A kind of active regulation method and terminal suitable for water conservancy project pressure steel pipe thickness

Abstract

The application discloses a kind of active regulation and control method and terminal suitable for water conservancy project pressure steel pipe thickness, it is related to water conservancy engineering design technical field, including the following steps: according to existing database to establish depth neural network model;Pressure steel pipe engineering calculation parameter is input into depth neural network model, to carry out intelligent optimization and prediction to the thickness of pressure steel pipe and its deformation, stress under different geological conditions;According to set requirement, whether the optimized pipe thickness and its corresponding deformation and stress prediction value meet the condition are autonomously discriminated, and the prediction database of pipe thickness and its corresponding deformation and stress that meet the condition is constructed;In prediction database, the pipe thickness is autonomously compared, and the optimal pipe thickness and its corresponding deformation and stress prediction value are output.The scheme is adopted, can be based on existing pressure steel pipe thickness design method, fusion intelligent optimization algorithm, realize the active regulation and control of pressure steel pipe thickness design, thereby reduce its engineering calculation amount, improve steel utilization rate.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of hydraulic engineering design, in particular to a method for actively regulating the thickness of a pressure steel pipe for hydraulic engineering and a terminal. BACKGROUND

[0002] In recent years, a large number of water diversion projects have been planned and built in China. As the core pressure-bearing water conveying facility, the pressure steel pipe bears the key task of energy transmission and accurate conveying of high water head and large flow water body. Its construction and operation play an irreplaceable important role in inter-basin water diversion and urban water supply. However, the traditional design method of the thickness of the pressure steel pipe mainly depends on the preliminary determination according to the geological conditions and the corresponding specifications, and then the economic and feasible final thickness is determined through multiple finite element simulation calculations and comparisons. This process is not only computationally intensive and time-consuming, but also highly dependent on the experience of engineers. Different designers may make different decisions due to their different experiences, which may lead to thickness decision deviation. This will inevitably affect the later unequal thickness butt joint construction and operation of the pressure steel pipe, and in severe cases, it will cause global thickness redundancy of the pressure steel pipe, resulting in excessive performance and a significant reduction in steel utilization, causing unnecessary material waste and economic loss. In addition, the layout of the pressure steel pipe often crosses different engineering geological regions along the water conveying route, and the dynamic changes of the boundary conditions such as surrounding rock parameters and seepage pressure make the original calculation model of the existing pressure steel pipe not applicable, and related calculations and model construction work need to be carried out again, which undoubtedly further increases the workload, prolongs the design cycle, and reduces the overall efficiency.

[0003] Currently, the acceleration of urbanization and the intensification of climate change have highlighted the problem of uneven spatial and temporal distribution of water resources, and regional droughts and floods have occurred frequently, making the rapid construction of water diversion projects more urgent. The pressure steel pipe has gradually become one of the important structural forms of modern water diversion project water conveying line planning and construction due to its short construction period, convenient installation, and strong adaptability to complex terrain. At the same time, with the development and progress of society, intelligent optimization algorithms have made breakthroughs in calculation accuracy, convergence speed, and global optimization ability. As a new technology, it has been widely used in various industries. Therefore, how to actively regulate the thickness of the pressure steel pipe based on the existing design method by integrating intelligent optimization algorithms has become a problem to be solved in the field of pressure steel pipe design technology in hydraulic engineering. SUMMARY

[0004] The application aims at providing a method and a terminal for actively regulating the thickness of a pressure steel pipe for water conservancy projects, which can reduce the engineering calculation amount and improve the utilization rate of steel by combining intelligent optimization algorithms with the existing pressure steel pipe thickness design method to actively regulate the thickness of the pressure steel pipe.

[0005] The application is achieved by the following technical solutions:

[0006] A method for actively regulating the thickness of a pressure steel pipe for water conservancy projects, comprising the following steps:

[0007] S1: establishing a deep neural network model according to an existing database of historical projects;

[0008] S2: inputting the calculation parameters of the pressure steel pipe project into the deep neural network model to intelligently optimize and predict the thickness of the pressure steel pipe and its deformation and stress under different geological conditions;

[0009] S3: autonomously determining whether the optimized pipe thickness and its corresponding deformation and stress prediction values meet the requirements according to the set requirements, and constructing a prediction database of the pipe thickness and its corresponding deformation and stress that meet the requirements;

[0010] S4: autonomously comparing the pipe thickness in the prediction database and outputting the optimal pipe thickness and its corresponding deformation and stress prediction values.

[0011] Further optimization, the step S1 further comprises a step of constructing an existing database before the step S1; the step of constructing the existing database comprises:

[0012] constructing a calculation parameter database according to the historical project design of the pressure steel pipe for water conservancy projects and the finite element simulation calculation parameters;

[0013] based on the calculation parameter database, constructing a pipe thickness and deformation and stress result database obtained by finite element simulation calculation of the pressure steel pipe under different calculation parameters, and combining the parameter database and the result database as an existing database.

[0014] Further optimization, the step of establishing a deep neural network model according to the existing database in the step S1 specifically comprises:

[0015] S11: building a deep neural network model based on the existing database;

[0016] S12: proportionally distributing the constructed parameter database and result database as a data training set and a validation set to jointly serve as input data for optimizing and training the built deep neural network model, and obtaining an optimized deep neural network model;

[0017] S13: Establish the yield strength index σ of steel pipes with different pipe thicknesses and buried pipe types. s The data discrimination index for its allowable stress coefficient α.

[0018] In a further optimization, step S2 also includes the following specific steps:

[0019] S21: Input the calculation parameters of the pressure steel pipe project into the deep neural network model;

[0020] S22: Using a deep neural network model to analyze the thickness of the pressure steel pipe, the maximum vertical deformation ∆, and the maximum Mises stress value of the pipe. max Perform optimized predictions;

[0021] S23: Based on the optimized pipe thickness and pressure, the buried pipe type in the steel pipe engineering calculation parameters, and the data discrimination index, the yield strength index σ is automatically determined. s And the allowable stress coefficient α.

[0022] In a further optimization, step S3 also includes the following specific steps:

[0023] S31: Based on the pipe diameter D input in step S21, the maximum vertical deformation ∆ predicted in step S22, and the maximum Mises stress value of the pipe. max And combined with the yield strength index σ obtained in step S23 s The allowable stress coefficient α is used to calculate ∆ / D and ασ. s The calculation results;

[0024] S32: Based on the data discrimination index, and according to the calculation results, perform data discrimination. If ∆ / D≤4% and Mises max ≤ɑσ s If the data for pipe thickness, deformation, and stress are not found, the process returns to step S22 for optimization and prediction.

[0025] For further optimization, if the predicted thickness of the pressure steel pipe in step S22 exceeds the maximum limit, the prediction is stopped, and a suitable pressure steel pipe thickness and ∆ are established with respect to Mises. max Predictive database.

[0026] Further optimization involves step S4, where the optimal output value is the minimum thickness of the pressure steel pipe and its corresponding Δ and Mises. max .

[0027] Further optimization, after outputting the optimal result in step S4, includes the following verification step:

[0028] S51: The optimal pipe thickness obtained in step S4 is numerically simulated using Midas gts to obtain the corresponding ∆ and Mises max The simulated value;

[0029] S52: Compare the calculated ∆ with Mises max And the ∆ and Mises obtained in step S4 max A comparative analysis was conducted to verify the deep neural network model.

[0030] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0031] 1. This invention provides an active control method and terminal for the thickness of pressure steel pipes in hydraulic engineering. This method constructs a database containing design and finite element simulation calculation parameters such as pipe diameter, surrounding rock parameters, design head, and steel strength by statistically analyzing historical engineering data on the design of pressure steel pipe thickness. It also establishes corresponding parameters for pipe thickness, Δ, and Mises. max Results Database. Based on traditional design methods, a deep neural network optimization algorithm is coupled and trained using the established database. This allows for the rapid output of the design thickness of the pressure steel pipe and its corresponding ∆ and Mises, simply by inputting relevant calculation parameters. max Predicted values. Subsequently, based on preset discrimination indicators, the above design thickness and predicted values ​​are automatically screened to select the most economical and safe thickness scheme and its corresponding deformation and stress prediction values. Finally, the input parameters and the selected design thickness are used to perform finite element simulation calculations using Midas GTS, and the simulation results (∆, Mises) are analyzed. max The optimal solution is verified by comparing the predicted value with the actual value.

