Identification of Steel Plate Damage Position Based on Particle Swarm Support Vector Machine

FBG No.	x-axis location /mm	y-axis location /mm	Center wavelength λ /nm	Grating length /mm	Reflectivity /%
1	50	50	1529.76	10	≥96
2	150	50	1532.94	10	≥96
3	250	50	1534.38	10	≥96
4	50	150	1535.38	10	≥96
5	150	150	1537.56	10	≥96
6	250	150	1540.64	10	≥96
7	250	250	1542.50	10	≥96
8	150	250	1545.24	10	≥96
9	50	250	1549.02	10	≥96

Table 2. Sample data for network training

View table

Table 2. Sample data for network training

FBG center wavelength change Δλ /pm									Location /mm
No. 1	No. 2	No. 3	No. 4	No. 5	No. 6	No. 7	No. 8	No. 9	x axis	y axis
-12	23	70	10	0	-1	3	6	17	40	20
-14	35	72	12	5	-1	-1	12	23	60	40
1	22	38	0	-2	-13	-1	28	21	60	100
8	30	30	1	-4	-17	-2	27	23	80	120
15	22	24	-10	-11	-28	-7	24	20	100	140
18	15	22	-4	-6	-24	-10	34	22	120	180
24	10	17	25	23	11	23	131	33	140	240
26	20	23	-8	6	-10	-6	60	21	160	200
31	18	19	2	5	4	5	74	20	180	220
21	12	17	19	33	36	46	88	14	200	260
20	11	17	11	21	23	42	60	11	220	260
8	10	19	16	28	19	90	46	13	240	280
33	31	13	-2	7	12	16	17	25	260	200

Table 3. Predictive results of BP and LSSVM
View table
Table 3. Predictive results of BP and LSSVM
Algorithm Maximum distance error /mm Variance of the distance error Sample forecast rate /%
BP 39.82 9.23 88.24
LSSVM 38.73 8.71 94.12

Table 4. Comparison of LSSVM and PSO-LSSVM predictive results
View table
Table 4. Comparison of LSSVM and PSO-LSSVM predictive results
Algorithm Maximum distance error /mm Variance of the distance error Sample forecast rate /%
LSSVM 38.73 8.71 94.12
PSO-LSSVM 29.29 7.44 97.06

Tools

Get Citation

Copy Citation Text

Zhang Yanjun, Wang Huimin, Fu Xinghu, Zhang Yinan. Identification of Steel Plate Damage Position Based on Particle Swarm Support Vector Machine[J]. Chinese Journal of Lasers, 2017, 44(10): 1006006

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites