CN115544874A - Real-time monitoring system, method, terminal and medium for shield hob abrasion loss - Google Patents
Real-time monitoring system, method, terminal and medium for shield hob abrasion loss Download PDFInfo
- Publication number
- CN115544874A CN115544874A CN202211185851.9A CN202211185851A CN115544874A CN 115544874 A CN115544874 A CN 115544874A CN 202211185851 A CN202211185851 A CN 202211185851A CN 115544874 A CN115544874 A CN 115544874A
- Authority
- CN
- China
- Prior art keywords
- hob
- shield
- abrasion
- abrasion loss
- loss
- 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
Images
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
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/04—Ageing analysis or optimisation against ageing
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Evolutionary Computation (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biophysics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computer Hardware Design (AREA)
- Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Biomedical Technology (AREA)
- Geometry (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Excavating Of Shafts Or Tunnels (AREA)
Abstract
The invention provides a shield hob abrasion loss real-time monitoring system, a shield hob abrasion loss real-time monitoring method, a shield hob abrasion loss real-time monitoring terminal and a shield hob abrasion loss real-time monitoring medium, wherein the shield hob abrasion loss real-time monitoring system comprises the following steps: the parameter acquisition module is used for acquiring hob abrasion parameters in the shield construction process to form hob abrasion parameter sets; the standard establishing module is used for establishing a hob service life evaluation standard based on the accumulated wear loss of the hob, the actual working time of the hob corresponding to the wear loss and the diameter of a hob ring of the hob; the model building module is used for building a CNN-RNN prediction model for predicting the abrasion loss of the hob; the training prediction module substitutes the hob wear parameter set and the hob life evaluation standard into the CNN-RNN prediction model, carries out iterative training on the CNN-RNN prediction model, obtains a final prediction model after training is finished, and obtains a hob wear loss prediction value by adopting the final prediction model; and the monitoring module is used for monitoring the shield hob abrasion loss in real time according to the hob abrasion loss prediction value of the prediction module. The invention improves the accuracy and reliability of the prediction result.
Description
Technical Field
The invention relates to a monitoring system in the field of tunnel construction, in particular to a shield hob abrasion loss real-time monitoring system, a shield hob abrasion loss real-time monitoring method, a shield hob abrasion loss real-time monitoring terminal and a shield hob abrasion loss real-time monitoring medium.
Background
In the process of underground tunnel excavation, the shield machine enables the hob to continuously extrude and cut the front stratum by applying top thrust and cutter head torque. The hob is gradually abraded under the abrasion action of the stratum, so that the diameter of a hob ring of the hob is reduced, the excavation efficiency is greatly reduced, and the hob is required to be periodically checked to replace the hob reaching an abrasion limit value. Due to complex geological conditions in shield construction, excavation of the soil cabin is often carried out under the condition of pressure, and higher risks are caused. Because of the limitation of the pressure environment in the cabin, the tool changing operation needs to be carried out in a relay mode, the tool changing efficiency is greatly reduced, and the risk of the tool changing operation is increased. In addition, constructors need to check the abrasion condition of each hob one by one, so that the hob needing to be replaced is determined, time and labor are wasted, and efficiency is low. Therefore, the abrasion condition of a single hob needs to be predicted, so that the hob needing to be replaced is accurately judged, and the hob replacing efficiency is improved. The existing hob abrasion loss prediction method mostly considers the overall abrasion condition of a hob of a cutter disc, however, the abrasion state of a certain hob needs to be determined in the hob changing operation so as to judge whether the hob needs to be replaced or not. Therefore, a method for predicting the abrasion loss of the shield hobbing cutter is urgently needed, the abrasion prediction of a single hobbing cutter is realized, the cabin opening and cutter changing operation time and times are reduced, the effective propulsion time in shield construction is increased, and the shield construction efficiency is improved.
The search of the prior art documents shows that the Chinese patent application publication number is CN108256168A, and the patent name is: a method for determining abrasion loss of a shield hob in a composite stratum comprises the steps of determining geological conditions of the composite stratum through geological investigation, establishing a hob abrasion prediction model according to a linear relation between friction energy and abrasion loss and tool changing data collected in early construction, calculating the accumulated abrasion loss of the hob by combining shield parameters, and determining tool changing time and a place for opening a warehouse according to a rated abrasion loss limit value of the shield hob, so that the purposes of reasonably arranging opening the warehouse for tool changing and reducing tool changing cost are achieved. The accuracy and the reliability of the prediction result of the method can be further improved.
Disclosure of Invention
Aiming at the defects in the existing method, the invention aims to provide a shield hob abrasion loss real-time monitoring system, a shield hob abrasion loss real-time monitoring method, a shield hob abrasion loss real-time monitoring terminal and a shield hob abrasion loss real-time monitoring medium, so that the abrasion loss of a certain hob in the shield tunneling process can be predicted in real time, and the accuracy and the reliability of a prediction result can be improved. The invention also considers the influence of stratum conditions in actual construction, and can further improve the accuracy of the prediction result.
