Font Size: a A A

Research On Trenchless Detection Of Buried Steel Pipeline

Posted on:2019-12-20Degree:MasterType:Thesis
Country:ChinaCandidate:S QinFull Text:PDF
GTID:2381330566973420Subject:Chemical Engineering and Technology
Abstract/Summary:PDF Full Text Request
Oil and natural gas have become an indispensable part of people's daily life.Therefore,the construction of underground pipelines for transporting these substances is also more and more intensive.Although these substances have brought convenience to people,but because of the long service time,the external anticorrosion layer of the pipeline may cause defects and breakages for various reasons,and eventually leads to the leakage of the steel pipeline.This will bring bad consequences to the safe,efficient and continuous transportation of the pipeline,and the serious one will explode.That can cause great environmental damage at the same time.Trenchless detection technology is commonly used in underground pipeline inspection at present.It is aimed at detecting whether the buried pipeline is corroded,leaking and determining the location of leakage points.In this paper,ANSYS finite element simulation and compensation fuzzy neural network were first used to conduct the trenchless simulation reduction of buried pipelines.Then the signal of the anticorrosion layer of buried steel pipe was detected in the field,and the relationship between the detection signal and the size and the position of the breakage points was analyzed.Finally,a trenchless vehicle for buried steel pipeline was designed:First,the communication between the hardware and software system of the vehicle is designed,and then the field experiment is carried out to prove the utility,operability and efficiency of the vehicle.The main contents are as follows:1.The ANSYS finite element analysis software was used to establish the buried steel pipeline model,and to simulate peak value and average value of potential of the pipeline under different influence factors.And the compensation fuzzy neural network model of 6 input and 1 output was constructed.Then the MATLAB software was programmed to train the data in the model to minimize the output error and verify theperformance of the network,so as to form a three-dimensional reduction database.The nonlinear interpolation method was applied to input 6 factors to calculate the output damage radius,and then the damaged pipeline model was constructed with ANSYS software.The conclusion was as follows: After the comparison and analysis of two standardized methods,the standardization method was more effective for the convergence of data when the compensation fuzzy neural network was trained for data pairs.When the compensation fuzzy neural network was used to train the data pairs while other parameters were fixed,the initial input membership function width had a certain influence on the training error and ultimately that it was more conducive to the convergence of the data when its value is determined as 0.1,and the initial output membership function width has no obvious influence on the convergence of the network data.The optimal parameters of the compensation fuzzy neural network were determined by orthogonal test,and the effect of the training error on the data was reduced to a minimum,and its performance was verified.The results showed that the output results were good.At the same time,it was proved that it was feasible to restore the buried pipe through this method.2.By setting up a reduced underground steel pipe model in the laboratory,the change of signals detected on the surface of the soil above the pipe anticorrosion layer was studied when the DC electrical signal was inputted.The conclusion was as follows: When the pipe anticorrosion layer was damaged in different areas,the surface potential above the damaged point of the pipe was greater than that of the other position,and the larger the damaged area was,the greater the peak of the curve,the greater the scope of the influence.The surface potential of pipeline anticorrosive coating was bigger than that of pipeline without breakage.The surface potential on both sides of the pipeline was basically consistent with the isoelectric potential distribution at the breakage point of the pipe anticorrosion layer.When soil resistivity was different,the more the resistivity was,the greater the surface potential was.In the range of error,the surface potential of pipeline anticorrosive coating is basically equal to equidistant equipotential distribution.When the damaged position of pipeline anticorrosive layer was different,The Influence caused by the sudden change ofsurface potential showed: the above>the left= the right >the below.3.The trenchless vehicle for buried steel pipeline were designed,including the design and selection of the electromagnetic detection module,automatic control system,data acquisition module and the communication processing system of the upper computer.The practicality,convenience and efficiency of the trenchless detecting trolley for buried steel pipeline were proved by field experiment.
Keywords/Search Tags:Buried steel pipeline, ANSYS, Compensation fuzzy neural network, Trenchless detection, Electromagnetic detection
PDF Full Text Request
Related items