Data-driven based analyzing and modeling of MIMO laser welding process by integration of six advanced sensors

作者：You, Deyong; Gao, Xiangdong^*; Katayama, Seiji

来源：International Journal of Advanced Manufacturing Technology, 2016, 82(5-8): 1127-1139.

DOI：10.1007/s00170-015-7455-x

• Laser welding
• MIMO nonlinear process
• Hammerstein-Wiener model
• Multiple-sensing system

Laser welding, as an advanced manufacturing technology, involves drastic and complicated interaction between laser energy and material. The whole process is a strong-coupling nonlinear system. Therefore, it is considered a great challenge to conduct accurate modeling on laser welding process. This paper proposes a novel data-driven based method for investigating and modeling nonlinear laser welding process. The system consists of six types of sensor, including visible sensing photodiode, laser reflection sensing photodiode, spectrometer, visible sensing camera, auxiliary illumination sensing camera, and X-ray sensing camera. The sensors help to synchronously obtain optical features, chemical properties, and physical characteristics during laser welding process. A multiple-input multiple-output (MIMO) Hammerstein-Wiener model was constructed and identified by the experimental data collected from the multiple-sensing system. It is concluded that the system designed in this research provides accurate and valid data for making comprehensive analyzing and modeling of laser welding process.

• 出版日期2016-2
• 单位广东工业大学