The invention provides a shield hob abrasion loss real-time monitoring system in a first aspect, which comprises:
the parameter acquisition module is used for acquiring hob abrasion parameters in the shield construction process to form a hob abrasion parameter set;
the standard establishing module establishes a hob service life evaluation standard based on the accumulated wear loss of the hob, the actual working time of the hob corresponding to the wear loss and the diameter of a hob ring of the hob;
the model building module is used for building a CNN-RNN prediction model for predicting the abrasion loss of the hob;
the training prediction module substitutes the hob wear parameter set and the hob life evaluation standard into a CNN-RNN prediction model, performs iterative training on the CNN-RNN prediction model, obtains a final prediction model after training is completed, and obtains a hob wear loss prediction value by adopting the final prediction model;
and the monitoring module is used for monitoring the shield hob abrasion loss in real time according to the hob abrasion loss prediction value of the prediction module.
Optionally, the standard establishing module is configured to establish a standard for the life evaluation of the hob: the life index of the hob cutter = stratum abrasiveness coefficient + accumulated wear amount of the hob cutter/(hob cutter ring diameter + actual working time of the hob cutter); the accumulated abrasion loss of the hob is the sum of the abrasion losses when the hob is replaced; the actual working time of the hob is the time of interaction between the hob and the stratum in the shield tunneling process, corresponding to the accumulated abrasion loss of the hob.
The second aspect of the invention provides a shield hob abrasion loss real-time monitoring method, which comprises the following steps:
acquiring hob abrasion parameters in the shield construction process to form a hob abrasion parameter set;
establishing a hob service life evaluation standard based on the accumulated wear loss of the hob, the actual working time of the hob corresponding to the wear loss and the diameter of a hob ring of the hob;
constructing a CNN-RNN prediction model for predicting the abrasion loss of the hob;
substituting the hob abrasion parameter set and the hob service life evaluation standard into a CNN-RNN prediction model, performing iterative training on the CNN-RNN prediction model, obtaining a final prediction model after training is completed, and obtaining a hob abrasion loss prediction value by adopting the final prediction model;
and monitoring the abrasion loss of the shield hob in real time according to the predicted value of the abrasion loss of the hob of the prediction module.
The invention also provides a real-time monitoring terminal for the abrasion loss of the shield hob, which comprises a memory, a processor and a computer program which is stored in the memory and can run on the processor, wherein the processor is used for running the real-time monitoring system for the abrasion loss of the shield hob or executing the real-time monitoring method for the abrasion loss of the shield hob when executing the program.
In a fourth aspect of the present invention, a computer-readable storage medium is provided, on which a computer program is stored, where the computer program is used to operate the real-time shield hob wear monitoring system or execute the real-time shield hob wear monitoring method when executed by a processor.
Compared with the prior art, the embodiment of the invention has at least one of the following beneficial effects:
the real-time monitoring system and the method for the abrasion loss of the shield hob provided by the invention are based on the evaluation standard of the hob service life in the shield tunneling construction process, the CNN-RNN combined deep learning network is adopted to realize the real-time determination of the abrasion loss of the shield hob, and the abrasion loss of a single hob can be predicted in real time along with the construction progress. The CNN-RNN model can effectively identify parameters with strong correlation with the hob life evaluation standard in the hob wear parameter set, more feature information of the parameters is reserved in model iterative calculation, and meanwhile, the participation degree of parameters with poor correlation in the iterative calculation process is filtered, so that the model convergence speed is increased, and the model calculation cost is reduced.
The invention provides a real-time monitoring system and a real-time monitoring method for the abrasion loss of a shield hob, wherein the influence of the change of tunneling parameters on the abrasion of the hob in shield construction is fully considered by the hob life evaluation standard, and the stratum abrasiveness coefficient is combined in the evaluation standard, so that the influence of the stratum abrasiveness of a hob changing position on the abrasion of the hob can be quantitatively reflected, the change condition of the abrasion loss of the hob under the actual application working condition can be accurately predicted in advance, shield site constructors can be helped to judge the cutter inspection time, the shield machine halt times and the shutdown time of each time are reduced, the construction efficiency is improved, and the construction cost is reduced. The method provided by the invention is simple, efficient, high in accuracy and high in application and popularization value.
Drawings
Other features, objects and advantages of the invention will become more apparent upon reading of the detailed description of non-limiting embodiments with reference to the following drawings:
FIG. 1 is a block diagram of a shield hob abrasion loss real-time monitoring system according to an embodiment of the present invention;
FIG. 2 is a flow chart of a shield hob abrasion loss real-time monitoring method according to an embodiment of the present invention;
FIG. 3 is a graph of cumulative wear of a hob as a function of operating time in accordance with one embodiment of the present invention;
FIG. 4 is a diagram of a CNN-RNN deep learning network structure according to an embodiment of the present invention;
FIG. 5 is a diagram illustrating the predicted results of the amount of wear of the hob according to an embodiment of the present invention.
Detailed Description
The present invention will be described in detail with reference to specific examples. The following examples will aid those skilled in the art in further understanding the present invention, but are not intended to limit the invention in any manner. It should be noted that variations and modifications can be made by persons skilled in the art without departing from the spirit of the invention. All falling within the scope of the present invention.
