Earth pressure balance shield hob abnormal wear quantity prediction method

Abstract

The invention discloses an earth pressure balance shield hob abnormal wear quantity prediction method. The method comprises the following steps: establishing a tunneling parameter, geological parameter, hob mechanical parameter and hob abnormal wear quantity data set; the data sets correspond to tunneling ring numbers, feature parameters are extracted from tunneling parameters, a hob mechanical parameter data set and a geological parameter data set are combined to serve as an input sample data set, and a hob abnormal wear quantity data set serves as an output sample data set; and establishing a hob abnormal wear prediction model, and performing multiple iterative calculations according to the sample data set until the model error meets the requirement to obtain a trained hob abnormal wear quantity prediction model. Feature parameters of tunneling parameters, geological parameters and hob mechanical parameters which need to be predicted serve as input data, the trained hob abnormal wear quantity prediction model is input, the hob abnormal wear quantity of the earth pressure balance shield is predicted in real time, and whether the hob needs to be maintained and replaced or not is judged.

Description

Technical Field

[0001] The invention belongs to the technical field of construction control of full-section tunnel boring machines, and in particular relates to a method for predicting the abnormal wear quantity of a cutter of an earth pressure balance shield. Background Art

[0002] Entering the 21st century, the speed of urban construction in my country has been accelerating. In large-scale urban construction and underground space development, the number of subway tunnel construction projects has been increasing, among which shield tunnel construction has been widely used. The shield machine is mainly composed of a roller cutter to break rock. During the excavation process, the roller cutter is squeezed into the rock, and the roller cutter is pressed into the stone. The rock is plastically deformed and broken by rotation, and then the slag system transmits the broken stone to the rear. The roller cutter is subjected to all-round, leaping and cyclic loads when breaking rock. Therefore, when the roller cutter works continuously for a long time, it will inevitably produce abnormal wear under the action of load, reducing the shield excavation efficiency. At present, the main method for judging abnormal wear of the roller cutter is still based on the experience of shield machine operators and experts. This method has a lot of uncertainties and it is difficult to accurately judge the health of the roller cutter. Therefore, a method for predicting the number of abnormal wear of the shield roller cutter will provide a theoretical basis for the research on efficient excavation and optimization control of the shield roller cutter, and it is of great significance.

[0003] At present, the earth pressure balance shield construction process uses the method of opening the cabin to check whether the roller cutter is abnormally worn, but it requires frequent shutdown to change the cutter, which makes it difficult to excavate efficiently. In addition, opening the cabin in poor geological conditions may cause the tunnel face to become unstable, increasing the risk of the cutter change operation. Therefore, a new prediction method for abnormal wear of the roller cutter of the earth pressure balance shield machine needs to be proposed. Summary of the invention

[0004] The problem to be solved by the present invention is to provide a method for predicting the abnormal wear quantity of earth pressure balance shield cutters, which can realize the prediction of the health status of the shield cutters.

[0005] In order to solve the above technical problems, the present invention adopts the following technical solutions:

[0006] In a first aspect, the present application provides a method for predicting the amount of abnormal wear of a cutter of an earth pressure balance shield, comprising:

[0007] The characteristic parameters of the excavation parameters that need to be predicted, geological parameters and mechanical parameters of the cutter are used as input data and input into a trained cutter abnormal wear quantity prediction model. The cutter abnormal wear quantity prediction model outputs the amount of abnormal wear of the earth pressure balance shield cutter, which is used to determine whether the cutter needs maintenance and replacement. Among them, the trained cutter abnormal wear quantity prediction model has the function of predicting and outputting the corresponding amount of abnormal wear of the earth pressure balance shield cutter according to the input data.

