrounding rock grade corresponding to the TBM tunneling related data can be output. Wherein, the surrounding rock grades corresponding to different lithology can be different. The surrounding rock grade prediction model is obtained through pre-training.
Fig. 4 is a flow chart of an intelligent decision method for adaptive rock mass lithology and grade TBM tunneling according to a fourth embodiment of the present invention, as shown in fig. 4, further, on the basis of the above embodiments, the step of training the lithology prediction model includes:
s401, acquiring rock-machine historical data and corresponding lithology;
specifically, rock-to-machine data during past tunneling of the TBM may be collected as rock-to-machine historical data. And acquiring rock mass parameters corresponding to the rock-machine history data. By clustering rock mass parameters corresponding to the rock-machine history data, lithology corresponding to the rock-machine history data can be obtained.
And clustering different rock parameters through a K-means clustering algorithm to obtain each lithology classification center. The specific clustering flow is as follows:
initializing classification, namely initializing K clustering centers, representing the similarity of each sample data (namely rock parameters) in the lithology sample data by using Euclidean distance, and calculating each sample data in the lithology sample data to each initial stage Initial cluster center C j Distance J (C) of (j=1, 2, …, k) j ) Each sample data s in the sexual sample data is judged according to the minimum distance judgment criterion i Divided into a certain class C j ；
Step two, updating a clustering center, and calculating each classification { C } 1 ,C 2 ,…,C k Mean value { u } of membership sample data in } 1 ,u 2 ,…,u k New cluster center C ' = { C ' as the category ' 1 ,C′ 2 ,…,C′ k Then re-acquiring the sum of squares J (C') of the distances from each sample data in the lithology sample data to the category cluster center where each sample data is located;
third, judging whether the values of the new clustering centers C ' and J (C ') change within the respective allowable fluctuation ranges, if the values of the new clustering centers and J (C ') change within the respective allowable fluctuation ranges, ending the clustering, wherein C ' = { C ' 1 ,C′ 2 ,…,C′ k As individual lithology classification centers. Otherwise, repeating the second step to continuously update the clustering center until the clustering is finished, and obtaining each lithology classification center.
S402, obtaining lithology prediction training data according to the rock-machine history data and the corresponding lithology;
specifically, a TBM historical tunneling parameter matrix is extracted from the rock-machine historical data and corresponds to lithology, and the TBM historical tunneling parameter matrix is used as lithology prediction training data. Wherein lithology is used as a training tag.
S403, respectively training to obtain a plurality of lithology predictor models according to the lithology prediction training data and a plurality of original models;
Specifically, training each original model according to the lithology prediction training data to obtain each lithology predictor model. The original model corresponds to the lithology predictor model one by one. The original model can be a neural network model, a support vector machine model, a least squares regression model and the like, and is set according to actual needs, and the embodiment of the invention is not limited.
S404, according to the evidence set and each lithology predictor model, obtaining a prediction error of each lithology predictor model; the evidence set is obtained based on random extraction of lithology prediction training data;
specifically, N subsets are randomly extracted from the lithology prediction training data as evidence sets, and according to the evidence sets and each lithology predictor model, a prediction error of each lithology predictor model can be obtained. Wherein N is a positive integer greater than or equal to 2.
S405, determining a weight coefficient of each lithology predictor model according to the prediction error of each lithology predictor model;
specifically, a smaller weight is given to a lithology predictor model with a larger prediction error, and a larger weight is given to a lithology predictor model with a smaller prediction error. The weight coefficient of each lithology predictor model may be determined according to the prediction error of the respective lithology predictor model. The specific mode for determining the weight coefficient based on the prediction error is selected according to actual needs, and the embodiment of the invention is not limited.
S406, obtaining the lithology prediction model according to each lithology prediction sub-model and the corresponding weight coefficient.
Specifically, each lithology prediction sub-model and the corresponding weight coefficient are fused to obtain the lithology prediction model.
For example, a lithology predictor model A is obtained based on neural network model training, a lithology predictor model B is obtained based on support vector machine model training, a lithology predictor model C is obtained based on least squares regression model training, the weight coefficient of the obtained lithology predictor model A is a, the weight coefficient of the lithology predictor model B is B, and the weight coefficient of the lithology predictor model C is C. The three lithology prediction sub-models a and the weight coefficients corresponding to each are fused, and the obtained lithology prediction model may be expressed as y=aa+bb+cc.
Fig. 5 is a flow chart of an intelligent decision method for adaptive rock mass lithology and grade TBM tunneling according to a fifth embodiment of the present invention, as shown in fig. 5, further, on the basis of the above embodiments, the step of training the lithology prediction model includes:
s501, acquiring rock-machine historical data and corresponding lithology;
specifically, the specific implementation process of this step is similar to step S301, and will not be described here.