Referring to fig. 1, in this embodiment, a system for monitoring wear loss of a shield hob in real time is provided, including: the system comprises a parameter acquisition module, a standard establishment module, a model establishment module, a training prediction module and a monitoring module, wherein the parameter acquisition module is used for acquiring hob abrasion parameters in the shield construction process to form a hob abrasion parameter set; the standard establishing module establishes a hob service life evaluation standard based on the accumulated wear loss of the hob, the actual working time of the hob corresponding to the wear loss and the diameter of a hob ring of the hob; the model construction module constructs a CNN-RNN prediction model for predicting the abrasion loss of the hob; substituting the hob abrasion parameter set and the hob service life evaluation standard into the CNN-RNN prediction model by the training prediction module, carrying out iterative training on the CNN-RNN prediction model, obtaining a final prediction model after the training is finished, and obtaining a hob abrasion loss prediction value by adopting the final prediction model; and the monitoring module monitors the wear loss of the shield hob in real time according to the hob wear loss prediction value of the prediction module.
The embodiment overcomes the defect that the health degree of the whole hob of the cutterhead is only considered in the prior art, realizes the real-time prediction of the abrasion loss of a certain hob in the shield tunneling process, improves the accuracy and reliability of the prediction result, reduces the times and time for examining and maintaining the hob in construction, effectively improves the construction efficiency of the shield tunnel, and reduces the risk of the hob changing process.
In the embodiment of the invention, the parameter acquisition module is used for acquiring a data sample of the time-varying tunneling parameters influencing the abrasion of a hob in shield construction, wherein the tunneling parameters comprise: shield machine power system parameters; a cutter head system parameter; parameters of a slag discharge system; adjusting system parameters by using the muck; and (4) shield tunnel geometric parameters. In some preferred embodiments, the parameter obtaining module further comprises a normalization processing sub-module, which performs non-dimensionalization processing on the hob wear parameter set and then inputs the processed hob wear parameter set into the training prediction module.
In the research process, the hob abrasion loss is influenced by different stratum conditions in actual construction, particularly when the stratum conditions are complex, and in order to evaluate the conditions more accurately, the stratum abrasiveness coefficient is considered in establishing an evaluation standard. Specifically, in a preferred embodiment of the present invention, the evaluation criterion of the lifetime of the hob established by the criterion establishing module is as follows: the service life index of the hob is = stratum abrasiveness coefficient and accumulated wear loss of the hob/(hob ring diameter and actual working time of the hob); the accumulated abrasion loss of the hob is the sum of abrasion losses of the hob during replacement; the actual working time of the hob is the time of interaction between the hob and the stratum in the shield tunneling process, corresponding to the accumulated abrasion loss of the hob. The evaluation standard of the service life of the hob in the embodiment fully considers the influence of the tunneling parameter change on the abrasion of the hob in shield construction, can predict the change condition of the abrasion quantity of the hob in advance, and can provide a prepared monitoring result, so that shield site constructors can judge the cutter inspection time conveniently, and the number of times of shield machine halt and the time length of each halt are reduced.
In some embodiments, the accumulated wear amount of the hob is obtained by a measuring module, and the measuring module outputs the obtained accumulated wear amount of the hob to a standard establishing module for establishing a life evaluation standard of the hob.
In some embodiments, the actual working time of the hob may be determined by the working time determination module, and the actual working time of the hob corresponding to the accumulated wear loss of the hob is obtained and output to the standard establishment module for establishing the evaluation standard of the lifetime of the hob.
Specifically, in a preferred embodiment, for a time period from installation start to current time period of the hob, the working time determination module further includes a shield non-working time filtering submodule, a shield stability stage data retention submodule and a hob actual working time determination submodule, wherein the shield non-working time filtering submodule is used for filtering out the non-working time of the shield machine in the ith abrasion loss record and the (i-1) th abrasion loss record section of the hob; the shield stability stage data retention submodule is used for filtering out the initial stage and end stage data of the shield machine in each ring tunneling process in the ith abrasion loss record and the (i-1) th abrasion loss record section and retaining the stable tunneling stage data; and after the actual working time determining submodule processes the shield machine non-working time filtering submodule and the shield stable stage data retaining submodule according to the shield machine non-working time, the ith abrasion loss record of the hob and the residual data sample amount in the i-1 th abrasion loss recording section, wherein the shield working time corresponding to the residual data sample amount is the actual working time corresponding to the accumulated abrasion loss of the hob. The state of the hob can be further accurately predicted through the actual working time corresponding to the accumulated abrasion loss of the hob, and conditions are improved for improving the accuracy and reliability of a prediction result.
In some embodiments, the model building module builds a CNN-RNN prediction model comprising an input layer, a convolutional layer, a pooling layer, an RNN network layer, a fully-connected layer, and an output layer connected in series in sequence, wherein:
the input layer inputs data of the hob abrasion parameter set;
the convolution layer is used for carrying out one-dimensional convolution processing on the input hob abrasion parameter set data;
the pooling layer is used for carrying out feature extraction processing on the hob abrasion parameter set data after convolution processing;
the RNN network layer updates historical information of the time sequence data of the hob abrasion parameter set after the pooling processing;
the full connection layer is used for calibrating the hob abrasion parameter set data after the time sequence conversion;
and outputting the standard data of the hob service life evaluation by the output layer.