[0008] In a possible implementation, the method further includes: training the abnormal wear quantity prediction model of the hob;

[0009] Training the abnormal wear quantity prediction model of the hob includes:

[0010] Acquire a tunneling parameter data set, which includes data samples of tunneling parameters that affect cutter wear during earth pressure balance shield tunneling over time;

[0011] Acquire a geological parameter data set, which includes data samples of geological parameters changing with excavation mileage during earth pressure balance shield excavation;

[0012] Obtaining a cutter mechanical parameter data set, the data set including data samples of cutter parameters on an earth pressure balance shield cutterhead;

[0013] Obtain a dataset of abnormal wear quantity of roller cutters, which includes data samples of the change of abnormal wear quantity of roller cutters with excavation mileage during earth pressure balance shield excavation;

[0014] The excavation parameters, geological parameters, hob mechanical parameters and the number of abnormal wear of the hob in the excavation parameter data set, geological parameter data set, hob mechanical parameter data set and the number of abnormal wear of the hob are matched according to the excavation ring number, and characteristic parameters are extracted from the excavation parameters according to the excavation ring number;

[0015] Constructing an input sample data set and an output sample data set; the input sample data set includes a plurality of input samples, each of which is a set of characteristic parameters, geological parameters and mechanical parameters of a set of excavation parameters corresponding to an excavation ring number; the output sample data set includes a plurality of output samples, each of which is a number of abnormal wear of a excavation cutter corresponding to an excavation ring number;

[0016] The above input sample data set and output sample data set are time series sample data sets.

[0017] A prediction model for the number of abnormal wear of hobs is established. During the training process, the input sample data set and the output sample data set are used to perform multiple iterative calculations, and the prediction accuracy of the prediction model for the number of abnormal wear of hobs is evaluated until the error of the prediction model for the number of abnormal wear of hobs meets the requirements, thereby obtaining a trained prediction model for the number of abnormal wear of hobs.

[0018] In a possible implementation, the excavation parameters in the excavation parameter data set include cutter head rotation speed, soil bin pressure, propulsion speed, cutter head torque, total thrust and pitch angle;

[0019] The geological parameters in the geological parameter data set include uniaxial compressive strength and rock mass integrity;

[0020] The hob parameters in the hob mechanical parameter data set include installation radius, hob size, material parameters and installation angle.

[0021] In a possible implementation, the step of obtaining a tunneling parameter data set includes:

[0022] The original excavation parameter data set is preprocessed to exclude abnormal values ​​and values ​​that are too large or too small in the original excavation parameter data set, and the statistical 3σ principle is used to eliminate values ​​that are more than 3 times the standard deviation from the mean value.

[0023] In a possible implementation, characteristic parameters are extracted from the excavation parameters according to the excavation ring number, including: calculating the average value and variance of each excavation parameter according to the excavation ring number, and using them as characteristic parameters.

[0024] In a possible implementation, the evaluation of the prediction accuracy of the hob abnormal wear quantity prediction model includes: using a correlation coefficient to evaluate the prediction accuracy of the hob abnormal wear quantity prediction model, and if the correlation coefficient reaches above a preset threshold, it is determined that the hob abnormal wear quantity prediction model error meets the requirements.

[0025] In some embodiments, the preset threshold is 0.9.

[0026] In a possible implementation, a prediction model for the abnormal wear quantity of a hob is established based on a back propagation (BP) neural network model optimized by a genetic algorithm (GA).

[0027] In a possible implementation, the method includes determining whether the cutter needs to be repaired and replaced based on whether the number of abnormal wear of the earth pressure balance shield cutter output by the cutter abnormal wear prediction model reaches a specified number.

[0028] In a second aspect, the present application provides an electronic device, including: a memory and a processor;

[0029] The memory is used to store computer programs;

[0030] The processor is used to call the computer program to execute the method as described above.

[0031] In a third aspect, the present application provides a computer-readable storage medium, in which a computer program is stored. When the computer program is executed on an electronic device, the electronic device implements the method as described above.

[0032] In a fourth aspect, the present application provides a computer program product, including a computer program, and when the computer program is executed on an electronic device, the electronic device implements the method as described above.

[0033] The specific implementation methods of the second to fifth aspects of the present application can refer to the implementation methods of the first aspect, which will not be repeated here.