S502, obtaining lithology prediction training data according to the rock-machine history data and the corresponding lithology;
specifically, the specific implementation process of this step is similar to step S302, and will not be described here.
S503, respectively training to obtain a plurality of lithology predictor models according to the lithology prediction training data and a plurality of original models;
specifically, the specific implementation process of this step is similar to step S304, and will not be described here.
S504, obtaining a basic probability value of each lithology predictor model according to the evidence set and each lithology predictor model; the evidence set is obtained based on random extraction of lithology prediction training data;
specifically, the evidence set is respectively input into each lithology predictor model, and a basic probability value of each lithology predictor model can be output. Wherein the evidence set is obtained based on random extraction of lithology predictive training data.
S505, obtaining a weight coefficient of each lithology predictor model based on the credibility factors and the basic probability value of each lithology predictor model; wherein the confidence factor is obtained based on the evidence set;
specifically, conventional D-S evidence theory may produce Zadeh paradox when synthesizing high-conflict evidence sources. In order to solve the problems, a reliability factor zeta is introduced to measure the reliability of different evidence sources, and the evidence sources are corrected according to zeta so as to reduce the conflict among the different evidence sources. The method comprises the steps of obtaining a credibility factor based on the evidence set, correcting the basic probability value of each lithology predictor model based on the credibility factor to obtain a corrected basic probability value of each lithology predictor model, and synthesizing the corrected basic probability values of different lithology predictor models through a Dempster synthesis rule to obtain a weight coefficient of each lithology predictor model so as to reduce conflict among different lithology predictor models.
The specific process of obtaining the credibility factor based on the evidence set is as follows:
an ith evidence set e defining different lithology predictor models i The distance between the base probability value of (c) and the base probability value of the remaining evidence set is:
wherein M is i Representing the ith evidence set e i Basic probability value, M of (2) j A base probability value representing a j-th evidence set of the remaining evidence sets. A is that k Is assumed to be space. k is a positive integer and k is less than or equal to n 1 J is a positive integer and j is less than or equal to n 2 。
Confidence factor ζ i The method comprises the following steps:
wherein a is a mapping function adjustment coefficient.
The Dempster synthesis rules are:
wherein,conflict of evidence, coefficientFor regularization factor, ++>Is assumed to be space.
Basic probability value after reliability factor correction is fixedMeaning asThe calculation formula is as follows:
s506, obtaining the lithology prediction model according to each lithology prediction sub-model and the corresponding weight coefficient.
Specifically, the specific implementation process of this step is similar to step S306, and will not be described here.
The lithology prediction model is established by improving the D-S evidence theory, so that fusion of prediction results of different models is realized, and the lithology prediction precision is improved.
Fig. 6 is a flow chart of an intelligent decision method for adaptive rock lithology and grade TBM tunneling according to a sixth embodiment of the present invention, as shown in fig. 6, further, on the basis of the foregoing embodiments, the step of training the surrounding rock grade prediction model includes:
S601, acquiring rock-machine historical data of the same lithology and corresponding surrounding rock grades;
specifically, rock-machine data in the past tunneling process of the TBM can be collected, rock-machine historical data with the same lithology can be obtained, and rock parameters corresponding to the rock-machine historical data with the same lithology can be obtained. By clustering rock parameters corresponding to the rock-machine history data of the same lithology, surrounding rock grades corresponding to the rock-machine history data of the same lithology can be obtained.
S602, according to rock-machine historical data of the same lithology and corresponding surrounding rock grades, obtaining surrounding rock grade prediction training data;
specifically, a TBM historical tunneling parameter matrix is extracted from rock-machine historical data with the same lithology and corresponds to the surrounding rock grade, and the TBM historical tunneling parameter matrix is used as surrounding rock grade prediction training data. Wherein the surrounding rock grade is used as a training label.
S603, respectively training to obtain a plurality of surrounding rock grade predictor models according to the surrounding rock prediction training data and a plurality of original models;
specifically, training each original model according to the surrounding rock prediction training data to obtain each surrounding rock grade predictor model. The original model corresponds to the surrounding rock grade predictor model one by one. The original model can be a neural network model, a support vector machine model, a least squares regression model and the like, and is set according to actual needs, and the embodiment of the invention is not limited.
S604, obtaining a basic probability value of each surrounding rock grade predictor model according to the evidence set and each surrounding rock predictor model; the evidence set is obtained based on random extraction of surrounding rock prediction training data;
specifically, the evidence set is respectively input into each surrounding rock grade predictor model, and the basic probability value of each surrounding rock grade predictor model can be output. Wherein the evidence set is obtained by randomly extracting predicted training data based on the surrounding rock grade.
