Acta Photonica Sinica, Volume. 48, Issue 12, 1206002(2019)
Temperature Error Modeling and Real-time Compensation of Fiber Optic Gyroscope Based on PSO-SVR
Figures & Tables(14)
Fig. 8. RBF神经网络实时补偿效果Real-time compensation effect of RBF neural networks
Table 1. PSO parameters setting
View tableView in ArticleTable 1. PSO parameters setting
Parameters Value Population size n 60 Number of iterations m 100 Learning factor c 12 Learning factor c 22
Table 2. Best parameters for SVR obtained by PSO
View tableView in ArticleTable 2. Best parameters for SVR obtained by PSO
Parameters ε C σ Value 3.59×10-5 51.86 88.22
Table 3. Comparison of results of threemodeling methods
View tableView in ArticleTable 3. Comparison of results of threemodeling methods
Schemes RMSE Maximum error/(°·h-1) Before compensation 8.09×10-2 1.28×10-1 Least squares 9.71×10-3 3.34×10-2 RBF neural networks 5.50×10-3 2.69×10-2 PSO-SVR 4.28×10-4 7.40×10-4
Table 4. Comparison of real-time compensation results
View tableView in ArticleTable 4. Comparison of real-time compensation results
Schemes RMSE Maximumerror/(°·h-1) Before compensation 6.99×10-2 1.24×10-1 Least squares 1.55×10-2 1.00×10-1 RBF neural networks 2.02×10-2 1.52×10-1 PSO-SVR 1.32×10-2 6.54×10-2
Table 5. Operation time
View tableView in ArticleTable 5. Operation time
Schemes Time of 10 000 compensation points/s Average time of per compensation point/s Least squares 0.15 1.50×10-5 RBF neural networks 88.60 8.86×10-3 PSO-SVR 1.11 1.11×10-4
Tools
Get Citation
Copy Citation Text
Chun-fu HUANG, An LI, Fang-jun QIN, Zhi WANG. Temperature Error Modeling and Real-time Compensation of Fiber Optic Gyroscope Based on PSO-SVR[J]. Acta Photonica Sinica, 2019, 48(12): 1206002
Paper Information
Received: Jul. 25, 2019
Accepted: Sep. 10, 2019
Published Online: Mar. 17, 2020
The Author Email: QIN Fang-jun (haig2005@126.com)
© Copyright 2018-2021 | Chinese Laser Press.
All Rights Reserved 沪ICP备15018463号-20