Prediction of CAT 6A U/FTP Cable Parameters Produced by Double Twist and Triple Twist machines Using Artificial Neural Networks, and Comparison of the Predicted Results

Yıl 2019, Cilt: 2 Sayı: 2, 41 - 51, 31.12.2019

Pelin Öztürk , Hafiz Alisoy , Reşat Mutlu

Öz

Data cables are hard to model due to their frequency dependent
parameters and manufacturing process variables. In order to obtain the desired
cable parameters, commonly trial and error methods in production are used.
Every trail has an associated cost and requires a production time. In a recent
study, it is shown that Artificial Neural Networks (ANNs) are able to predict
parameters of CAT 6A U/FTP cable such as Characteristic impedance, Near-end
cross-talk (NEXT) and Return loss (RL) well. There are different types of
machines used in manufacturing CAT 6A U/FTP Data cables; Double
Twist and Triple Twist machines. The performance of the ANN’s parameter
prediction change with the machine type used in manufacturing process. In this
study, using MATLAB^TM’s NNTool toolbox, the performance of the ANN’s
prediction of parameters such as impedance, NEXT and Return Loss (RL) are
examined for the Double Twist and Triple Twist machines. It has been found that
the ANNs are able to predict the cable parameters manufactured with the Double
Twist machine more accurately.

Anahtar Kelimeler

ANN, Parameter prediction, Data cables, Cable modeling

Yapay Sinir Ağları Kullanarak İkili ve Üçlü Büküm Makinaların Ürettiği CAT 6A U/FTP Kabloların Parametrelerinin Tahmini ve Tahmin Edilen Sonuçların Karşılaştırılması

Öz

Data kablolarının frekans bağımlılıklarından ve üretim süreçlerindeki değişkenlerden dolayı modellenmeleri oldukça zordur. İstenilen Data kablosu parametrelerini elde etmek için üretimde yaygınca deneme ve yanılma yöntemleri kullanılır. Her bir denemenin ayrı bir masrafı olmakta ve her bir kablo üretimi ayrı bir üretim zamanı gerektirmektedir. Yapay Sinir Ağlarının (YSA) CAT 6A U/FTP data kablolarının karakteristik empedans, yakın-uç çapraz-atlama gürültüsü (NEXT) ve Dönüş Kaybı (RL) gibi parametrelerini iyi bir şekilde tahmin edebildikleri son zamanlarda yapılan bir çalışmada gösterilmiştir.  CAT 6A U/FTP data kablosu üretiminde kullanılan farklı tipte makinalar; İkili büküm (Double Twist) ve Üçlü büküm (Triple Twist) makinaları mevcuttur. YSA’larının data kablosu parametrelerinin tahmin başarımı kullanılan makine tipine göre değişebilir. Bu çalışmada YSA’ların İkili büküm (Double Twist) ve Üçlü büküm makinalarında üretilen CAT 6A U/FTP data kablolarının karakteristik empedans, yakın-uç çapraz-atlama gürültüsü (NEXT) ve Dönüş Kaybı (RL) gibi parametrelerini tahmin başarımı MATLAB^TM’in NNTool paket programı kullanılarak incelenmiştir. YSA’ların İkili büküm makinası tarafından kullanılan kablo parametrelerini daha hassas bir şekilde tahmin ettiği görülmüştür.