[0032] 2. This invention provides an active control method and terminal for the thickness of pressure steel pipes in water conservancy projects. By simply inputting relevant parameters, the corresponding optimized design value of the steel pipe thickness and its corresponding ∆ and Mises can be directly obtained. max The predicted value, verified through a single finite element simulation, allows for the design of the pressure steel pipe thickness. This significantly reduces reliance on engineers' experience, substantially shortens the engineering calculations and design cycle, and avoids material waste and economic losses caused by redundant steel properties, thereby improving steel utilization. Furthermore, this optimization mechanism can adapt to dynamic parameter changes under different geological conditions along water transmission lines, enabling automatic design of pipe thickness that varies with design parameters.

[0033] 3.The application provides a method and a terminal for actively regulating the thickness of a pressure steel pipe for water conservancy projects, provides a complete process for actively regulating the thickness of a pressure steel pipe, obtains a large amount of design and finite element simulation calculation parameter data of the thickness of a pressure steel pipe, and accumulates a large amount of experience in actively regulating the thickness of a pressure steel pipe, thereby providing technical support for promoting the standardization of the thickness design of a pressure steel pipe and filling the gap in the field of actively regulating the thickness of a pressure steel pipe for water conservancy projects in China, and accumulating experience for realizing the paradigm shift from "manual trial and error design" to "intelligent convergent design". BRIEF DESCRIPTION OF DRAWINGS

[0034] In order to more clearly illustrate the technical solutions of the exemplary embodiments of the present application, the drawings needed in the examples will be briefly introduced as follows, and it should be understood that the following drawings only show some embodiments of the present application, and therefore should not be regarded as a limitation on the scope, and for those skilled in the art, other related drawings can also be obtained without creative labor on the basis of these drawings. In the drawings:

[0035] Figure 1 a deep neural network model diagram provided by the present application;

[0036] Figure 2 an active regulation method flowchart provided by the present application;

[0037] Figure 3 a pressure steel pipe simulation calculation finite model diagram for verifying the preferred scheme provided by the present application;

[0038] Figure 4 a pressure steel pipe vertical deformation cloud chart in the verification calculation result provided by the present application;

[0039] Figure 5 a pressure steel pipe Mises stress cloud chart in the verification calculation result provided by the present application. DETAILED DESCRIPTION

[0040] In order to make the purpose, technical solutions and advantages of the present application more clear and obvious, the present application will be further described in detail below in combination with examples and drawings, and the exemplary embodiments of the present application and their descriptions are only used to explain the present application, and do not limit the present application.

[0041] Example 1: The present example 1 provides an active regulation method for the thickness of a pressure steel pipe for water conservancy projects, as shown in Figures 1-5 , which includes the following specific steps:

[0042] (1) Construct a database of historical engineering design and finite element simulation calculation parameters of a pressure steel pipe for water conservancy projects and a corresponding result database. It includes the following specific steps:

[0043] (1-1) In the history of the design of water conservancy pressure steel pipe engineering, geological parameters, pipe diameter, water head design, steel properties, buried pipe type and other design and finite element simulation calculation parameters are used to build a calculation parameter database;

[0044] (1-2) On the basis of the parameter database in step (1-1), a pipe thickness and pressure steel pipe finite element simulation calculation result database corresponding to different calculation parameters is constructed.

[0045] (2) On the basis of the existing database, a neural network optimization algorithm is built, and the existing database is used as input data for optimization training. It includes the following specific steps:

[0046] (2-1) A deep neural network (DNN) model is built using commercial programming software, which includes an input layer, three hidden layers, and an output layer, and all layers are fully connected layers;

[0047] (2-2) The parameter database built in step (2-1) and the result database established in step (1-2) are divided into data training set and validation set according to 7:3, which are used as input data to optimize and train the DNN model built in step (2-1);

[0048] (2-3) On the basis of step (2-2), further use commercial programming software to build data discriminant indicators of yield strength index σ s and allowable stress coefficient a of steel pipe corresponding to different pipe thicknesses and buried pipe types according to the specified requirements.

[0049] (3) The pressure steel pipe engineering calculation parameters are used as input data, and the optimized and trained neural network is used to intelligently optimize and predict the pressure steel pipe thickness and its deformation and stress under different geological conditions. It includes the following specific steps:

[0050] (3-1) Set the pipe diameter, geological parameters, water head design, steel properties, buried pipe type and other calculation parameters of the pressure steel pipe to be designed as input data;

[0051] (3-2) Based on the input data in step (3-1), use the DNN model optimized and trained in step (2-2) to optimize and predict the pressure steel pipe thickness, maximum vertical deformation ∆ and maximum Mises stress value Mises max of the pipe;

[0052] (3-3) According to the buried pipe type input in step (3-1) and the pipe thickness optimized in step (3-2), use the discriminant indicators built in step (2-3) to automatically discriminate the yield strength index σ s and the allowable stress coefficient a of the steel pipe.