S605, obtaining a weight coefficient of each surrounding rock grade predictor model based on the credibility factors and the basic probability value of each surrounding rock grade predictor model; wherein the confidence factor is obtained based on the evidence set;
specifically, conventional D-S evidence theory may produce Zadeh paradox when synthesizing high-conflict evidence sources. In order to solve the problems, a reliability factor zeta is introduced to measure the reliability of different evidence sources, and the evidence sources are corrected according to zeta so as to reduce the conflict among the different evidence sources. The method comprises the steps of obtaining a credibility factor based on an evidence set, correcting a basic probability value of each surrounding rock grade prediction sub-model based on the credibility factor, obtaining a corrected basic probability value of each surrounding rock grade prediction sub-model, and synthesizing corrected basic probability values of different surrounding rock grade prediction sub-models through a Dempster synthesis rule to obtain a weight coefficient of each surrounding rock grade prediction sub-model so as to reduce conflict among different surrounding rock grade prediction sub-models.
S606, obtaining the surrounding rock grade prediction model according to each surrounding rock grade prediction sub-model and the corresponding weight coefficient.
Specifically, the surrounding rock grade prediction sub-models and the weight coefficients corresponding to the surrounding rock grade prediction sub-models are fused to obtain the surrounding rock grade prediction model.
Fig. 7 is a flow chart of an intelligent decision method for adaptive rock mass lithology and grade TBM tunneling according to a seventh embodiment of the present invention, as shown in fig. 7, further, on the basis of the foregoing embodiments, the step of training the surrounding rock grade prediction model includes:
s701, acquiring rock-machine historical data of the same lithology and corresponding surrounding rock grades;
specifically, the specific implementation process of this step is similar to step S501, and will not be described here.
S702, according to rock-machine historical data of the same lithology and corresponding surrounding rock grades, obtaining surrounding rock grade prediction training data;
specifically, the specific implementation process of this step is similar to step S502, and will not be described here.
S703, respectively training to obtain a plurality of surrounding rock grade predictor models according to the surrounding rock prediction training data and a plurality of original models;
specifically, the specific implementation procedure of this step is similar to step S503, and will not be described here in detail.
S704, according to the evidence set and each surrounding rock predictor model, obtaining a prediction error of each surrounding rock grade predictor model; the evidence set is obtained based on random extraction of lithology prediction training data;
specifically, N subsets are randomly extracted from surrounding rock grade prediction training data to serve as evidence sets, and according to the evidence sets and each surrounding rock grade prediction sub-model, prediction errors of each surrounding rock grade prediction sub-model can be obtained. Wherein N is a positive integer greater than or equal to 2.
S705, determining a weight coefficient of each surrounding rock grade predictor model according to the prediction error of each surrounding rock grade predictor model;
specifically, a lower weight is given to the surrounding rock grade predictor model with a larger prediction error, and a higher weight is given to the surrounding rock grade predictor model with a smaller prediction error. The weight coefficient of each surrounding rock grade predictor model can be determined according to the prediction error of each surrounding rock grade predictor model. The specific mode for determining the weight coefficient based on the prediction error is selected according to actual needs, and the embodiment of the invention is not limited.
S706, obtaining the surrounding rock grade prediction model according to each surrounding rock grade prediction sub-model and the corresponding weight coefficient.
Specifically, the surrounding rock grade prediction sub-models and the weight coefficients corresponding to the surrounding rock grade prediction sub-models are fused to obtain the surrounding rock grade prediction model.
And the surrounding rock grade prediction model is established by improving the D-S evidence theory, so that fusion of prediction results of different models is realized, and the prediction precision of the surrounding rock grade is improved. Based on the above embodiments, further, the preprocessing the TBM tunneling related data to obtain TBM tunneling feature data includes:
and carrying out outlier processing, noise reduction processing and standardization processing on the TBM tunneling related data.
Specifically, in the acquired TBM tunneling related data, there may be individual abnormal data which deviate from the normal data obviously and do not conform to the statistical rule, in order to reduce the probability of false alarm of TBM tunneling parameters caused by the abnormal data, abnormal value processing may be performed on the TBM tunneling related data, and the abnormal data processing may be performed by adopting a 3-fold standard deviation method.
In order to reduce the influence of random noise in TBM tunneling related data, noise reduction processing can be carried out on the TBM tunneling related data, a wavelet packet noise reduction algorithm can be adopted in the noise reduction processing, and the data quality can be improved through the noise reduction processing.
