Stability Analysis Of Hyperbaric GMAW Welding Process Based On Multi-factor Sensitive Weight Method

The ocean is rich in fossil resources such as oil and gas.Along with the development of marine resources,a large number of offshore oil platforms,submarine pipelines and underwater structures are used in the exploitation and transportation of fossil fuels such as oil and gas.Due to poor underwater environment,a large number of structural parts of the erosion by the erosion of long-term,requiring a lot of welding repair.Therefore,underwater welding research has been paid more and more attention by all countries in the world.High-pressure dry GMAW welding has the advantage of high welding efficiency,easy to achieve the characteristics of welding process control,as the preferred welding method of underwater welding.Research on high pressure GMAW welding is of great significance to promote the development of marine resources.Due to the influence of environmental pressure,the factors influencing the welding process of underwater high pressure GMAW are more complicated and coupled with each other.It is very important to study the inherent law of welding GMAW welding,reduce the splash and improve the welding quality.The main contents of the thesis are as follows:（1）First,the orthogonal test scheme and the welding splash rate measurement scheme were developed.The high pressure GMAW welding test was carried out based on the orthogonal test scheme and the splashing rate of the welding process was determined.Then,through the welding analyzer to obtain the welding process signal,get the arc voltage and welding current waveform and probability density map.The effects of ambient pressure,welding current,arc voltage,welding speed,wire length and CO₂ ratio on the stability of the welding process were analyzed.By analyzing the probability density distribution and the fluctuation amplitude of the current and voltage,the stability of the welding process of the high pressure GMAW is analyzed,and the welding process evaluation model of the high pressure GMAW is prepared.（2）Based on BP and GA-BP neural network,the prediction model of welding process stability is established respectively.By comparing the mean square error of the two models,it shows that the GA-BP network has better prediction accuracy.Therefore,the GA-BP neural network is used to establish the welding process stability evaluation model.The stability evaluation model of welding process was validated by high pressure GMAW welding test.（3）Extract the weights and moments of the matrix in the GA-BP network model.The Garson algorithm is used to deal with the matrix to obtain the influence coefficient of welding parameters on the stability of the welding process.In this paper,GA-BP neural network is used to construct the stability evaluation model of welding process,and the sensitive weight coefficient of input parameters is determined.When the welding plan is carried out under high pressure environment,the above-mentioned welding parameter sensitivity weight coefficient can be used to obtain the expected stability of the welding process,which has certain reference value for high pressure GMAW welding.