[0053] (4) According to the set requirements, the optimized pipe thickness and its corresponding deformation and stress prediction values are autonomously judged whether they meet the conditions, and the pipe thickness and its corresponding deformation and stress prediction database that meet the conditions are constructed. It includes the following specific steps:

[0054] (4-1) Data calculation is performed in combination with the obtained ∆, Mises max , σ s , ɑ and the pipe diameter D input in step (3-1), to obtain the calculation results of ∆ / D and ɑσ s ;

[0055] (4-2) Data judgment is performed in combination with the calculation results obtained in step (4-1) by using the data judgment index built in step (2-3). If ∆ / D≤4% and Mises max ≤ɑσ s , the pipe thickness, deformation and stress data are saved; if not, return to step (3-2) for optimization prediction; if the predicted pipe thickness exceeds the maximum limit value, stop the prediction, thereby establishing the pressure pipe thickness and ∆ and Mises max prediction database that meet the conditions.

[0056] (5) According to the pipe thickness database created by optimization, autonomous comparison is performed, and the optimal pipe thickness and its corresponding deformation and stress prediction values are output;

[0057] Among the data established in step (4-2), the thickness of the pressure pipe is autonomously compared, and the minimum value of the pipe thickness and its corresponding ∆ and Mises max are obtained.

[0058] (6) Numerical simulation calculation is performed on the optimized pressure pipe thickness by using Midas gts, to further verify the effectiveness and reliability of the neural network optimization algorithm. It includes the following specific steps:

[0059] (6-1) In combination with the pressure pipe thickness obtained in step (5-1), Midas gts is used for further verification calculation, to obtain the simulation values of ∆ and Mises max ;

[0060] (6-2) The simulation values of the pressure pipe ∆ and Mises max obtained in step (6-1) are compared with the prediction values obtained in step (5), to verify the effectiveness and reliability of the DNN model used.

[0061] In summary, the application provides a new method for active regulation of the thickness of a pressure steel pipe for water conservancy projects. By constructing a historical engineering database for the design of the thickness of a pressure steel pipe, and combining a deep neural network optimization algorithm, the thickness of the pressure steel pipe is autonomously designed according to changes in design parameters, and autonomous discrimination indicators for pipe deformation and stress under different design thicknesses are established according to specified requirements, thereby achieving autonomous regulation of the thickness of the pressure steel pipe, improving design efficiency, effectively reducing the dependence on artificial experience of traditional technical means, and providing technical support for promoting the standardization of the thickness design of pressure steel pipes.

[0062] In some example embodiments, the present embodiment also provides a terminal device for active regulation of the thickness of a pressure steel pipe for water conservancy projects, which comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the minimum technical solution of the active regulation method for the thickness of a pressure steel pipe for water conservancy projects as described in Embodiment 1 when executing the computer program, which is used to achieve the purpose of actively regulating the thickness design of a pressure steel pipe based on existing methods for the thickness design of a pressure steel pipe, combining intelligent optimization algorithms, thereby reducing the engineering calculation amount and improving the utilization rate of steel.

[0063] Those skilled in the art will understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer usable program code.

[0064] The present application is described with reference to flowcharts and / or block diagrams according to the methods, devices (systems), and computer program products of the embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of flows and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to produce a machine, so that the instructions executed by the computer or other programmable data processing devices produce a device that implements the functions specified in the flowcharts and / or block diagrams. Figure 1 The functions specified in one or more flows and / or blocks Figure 1 The means for performing the functions specified in one or more flows and / or blocks.

[0065] These computer program instructions can 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 instructions which implement the Figure 1 function specified in the flow or flows and / or blocks Figure 1 of the block or blocks.

[0066] These computer program instructions can 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 Figure 1 function specified in the flow or flows and / or blocks Figure 1 of the block or blocks.

[0067] Those of ordinary skill in the art can understand that all or part of the steps of the above-mentioned facts and methods can be completed by programs instructing relevant hardware, and the programs involved or the programs mentioned can be stored in a computer-readable storage medium. When the program is executed, it includes the following steps: at this time, the corresponding method steps are derived, and the storage medium can be ROM / RAM, a magnetic disc, an optical disc, etc.

