Study On The Acoustic-magnetic Properties And Experimental Method Of SPV490Q Steel Welding Cold Crack

In order to meet the requirements of high strength and light weight,high strength low alloy steels are widely used in the manufacturing process of high-parameter pressure equipment in petroleum and petrochemical industry.However,with the increase of the strength of the material,cold cracks are more likely to appear in the welding structures.And due to its delay characteristics,it is difficult to detect and prevent the cold cracks,and it can also lead to some catastrophic accidents in some important welding structures.Therefore,studying on the mechanism of welding cold crack,establishing a series of methods for detecting and distinguish cold crack is significant to guarantee the quality and long-period safe operation of welding structures and pressure vessels.In this paper,SPV490 Q,a commonly used high strength low alloy steel for large storage tanks,is taken as an object.According to the characteristics of elastic stress wave and residual stress release during the formation of cold crack,the test method of acoustic emission and magnetic memory technology was used to collect the sound and magnetic signal in the formation of welding cold crack.Cold crack formation mechanism is revealed through in-depth analyzing from aspects of welding crack micro morphology,which provides the basis for welding process improvement and weld seam quality testing.The main research contents of this paper include:1.Combine the cold crack formation mechanism with the transient stress wave propagation theory to find out mechanism of different acoustic emission sources in the process of cold crack formation and propagation.Then put forward the method for analyzing AE information through AE Hit-Time and AMP-Time process diagram.The law for AE signal during the whole process and on different stages can be shown through a variety of parameters as count,amplitude,energy,rise time,duration etc and the change of waveform.Due to the change of the stress in the process of cold crack formation,the variation of the magnetic flux leakage in the welding seam is inevitable.Two magnetism characteristic parameters,the average leakage magnetic field intensity gradient value Kmed and maximum leakage magnetic field intensity gradient value Kmax,were selected to indicate the stress variation of the weld surface.2.Through the analysis of AE monitoring data during SPV490 Q High Strength Low Alloy Steel oblique Y type groove welding experiment,information for cold crack formation and expansion during the whole process of acoustic emission can be acquired and the evolution law can be deduced.Magnetic memory testing technology was used to test SPV490 Q high strength low alloy steel oblique Y groove welding cold crack specimens intermittently.Db4 wavelet was used to decompose and reconstruct the magnetic memory signals and extracting magnetic memory leakage magnetic field intensity gradient value K at different time after welding,thus the curve of K value over time can be drawn.Through the analysis of the acoustic emission and magnetic memory characteristics of the cold crack of different specimens,we came to the conclusion that with the increase of weld thickness,weld length,binding electrode humidity and steel strength,the initiation time comes early,the intensity and activity of acoustic emission signal increases,the average leakage field intensity gradient value Kmed increases,and the sensitivity of cold crack tends to increase.By adopting the law,measures can be taken to prevent the formation of cold crack.3.Acoustic emission source positioning method was used to analyze the crack initiation and propagation direction of welding cold crack.According to the mechanism that stress concentration and energy release often accompany the generation and propagation of the cold crack,we use magnetic memory technology to get the stress distribution of welding cold crack at different time.By using the method of segmented Kmax,the crack initiation and propagation direction of the cold crack can be determined.By using the method of numerical simulation,the cold crack initiation position and propagation direction of the simulation on SPV490 Q steel oblique Y groove welding can be analyzed and the results conform with the ones we get through adopting the method of acoustic emission source position and the maximum magnetic flux leakage field intensity gradient method.For steel SPV490 Q,the welding cold crack starts from the relative end position of the weld,and gradually extends to the starting end of the weld.And the combination of sound and magnetic positioning detection method can not only determine the time,position and direction of the occurrence of cold crack accurately,but also observe the dynamic crack growth in real time,this can help in taking measures to prevent the expansion of the cold crack.4.By using the method of cluster analysis,the acoustic emission signal of weld cold crack was divided into initiation and propagation signals,and the characteristic parameters of different types of AE signals were obtained.Wavelet transform analysis method,waveform analysis method and high order spectrum analysis method were used to obtain the real waveform of the signal,the waveform for initiation and propagation of AE signal and the frequency distribution law.The differences in parameters and waveform between the initiation signals and propagation signals can provide support for the establishment of the BP neural network.5.On the basis of obtaining different typical characteristics of the initiation signal and the noise signal,a 3-layer BP neural network model was established.The 5 parameters,namely the rise time,count,energy,duration and amplitude,were selected as input units of the neural network.The number of nodes in the hidden layer is 10.The signal pattern is the output unit,and the output layer node number is 1.The typical initiation signals and noise signals which are have obvious differences in parameters and waveform are selected as samples.Train the BP neural network,so that it can intelligently identify the signal.Satisfactory recognition result is acquired through testing.6.Through SPV490 Q plate rigid restraint welding cold crack experiment,the different state of the experimental control group was set,the above-mentioned evaluation method was verified from aspects of cold crack judgment,cold crack sensitivity evaluation and the crack initiation position and propagation direction dertermination etc.Feasibility and accuracy of using it as the criterion were proved.