| Polyethylene pipe is one of the most widely used plastic pipes,with its excellent corrosion resistance,non-pollution,high flexibility,easy maintenance and other characteristics are widely used in production and life of the pipeline transportation field.Polyethylene pipes are usually welded on site by hot melt or electrofusion when laid.In the process of welding,defects are easily formed at the pipeline joints due to improper human operation,environmental impact and other factors.The defects and the aging of the pipeline will lead to the reduction of the comprehensive performance of the pipeline,which poses a serious threat to the safe operation of its application field.Terahertz time-domain spectroscopy has good penetration for polyethylene and other materials,which can be effectively used for the detection of polyethylene welded joints and aging.In this paper,the intelligent high-precision identification and aging characterization prediction of polyethylene pipe hot melt welding defects are carried out.The main work is as follows:1)Study on aging characteristics evaluation and prediction method of polyethylene pipe.Firstly,the terahertz spectral characteristics of PE samples with different compressive strength and different components under different photothermal aging degrees were studied.The spectral screening algorithm was used to extract the characteristics of the absorption spectrum,and the spectral slope characteristics of the aging sensitive band were selected as the evaluation index of the aging degree of polyethylene.Based on the partial least squares model,the aging prediction of artificial aging samples was established.The prediction accuracy of aging degree of different kinds of pipes was above 93.16 %,and the accuracy of validation set is above 92.1 %.2)Analysis of pulsed terahertz characteristics of defects in hot-melt joints of polyethylene pipes.According to the types of defects existing in the actual hot-melt welding of polyethylene pipes,a variety of defect simulation specimens were designed and fabricated,and the pulsed terahertz time-domain spectroscopy imaging technology of polyethylene pipe hot-melt welding defects was studied.The terahertz waveform characteristics of different kinds of defects are analyzed.There are obvious differences in the waveform between the samples with incomplete fusion and inclusion hot melt defects and the standard welded samples.Cold welding,over welding,incomplete fusion and inclusion hot melt defects can be seen in the kurtosis imaging diagram,and the terahertz kurtosis imaging and industrial CT detection imaging were compared and analyzed.3)Research on defect identification method of pulsed terahertz detection for polyethylene pipe hot melt joint.A wavelet scattering network-convolution neural network defect recognition model was constructed according to the waveform differences of different hot melt welding defect types.Using the high sensitivity of industrial CT imaging technology to inclusion defects,the defect area calculation of industrial CT detection is compared with the defect area of terahertz recognition to verify the accuracy of terahertz detection.The relative errors of terahertz and industrial CT detection and recognition are all below 8.02 %.The application technology of the actual cylindrical polyethylene pipeline was studied.The robot equipped with terahertz detector intelligent system was used to realize the surface detection of samples with inclusions and incomplete fusion defects.The constructed defect recognition network was applied to the actual defect recognition.The recognition accuracy is as high as 91.05 %,which confirms the feasibility of pulsed terahertz for the welding quality detection of polyethylene pipelines. |
PDF Full Text Request