In order to eliminate the uncertain influence caused by the singular samples, the TBM tunneling related data can be standardized, and a minimum-maximum-standardization method can be adopted in the standardization process, so that the TBM tunneling related data is limited in a certain range, and the data convergence is enhanced.
Compared with the prior art, the invention has the following beneficial effects:
according to the TBM tunneling method, the complete tunneling cycle data is selected to predict TBM tunneling parameters, so that complexity and difficulty in processing huge amount of data generated in the TBM tunneling process can be effectively reduced, and the preprocessing efficiency of massive TBM tunneling data is greatly improved;
according to the invention, the problem that the tunneling targets focused by TBM construction under different surrounding rock grade conditions have differences is solved by combining the intelligent decision model library based on deep learning and the multi-mode control idea, so that the tunneling operation state of the TBM can be adaptively adjusted along with the change of geological conditions, and the problem that the tunneling targets focused by TBM construction under different geological conditions have differences can be solved by remarkably improving the single and subjective defects of the traditional decision rule and artificial judgment.
Fig. 8 is a schematic structural diagram of an adaptive rock mass lithology and grade TBM tunneling intelligent decision device provided by an eighth embodiment of the present invention, and as shown in fig. 8, the adaptive rock mass lithology and grade TBM tunneling intelligent decision device provided by the embodiment of the present invention includes a first obtaining unit 801, a preprocessing unit 802, a second obtaining unit 803, an obtaining unit 804 and a predicting unit 805, where:
The first acquiring unit 801 is configured to acquire TBM tunneling related data; the preprocessing unit 802 is configured to preprocess the TBM tunneling related data to obtain TBM tunneling feature data; the second obtaining unit 803 is configured to obtain a surrounding rock grade corresponding to the TBM tunneling related data; the obtaining unit 804 is configured to obtain a decision model corresponding to the TBM tunneling related data according to the surrounding rock grade corresponding to the TBM tunneling related data and an intelligent decision model library; the prediction unit 805 is configured to predict TBM tunneling parameters according to the TBM tunneling feature data and a decision model corresponding to the TBM tunneling related data; the decision model corresponding to the TBM tunneling related data is obtained through pre-training.
Specifically, the first obtaining unit 801 may obtain TBM tunneling related data from a data collecting system in the on-site TBM monitoring platform, where the TBM tunneling related data may include status information of surrounding rock of a tunnel, geological forecast information of the tunnel, fine exploration information, accurate lining information, real-time tunneling effect information, and the like, and is set according to actual needs, and the embodiment of the present invention is not limited.
The preprocessing unit 802, after obtaining the TBM tunneling-related data, preprocesses the TBM tunneling-related data to obtain TBM tunneling feature data. The TBM tunneling characteristic data is used for predicting TBM tunneling parameters. The preprocessing may include outlier processing, noise reduction processing, normalization processing, and the like, and is set according to actual needs, which is not limited by the embodiment of the present invention.
The second obtaining unit 803 may obtain the surrounding rock grade corresponding to the TBM tunneling related data from the pre-geological exploration result. In order to obtain the surrounding rock grade more accurately, the second obtaining unit 803 may also predict the surrounding rock grade corresponding to the TBM tunneling related data based on the rock-machine data.
The obtaining unit 804 queries in an intelligent decision model library according to the surrounding rock grade corresponding to the TBM tunneling related data, and uses a decision model corresponding to the surrounding rock grade corresponding to the TBM tunneling related data as a decision model corresponding to the TBM tunneling related data. The intelligent decision model library is preset and comprises decision models corresponding to all surrounding rock grades. Through different surrounding rock grades, the corresponding decision model is selected, so that the requirement of using working condition diversity under complex geological conditions in the TBM tunneling process can be met. The setting of the intelligent decision model library is beneficial to realizing the selection of proper decision models according to different surrounding rock grades.
The prediction unit 805 inputs the TBM tunneling feature data into a decision model corresponding to the TBM tunneling related data, and outputs TBM tunneling parameters after processing the decision model corresponding to the TBM tunneling related data. The decision model corresponding to the TBM tunneling related data is obtained through pre-training.
According to the intelligent decision device for self-adaptive rock mass lithology and grade TBM tunneling, TBM tunneling related data is obtained, TBM tunneling related data is preprocessed, TBM tunneling characteristic data is obtained, surrounding rock grades corresponding to the TBM tunneling related data are obtained, decision models corresponding to the TBM tunneling related data are obtained according to the surrounding rock grades corresponding to the TBM tunneling related data and an intelligent decision model library, TBM tunneling parameters are predicted according to the TBM tunneling characteristic data and the decision models corresponding to the TBM tunneling related data, and accuracy of TBM tunneling parameter prediction can be improved. In addition, different decision models can be selected for TBM tunneling parameter prediction according to surrounding rock grades of different geological conditions, and the reliability of TBM tunneling parameter prediction is improved. The more accurate and reliable the TBM tunneling parameter prediction, the more accurate and reliable the TBM control strategy obtained.
