TY - JOUR
T1 - Self-constructing sugeno type adaptive fuzzy neural network for two-axis motion control system
AU - Lin, Faa Jeng
AU - Chou, Po Huan
N1 - Funding Information:
financial support of the National Science Council of Taiwan, R.O.C. through its grant NSC 95-2221-E-259-042-MY3.
PY - 2007
Y1 - 2007
N2 - A self-constructing Sugeno type adaptive fuzzy neural network (SAFNN) control system is proposed in this study for the contour tracking control of a two-axis motion control system. The adopted two-axis motion control system is composed of two permanent magnet linear synchronous motors (PMLSMs). The proposed SAFNN combines the merits of a self-constructing fuzzy neural network (SCFNN) and a TSK-type fuzzy inference mechanism. Moreover, the structure and the parameter learning phases are performed concurrently and on line in the SAFNN. The structure learning is based on the partition of input space, and the parameter learning is based on the supervised gradient descent method using a delta adaptation law. Furthermore, the proposed control algorithms are implemented in a TMS320C32 DSP-based control computer. From the simulated and experimental results, the contour tracking performance of the two-axis motion control system is significantly improved and robustness can be obtained as well using the proposed SAFNN control system.
AB - A self-constructing Sugeno type adaptive fuzzy neural network (SAFNN) control system is proposed in this study for the contour tracking control of a two-axis motion control system. The adopted two-axis motion control system is composed of two permanent magnet linear synchronous motors (PMLSMs). The proposed SAFNN combines the merits of a self-constructing fuzzy neural network (SCFNN) and a TSK-type fuzzy inference mechanism. Moreover, the structure and the parameter learning phases are performed concurrently and on line in the SAFNN. The structure learning is based on the partition of input space, and the parameter learning is based on the supervised gradient descent method using a delta adaptation law. Furthermore, the proposed control algorithms are implemented in a TMS320C32 DSP-based control computer. From the simulated and experimental results, the contour tracking performance of the two-axis motion control system is significantly improved and robustness can be obtained as well using the proposed SAFNN control system.
KW - Gradient descent method
KW - Permanent magnet synchronous motor
KW - Self-constructing
KW - Sugeno type fuzzy neural network
KW - Two-axis motion control
UR - http://www.scopus.com/inward/record.url?scp=37049032648&partnerID=8YFLogxK
U2 - 10.1080/02533839.2007.9671343
DO - 10.1080/02533839.2007.9671343
M3 - 期刊論文
AN - SCOPUS:37049032648
SN - 0253-3839
VL - 30
SP - 1153
EP - 1166
JO - Journal of the Chinese Institute of Engineers, Transactions of the Chinese Institute of Engineers,Series A/Chung-kuo Kung Ch'eng Hsuch K'an
JF - Journal of the Chinese Institute of Engineers, Transactions of the Chinese Institute of Engineers,Series A/Chung-kuo Kung Ch'eng Hsuch K'an
IS - 7
ER -