Font Size: a A A

Research On Trenchless Reconstruction Of Buried Steel Pipeline

Posted on:2018-01-10Degree:MasterType:Thesis
Country:ChinaCandidate:M HeFull Text:PDF
GTID:2321330536488566Subject:Chemical Engineering
Abstract/Summary:PDF Full Text Request
In order to present the 3D shape of the pipeline in the pipeline inspection process,in this paper,by using the finite element simulation software ANSYS,the potential change of the pipeline anti-corrosion layer is studied theoretically through the establishment of three-dimensional model of steel pipeline,soil,air and anti-corrosion layer,and the potential change diagram is obtained.Then the experiments are carried out to verify the correctness of the simulation results.Secondly,the model of the reconstruction of the fuzzy neural network is constructed by using the model of pipeline and the change of potential.The model of the pipeline is output by inputting the potential map of the pipeline,so as to realize the non-excavation reconstruction of buried steel pipeline.The main contents are summarized into 4points:1?The ANSYS finite element method is used to simulate and analyze the factors affecting the signal transmission in the pipeline detection process.Results show:(1)When there is stray current around soil environment or pipeline,The distribution of surface potential is equal to the equipotential distribution and the distribution trend is related to the position of stray current source and the value of stray current;(2)The distribution of the surface potential is related to the crossing angle of the pipeline and the pipeline.When the intersection angle is greater than 0 degree,the distribution of surface potential of the pipeline is related to the intersection angle of underground flow line and the pipeline.When he crossing angle is greater than 0 °,the surface potential equidistant equipotential distribution centered on the projection of the current carrying line and the target pipe;(3)When the soil resistivity is relatively small,the potential of the ground surface is equal to the equipotential distribution at the center of the damage;When the soil resistivity is large,the distribution pattern of the potential leakage on the surface of the earth is not consistent with the distribution of equal interval potential and the potential of the soil and the surface of the soil showed a downward trend,and the potential of the damaged layer was not obvious;(4)When the soil medium is unevenly distributed,the change rate of the surface potential at different soil junction became larger.The potential does not conform to the distribution law of the leakage potential at the center of the damage distribution with equal equipotential distribution;(5)When the depth of the pipeline is not the same,at the junction of different soil thickness,the potential of the radiation from the pipeline to the surface of the earth is suddenly changed;(6)The maximum position of the surface potential is different with the location of the damaged points.2?The experiments were carried out to verify the correctness and feasibility of the finite element simulation results when the damaged area and the soil resistivity are different And the results showed similar distribution trend of experimental results with simulation results;3?Using the fuzzy neural network,the model of pipeline and the potential curve by the ANSYS simulation are built into network data model of the 6 input and1 output.Using MATLAB to write the program on the data in the model for network training and performance verification,network database is formed.The damage radius output by the database is combined with the ANSYS software,and the damaged pipeline model is output to realize the non excavation reconstruction of the buried steel pipeline.4?The paper constructed the fuzzy neural network,among of that,the network error was set as 10-4,and the membership function was normal distribution,the number of fuzzy rules was 26;the center of initial input membership function was 0.5+0.55*i/25 and the width was 0.5,The center of initial output membership function was 0.5,the width was 0.3;The initial learning rate was 0.6,and the final learning rate was 2.
Keywords/Search Tags:Steel pipeline, Trenchless reconstruction, ANSYS simulation, Neural network, Database
PDF Full Text Request
Related items