[0034] Beneficial effects: This application studies the causes of abnormal conditions of the cutters, analyzes the key influencing factors that cause abnormal wear of the cutters, and then combines machine learning to build and train a model for the quantity of abnormal wear of the cutters. Based on the trained model, real-time prediction of the quantity of abnormal wear of the cutters of the earth pressure balance shield is performed. The health status of the cutters can be predicted in time, and it can be determined whether the cutters need maintenance and replacement, and then personnel can be organized to carry out repairs and replacements to improve excavation efficiency, reduce operational risks, and achieve refined management. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 This is a flowchart of the application example of the present application;

[0036] Figure 2 This is a schematic diagram of the correlation analysis of the average values ​​of the excavation parameters in the embodiment of the present application;

[0037] Figure 3 This is a schematic diagram of the variance correlation analysis of tunneling parameters in the embodiment of the present application;

[0038] Figure 4 This is a schematic diagram of a tunneling parameter data set in an embodiment of the present application;

[0039] Figure 5 This is a schematic diagram of the GA-BP neural network model in the embodiment of the present application. DETAILED DESCRIPTION

[0040] The technical solution of the present invention will be clearly and completely described below through specific embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0041] like Figure 1 As shown, in the first aspect, the present application provides a method for predicting the abnormal wear quantity of earth pressure balance shield cutter, comprising:

[0042] The characteristic parameters of the excavation parameters that need to be predicted, geological parameters and mechanical parameters of the cutter are used as input data and input into a trained cutter abnormal wear quantity prediction model. The cutter abnormal wear quantity prediction model outputs the amount of abnormal wear of the earth pressure balance shield cutter, which is used to determine whether the cutter needs maintenance and replacement. Among them, the trained cutter abnormal wear quantity prediction model has the function of predicting and outputting the corresponding amount of abnormal wear of the earth pressure balance shield cutter according to the input data.

[0043] The characteristic parameters of the tunneling parameters, geological parameters and roller cutter mechanical parameters that need to be predicted may be the characteristic parameters of the tunneling parameters, geological parameters and roller cutter mechanical parameters collected in real time.

[0044] Based on this solution, the health of the hob can be evaluated to determine whether the hob needs maintenance and replacement, so as to adjust the hob in real time.

[0045] In some embodiments, before applying the trained hob abnormal wear quantity prediction model for prediction, the method further comprises: training the hob abnormal wear quantity prediction model;

[0046] Training the abnormal wear quantity prediction model of the hob includes:

[0047] Acquire a tunneling parameter data set, which includes data samples of tunneling parameters that affect cutter wear during earth pressure balance shield tunneling over time;

[0048] Acquire a geological parameter data set, which includes data samples of geological parameters changing with excavation mileage during earth pressure balance shield excavation;

[0049] Obtaining a cutter mechanical parameter data set, the data set including data samples of cutter parameters on an earth pressure balance shield cutterhead;

[0050] Obtain a dataset of abnormal wear quantity of roller cutters, which includes data samples of the change of abnormal wear quantity of roller cutters with excavation mileage during earth pressure balance shield excavation;

[0051] The excavation parameters, geological parameters, hob mechanical parameters and the number of abnormal wear of the hob in the excavation parameter data set, geological parameter data set, hob mechanical parameter data set and the number of abnormal wear of the hob are matched according to the excavation ring number, and characteristic parameters are extracted from the excavation parameters according to the excavation ring number;

[0052] Since the excavation parameters are a set of time series data, the characteristic parameters of the excavation parameters are extracted according to the excavation ring number, and the characteristic parameters of the excavation parameters corresponding to each excavation ring number can be obtained, which are used for prediction by the prediction model.

[0053] Constructing an input sample data set and an output sample data set; the input sample data set includes a plurality of input samples, each of which is a set of characteristic parameters, geological parameters and mechanical parameters of a set of excavation parameters corresponding to an excavation ring number; the output sample data set includes a plurality of output samples, each of which is a number of abnormal wear of a excavation cutter corresponding to an excavation ring number;

[0054] A prediction model for the number of abnormal wear of hobs is established. During the training process, the input sample data set and the output sample data set are used to perform multiple iterative calculations, and the prediction accuracy of the prediction model for the number of abnormal wear of hobs is evaluated until the error of the prediction model for the number of abnormal wear of hobs meets the requirements, thereby obtaining a trained prediction model for the number of abnormal wear of hobs.

[0055] In some embodiments, the excavation parameters in the excavation parameter data set may include cutter head rotation speed, cylinder thrust pressure, soil bin pressure, thrust speed, cutter head torque, total thrust, pitch angle and penetration.