On the basis of the above embodiments, further, the intelligent decision device for adaptive rock lithology and grade TBM tunneling provided by the embodiment of the present invention further includes:
the first historical data acquisition unit is used for acquiring the rock-machine historical data and the corresponding lithology; the dimension reduction unit is used for extracting key parameters of TBM tunneling historical data corresponding to each surrounding rock grade and obtaining TBM tunneling training data corresponding to each surrounding rock grade; the first training unit is used for training and obtaining a decision model corresponding to TBM tunneling related data according to TBM tunneling training data corresponding to each surrounding rock grade and the initial model.
On the basis of the above embodiments, further, the intelligent decision device for adaptive rock lithology and grade TBM tunneling provided by the embodiment of the present invention further includes:
the second historical data acquisition unit is used for acquiring the rock-machine historical data and the corresponding lithology; the first training data obtaining unit is used for obtaining lithology prediction training data according to the rock-machine historical data and the corresponding lithology; the second training unit is used for respectively training and obtaining a plurality of lithology predictor models according to the lithology prediction training data and a plurality of original models; a third obtaining unit, configured to obtain a prediction error of each lithology predictor model according to the evidence set and each lithology predictor model; the evidence set is obtained based on random extraction of lithology prediction training data; a first determining unit, configured to determine a weight coefficient of each lithology predictor model according to the basic probability value of each lithology predictor model; and the fourth obtaining unit is used for obtaining the lithology prediction model according to each lithology prediction sub-model and the corresponding weight coefficient.
On the basis of the above embodiments, further, the intelligent decision device for adaptive rock lithology and grade TBM tunneling provided by the embodiment of the present invention further includes:
The TBM tunneling data acquisition unit is used for acquiring TBM tunneling data based on the rock-machine data; the lithology obtaining unit is used for obtaining lithology corresponding to the TBM tunneling prediction data according to the TBM tunneling data and the lithology prediction model; wherein the lithology prediction model is obtained by pre-training; the surrounding rock grade obtaining unit is used for obtaining the surrounding rock grade corresponding to the TBM tunneling prediction data according to the TBM tunneling data and a surrounding rock grade prediction model of lithology corresponding to the TBM tunneling prediction data; the surrounding rock grade prediction model is obtained through pre-training.
On the basis of the above embodiments, further, the intelligent decision device for adaptive rock lithology and grade TBM tunneling provided by the embodiment of the present invention further includes:
the third historical data acquisition unit is used for acquiring the rock-machine historical data and the corresponding lithology; the second training data obtaining unit is used for obtaining lithology prediction training data according to the rock-machine historical data and the corresponding lithology; the third training unit is used for respectively training and obtaining a plurality of lithology predictor models according to the lithology prediction training data and a plurality of original models; a fourth obtaining unit, configured to obtain a basic probability value of each lithology predictor model according to the evidence set and each lithology predictor model; the evidence set is obtained based on random extraction of lithology prediction training data; the correction unit is used for obtaining the weight coefficient of each lithology predictor model based on the basic probability value of each lithology predictor model by the credibility factor; wherein the confidence factor is obtained based on the evidence set; and a fifth obtaining unit, configured to obtain the lithology prediction models according to each lithology prediction sub-model and each corresponding weight coefficient.
On the basis of the above embodiments, further, the intelligent decision device for adaptive rock lithology and grade TBM tunneling provided by the embodiment of the present invention further includes:
the fourth historical data acquisition unit is used for acquiring the rock-machine historical data of the same lithology and corresponding surrounding rock grades; the third training data obtaining unit is used for obtaining surrounding rock grade prediction training data according to the rock-machine history data of the same lithology and the corresponding surrounding rock grade; the fourth training unit is used for respectively training and obtaining a plurality of surrounding rock grade predictor models according to the surrounding rock prediction training data and a plurality of original models; a sixth obtaining unit, configured to obtain a basic probability value of each surrounding rock level predictor model according to the evidence set and each surrounding rock predictor model; the evidence set is obtained based on random extraction of surrounding rock prediction training data; a seventh obtaining unit, configured to obtain a weight coefficient of each surrounding rock grade predictor model based on the reliability factor and the basic probability value of each surrounding rock grade predictor model; wherein the confidence factor is obtained based on the evidence set; and an eighth obtaining unit, configured to obtain the surrounding rock grade prediction model according to each surrounding rock grade prediction sub-model and each corresponding weight coefficient.
