Determination of distance between DC traction power centers in a 1500-V DC subway line with artificial intelligence methods

The electrification system in rail systems is designed with regard to the operating data and design parameters.While the electrification system is formed, the minimum voltage rating that the traction force requires during theoperation needs to be provided. The highest value of the voltage drop occurring on the line is determined by the distancebetween power centers. This value needs to be kept within certain limits for the continuity of operation. In this study,the determination of the distance between DC traction power centers for a 1500-V DC-fed rail system is done by meansof the adaptive neuro-fuzzy inference system (ANFIS), support vector machines (SVMs), and artificial neural networks(ANNs). The distance occurring on the line is calculated with regard to the operating parameters by means of theANFIS, SVMs, and ANNs. The ANFIS, SVMs, and ANNs are explained and a comparison is made. The data createdregarding one-way and two-way supply conditions are examined for simulation. The main contribution of this paper isthe determination of the distance between railway traction power centers with artificial intelligence methods.

PDF

___

[1] Limbong FG. The use of neural network (NN) to predict voltage drop during starting of medium voltage induction motor. In: 3rd International Conference on Information Technology, Computer, and Electrical Engineering; 19–20 October 2016; Semarang, Indonesia. New York, NY, USA: IEEE. pp. 156-160.
[2] Nuzzo S, Galea M, Gerada C, Brown NL. Prediction of the voltage drop due to the diode commutation process in the excitation system of salient-pole synchronous generators. In: 19th International Conference on Electrical Machines and Systems; 13–16 November 2016; Chiba, Japan. New York, NY, USA: IEEE. pp. 1-6.
[3] Ibrahem A, Elrayyah A, Sozer Y, Garcia JAA. DC railway system emulator for stray current and touch voltage prediction. IEEE T Ind Appl 2017; 53: 439-446.
[4] Meghwani A, Chakrabarti S, Srivastava SC. A fast scheme for fault detection in DC microgrid based on voltage prediction. In: National Power Systems Conference; 19–21 December 2016; Bhubaneswar, India. New York, NY, USA: IEEE. pp. 1-6.
[5] Huh JS, Shin HS, Moon WS, Kang BW, Kim JC. Study on voltage unbalance improvement using SFCL in power feed network with electric railway system. IEEE T Appl Supercond 2013; 3: 3601004.
[6] Goodman CJ, Chymera M. Modelling and simulation. In: REIS 2013 Railway Electrification Infrastructure and Systems Conference; 3–6 June 2013; London, UK. New York, NY, USA: IEEE. pp. 16-25.
[7] Ladoux P, Raimondo G, Caron H, Marino P. Chopper-controlled Steinmetz circuit for voltage balancing in railway substations. IEEE T Power Electron 2013; 28: 5813-5822.
[8] Shin, HS, Cho SM, Kim JC. Protection scheme using SFCL for electric railways with automatic power changeover switch system. IEEE T Appl Supercond 2012; 20: 5600604.
[9] Shin HS, Cho SM, Huh JS, Kim, JC, Kweon, DJ. Application on of SFCL in automatic power changeover switch system of electric railways. IEEE T Appl Supercond 2012; 22: 5600704.
[10] Kolar V, Hrbac R, Mlcak T. Measurement and simulation of stray currents caused by AC railway traction. In: EPE 2015 Electric Power Engineering Conference; 20–22 May 2015; Prague, Czech Republic. New York, NY, USA: IEEE. pp. 764-768.
[11] Chen M, Jiang W, Luo J, Wen T. Modelling and simulation of new traction power supply system in electrified railway. In: ITSC 2015 IEEE 18th International Conference on Intelligent Transportation Systems; 15–18 September 2015; Las Palmas, Spain. New York, NY, USA: IEEE. pp. 1345-1350.
