Research On The Acoustic Emission Characteristics Based On The Tensile Deformation Process Of Metallic Materials

With the continuous improvement of the performance of acoustic emission equipment and the improvement of signal processing methods,acoustic emission technology,as a dynamic nondestructive testing technology,has been widely used in oil,bridge,concrete,construction,pipeline,aviation and other industries,realizing real-time dynamic online monitoring and protecting people's life and property safety.Metal material HRB400 is widely used in construction,bridge and steel structure engineering.Due to the need of engineering,it is unavoidable to weld it in the process of use.The quality of the welding is greatly influenced by the quality,current,voltage and other external factors of the operator.It is often easy to produce the defects quality problems such as bite,slag,gas hole and non penetration.Its safety and use function are inevitable.Therefore,it is an important basis for safety operation to evaluate scientifically and reasonably the safety and reliability of them.At present,the most common thread steel in building,bridge and steel structure is HRB400 section steel.Therefore,it is significant to study the acoustic emission signals during the tensile damage of HRB400 steel.The main research work of this paper is summarized as follows:1.The tensile tests of different HRB400 welding defects and complete samples were carried out.Acoustic emission signals were collected successfully.The distribution characteristics of acoustic emission signals of different samples during tensile process were analyzed,and the counting rate of acoustic emission characteristic parameters was analyzed in detail.2.According to the characteristics and load conditions of acoustic emission signals,the material damage is divided into different stages,and the S transform and pseudo Margenau-Hill(PMH)distribution are used to analyze the signal of different stages of each sample.The characteristics of acoustic emission signals corresponding to different damage degrees of different specimens are expounded through the aspects of time and frequency aggregation,cross interference and frequency band.3.The wavelet packet algorithm is used to analyze the different phases of the damaged samples,and the energy distribution characteristics of each stage are found through the energy analysis.4.The wavelet packet algorithm is used to extract the energy characteristics of each phase acoustic emission signal and to form the eigenvector group,and the pattern recognition of different stages of different samples is carried out with the probabilistic neural network.