Design And Data Analysis Of Pipeline Defect Detection Device Based On Multi-frequency Electromagnetic

The safe transportation of pipelines is related to the lifeline of the national economy and plays an important role in the development of the national economy.However,once a pipeline leaks,it will not only cause environmental pollution and economic losses,but also threaten people's lives and safety when it is serious.It is of great significance to detect defects in pipelines.Most of the existing devices for pipeline detection have high cost and low detection sensitivity,and it is of great significance to design a detection device with low cost and high detection sensitivity.The main work of this paper is as follows:Firstly,based on the single-frequency AC electromagnetic detection principle,the difference of the detection effect of different frequency signals on the defect is fully considered,and the multi-frequency electromagnetic detection theory is deeply analyzed,which is used as the theoretical basis for the design of multi-frequency electromagnetic defect detection device.Secondly,the finite element modeling analysis of the multi-frequency electromagnetic detection technology,by changing the key parameters in the model,including the frequency of the excitation signal,the detection probe and the metal test piece to be tested in the model,the detection of the outer diameter of the coil,compare the magnetic displacement vector and magnetic field strength of the metal specimen to be tested,determine the optimal parameters of the model,and use this result as the basis for the hardware design of the device.Thirdly,the design of the hardware circuit and the corresponding software programming are performed on the excitation signal module,the power amplification module and the data acquisition module of the multi-frequency electromagnetic detecting device,and the frequency of the defect is further verified by analyzing the electrical signal of the defect detected by the detecting device.The difference between the width and the depth of the detection,and the size of the defect is measured by detecting the voltage signal.Fourthly,this paper proposes a KNN algorithm based on BP neural network decision-making.The algorithm is driven by a large amount of experimental data acquired by the detection device.The defect inversion model based on BP neural network decision KNN algorithm is established,and the traditional KNN algorithm inversion is improved.The effect is greatly affected by the K value.The model predicts the defect size based on the detection signal obtained by the multi-frequency electromagnetic defect detecting device.