In some embodiments, the training prediction module substitutes the hob wear parameter set into the input layer of the CNN-RNN prediction model, substitutes the hob life evaluation standard into the output layer of the CNN-RNN prediction model, performs iterative training on the CNN-RNN prediction model, updates the weight matrix and the offset matrix in the model, and obtains the final prediction model after the training is completed. The hob life evaluation standard is output of the CNN-RNN prediction model, and the difference between the output value of the CNN-RNN prediction model and the output value of the CNN-RNN prediction model in the iteration process is used as model error back propagation, so that the weight matrix and the bias matrix are updated to obtain an optimized model, and a final prediction result is obtained by adopting the optimized model.
The above parts not specifically described may all be implemented by using the prior art, and are not described herein again.
Based on the same technical concept, the invention further provides a real-time monitoring method for the abrasion loss of the shield hob in another embodiment. Specifically, referring to fig. 2, the method for monitoring the wear loss of the shield hob in real time in the embodiment includes the following steps:
s1, acquiring hob abrasion parameters in a shield construction process to form a hob abrasion parameter set;
s2, establishing a hob service life evaluation standard based on the accumulated wear loss of the hob, the actual working time of the hob corresponding to the wear loss and the diameter of a hob ring of the hob;
s3, constructing a CNN-RNN prediction model for predicting the abrasion loss of the hob;
s4, substituting the hob abrasion parameter set and the hob service life evaluation standard into the CNN-RNN prediction model, performing iterative training on the CNN-RNN prediction model, obtaining a final prediction model after the training is completed, and obtaining a hob abrasion loss prediction value by adopting the final prediction model;
and S5, monitoring the abrasion loss of the shield hob in real time according to the hob abrasion loss prediction value of the prediction module.
In the above embodiment, before S2, the measurement of the accumulated wear amount of the hob and the determination of the actual working time of the shield may be further included, which are used to obtain the accumulated wear amount of the hob and the actual working time of the hob corresponding to the wear amount, respectively.
Specifically, the cumulative wear amount of the hob refers to the sum of the wear amounts at the time of replacement of the hob at a certain mounting position. In some embodiments, the amount of wear is obtained by manual measurement or other image processing methods. For example, the manual measurement can adopt a measuring knife caliper matched with the size of a hob ring to measure the radial abrasion of the hob when the shield machine is opened for inspection. The image processing can be performed by acquiring images of the front hob and the rear hob and acquiring hob contour lines in the front hob and the rear hob, so as to determine the radial abrasion loss of the hob.
Specifically, the actual working time refers to the time when the hob at a certain installation position interacts with the stratum in the shield tunneling process, and the method can specifically determine the working time determination module in the system.
Specifically, the life evaluation criterion of the hob is established in S2, wherein the wear state evaluation criterion of the hob can be specifically expressed by the following formula:
CLI＝αW a /(d·t w )
in the formula, CLI is the index of the abrasion state of the hob, and the unit is 1/min; alpha is the abrasiveness coefficient of the stratum at the tool changing position, 0.5-1 is taken, and the abrasiveness coefficient is obtained by inquiring from an engineering geological exploration report; w a The accumulated abrasion loss of the hob is in mm; t is t w The unit of the actual working time corresponding to the accumulated abrasion loss of the hob is min; d is the diameter of the hob ring and the unit is mm. Specifically, the abrasiveness coefficient of the stratum at the tool changing position can be obtained according to an engineering geological exploration report.
According to the hob abrasion state evaluation standard, the hob abrasion loss and the corresponding actual working time of the hob are considered, the evaluation accuracy can be improved, meanwhile, the abrasion coefficient of a stratum at a hob changing position is combined, the environmental parameters of the actual working of a shield machine are considered, the effect of predicting the hob abrasion loss can be further improved, and the prediction effect is more consistent with the actual situation of the hob.
In another embodiment of the present invention, a terminal for monitoring the wear loss of a shield hob in real time is further provided, which includes a memory, a processor, and a computer program stored in the memory and capable of being executed on the processor, where the processor is configured to execute the system for monitoring the wear loss of a shield hob in real time or execute the method for monitoring the wear loss of a shield hob in real time in the above embodiment when executing the program.
In another embodiment of the present invention, a computer-readable storage medium is further provided, on which a computer program is stored, where the computer program is used to execute the system for monitoring the wear of the shield hob in real time or the method for monitoring the wear of the shield hob in real time in the above embodiments.
In order to better understand the technical scheme of the invention, the following detailed description is taken as an example in connection with a specific application test.
Taking the construction of a slurry shield tunnel penetrating through a core area of a city as an example, the total length of the tunnel is 3.603km, the burial depth is 22-62m, the excavation diameter of the slurry balance shield machine is 15.81m, the inner diameter and the outer diameter of each pipe piece are respectively 13.9m and 15.2m, and the width is 2.0m. The opening rate of the cutter head is about 29 percent, 82 disc cutters are installed together, a double-shaft double-edge cutter with the diameter of 17 inches is adopted in the center area, a double-shaft double-edge cutter with the diameter of 19 inches is adopted in the front and edge areas, and a single-edge cutter with the diameter of 19 inches is adopted in the outermost side. In this embodiment, the method for monitoring the wear of the shield hob in real time is described in detail by taking the prediction of the wear of the outermost 81 # single-blade hob as an example.