On the basis of the above embodiments, further, the intelligent decision device for adaptive rock lithology and grade TBM tunneling provided by the embodiment of the present invention further includes:
a fifth historical data acquisition unit, configured to acquire rock-machine historical data of the same lithology and corresponding surrounding rock grades; the fourth training data obtaining unit is used for obtaining surrounding rock grade prediction training data according to the rock-machine history data of the same lithology and the corresponding surrounding rock grade; the fifth training unit is used for respectively training and obtaining a plurality of surrounding rock grade predictor models according to the surrounding rock prediction training data and a plurality of original models; a ninth obtaining unit, configured to obtain a prediction error of each surrounding rock grade predictor model according to the evidence set and each surrounding rock predictor model; the evidence set is obtained based on random extraction of lithology prediction training data; the second determining unit is used for determining the weight coefficient of each surrounding rock grade predictor model according to the prediction error of each surrounding rock grade predictor model; and a tenth obtaining unit, configured to obtain the surrounding rock grade prediction model according to each surrounding rock grade prediction sub-model and each corresponding weight coefficient.
Further, based on the above embodiments, the preprocessing unit is specifically configured to
And carrying out outlier processing, noise reduction processing and standardization processing on the TBM tunneling related data.
The embodiment of the apparatus provided in the embodiment of the present invention may be specifically used to execute the processing flow of each method embodiment, and the functions thereof are not described herein again, and may refer to the detailed description of the method embodiments.
Fig. 9 is a schematic physical structure of an electronic device according to a ninth embodiment of the present invention, as shown in fig. 9, the electronic device may include: processor 901, communication interface (Communications Interface) 902, memory 903 and communication bus 904, wherein processor 901, communication interface 902 and memory 903 communicate with each other via communication bus 904. The processor 901 may call logic instructions in the memory 903 to perform the following method: acquiring TBM tunneling related data; preprocessing TBM tunneling related data to obtain TBM tunneling characteristic data; acquiring surrounding rock grades corresponding to the TBM tunneling related data; obtaining a decision model corresponding to the TBM tunneling related data according to the surrounding rock grade corresponding to the TBM tunneling related data and an intelligent decision model library; predicting TBM tunneling parameters according to the TBM tunneling characteristic data and a decision model corresponding to the TBM tunneling related data; the decision model corresponding to the TBM tunneling related data is obtained through pre-training.
Further, the logic instructions in the memory 903 described above may be implemented in the form of software functional units and may be stored in a computer readable storage medium when sold or used as a stand alone product. Based on this understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
The present embodiment discloses a computer program product comprising a computer program stored on a non-transitory computer readable storage medium, the computer program comprising program instructions which, when executed by a computer, are capable of performing the methods provided by the above-described method embodiments, for example comprising: acquiring TBM tunneling related data; preprocessing TBM tunneling related data to obtain TBM tunneling characteristic data; acquiring surrounding rock grades corresponding to the TBM tunneling related data; obtaining a decision model corresponding to the TBM tunneling related data according to the surrounding rock grade corresponding to the TBM tunneling related data and an intelligent decision model library; predicting TBM tunneling parameters according to the TBM tunneling characteristic data and a decision model corresponding to the TBM tunneling related data; the decision model corresponding to the TBM tunneling related data is obtained through pre-training.
The present embodiment provides a computer-readable storage medium storing a computer program that causes the computer to execute the methods provided by the above-described method embodiments, for example, including: acquiring TBM tunneling related data; preprocessing TBM tunneling related data to obtain TBM tunneling characteristic data; acquiring surrounding rock grades corresponding to the TBM tunneling related data; obtaining a decision model corresponding to the TBM tunneling related data according to the surrounding rock grade corresponding to the TBM tunneling related data and an intelligent decision model library; predicting TBM tunneling parameters according to the TBM tunneling characteristic data and a decision model corresponding to the TBM tunneling related data; the decision model corresponding to the TBM tunneling related data is obtained through pre-training.
It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
In the description of the present specification, reference to the terms "one embodiment," "one particular embodiment," "some embodiments," "for example," "an example," "a particular example," or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, schematic representations of the above terms do not necessarily refer to the same embodiments or examples. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
The foregoing description of the embodiments has been provided for the purpose of illustrating the general principles of the invention, and is not meant to limit the scope of the invention, but to limit the invention to the particular embodiments, and any modifications, equivalents, improvements, etc. that fall within the spirit and principles of the invention are intended to be included within the scope of the invention.