[0056] In some embodiments, the above-mentioned different types of excavation parameters may be first analyzed for correlation. For the excavation parameter time series data, mean value correlation analysis and variance correlation analysis may be performed according to the excavation ring number, such as Figure 2 and 3 As shown. For two tunneling parameters with high correlation, that is, high correlation coefficient, only one of them can be selected. For example, according to the correlation analysis results, the correlation coefficient between penetration and propulsion speed is high, and the correlation coefficient between cylinder propulsion pressure and total thrust is high. Therefore, the cutterhead speed, soil bin pressure, propulsion speed, cutterhead torque, total thrust and pitch angle can be selected as tunneling parameters to establish a tunneling parameter data set.

[0057] In some embodiments, the geological parameters in the geological parameter data set may include uniaxial compressive strength and rock mass integrity.

[0058] In some embodiments, the hob parameters in the hob mechanical parameter data set may include an installation radius, a hob size, a material parameter, and an installation angle.

[0059] In some embodiments, the step of obtaining a tunneling parameter data set includes:

[0060] The original excavation parameter data set is preprocessed to exclude abnormal values ​​and values ​​that are too large or too small in the original excavation parameter data set, and the statistical 3σ principle is used to eliminate values ​​that are more than 3 times the standard deviation from the mean value.

[0061] In some embodiments, characteristic parameters are extracted from the excavation parameters according to the excavation ring number, including: calculating the average value and variance of each excavation parameter according to the excavation ring number as the characteristic parameters.

[0062] For example, the mean and variance of the cutterhead speed, soil bin pressure, propulsion speed, cutterhead torque, total thrust and pitch angle are calculated to obtain the characteristic parameters of the excavation parameters, such as Figure 4As shown, from left to right are: the mean value and variance of the cutterhead speed, the mean value and variance of the soil bin pressure, the mean value and variance of the propulsion speed, the mean value and variance of the cutterhead torque, the mean value and variance of the total thrust, and the mean value and variance of the pitch angle.

[0063] In some embodiments, the evaluation of the prediction accuracy of the hob abnormal wear quantity prediction model includes: using a correlation coefficient to evaluate the prediction accuracy of the hob abnormal wear quantity prediction model, and if the correlation coefficient reaches above a preset threshold, it is determined that the hob abnormal wear quantity prediction model error meets the requirements.

[0064] In some embodiments, the preset threshold is 0.9.

[0065] In some embodiments, a prediction model for the abnormal wear quantity of a hob may be established based on a back propagation (BP) neural network model optimized by a genetic algorithm (GA).

[0066] For example, Figure 5 As shown, a hidden layer of 15 neurons and an output layer of 1 neuron can be set. The number of input parameters of the application network is 18, including 12 characteristic parameters of tunneling parameters, 2 geological parameters and 4 mechanical parameters of the cutter. The number of output parameters is 34, which correspond to the 37 cutters of the earth pressure balance shield and are used to characterize whether each cutter has abnormal wear.

[0067] In some embodiments, the method includes determining whether the cutter needs to be repaired and replaced based on whether the amount of abnormal wear of the earth pressure balance shield cutter output by the cutter abnormal wear amount prediction model reaches a defined amount.

[0068] The embodiment of the present application also provides an electronic device, including: a memory and a processor;

[0069] The memory is used to store computer programs;

[0070] The processor is used to call the computer program to execute the method as described above.

[0071] An embodiment of the present application further provides a computer-readable storage medium, in which a computer program is stored. When the computer program is executed on an electronic device, the electronic device implements the method as described above.

[0072] An embodiment of the present application further provides a computer program product, including a computer program. When the computer program is executed on an electronic device, the electronic device implements the method as described above.

[0073] The embodiments of the present application also provide a system, an electronic device, a computer-readable storage medium, and a computer program product. The specific implementation methods can refer to the specific embodiments of the above methods and will not be repeated here.

[0074] Obviously, those skilled in the art should understand that the above-mentioned units or steps of the present application can be implemented by a general-purpose computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, and optionally, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. In this way, the present application is not limited to any specific combination of hardware and software.

[0075] The above embodiments are only examples to clearly illustrate the present invention, rather than limiting the implementation methods. For ordinary technicians in the field, other different forms of changes or modifications can be made on the basis of the above description. It is not necessary and impossible to list all implementation cases here. The obvious changes or modifications derived from this are still within the protection scope of the present invention.