As shown in fig. 3 to fig. 5, the present embodiment provides a shield hob abrasion real-time monitoring method based on a CNN-RNN combined deep learning network, which may be performed according to the following specific steps:
step one, acquiring hob abrasion parameters in a shield construction process by adopting a parameter acquisition module to form hob abrasion parameter sets;
the hob abrasion parameter set refers to a data sample of the time-varying excavation parameters influencing hob abrasion in shield construction.
In this embodiment, the tunneling parameters affecting the wear of the hob include a shield machine power system parameter, a cutter head system parameter, a slag discharge system parameter, a slag soil adjustment system parameter, and a shield tunnel geometric parameter. The power system parameters comprise shield thrust, propulsion speed, penetration degree, soil cabin pressure and six groups of shield hydraulic oil cylinder jacking force; the cutter parameters comprise the rotating speed of the cutter and the torque of the cutter; the slag discharge system parameters comprise the rotating speed of the screw machine, the torque of the screw machine and the soil pressure of the screw machine; the parameters of the muck adjusting system comprise nine groups of foam pipe pressure, nine groups of foam flow rate and nine groups of air flow rate; the geometric parameters of the shield tunnel are taken as the distance from the tunnel top plate to the ground.
Further, a normalization processing submodule is adopted to carry out non-dimensionalization processing on the hob abrasion parameter set. In this embodiment, the normalization processing refers to performing non-dimensionalization processing on the hob wear parameter set, and the normalization formula is shown as follows:
x is a data sample after normalization processing; x is a radical of a fluorine atom min And x max Respectively setting the minimum value and the maximum value of each variable in the hob abrasion parameter set; x is the original data sample. Normalization can make the subsequent model easier to converge and prevent overfitting.
Secondly, a standard establishing module is adopted, and a hob service life evaluation standard is established based on the accumulated abrasion loss of the hob, the actual working time of the hob corresponding to the abrasion loss and the diameter of a hob ring of the hob;
in the step, a service life evaluation standard of the No. 81 hob is established through a standard establishing module. The cumulative wear of the 81 hob as a function of operating time is shown in fig. 3. In this step, the service life evaluation standard of the 81 # hob is an index for measuring the wear condition of the 81 # hob in the shield tunneling process, and is determined by the following formula:
CL＝αW a /(d·t w )
in the formula, CL is the service life index of a No. 81 hob and the unit is mm/min; alpha is the abrasiveness coefficient of the stratum at the cutter changing position of the No. 81 hob,taking 0.5-1; w is a group of a The accumulated abrasion loss of the No. 81 hob is in mm; t is t w The working time corresponding to the accumulated abrasion loss of the No. 81 hob is min; d is the diameter of the hob ring of No. 81 hob with unit of mm. In the engineering of the embodiment, the diameter of the No. 81 hob ring is 483mm, and the formation abrasiveness coefficient is 1.
In the embodiment, in order to evaluate and monitor the service life of the hob more accurately, the abrasiveness coefficient of the stratum at the tool changing position of the hob is considered in the established hob life evaluation standard, so that the influence of the abrasiveness of the stratum at the tool changing position on the abrasion of the hob can be quantitatively reflected, and the shield machine is favorable for practical application in a shield construction site. In addition, the diameters of the hobbing cutters on the cutter head of the shield machine in the construction site are different, the geometrical size of the hobbing cutter needing to predict the abrasion condition is considered by the hobbing cutter service life evaluation standard, the actual condition of the shield construction in the construction site is more met, and the method has strong practicability.
In this step, the measurement module is adopted to measure the cumulative wear loss of the No. 81 hob, namely: when the No. 81 hob is replaced in the shield tunneling process, the sum of the radial abrasion loss of the hob is obtained by measuring with a special measuring cutter caliper for the abrasion loss of the 19-inch hob.
In the step one, an actual working time determining module is adopted to determine the actual working time corresponding to the accumulated abrasion loss of the No. 81 hob in the step one. Specifically, the actual working time refers to the interaction time of the 81 # hob with the stratum in the shield tunneling process, and is determined by the following formula:
in the formula, t w Representing the actual working time of the No. 81 hob; t is t i Representing the interaction time of the hob and the stratum in the ith-1 th abrasion loss recording section and the ith abrasion loss recording section of the No. 81 hob; n represents the number of times of recording the wear amount of the # 81 hob, and in this embodiment, the number of times of recording the wear amount of the # 81 hob, n, is 28.