[0068] The above specific embodiments further illustrate the purpose, technical solutions and beneficial effects of the present application. It should be understood that the above description is only a specific embodiment of the present application and is not intended to limit the protection scope of the present application. Any modification, equivalent replacement, improvement, etc. within the spirit and principles of the present application should be included in the protection scope of the present application.

Claims

1. A method for active regulation of the thickness of a pressure steel pipe suitable for hydraulic engineering, characterized in that, Comprise the following steps: S1: establishing a deep neural network model according to an existing database of historical projects; S2: inputting pressure steel pipe engineering calculation parameters into the deep neural network model to intelligently optimize and predict the thickness of the pressure steel pipe and its deformation and stress under different geological conditions; S3: according to the set requirements, autonomously judging whether the optimized pipe thickness and its corresponding deformation and stress prediction values meet the conditions, and constructing a prediction database of pipe thicknesses and their corresponding deformations and stresses that meet the conditions; S4: autonomously comparing the pipe thicknesses in the prediction database and outputting the optimal pipe thickness and its corresponding deformation and stress prediction values; The step S2 further comprises the following specific steps: S21: inputting pressure steel pipe engineering calculation parameters into the deep neural network model; S22: optimizing and predicting the thickness of the penstock, the maximum vertical deformation and the maximum Mises stress value Mises of the penstock by the deep neural network model max ​​ S23: According to the optimized pipe thickness pressure and the pipe burying type in the steel pipe engineering calculation parameters, and based on the data discrimination index, the yield strength index σ s and the allowable stress coefficient a; The step S3 further comprises the following specific steps: S31: according to the pipe diameter D input in the step S21, the maximum vertical deformation amount predicted in the step S22 and the maximum Mises stress value Mises of the pipe max , and the yield strength index σ s obtained in the step S23, and the allowable stress coefficient α, data calculation is performed to obtain the calculation results of D and ασ s ; S32: Based on the data discrimination index, and according to the calculation result, if / D≤4% and Mises max ≤ɑσ s , then the data of pipe thickness, deformation and stress are saved; if not, return to step S22 for optimization prediction.

2. The method according to claim 1, characterized in that, The step S1 further comprises a step of constructing an existing database before the step S1; The step of constructing an existing database comprises: According to the water conservancy pressure steel pipe historical project design and finite element simulation calculation parameters, a calculation parameter database is constructed; Based on the calculation parameter database, a pipe thickness and deformation and stress result database obtained by finite element simulation calculation of the pressure steel pipe under different calculation parameters are constructed, and the parameter database and the result database are combined as an existing database.

3. The method according to claim 2, wherein, The step of establishing a deep neural network model according to the existing database in the step S1 specifically comprises: S11: based on the existing database, a deep neural network model is built; S12: the constructed parameter database and result database are proportionally allocated as a data training set and a validation set, which are used as input data to optimize and train the built deep neural network model, to obtain an optimized deep neural network model; S13: Build the yield strength index σ of the steel pipe corresponding to different pipe thickness and buried pipe type s and its allowable stress coefficient a data discrimination index.

4. The method according to claim 1, wherein, If the predicted thickness of the penstock exceeds the maximum limit in step S22, the prediction is stopped, and the thickness of the penstock that meets the conditions is established and Mises max prediction database.

5. The method according to claim 1, wherein, The optimal value output in the step S4 is the minimum value of the thickness of the penstock and its corresponding Mises max .

6. The method according to claim 1, wherein, After the step S4 outputs the optimal result, the following verification step is further included: S51: numerical simulation calculation is performed on the output optimal pipe thickness obtained in step S4 by Midas gts, to obtain corresponding simulation value of Mises max ​ S52: comparing the calculated with Mises max and the result of step S4, thereby verifying the deep neural network model. with Mises max and the result of step S4, thereby verifying the deep neural network model.

7. A terminal, characterized by comprising: The computer program is executed by the processor to realize the active regulation method for the thickness of the water conservancy pressure steel pipe according to any one of claims 1 to 6.

8. A storage medium storing a computer program, characterized by The computer program is executed by the processor to realize the active regulation method for the thickness of the water conservancy pressure steel pipe according to any one of claims 1 to 6.

Images