Adaptive control for laser welding with filler wire of marine high strength steel with tight butt joints for large structures

作者：Zhang, Ke^*; Li, Danyang; Gui, Hao; Li, Zhuguo

来源：Journal of Manufacturing Processes, 2018, 36: 434-441.

DOI：10.1016/j.jmapro.2018.10.042

• Adaptive control
• BP neural network
• Genetic algorithm
• Laser welding with filler wire
• Tight butt joints

To large structures, ensuring a uniform joint gap without mismatch over an entire seam remains a challenge. Varying gaps and mismatches significantly affect the welding qualities, especially for sheet metal. In the paper, an adaptive filling model for process parameters based on back propagation neural network (BPNN) combined with genetic algorithm (GA) is used as an adaptive controller to solve the problems. First, a real-time closed loop feedback control from groove information collection, processing to model prediction and control output was established. Then, an adaptive control approach was investigated, a precision 3D laser sensor was used for measuring the size of gap and mismatch, and an adaptive parameters table was designed as an adaptive controller based on the optimal BPNN, an interpolation operation was introduced for an intermediate value. The results of the experiment showed that the welding parameters can be continuously adjusted in real time despite the variation in the gap and mismatch according to the adaptive filling algorithm and the laser sensor; the welds with desirable uniform weld appearance was achieved despite the changing gap, from 0 to 1.0 mm, and mismatch, from 0 to 1.2 mm, meet the requirement of practical industrial application.

• 出版日期2018-12
• 单位上海交通大学