In this embodiment, the shield non-operating time filtering submodule is used to filter the non-operating time of the shield machine in the i-th wear loss recording section and the i-1 st wear loss recording section of the 81 # hob, and the specific non-operating time of the shield machine is determined by the following formula:
F＝f(AR)×f(PE)×f(CRS)
in the formula, if the value F is zero, the shield machine is in a non-working state, AR is the tunneling speed, PE is the penetration degree, and CRS is the cutter head rotating speed. f (x) is a function for judging whether the tunneling speed, the penetration degree and the cutter head rotating speed contain zero values or not, and is shown as the following formula:
further, a shield steady stage data retention submodule is adopted to filter the data of the initial section and the final section of the shield machine in each ring tunneling process of the shield machine in the ith abrasion loss recording section and the (i-1) th abrasion loss recording section of the 81 # hob, and the data of the steady tunneling section is retained. The criteria met by the stability phase data are determined by:
in the formula (I), the compound is shown in the specification,
the average value of the variable j in the k-th loop tunneling process is obtained;
the standard deviation of a variable j in the k-th loop tunneling process is obtained; g j Setting the value of a variable j in the data of each stable tunneling section of the shield tunneling machine; m is the number of tunneling rings; TF is shield thrust; TO is the torque of the shield cutter head; and CRS is the rotating speed of the shield cutter head.
Determining interaction time t of No. 81 hob and stratum by a hob actual working time determining submodule i Sub-module equal to sub-module filtered by shield non-working time and sub-module reserved by shield stable stage dataAfter the processing, the ith abrasion loss of the No. 81 hob records the residual data sample size in the section of the ith-1 abrasion loss recording, and the shield working time corresponding to the residual data sample size is the actual working time corresponding to the cumulative abrasion loss of the No. 81 hob.
And step three, establishing a CNN-RNN deep learning network model by adopting a model construction module.
In this step, the CNN-RNN deep learning network model network structure is shown in fig. 4, and is composed of an input layer, a convolution layer, a pooling layer, an RNN network layer, a full connection layer, and an output layer.
The model parameters include the number of neurons in the input layer, the fully-connected layer, and the output layer, the number of iterations, the cost function, and the optimizer.
And the number of neurons in the input layer is equal to the number of parameters in the hob abrasion parameter set. In this example, the number of input layer neurons is 43.
The convolutional layer is determined by the convolutional kernel depth, convolutional kernel size, and convolutional step size, and the pooling layer is determined by the pooling region size and pooling step size. In this embodiment, the depth of the convolution kernel in the convolution layer is 24, the size of the convolution kernel is 6, and the step size is 2. The size of the pooling area in the pooling layer is 2 and the step size is 1.
The calculation formula of the RNN deep learning network is as follows:
O t ＝tanh(VS t )
S t ＝σ(Ux t +WS t-1 )
in the formula, U, V and W are weight matrixes of current input, a hidden state at the current moment and a hidden state at the last moment respectively; s. the t Is a hidden state at the moment t; s t-1 Is a hidden state at the moment of t-1; o is t Is the output of RNN unit at time t; tan h is a hyperbolic tangent function; sigma is a sigmoid function.
The iteration times are calculated by substituting the hob wear parameter set and the hob life evaluation standard into the CNN-RNN deep learning network model, and in this embodiment, the iteration times are set to 200 times.
The cost function is a function for measuring the error magnitude of the predicted value and the measured value of the CNN-RNN deep learning network model. In this embodiment, a mean square error function (MSE) is used as the cost function.
The optimizer is an algorithm for optimizing a weight matrix and a bias matrix in the CNN-RNN deep learning network model in each iteration process. In this embodiment, the Adam algorithm is used as an optimization algorithm for the weight matrix and the bias matrix.
And step four, substituting the input hob wear parameter set and the hob life evaluation standard into the CNN-RNN deep learning network model by adopting a training prediction module, and updating model training parameters by utilizing an optimizer.
In the step, after the input hob abrasion parameter set and the hob service life evaluation standard are substituted into the CNN-RNN deep learning network model, the prediction mode, the number of neurons of a full connection layer and an output layer are further determined, wherein the prediction mode refers to historical information and output information considered by the CNN-RNN model for predicting hob abrasion. Specifically, the amount of history information and the amount of output information are determined by the following formula:
in the formula, t i Is the historical information amount; t is t o Is the output information quantity; and T is the average time required by the shield to tunnel a ring.
In the embodiment, the average time T =200min for one ring of shield tunnel tunneling; amount of history information t i =5min; quantity of output information t o ＝3min。
The number of output layer neurons equals the amount of output information. The number of output layer neurons in this example is 3.
The number of full-junction neurons was determined by the following formula:
q＝p 2 -1
in the formula, q is the number of neurons in the full connecting layer, and p is the number of neurons in the output layer. In this example, the number of full-connectivity layer neurons q =8.
And step five, repeating the step four until the iteration times of the CNN-RNN deep learning model reach 200 times, and outputting a predicted value of the abrasion loss of the hob. Real-time monitoring of shield hob abrasion loss is realized according to predicted value of hob abrasion loss
As shown in fig. 5, it is a diagram of the prediction result of the abrasion loss of the hob in this embodiment 81, and according to the result, the state of the hob can be monitored, for example, the result is sent to a display interface for the reference of a worker, so as to predict the change of the abrasion loss of the hob in advance, or when the abrasion loss of the hob reaches or exceeds the threshold of the abrasion loss of the hob, an alarm is given, which can help shield site constructors to judge the time for checking the hob, and reduce the number of times of shutdown of the shield machine and the time length of each shutdown, thereby improving the construction efficiency and reducing the construction cost.