Anahtar Kelimeler

YSA, Parametre tahmini, Data kabloları, Kablo modelleme

Kaynakça

• [1] Nevosad M. & Lafata P, Modelling of Propagation Constant of Twisted Pairs and Its Temperature Dependence at G. fast Frequencies (2016). Elektronika ir Elektrotechnika, 22(2), 107-113.[2] Lafata, P. (2015). Simple Attenauation Models of Metallic Cables Suitable for G. fast Frequencies. Advances in Electrical and Electronic Engineering, 13(2), 147-155.[3] Cecchi, V., Leger, A. S., Miu, K., Nwankpa, C. O., Incorporating temperature variations into transmission-line models (2011). IEEE Transactions on Power Delivery, 26(4), 2189-2196.[4] Hoshmeh, A., Schmidt, U. (2017). A full frequency-dependent cable model for the calculation of fast transients. Energies, 10(8), 1158.[5] Kirawanich, P., Islam, N. E., Yakura, S. J. (2006). An electromagnetic topology approach: Crosstalk characterizations of the unshielded twisted-pair cable. Progress In Electromagnetics Research, 58, 285-299.[6] Nevosad, M., Lafata, P., Jares, P. (2013). Modeling of telecommunication cables for gigabit DSL application. Advances in Electrical and Electronic Engineering, 11(5), 336-341.[7] Celozzi, S., Feliziani, M. (1990). EMP-coupling to twisted-wire cables, in IEEE International Symposium on Electromagnetic Compatibility pp. 85-89.[8] Komisarek, K. S., Chamerberlin, K. A., Sivaprasad, K.(1993). A method of moment analysis of a twisted-pair transmission line, in Proceedings of IEEE Antennas and Propagation Society International Symposium. (s. 64-67).[9] Joffe, E. B., Axelrod, A. (1994). On the benefits (if any) of pair twisting in reducing radiated emissions from two-wire cables, in Proceedings of IEEE Symposium on Electromagnetic Compatibility, 474-478.[10] Im, G. H., Werner, J. (1993). Bandwidth-efficient digital transmission up to 155 Mb/s over unshielded twisted pair wiring, in Proceedings of ICC'93-IEEE International Conference on Communications (s.1797-1803).[11] Diakun, P. C., Derewiany, C. F. (1993). Magnetic field shielding effectiveness of a high-permeability shield on twisted pair and coaxial cables, in 1993 International Symposium on Electromagnetic Compatibility, (s. 170-175).[12] Piper, G. R., Prata, A. (1996). Magnetic flux density produced by finite-length twisted-wire pairs, IEEE transactions on electromagnetic compatibility, 38(1), 84-92.[13] Roden, J. A., Gedney, S. D., Paul, C. R. (1996). A rigorous analysis of twisted pair transmission lines using non-orthogonal FDTD and the PML absorbing boundary condition, in Proceedings of Symposium on Electromagnetic Compatibility, IEEE, 254-258.[14] Poltz, J., Gleich, D., Josefsson, M., Lindstrom, M. (2000). Electromagnetic modeling of twisted pair cables, In Proceedings of the 49th International Wire and Cable symposium.[15] Shao, J., Nitta, S., Mutoh, A. (1999). Study on the influence of ground on crosstalk reduction characteristics of twisted-pair-wire. The case of capacitive coupling, in 1999 International Symposium on Electromagnetic Compatibility (IEEE Cat. No. 99EX147), 730-733.[16] Umek, A. (2000). Modeling the structural return loss in twisted pair cables, in 2000 10th Mediterranean Electrotechnical Conference. Information Technology and Electrotechnology for the Mediterranean Countries. Proceedings. MeleCon 2000 (Cat. No. 00CH37099), vol. 1, pp. 173-176.[17] Liu, X.(2006). Low pressure partial discharge investigation with FEM modeling for a twisted pair of insulated conductors, in 2006 IEEE Conference on Electrical Insulation and Dielectric Phenomena, 611-614.[18] Belkhelfa, S., Lefouili, M., Drissi, K. E. K. (2015). Frequency domain analysis of EM crosstalk problem in a quad by the equivalent cable bundle method among twisted-wire pairs cable bundle. IEEE Transactions on Magnetics, 51(11), 1-4.[19] Tatematsu, A., Rachidi, F., Rubinstein, M. (2017). A technique for calculating voltages induced on twisted-wire pairs using the FDTD method, IEEE Transactions on Electromagnetic Compatibility, 59(1), 301-304.[20] Shang, Y., Fei, W., Yu, H.(2013). A fractional-order RLGC model for terahertz transmission line. in 2013 IEEE MTT-S International Microwave Symposium Digest (MTT) (s. 1-3). IEEE.[21] Limei Y., Yusong Z., Jianjun X., Weijian R., Qiong W., Zhigang S. (2014). Transmission lines modeling method based on fractional order calculus theory, Transactions of China Electrotechnical Society, 29, 260-268.[22] Liang, G., & Liu, X., A reduction algorithm for fractional order transmission line modeling with skin effect, International Journal of u-and e-Service, Science and Technology, 239-250, 2015.[23] Pozar, D. M. (2014). Mikrodalga mühendisliği, Palme Yayıncılık.[24]Google, Reçber Kablo Datasheet, https://www.recber.com.tr , 2019.[25] Haykin, S and Network, N. (2004). Neural networks-A Comprehensive Foundation.[26] Bose, N. K., Liang, P. (1996). Neural Network Fundamentals with Graphs, Algorithms, and Applications, McGraw-Hill Series in Electrical Computer Engineering. [27] Monostori L., Prohaszka J. (1993). A step towards intelligent manufacturing: Modelling and monitoring of manufacturing processes through artificial neural networks, CIRP Annals-Manufacturing Technology, 42(1), 485-488.[28] Rajagopalan R., Rajagopalan P.(1999) Applications of Neural Network in Manufacturing, in Proceedings of the 29th Annual Hawaii International Conference on System Sciences.[29] Cho, H.S., Leu, M. C. (1998). Artificial Neural Networks in Manufacturing Processes Monitoring and Control, IFAC Proceedings Volumes.[30] Kumar, A., Chauhan, V., Bist, A. S. (2013). Role of artificial neural network in welding technology: a survey network, 67(1).[31] Smail M.K., Le Bihan Y., Pichon L. (2012). Fast Diagnosis of Transmission Lines using Neural Networks and Principal Component Analysis, International Journal of Applied Electromagnetics and Mechanics, IJAEM, Vol. 39, Issue: 1, 2012, 435-441.[32] Chauhan, N., Yadav, N., Arya, N. (2018). Applications of Artificial Neural Network in Textiles, International Journal of Current Microbiology and Applied Sciences, 7(4), 3134-3143.[33] Pelin Ö., Hafız A., and Reşat M. (2019). CAT 6A U/FTP Data Kablosunun Yüksek Frekans Paremetrelerinin YSA ile Tahmin Modeli. 4nd International Conference on Material Science and Technology in Kızılcahamam (IMSTEC’19) October 18-20 2019 (s. 604-612). [34] Öztürk P., Yapay Sinir Ağı Tabanlı Data Kablosu Parametreleri Tahmini, (2019) Yüksek Lisans Tezi, Tekirdağ Namık Kemal Üniversitesi, Tekirdağ. [35] Lewis, C. D. (1982). Industrial and business forecasting methods: A practical guide to exponential smoothing and curve fitting, Butterworth-Heinemann.