[12] Soler M, Lopez J, Manuel J, Pedro MS, Maroto J. Methodology for multiobjective optimization of the AC railway power supply system. IEEE T Intell Transp Syst 2015; 16: 2531-2542.
[13] Ozkan IA, Saritas I, Herdem S. Modeling of magnetic filtering with ANFIS. In: 12th National Conference on Electrical, Electronics, Computer and Biomedical Engineering; 14–18 November 2007; Eskişehir, Turkey. Ankara, Turkey: CEE. pp. 415-418.
[14] Şit S, Özçalık HR, Kılıç E, Doğmuş O, Altun M. Investigation of performance based on online adaptive neurofuzzy inference system (ANFIS) for speed control of induction motors. Journal of the Faculty of Engineering and Architecture of Çukurova University 2016; 31: 33-42 (in Turkish with abstract in English).
[15] Jang JSR. ANFIS: Adaptive-network-based fuzzy inference system. IEEE T Syst Man Cyb Syst 1993; 23: 665-685.
[16] Ayhan S, Erdoğmuş S. Kernel function selection for the solution of classification problems via support vector machines. University Journal of Economics and Administrative Sciences of Çukurova University 2014; 9: 175-198.
[17] Yakut E, Elmas B, Yavuz S. Predictıng stock-exchange index using methods of neural networks and support vector machines. Journal of Faculty of Economics and Administrative Sciences of Süleyman Demirel University 2014; 19: 139-157.
[18] Kavzoğlu T, Çölkesen İ. Investigation of the effects of kernel functions in satellite image classification using support vector machines. Journal of Map of Gebze High Technology Institute 2010; 144: 73-82 (in Turkish with abstract in English).
[19] Guran A, Uysal M, Dogrusoz O. Effects of support vector machines parameter optimization on sentiment analysis. Journal of Engineering Sciences of DEU Faculty of Engineering 2014; 16: 86-93.
[20] Ozdemir H. Artificial neural networks and their usage in weaving technology. Electronic Journal of Textile Technologies 2013; 7: 51-68.
[21] Sahin M, Buyuktumturk F, Oguz Y. Light quality control with artificial neural networks. AKU-J Sci Eng 2013; 13: 1-10.
[22] Bayındır R, Sesveren Ö. Design of a visual interface for ANN based systems. Journal of Engineering Science of Pamukkale University Faculty of Engineering 2008; 14: 101-109.
[23] Aşkın D, İskender İ, Mamizadeh A. Dry type transformer winding thermal analysis using different neural network methods. Journal of the Faculty of Engineering and Architecture of Gazi University 2011; 26: 905-913 (in Turkish with abstract in English).
[24] Çavuşlu MA, Becerikli Y, Karakuzu C. Hardware implementation of neural network training with LevenbergMarquardt algorithm. Turk J Elec Eng & Comp Sci 2012; 5: 31-38.
[25] Dalkıran I, Danışman K. Artificial neural network based chaotic generator for cryptology. Turk J Elec Eng & Comp Sci 2010; 18: 225-240.
[26] Ceylan M, Özbay Y, Uçan ON, Yıldırım E. A novel method for lung segmentation on chest CT images: complexvalued artificial neural network with complex wavelet transform. Turk J Elec Eng & Comp Sci 2010; 18: 613-623.
[27] Partal S, Şenol İ, Bakan AF, Bekiroğlu KN. Online speed control of a brushless AC servomotor based on artificial neural networks. Turk J Elec Eng & Comp Sci 2011; 19: 373-383.
[28] Jashfar S, Esmaeili S, Jahromi MZ, Rahmanian M. Classification of power quality disturbances using s-transform and tt-transform based on the artificial neural network. Turk J Elec Eng & Comp Sci 2013; 21: 1528-1538.
[29] Afsharizadeh M, Mohammadi M. Prediction-based reversible image watermarking using artificial neural networks. Turk J Elec Eng & Comp Sci 2016; 24: 896-910.
[30] Minaz MR, Gün A, Kurban M, İmal N. Estimation of pressure, temperature and wind speed of Bilecik using different methods. Gaziosmanpaşa Journal of Scientific Research 2013; 3: 100-111 (in Turkish with abstract in English)