LS-SVM-based surface roughness prediction model for a reflective fiber optic sensor

Li Fu; Jun Luo; Weimin Chen; Xueming Liu; Dong Zhou; Zhongling Zhang; Sheng Li

doi:10.3788/COL201715.091201

Journals >Chinese Optics Letters >Volume 15 >Issue 9 >Page 091201 > Article

Chinese Optics Letters
Vol. 15, Issue 9, 091201 (2017)

LS-SVM-based surface roughness prediction model for a reflective fiber optic sensor

Li Fu^1、*, Jun Luo¹, Weimin Chen¹, Xueming Liu², Dong Zhou¹, Zhongling Zhang¹, and Sheng Li¹

Author Affiliations

¹Key Lab of Optoelectronic Technology & Systems of Ministry of Education, Chongqing University, Chongqing 400044, China

²5011 District Measurement Station of Weapon Industry, Chongqing 400050, China

show less

DOI: 10.3788/COL201715.091201 Cite this Article Set citation alerts

Li Fu, Jun Luo, Weimin Chen, Xueming Liu, Dong Zhou, Zhongling Zhang, Sheng Li. LS-SVM-based surface roughness prediction model for a reflective fiber optic sensor[J]. Chinese Optics Letters, 2017, 15(9): 091201 Copy Citation Text

EndNote(RIS)

BibTex

Plain Text

show less

Fig. 1. Schematic diagram of surface roughness measurement using reflective fiber optic sensors.

Download full size | View in the Article

Fig. 2. Measuring system.

Download full size | View in the Article

Fig. 3. End-face structure of fiber optic probe.

Download full size | View in the Article

		LS-SVM			BPNN			SVM
Processing Methods	Ra (μm)	MRa (μm)	MRE (%)	SD (μm)	MRa (μm)	MRE (%)	SD (μm)	MRa (μm)	MRE (%)	SD (μm)
Internal Grinded	1.6	1.5963	−0.2280	0.0137	1.6061	0.3870	0.0125	1.6005	0.0312	0.0015
	0.8	0.7957	$- 0.5340$	0.0096	0.7896	−1.2963	0.0274	0.7416	−7.2940	0.0879
	0.4	0.4002	0.0534	0.0006	0.3882	−2.9266	0.0164	0.4098	2.4622	0.0151
	0.2	0.1999	$- 0.0440$	0.0002	0.2079	3.9568	0.0073	0.2002	0.1135	0.0039
Circular Grinded	0.8	0.8008	0.1034	0.0470	0.8100	1.2560	0.0539	0.7682	−3.9790	0.0731
	0.4	0.4102	2.5709	0.0855	0.4231	5.7797	0.1018	0.4204	5.0916	0.0578
	0.2	0.1915	$- 4.2320$	0.0108	0.2109	5.4714	0.0118	0.2215	10.7345	0.0520
	0.1	0.1000	0.0006	0.0002	0.0996	−0.3088	0.0064	0.1069	6.8799	0.0039

Table 1. Experimental Results Comparison

View in the Article

Processing Methods	Parameter	LS-SVM (s)	BPNN (s)	SVM (s)
Internal Grinding	POT	2.135	16.938	399.346
	OT	0.026	1.717	0.167
	TPT	2.161	18.655	399.513
Circular Grinding	POT	2.047	17.406	936.231
	OT	0.052	1.947	0.573
	TPT	2.099	19.353	936.804