Claims (11)
1. An intelligent decision-making method for self-adaptive rock lithology and grade TBM tunneling is characterized by comprising the following steps:
acquiring TBM tunneling related data;
preprocessing TBM tunneling related data to obtain TBM tunneling characteristic data;
acquiring surrounding rock grades corresponding to the TBM tunneling related data;
obtaining a decision model corresponding to the TBM tunneling related data according to the surrounding rock grade corresponding to the TBM tunneling related data and an intelligent decision model library;
predicting TBM tunneling parameters according to the TBM tunneling characteristic data and a decision model corresponding to the TBM tunneling related data; the decision model corresponding to the TBM tunneling related data is obtained through pre-training.
2. The method of claim 1, wherein training a decision model corresponding to TBM tunneling-related data comprises:
acquiring TBM tunneling historical data corresponding to various surrounding rock grades;
extracting key parameters of TBM tunneling historical data corresponding to each surrounding rock grade, and obtaining TBM tunneling training data corresponding to each surrounding rock grade;
and training to obtain a decision model corresponding to the TBM tunneling related data according to TBM tunneling training data corresponding to each surrounding rock grade and the initial model.
3. The method of claim 1, wherein the obtaining the surrounding rock grade corresponding to the TBM tunneling-related data comprises:
acquiring TBM tunneling data based on the rock-machine data;
according to TBM tunneling data and lithology prediction models, lithology corresponding to TBM tunneling related data is obtained; wherein the lithology prediction model is obtained by pre-training;
acquiring surrounding rock grades corresponding to TBM tunneling related data according to TBM tunneling data and a surrounding rock grade prediction model of lithology corresponding to TBM tunneling related data; the surrounding rock grade prediction model is obtained through pre-training.
4. A method according to claim 3, wherein the step of training the lithology predictive model comprises:
acquiring rock-machine historical data and corresponding lithology;
according to the rock-machine history data and the corresponding lithology, lithology prediction training data are obtained;
respectively training to obtain a plurality of lithology prediction sub-models according to the lithology prediction training data and a plurality of original models;
obtaining a prediction error of each lithology predictor model according to the evidence set and each lithology predictor model; the evidence set is obtained based on random extraction of lithology prediction training data;
Determining a weight coefficient of each lithology predictor model according to the prediction error of each lithology predictor model;
and obtaining the lithology prediction model according to each lithology prediction sub-model and the corresponding weight coefficient.
5. A method according to claim 3, wherein the step of training the lithology predictive model comprises:
acquiring rock-machine historical data and corresponding lithology;
according to the rock-machine history data and the corresponding lithology, lithology prediction training data are obtained;
respectively training to obtain a plurality of lithology prediction sub-models according to the lithology prediction training data and a plurality of original models;
according to the evidence set and each lithology predictor model, obtaining a basic probability value of each lithology predictor model; the evidence set is obtained based on random extraction of lithology prediction training data;
obtaining a weight coefficient of each lithology predictor model based on the basic probability value of the credibility factor for each lithology predictor model; wherein the confidence factor is obtained based on the evidence set;
and obtaining the lithology prediction model according to each lithology prediction sub-model and the corresponding weight coefficient.
6. A method according to claim 3, wherein the step of training the surrounding rock grade prediction model comprises:
acquiring rock-machine historical data of the same lithology and corresponding surrounding rock grades;
according to the rock-machine history data of the same lithology and the corresponding surrounding rock grade, obtaining surrounding rock grade prediction training data;
respectively training to obtain a plurality of surrounding rock grade predictor models according to the surrounding rock prediction training data and a plurality of original models;
according to the evidence set and each surrounding rock predictor model, obtaining a basic probability value of each surrounding rock grade predictor model; the evidence set is obtained based on random extraction of surrounding rock prediction training data;
based on the credibility factors and the basic probability value of each surrounding rock grade predictor model, obtaining the weight coefficient of each surrounding rock grade predictor model; wherein the confidence factor is obtained based on the evidence set;
and obtaining the surrounding rock grade prediction model according to each surrounding rock grade prediction sub-model and the corresponding weight coefficient.
7. A method according to claim 3, wherein the step of training the surrounding rock grade prediction model comprises:
Acquiring rock-machine historical data of the same lithology and corresponding surrounding rock grades;
according to the rock-machine history data of the same lithology and the corresponding surrounding rock grade, obtaining surrounding rock grade prediction training data;
respectively training to obtain a plurality of surrounding rock grade predictor models according to the surrounding rock prediction training data and a plurality of original models;
according to the evidence set and each surrounding rock predictor model, obtaining a prediction error of each surrounding rock grade predictor model; the evidence set is obtained based on random extraction of lithology prediction training data;
determining a weight coefficient of each surrounding rock grade predictor model according to the prediction error of each surrounding rock grade predictor model;
and obtaining the surrounding rock grade prediction model according to each surrounding rock grade prediction sub-model and the corresponding weight coefficient.