Claims

1. A method for predicting the abnormal wear quantity of earth pressure balance shield cutter, characterized in that: include: The characteristic parameters of the excavation parameters that need to be predicted, the geological parameters and the mechanical parameters of the cutter are used as input data, and input into a trained cutter abnormal wear quantity prediction model, and the cutter abnormal wear quantity prediction model outputs the amount of abnormal wear of the earth pressure balance shield cutter; wherein the trained cutter abnormal wear quantity prediction model has the function of predicting and outputting the corresponding amount of abnormal wear of the earth pressure balance shield cutter according to the input data.

2. The method according to claim 1, characterized in that The method further comprises: training the abnormal wear quantity prediction model of the hob; The training of the abnormal wear quantity prediction model of the hob includes: Acquire a tunneling parameter data set; the tunneling parameter data set includes data samples of tunneling parameters that affect cutter wear during earth pressure balance shield tunneling over time; Acquire a geological parameter data set; the geological parameter data set includes data samples of geological parameters changing with excavation mileage during earth pressure balance shield excavation; Acquire a roller cutter mechanical parameter data set; the roller cutter mechanical parameter data set includes data samples of roller cutter parameters on an earth pressure balance shield cutterhead; Acquire a data set of abnormal wear quantity of roller cutters; the data set of abnormal wear quantity of roller cutters includes data samples of the abnormal wear quantity of roller cutters changing with the excavation mileage during the excavation of an earth pressure balance shield; The excavation parameters, geological parameters, hob mechanical parameters and the number of abnormal wear of the hob in the excavation parameter data set, geological parameter data set, hob mechanical parameter data set and the number of abnormal wear of the hob are matched according to the excavation ring number, and characteristic parameters are extracted from the excavation parameters according to the excavation ring number; Constructing an input sample data set and an output sample data set; the input sample data set includes a plurality of input samples, each of which is a set of characteristic parameters, geological parameters and mechanical parameters of a set of excavation parameters corresponding to an excavation ring number; the output sample data set includes a plurality of output samples, each of which is a number of abnormal wear of a excavation cutter corresponding to an excavation ring number; A prediction model for the number of abnormal wear of hobs is established; the established prediction model for the number of abnormal wear of hobs is trained based on an input sample data set and an output sample data set; during the training process, multiple iterative calculations are performed, and the prediction accuracy of the prediction model for the number of abnormal wear of hobs is evaluated until the error of the prediction model for the number of abnormal wear of hobs meets the requirements, thereby obtaining a trained prediction model for the number of abnormal wear of hobs.

3. The method according to claim 1, characterized in that The excavation parameters in the excavation parameter data set include cutter head rotation speed, soil bin pressure, propulsion speed, cutter head torque, total thrust and pitch angle; The geological parameters in the geological parameter data set include uniaxial compressive strength and rock mass integrity; The hob parameters in the hob mechanical parameter data set include installation radius, hob size, material parameters and installation angle.

4. The method according to claim 2, characterized in that: The step of obtaining a tunneling parameter data set includes: The original excavation parameter data set is preprocessed to exclude abnormal values ​​and values ​​that are too large or too small in the original excavation parameter data set, and the statistical 3σ principle is used to eliminate values ​​that are more than 3 times the standard deviation from the mean value.

5. The method according to claim 1, characterized in that: And extract characteristic parameters from the excavation parameters according to the excavation ring number, including: calculating the average value and variance of each excavation parameter according to the excavation ring number, and taking them as characteristic parameters.

6. The method according to claim 1, characterized in that The method of evaluating the prediction accuracy of the hob abnormal wear quantity prediction model includes: using a correlation coefficient to evaluate the prediction accuracy of the hob abnormal wear quantity prediction model, and if the correlation coefficient reaches a preset threshold, it is determined that the hob abnormal wear quantity prediction model error meets the requirement.

7. The method according to claim 1, characterized in that The method of establishing a prediction model for the number of abnormal wear of the hob includes: establishing a prediction model for the number of abnormal wear of the hob based on a back propagation neural network model optimized by a genetic algorithm.

8. An electronic device, characterized in that: include: Memory and processor; The memory is used to store computer programs; The processor is configured to call the computer program to execute the method according to any one of claims 1 to 7.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed on an electronic device, the electronic device implements the method according to any one of claims 1 to 7.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed on an electronic device, the electronic device implements the method according to any one of claims 1 to 7.

Images