| Double pulse MIG is an efficient welding process for aluminum alloy.In the welding process,because the equipment operation is affected by the external interference factors,the welding process parameters fluctuate greatly,the parameters obtained from the traditional experience test welding are difficult to meet the actual welding process requirements,so it is difficult to ensure the welding quality.The electric signal generated in the welding process directly reflects the arc stability and weld quality.The main method of online welding quality monitoring is to extract the dynamic information of the electric signal in the welding process,and use the dynamic information to realize the arc stability,process rationality evaluation and process optimization in the welding process.In order to ensure the quality of double pulse MIG welding of aluminum alloy,it provides prior information and technical means,which has important theoretical guidance value and engineering application significance.The aim of this paper is to optimize the process parameters of aluminum alloy double pulse MIG welding.On the basis of the aluminum alloy double pulse MIG welding experimental test platform,the electric arc signal of aluminum alloy double pulse MIG welding is acquired in real time.Through the study of the methods of noise reduction,feature extraction and the reasonable evaluation and optimization method of process parameter combination,the quantitative extraction of current signal feature information and optimization of welding process parameters are realized.The main research contents are as follows:(1)Combined with ABB IRB1410 industrial robot,DPMIG-500 aluminum welding machine and data acquisition system,the experimental test platform of robot aluminum alloy double pulse MIG welding is built,and the data acquisition experiment of robot aluminum alloy double pulse MIG welding electric signal is designed.Through the data acquisition experiment,the arc current and voltage signals produced in the process of welding are obtained,and the real-time online collection and storage of arc current and voltage signals in the double pulse MIG welding of aluminum alloy are realized.(2)This paper studies the method of welding current signal denoising which combines wavelet denoising and improved harmonic wavelet packet decomposition.Firstly,after the welding current signal is denoising by wavelet,according to the characteristics of the double pulse MIG welding current signal pulse frequency,the harmonic wavelet packet decomposition is improved,the coefficients of the corresponding frequency band are extracted and the signal is reconstructed,the effective output current signal of the equipment is obtained,and the effective denoising of the welding current signal is realized.(3)The characteristic construction method of current signal in double pulse MIG welding of aluminum alloy was studied.Firstly,the effective welding current signal is transformed by HHT to obtain the energy distribution characteristics in time and frequency domain,and the multiscale permutation entropy algorithm is introduced to calculate the multi-parameter eigenvalues of the signal energy distribution.The multi-scale entropy eigenvector of currentenergy signal,which can describe and describe the characteristics of welding arc quantitatively,is constructed with PCA as the numerical index to identify the stability characteristics of welding arc quantitatively.(4)An optimization method of double pulse MIG welding process parameters for aluminum alloy based on BBA-BP was studied.BP neural network model is used to establish the mapping model between process parameters and multi-scale arrangement entropy.The optimization model of process parameters was established with the multi-scale entropy of arc electric signals as the optimization objective,and the binary bat algorithm was used to optimize the process parameters,so as to realize the automatic optimization of the process parameters of aluminum alloy double pulse MIG welding. |
PDF Full Text Request