For example, as can be seen from fig. 5, when the shield advance time is 9000min, the predicted value of the cumulative wear loss of the 81 st hob is 129mm, the measured value of the cumulative wear of the last hob at that time is 127mm, the difference between the two is 2mm, and is less than 5mm, which is the wear limit of the 81 st hob, so that the 81 st hob does not need to be replaced when the shield machine operation time reaches 9000min, and if the difference between the two is greater than or equal to 5mm, the 81 st hob needs to be replaced. If the difference between the two approaches 5mm (for example, reaches 4 mm), the tool change can be prepared in advance according to the field situation.
The foregoing description of specific embodiments of the present invention has been presented. It is to be understood that the present invention is not limited to the specific embodiments described above, and that various changes or modifications may be made by one skilled in the art within the scope of the appended claims without departing from the spirit of the invention.
Claims (10)
1. The utility model provides a shield constructs hobbing cutter wearing and tearing volume real-time monitoring system which characterized in that includes:
the parameter acquisition module is used for acquiring hob abrasion parameters in the shield construction process to form hob abrasion parameter sets;
the standard establishing module establishes a hob service life evaluation standard based on the accumulated wear loss of the hob, the actual working time of the hob corresponding to the wear loss and the diameter of a hob ring of the hob;
the model building module is used for building a CNN-RNN prediction model for predicting the abrasion loss of the hob;
the training prediction module substitutes the hob abrasion parameter set and the hob service life evaluation standard into a CNN-RNN prediction model to perform iterative training on the CNN-RNN prediction model, obtains a final prediction model after the training is finished, and obtains a hob abrasion loss prediction value by adopting the final prediction model;
and the monitoring module is used for monitoring the shield hob abrasion loss in real time according to the hob abrasion loss prediction value of the prediction module.
2. The system for monitoring the abrasion loss of the shield hob according to claim 1, wherein the parameter acquisition module acquires data samples of the time-varying tunneling parameters affecting hob abrasion in shield construction, wherein the tunneling parameters include:
parameters of a shield machine power system;
a cutter head system parameter;
parameters of a slag discharge system;
adjusting system parameters by using the muck;
geometric parameters of the shield tunnel.
3. The shield hobbing cutter wear loss real-time monitoring system of claim 1, wherein the parameter acquisition module further comprises a normalization processing sub-module, wherein:
and the normalization processing sub-module carries out non-dimensionalization processing on the hob wear parameter set and then inputs the processed hob wear parameter set into a training prediction module.
4. The shield hob abrasion loss real-time monitoring system according to claim 1, wherein the standard establishing module is configured to:
the evaluation standard of the service life of the hob is as follows: the service life index of the hob is = stratum abrasiveness coefficient and accumulated wear loss of the hob/(hob ring diameter and actual working time of the hob);
the accumulated abrasion loss of the hob is the sum of the abrasion losses when the hob is replaced;
the actual working time of the hob is the time corresponding to the accumulated abrasion loss of the hob and the interaction between the hob and the stratum in the shield tunneling process.
5. The shield hob abrasion loss real-time monitoring system according to claim 4, further comprising:
the measuring module is used for measuring the accumulated abrasion loss of the hob and outputting the accumulated abrasion loss of the hob to the standard establishing module;
and the working time determining module is used for determining the actual working time of the hob corresponding to the accumulated abrasion loss of the hob and outputting the actual working time of the hob to the standard establishing module.
6. The shield hob abrasion loss real-time monitoring system according to claim 5, wherein the working time determining module includes:
the shield non-working time filtering submodule filters out the non-working time of the shield machine in the ith abrasion loss record and the (i-1) th abrasion loss record section of the hob;
the module filters the initial segment and the end segment data of the shield machine in each ring tunneling process in the ith abrasion loss record and the (i-1) th abrasion loss record section and reserves the stable tunneling segment data;
and after the actual working time of the hob is processed by the shield machine non-working time filtering submodule and the shield stable stage data retaining submodule, the ith abrasion loss record of the hob and the residual data sample size in the (i-1) th abrasion loss record section, wherein the shield working time corresponding to the residual data sample size is the actual working time corresponding to the accumulated abrasion loss of the hob.
7. The shield hobbing cutter abrasion loss real-time monitoring system according to claim 1, wherein the CNN-RNN prediction model constructed by the model construction module comprises an input layer, a convolution layer, a pooling layer, an RNN network layer, a full connection layer and an output layer which are sequentially connected in series, wherein:
the input layer inputs data of hob abrasion parameter sets;
the convolution layer is used for carrying out one-dimensional convolution processing on the input hob abrasion parameter set data;
the pooling layer is used for carrying out feature extraction processing on the hob abrasion parameter set data after convolution processing;
the RNN network layer is used for carrying out time sequence transformation processing on the hob abrasion parameter set data after the pooling processing;
the full connection layer is used for calibrating the hob abrasion parameter set data after the time sequence transformation;
and outputting the standard data of the hob service life evaluation by the output layer.
