A novel hybrid intelligent optimization model for twin wire tandem co-pool high-speed submerged arc welding of steel plate

作者：He, Kuanfang^*; Xiao, Dongming

来源：Journal of Advanced Mechanical Design, Systems and Manufacturing, 2015, 9(2): 14-00319.

DOI：10.1299/jamdsm.2015jamdsm0017

• Twin wire tandem co-pool high-speed submerged arc welding
• Welding parameters
• LMD
• Energy entropy
• BPNN
• Hybrid intelligent optimization model

In order to solve the welding formation defects of undercut and hump caused by the irrationality selection of parameters at a high welding speed more than 100cm/min in twin wire tandem co-pool submerged arc welding, a novel hybrid intelligent optimization model for twin wire tandem co-pool high speed submerged arc welding is proposed. This model combines local mean decomposition (LMD), energy entropy, back propagation neural network (BPNN) and particle swarm optimization algorithm (PSO). LMD is employed to decompose the collected welding current signal, excavate the underlying arc feature information related to the rationality of parameters and welding quality of welding seam formation appearance and sectional morphology. The energy entropy is used as the quantificational parameter to describe the rationality of parameters and welding quality. The relationship between the welding parameters and the energy entropy is established by BPNN, and the welding parameters are automatically obtained by the PSO. The application shows that the model is able to reliably achieve the optimization selection of welding parameters to guarantee welding quality.

• 出版日期2015
• 单位湖南科技大学