TY - GEN
T1 - CMAC based modeling and closed loop control of HPDD laser welding system
AU - Duan, Peiyong
AU - Zhang, Yu Ming
PY - 2006
Y1 - 2006
N2 - The development of a CMAC (Cerebellar Model articulation Controller) based modeling method and closed loop control system are presented to estimate and control the weld fusion specified by the topside and backside bead widths of the weld pool using HPDD (High Power Direct Diode) laser as heat source. Because of the Inconvenience of using backside sensor, a topside vision sensor is applied to measure the topside and estimate the backside widths respectively by on-line image processing. The laser driver current and welding speed are selected as the control variables. Basis function CMAC based steady-state models, easier to be obtained than dynamic ones, of the nonlinear controlled process can predict the control variables in a satisfactory accuracy to shorten output response transient time, and two simple PT controllers are included in the control system to adjust the control variables to maintain outputs at the desired levels. The results of real time closed-loop control experiments of laser welding process demonstrate that the control performance can be guaranteed when fluctuations or variations in welding conditions which lead to the changing parameters of the nonlinear model.
AB - The development of a CMAC (Cerebellar Model articulation Controller) based modeling method and closed loop control system are presented to estimate and control the weld fusion specified by the topside and backside bead widths of the weld pool using HPDD (High Power Direct Diode) laser as heat source. Because of the Inconvenience of using backside sensor, a topside vision sensor is applied to measure the topside and estimate the backside widths respectively by on-line image processing. The laser driver current and welding speed are selected as the control variables. Basis function CMAC based steady-state models, easier to be obtained than dynamic ones, of the nonlinear controlled process can predict the control variables in a satisfactory accuracy to shorten output response transient time, and two simple PT controllers are included in the control system to adjust the control variables to maintain outputs at the desired levels. The results of real time closed-loop control experiments of laser welding process demonstrate that the control performance can be guaranteed when fluctuations or variations in welding conditions which lead to the changing parameters of the nonlinear model.
KW - Laser welding
KW - Modeling
KW - Neurocontrol
KW - Nonlinear systems
UR - http://www.scopus.com/inward/record.url?scp=34047199224&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=34047199224&partnerID=8YFLogxK
U2 - 10.1109/WCICA.2006.1713090
DO - 10.1109/WCICA.2006.1713090
M3 - Conference contribution
AN - SCOPUS:34047199224
SN - 1424403324
SN - 9781424403325
T3 - Proceedings of the World Congress on Intelligent Control and Automation (WCICA)
SP - 3839
EP - 3843
BT - Proceedings of the World Congress on Intelligent Control and Automation (WCICA)
T2 - 6th World Congress on Intelligent Control and Automation, WCICA 2006
Y2 - 21 June 2006 through 23 June 2006
ER -