Research Of Corrosion Damage Monitoring In Pipeline Structure By Acoustic Emission Technique

Pipeline is widely used as an engineering structure, which undertakes the important missi-on of production and usually takes a long time in service. If the pipeline structure was destroye-d, there would be very serious negative effects and even lead to disasters, to ensure the safety of the pipeline structure is significant, pipeline structure health monitoring is the most effecti-ve means to ensure the normal use of pipeline engineering. Pipe corrosion damage is one of t-he major reason for pipeline structural failure, which not only lead to the carrying capacity of the pipeline structure reduced, and even lead pipeline leaking by corrosion perforation directly, therefore pipeline corrosion damage is the main focus in all kinds of pipeline engineering acc-ident prevention. This dissertation focuses on the research of corrosion damage monitoring in pipeline structure based on acoustic emission testing technology, The main contents are as fo-llows.(1) The experimental study of corrosion damage monitoring in steel pipeline by acoustic emission. After introducing the pipeline corrosion mechanism, we designed an accelerated el-ectrochemical corrosion experiment in steel pipeline, and carried out dynamic monitoring dur-ing the entire course of the experiment by acoustic emission detection system, and finally obt-ained all the AE signals. Through the characteristic parameter analysis of those AE signals, we got the characteristics of AE signals generated by pipeline corrosion and divided the pipel-ine corrosion process into four stages, including the incubation period, the accelerated phase, the stable phase and the stagnation period. By waveform analysis, we distinguished the wave-form and spectrum with different peak frequency.(2) In order to distinguish the different AE sources caused by pipeline corrosion damage, cluster analysis of AE signals generated by pipeline corrosion was done. Firstly, the principal component analysis was done for pipeline corrosion data composed by AE signals, reducing t-he amount of redundant information. Then K-means algorithm and fuzzy C-means algorithm were used in cluster analysis for pipeline corrosion acoustic emission signals respectively, and the comparative analysis between the results was done. Through the two clustering validity a-nalysis, we got the optimal clustering categories. The classification and recognition results of AE signals generated by pipeline corrosion were exactly the same with the two methods, and both of the two nethods distinguished the pipeline corrosion damage sources and noise sourc-es effectively. Through the analysis of the cumulative curves which are according to some ch- aracteristic parameters of different signals, we reveals the occurrence and development of var-ious types of signals, and recognised the AE signals corresponding to different damage sourc-es in the process of pipeline corrosion.(3) The leakage source localization research based on acoustic emission testing technolo-gy in a pipeline which happened corrosion perforation leak was done. Firstly, we studied the attenuation characteristics of AE signals propagating along the pipeline, verifying that AE si-gnals have a good propagation performance. By simulating the natural gas pipeline leakage p-rocess, we designed the high-pressure gas pipeline leak detection and location experiment. A-nd analyzed the characteristics of leakage AE signals under different pipeline pressure. Using wavelet analysis technique to reduce the noise and calculating the time delay between the de--noised signals by the cross-correlation function method, we finally calculated the position of leakage source by time difference location method and achieved good positioning accuracy.