8. A method according to any one of claims 1 to 7, wherein pre-processing TBM ripping related data to obtain TBM ripping feature data comprises:
and carrying out outlier processing, noise reduction processing and standardization processing on the TBM tunneling related data.
9. An intelligent decision making device for self-adaptive rock mass lithology and grade TBM tunneling, which is characterized by comprising:
The first acquisition unit is used for acquiring TBM tunneling related data;
the pretreatment unit is used for carrying out pretreatment on TBM tunneling related data to obtain TBM tunneling characteristic data;
the second acquisition unit is used for acquiring surrounding rock grades corresponding to the TBM tunneling related data;
the obtaining unit is used for obtaining a decision model corresponding to the TBM tunneling related data according to the surrounding rock grade corresponding to the TBM tunneling related data and an intelligent decision model library;
the prediction unit is used for predicting TBM tunneling parameters according to the TBM tunneling characteristic data and a decision model corresponding to the TBM tunneling related data; the decision model corresponding to the TBM tunneling related data is obtained through pre-training.
10. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the steps of the method of any one of claims 1 to 8 when the computer program is executed by the processor.
11. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method according to any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210849366.0A CN117473689A (en) | 2022-07-19 | 2022-07-19 | Intelligent decision-making method and device for self-adaptive rock lithology and grade TBM tunneling |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210849366.0A CN117473689A (en) | 2022-07-19 | 2022-07-19 | Intelligent decision-making method and device for self-adaptive rock lithology and grade TBM tunneling |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117473689A true CN117473689A (en) | 2024-01-30 |
Family
ID=89622516
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210849366.0A Pending CN117473689A (en) | 2022-07-19 | 2022-07-19 | Intelligent decision-making method and device for self-adaptive rock lithology and grade TBM tunneling |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117473689A (en) |
-
2022
- 2022-07-19 CN CN202210849366.0A patent/CN117473689A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109635461A (en) | A kind of application carrys out the method and system of automatic identification Grades of Surrounding Rock with brill parameter | |
| US20060142990A1 (en) | Information processor, state judging unit and diagnostic unit, information processing method, state judging method and diagnosing method | |
| CN110561191B (en) | Numerical control machine tool cutter abrasion data processing method based on PCA and self-encoder | |
| CN110210169B (en) | LSTM-based shield tunneling machine fault prediction method | |
| CN110794308B (en) | Method and device for predicting train battery capacity | |
| CN114418469B (en) | LGBM-NSGA-III-based shield proximity construction parameter multi-objective optimization method and device | |
| CN112879024A (en) | Dynamic prediction method, system and equipment for shield attitude | |
| CN112464996A (en) | Intelligent power grid intrusion detection method based on LSTM-XGboost | |
| CN108961460B (en) | Fault prediction method and device based on sparse ESGP (Enterprise service gateway) and multi-objective optimization | |
| CN115081749A (en) | Bayesian optimization LSTM-based shield tunneling load advanced prediction method and system | |
| CN113459867A (en) | Electric vehicle charging process fault early warning method based on adaptive deep confidence network | |
| CN115049627B (en) | Steel surface defect detection method and system based on domain self-adaptive depth migration network | |
| CN114897007A (en) | Drilling machine health condition evaluation method of composite information layered convolution neural network | |
| CN110633504A (en) | Prediction method for coal bed gas permeability | |
| CN104732067A (en) | Industrial process modeling forecasting method oriented at flow object | |
| CN114254915A (en) | Method for deciding and optimizing qualified state of full-flow processing quality of shaft parts | |
| CN115049093B (en) | Yield stress prediction method and system based on ensemble learning algorithm | |
| CN112149896A (en) | Attention mechanism-based mechanical equipment multi-working-condition fault prediction method | |
| CN114065639A (en) | Closed-loop real-time inversion method for construction parameters of dredger | |
| CN118035866A (en) | Fault diagnosis method for aluminum extrusion equipment based on expert network | |
| CN117473689A (en) | Intelligent decision-making method and device for self-adaptive rock lithology and grade TBM tunneling | |
| CN115329219A (en) | Complex equipment abnormity detection method based on prediction | |
| CN115138472A (en) | Sorting suspension density setting method and system | |
| CN112348275A (en) | Regional ecological environment change prediction method based on online incremental learning | |
| CN118568667B (en) | Tunnel surrounding rock identification method, system, equipment and medium based on multi-model fusion |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication |