Time Series Artificial Neural Network Approach for Prediction of Optical Lens Properties

• Authors: A. Isik , N. Isik
• Languages of publication: EN

Abstracts

EN

Well-designed electrostatic cylindrical lenses are commonly used to control charged particles in atomic and molecular physics instruments such as electron guns and electron microscopes. The most commonly used of these, three-element electrostatic lenses are capable of keeping magnification constant for definite image position. The correct determination of focal and aberration characteristics of these lenses is very important for experimental studies. In this study, motions of electrons in three-element electrostatic cylindrical lenses have been investigated with nonlinear autoregressive exogenous based time series artificial neural network technique. The spherical and chromatic aberrations which affect the beam are also predicted with time series artificial neural network technique. This method is a mathematical model that emulates the biological neural networks. The basic working principle of time series artificial neural network technique is training of network with the known data and then prediction of the unknown data. Simulation results from SIMION 8.1 ray-tracing program are used as training and test data set. According to the results obtained from time series artificial neural network technique technique, a considerably agreement is found between simulation and artificial neural network technique prediction results. The study shows that such an artificial neural network model which has time advantage can be applicable to various electron and ion beam apparatus.

Keywords

EN

41.85.Ne; 07.05.Tp; 84.35.+I; 07.05.Mh

Discipline

• 07.05.Mh: Neural networks, fuzzy logic, artificial intelligence
• 41.85.Ne: Electrostatic lenses, septa
• 07.05.Tp: Computer modeling and simulation

• Publisher: Institute of Physics, Polish Academy of Sciences
• Journal: Acta Physica Polonica A
• Year: 2016
• Volume: 129
• Issue: 4
• Pages: 514-516

Dates

published

2016-04

Contributors

author

A. Isik

• Mehmet Akif Ersoy University, Department of Computer Engineering, Burdur, Turkey

author

N. Isik

• Mehmet Akif Ersoy University, Department of Science Education, Burdur, Turkey

References

• [1] N. Mirsaleh-Kohan, W.D. Robertson, R.N. Compton, Mass Spectrom. Rev. 27, 237 (2008), doi: 10.1002/mas.20162
• [2] M. Yavuz, N. Okumus, Z.N. Ozer, M. Ulu, M. Dogan, Acta Phys. Pol. A 125, 442 (2014), doi: 10.12693/aphyspola.125.442
• [3] O. Sise, M. Ulu, M. Dogan, Nucl. Instrum. Methods Phys. Res. A 554, 114 (2005), doi: 10.1016/j.nima.2005.08.068
• [4] O. Sise, N. Okumus, M. Ulu, M. Dogan, J. Electron. Spectrosc. Relat. Phenom. 175, 76 (2009), doi: 10.1016/j.elspec.2009.08.004
• [5] D.W.O. Heddle, Electrostatic Lens Systems, IOP Press, London, UK 2000
• [6] O. Sise, J. Electron. Spectrosc. Relat. Phenom. 197, 7 (2014), doi: 10.1016/j.elspec.2014.07.014
• [7] O. Sise, M. Ulu, M. Dogan, Nucl. Instrum. Methods Phys. Res. A 573, 329 (2007), doi: 10.1016/j.nima.2007.01.051
• [8] A.H. Isik, Acta Phys. Pol. A 127, 1317 (2015), doi: 10.12693/aphyspola.127.1317
• [9] A.H. Isik, Acta Phys. Pol. A 127, 1717 (2015), doi: 10.12693/APhysPolA.127.1717
• [10] H.C. Chen, Y.J. Chen, C.Y. Chang, Y.A. Lin, J.C. Chien, R.C. Hwang, Adv. Mech. Electron. Eng. 177, 209 (2012), doi: 10.1007/978-3-642-31516-9_35
• [11] D. Marquardt, SIAM J. Appl. Math. 11, 431 (1963), doi: 10.1137/0111030

Identifiers

DOI

10.12693/APhysPolA.129.514