Email this article 
Desain dan simulasi adaptive high power LED driver menggunakan feedforward backpropagation neural network
Abstract
Lampu LED (light emitting diode) merupakan salah satu jenis lampu hemat energi yang mempunyai lifetime yang panjang. Lampu LED memiliki dua bagian utama, yaitu LED modul dan LED driver modul. LED modul dan LED driver keduanya harus memiliki sepesifikasi yang sama, apabila tidak sesuai maka LED modul tidak bisa menyala karena tegangan atau arus yang kurang dan bisa mengalami kerusakan karena melebihi batas arus dan tegangan. LED driver yang umum dipasaran hanya dapat digunakan untuk satu jenis LED modul saja, sehingga setiap LED modul dengan daya yang berbeda memerlukan LED driver yang berbeda-beda. Penelitian ini membahas tentang adaptive LED driver yang mampu menyesuakain dengan kebutuhan LED modul. Metode yang digunakan adalah feed-forward backpropagation neural network (FF-BPNN) yang diadopsi dari cara kerja sistem saraf biologis. FFBPNN terdiri dari layer input, hidden layer, dan layer output. Penggunaan metode ini berfungsi sebagai kontrol driver LED agar didapatkan daya yang sesuai dengan kebutuhan tiap lampu LED modul sehingga tidak terjadi over current dan over voltage. Pengujian simulasi adaptif LED driver dilakukan dengan 3 variasi daya LED modul yaitu sebesar 50 watt, 70 watt dan 100 watt. Hasil simulasi menunjukkan bahwa driver LED mampu menyesuaikan rating dari daya led yaitu sebesar 49.89 watt, 69.94 watt dan 99.42 watt.
LED (light-emitting diode) lamps are one type of energy-saving lamp that has a long lifetime. The LED lamp has two main parts, i.e the LED module and the LED driver module. Both module LEDs and driver LEDs must have the same specifications, if they do not match, the module LEDs cannot turn on due to insufficient voltage or current and can be damaged because they exceed the current and voltage limits. The general LED driver in the market can only be used for one type of LED module, so each LED module with different power requires a different LED driver. This research discusses the adaptive LED driver that can suit the needs of the LED module. The method used is a feed-forward backpropagation neural network (FF-BPNN) which is adopted from the workings of the biological nervous system. FF-BPNN consists of an input layer, a hidden layer, and an output layer. The use of this method functions as an LED driver control to obtain power that is under the needs of each LED module lamp so that over current and over voltage do not occur. Adaptive LED driver simulation testing is done with 3 variations of LED module power, i.e 50 watts, 70 watts, and 100 watts. The simulation results show that the LED driver can adjust the rating of the led power which is 49.89 watts, 69.94 watts, and 99.42 watts.
Keywords
Full Text:
PDF (Indonesian)References
Suhardi, D. (2014). Prototipe controller lampu penerangan LED (light emitting diode) independent bertenaga surya. Jurnal Gamma, no. 10, vol. 1, pp. 116-122.
Bakin, N. N., Tuyev, V. I., & Yauk, E. F. (2011). LED lighting. 2011 International Conference and Seminar on Micro/Nanotechnologies and Electron Devices Proceedings, Erlagol, Altai, IEEE proceeding, pp. 346-348. doi: 10.1109/EDM.2011.6006944.
Yudong, Y., & Liqun, H. (2012). Design of high power LED driver power supply. 2012 Fifth International Conference on Information and Computing Science, Liverpool, IEEE proceeding, vol. 1, pp. 99-102. doi: 10.1109/ICIC.2012.19.
Lun, W. K., Loo, K. H., Tan, S. C., Lai, Y. M., & Tse, C. K. (2009). Bilevel current driving technique for LEDs. IEEE Trans. Power Electron, vol. 24, no. 12, pp. 2920–2932. doi: 10.1109/TPEL.2009.2021687.
Ng, S. K., Loo, K. H., Ip, S. K., Lai, Y. M., Tse, C. K., & Mok, K. T. (2012). Sequential variable bilevel driving approach suitable for use in highcolor-precision LED
display panels. IEEE Transactions on Industrial Electronics, vol. 59, no. 12, pp. 4637-4645. doi: 10.1109/TIE.2011.2180490.
Hsia, S., Sheu, M., & Lai, S. (2015). Chip implementation of high-efficient light-emitting diode dimming driver for high-power light-emitting diode lighting system. IET Power Electronics, vol. 8, no. 6, pp. 1043-1051. doi: 10.1049/iet-pel.2013.0805.
Hsia, S., Sheu, M., Hsiao, P., & Chen, K. (2017). Auto tracking DC/DC converter for adaptive LED driving system. 2017 IEEE Wireless Power Transfer Conference (WPTC), Taipei, pp. 1-4. doi: 10.1109/WPT.2017.7953807.
Mehrotra, K., Mohan, C. K., & Ranka, S. (1997). Elements of Artificial Neural Networks. Cambridge: MIT Press.
Hart, D. W. (2002). Power Electronic. New York: McGraw - Hill Companies, Inc.
Salam, Z. (2003). Power electronics and Drives (version 3). UTMJB. https://documents.pub/document/power-electronics-and-drives-version-3-2003- dr-zainal-salam-utmjb-1-dc.html. Accessed on 15th June 2020.
Anggriawan, D. O., Satriawan, A. L., Sudiharto, I., Wahjono, E., Prasetyono, E., & Tjahjono, A. (2018). Levenberg Marquardt backpropagation neural network for harmonic detection. International Electronics Symposium on Engineering Technology and Applications (IES-ETA), Bali, pp. 129132. doi: 10.1109/ELECSYM.2018.8615531.
Tjahjono, A., Anggriawan, D. O., Priyadi, A., Pujiantara, M., & Purnomo, M. H. (2015). Digital overcurrent relay with conventional curve modeling using Levenberg-Marquardt backpropagation. 2015 International Seminar on Intelligent Technology and Its Applications (ISITIA), Surabaya, pp. 171-176. doi: 10.1109/ISITIA.2015.7219974.
Zhan, X., Wang, W., & Chung, H. S. (2017). Application of artificial neural-network to control the light of multi-color LED system. 2017 IEEE Energy Conversion Congress and Exposition (ECCE), Cincinnati, OH, pp. 3669-3675. doi: 10.1109/ECCE.2017.8096650.
DOI: http://dx.doi.org/10.36055/tjst.v16i2.8411
Refbacks
- There are currently no refbacks.
Copyright (c) 2020 Teknika: Jurnal Sains dan Teknologi
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
Teknika: Jurnal Sains dan Teknologi is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.