8. A shield hob abrasion loss real-time monitoring method is characterized by comprising the following steps:
acquiring hob abrasion parameters in the shield construction process to form a hob abrasion parameter set;
establishing a hob service life evaluation standard based on the accumulated wear loss of the hob, the actual working time of the hob corresponding to the wear loss and the diameter of a hob ring of the hob;
constructing a CNN-RNN prediction model for predicting the abrasion loss of the hob;
substituting the hob abrasion parameter set and the hob service life evaluation standard into a CNN-RNN prediction model, performing iterative training on the CNN-RNN prediction model, obtaining a final prediction model after the training is completed, and obtaining a hob abrasion loss prediction value by adopting the final prediction model;
and realizing real-time monitoring on the shield hob abrasion loss according to the hob abrasion loss predicted value.
9. A real-time monitoring terminal for the abrasion loss of a shield hob, which comprises a memory, a processor and a computer program which is stored in the memory and can run on the processor, and is characterized in that the processor is used for running the real-time monitoring system for the abrasion loss of the shield hob according to any one of claims 1 to 8 or executing the real-time monitoring method for the abrasion loss of the shield hob according to claim 9 when executing the program.
10. A computer-readable storage medium, on which a computer program is stored, wherein the program, when executed by a processor, is configured to operate the system for monitoring the wear of the shield hob according to any one of claims 1 to 8 in real time, or to perform the method for monitoring the wear of the shield hob according to claim 9 in real time.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211185851.9A CN115544874A (en) | 2022-09-27 | 2022-09-27 | Real-time monitoring system, method, terminal and medium for shield hob abrasion loss |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211185851.9A CN115544874A (en) | 2022-09-27 | 2022-09-27 | Real-time monitoring system, method, terminal and medium for shield hob abrasion loss |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115544874A true CN115544874A (en) | 2022-12-30 |
Family
ID=84730473
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211185851.9A Pending CN115544874A (en) | 2022-09-27 | 2022-09-27 | Real-time monitoring system, method, terminal and medium for shield hob abrasion loss |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115544874A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115839344A (en) * | 2023-02-17 | 2023-03-24 | 石家庄宏昌泵业有限公司 | Wear monitoring method, device, equipment and storage medium for slurry pump |
-
2022
- 2022-09-27 CN CN202211185851.9A patent/CN115544874A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115839344A (en) * | 2023-02-17 | 2023-03-24 | 石家庄宏昌泵业有限公司 | Wear monitoring method, device, equipment and storage medium for slurry pump |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113886917B (en) | CNN-LSTM model-based railway line region ground settlement prediction early warning method | |
| WO2006001121A1 (en) | Data compressing device and method, data analyzing device and method, and data managing system | |
| CN110210169B (en) | LSTM-based shield tunneling machine fault prediction method | |
| CN112966425B (en) | Slope stability prediction and evaluation method | |
| CN113404502B (en) | Shield hob abrasion monitoring device and method based on ballast piece morphology | |
| CN115600654A (en) | Convolution LSTM-based shield hob abrasion loss real-time prediction method and system | |
| CN115544874A (en) | Real-time monitoring system, method, terminal and medium for shield hob abrasion loss | |
| CN114417697A (en) | Neural network-based TBM hob abrasion real-time prediction method and system | |
| CN115563863A (en) | Method and terminal for determining abrasion loss of slurry shield hobbing cutter in real time based on deep learning | |
| CN112131796B (en) | Real-time hard rock TBM hob abrasion evaluation method based on gene expression programming | |
| CN115544753A (en) | Real-time assessment method, system, terminal and medium for shield hob abrasion state | |
| CN115234220A (en) | Method and device for identifying underground stick-slip vibration in real time by using intelligent drill bit | |
| CN116245020A (en) | Cutter head abrasion and energy consumption optimization method and system based on shield tunneling machine | |
| CN113901609A (en) | Method for analyzing relevance of multi-source tunneling parameters of shield tunneling machine | |
| CN117934244A (en) | System and method for tough urban ecological planning and design | |
| CN111340275B (en) | Tunnel support mode selection real-time prediction method based on detection while drilling technology | |
| CN115563872B (en) | TBM hob abrasion loss prediction method based on DE-SVR algorithm | |
| CN115563760A (en) | Shield hob service life assessment index construction method, system, terminal and medium | |
| CN113158562B (en) | TBM rock machine mapping construction method and system based on physical constraint and data mining | |
| CN114382490A (en) | Shield tunneling machine cutter head wear assessment and prediction method | |
| CN107679330A (en) | A kind of real-time estimating method of TBM cutter disc systems broken rock performance extent of deterioration | |
| CN114722924A (en) | Pressure prediction method and device, storage medium and terminal | |
| CN115495987A (en) | Hobbing cutter actual working time determining and abrasion loss real-time predicting method and terminal | |
| CN117851761B (en) | Method and system for evaluating states of cutterheads of shield machine | |
| CN113139340B (en) | Real-time prediction method, terminal and medium for shield tunneling line |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |