Zaman Serileri Tahminlenmesinde Makine Öğrenimi ve Derin Öğrenme Tekniklerinin Kıyaslanması: Türkiye Elektirik Üretimi için En İyi Tahmin Modelinin Seçilmesine Yönelik Bir Vaka Çalışması

Yıl 2019, , 635 - 646, 25.08.2019

Ramazan Ünlü

https://doi.org/10.19113/sdufenbed.494396

Cited By: 7

Öz

Son yıllarda Türkiye ihtiyaçlarını karşılayabilmek adına elektrik üretimine yoğun bir şekilde dikkat vermektedir. Araştırmacılar elektrik üretim, tüketim ve talep miktarını doğru bir şekilde tahmin etmek için istatistik ve yapay zeka tabanlı yöntemleri de içeren birçok farklı metod uygulamışlardır. Sınırlı sayıda araştırmacı Türkiye’nin elektrik üretim tahminleme problemini bir zaman serisi analizi olarak irdelemiştir. Bu nedenle bu çalışmada söz konusu problem zaman serileri analizi olarak ele alınmıştır. Bu açıdan çalışmada hem Destek Vektör Makineleri (DVM) ve Çok Katmanlı Nöronlar (ÇKN) gibi klasik makine öğrenimi yöntemleri hem de Uzun Kısa Dönemli Hafıza (UKDH) yöntemi gibi derin öğrenme yöntemi Türkiye’nin üretmesi gereken aylık elektrik üretim miktarını tahmin etmek için kullanılmıştır. Çalışmanın bulgularına dayalı olarak derin öğrenme algoritması istatistiksel hata oranlarına göre diğer klasik makine öğrenimi yöntemlerinden daha başarılı sonuçlar vermektedir.

Anahtar Kelimeler

Makine öğrenimi, Derin öğrenme, UKDH, Zaman serileri

A Comparative Study of Machine Learning and Deep Learning for Time Series Forecasting: A Case Study of Choosing the Best Prediction Model for Turkey Electricity Production

Öz

Over the last decades, Turkey pays special attention to electricity productionbto afford its needs. Researchers applied different methodologies including statisticalbased and artificial intelligence-based to correctly predict the future amount of electricity production, consumption, and demand. However,limited researchers focused on Turkey’s electricity production prediction problem as a time series analysis. For this reason, we tackle this problem by considering it as a time series analysis in this study. We have used different methods including traditional machine learning algorithms Support Vector Regression (SVR) and Multilayer Perceptrons (MLP) and a deep learning algorithm Long Short-Term Memory (LSTM) to create a better model for Turkey monthly electricity production dataset. Based on our findings LSTM outperforms SVR and MLP approaches in terms of commonly used statistical error evaluation metrics.

Anahtar Kelimeler

Machine learning, Deep learning, LSTM, Time series

Kaynakça

• [1] Zafer Dilaver and Lester C Hunt. Industrial electricity demand for turkey: a structural time series analysis. Energy Economics, 33(3):426–436, 2011.
• [2] Alper Ünler. Improvement of energy demand forecasts using swarm intelligence: The case of turkey with projections to 2025. Energy Policy, 36(6):1937– 1944, 2008.
• [3] M Duran Toksarı. Estimating the net electricity energy generation and demand using the ant colony optimization approach: case of turkey. Energy Policy, 37(3):1181–1187, 2009.
• [4] Yi Lin, Mian-yun Chen, and Sifeng Liu. Theory of grey systems: capturing uncertainties of grey information. Kybernetes, 33(2):196–218, 2004.
• [5] Diyar Akay and Mehmet Atak. Grey prediction with rolling mechanism for electricity demand forecasting of turkey. Energy, 32(9):1670–1675, 2007.
• [6] Coskun Hamzacebi and Huseyin Avni Es. Forecasting the annual electricity consumption of turkey using an optimized grey model. Energy, 70:165–171, 2014.
• [7] Volkan ¸S Ediger and Sertac Akar. Arima forecasting of primary energy demand by fuel in turkey. Energy Policy, 35(3):1701–1708, 2007.
• [8] Erkan Erdogdu. Natural gas demand in turkey. Applied Energy, 87(1):211–219, 2010.
• [9] Jun-song Jia, Jing-zhu Zhao, Hong-bing Deng, and Jing Duan. Ecological footprint simulation and prediction by arima modelâC”a case study in henan province of china. Ecological Indicators, 10(2):538–544, 2010.
• [10] Ali Sait Albayrak. Arima forecasting of primary energy production and consumption in turkey: 1923-2006. Enerji, piyasa ve düzenleme, 1(1):24–50, 2010.
• [11] Samuel Asuamah Yeboah, Manu Ohene, TB Wereko, et al. Forecasting aggregate and disaggregate energy consumption using arima models: a literature survey. Journal of Statistical and Econometric Methods, 1(2):71–79, 2012.
• [12] Kadir Kavaklioglu. Modeling and prediction of Turkeys electricity consumption using support vector regression. Applied Energy, 88(1):368–375, 2011.
• [13] Yetis Sazi Murat and Halim Ceylan. Use of artificial neural networks for transport energy demand modeling. Energy policy, 34(17):3165–3172, 2006.
• [14] Adnan Sozen, Erol Arcaklioglu, and Mehmet Ozkaymak. Modelling of turkey’s net energy consumption using artificial neural network. International Journal of Computer Applications in Technology, 22(2-3):130–136, 2005.
• [15] Serhat Kucukali and Kemal Baris. Turkeys short-term gross annual electricity demand forecast by fuzzy logic approach. Energy policy, 38(5):2438– 2445, 2010.
• [16] Coskun Hamzaçebi. Forecasting of Turkey’s net electricity energy consumption on sectoral bases. Energy Policy, 35(3):2009–2016, 2007.
• [17] Ujjwal Kumar and VK Jain. Time series models (grey-markov, grey model with rolling mechanism and singular spectrum analysis) to forecast energy consumption in india. Energy, 35(4):1709–1716, 2010.
• [18] Ramazan Ünlü and Petros Xanthopoulos. Estimating the number of clusters in a dataset via consensus clustering. Expert Systems with Applications, 125:33– 39, 2019.
• [19] Ramazan Ünlü and Petros Xanthopoulos. A weighted framework for unsupervised ensemble learning based on internal quality measures. Annals of Operations Research, 276(1-2):229–247, 2019.
• [20] Amanpreet Singh, Narina Thakur, and Aakanksha Sharma. A review of supervised machine learning algorithms. In 2016 3rd International Conference on Computing for Sustainable Global Development (INDIACom), pages 1310–1315. IEEE, 2016.
• [21] Jürgen Schmidhuber. Deep learning in neural networks: An overview. Neural networks, 61:85–117, 2015.
• [22] Yuming Hua, Junhai Guo, and Hua Zhao. Deep belief networks and deep learning. In Proceedings of 2015 International Conference on Intelligent Computing and Internet of Things, pages 1–4. IEEE, 2015.
• [23] Geoffrey E Hinton, Simon Osindero, and Yee-Whye Teh. A fast learning algorithm for deep belief nets. Neural computation, 18(7):1527–1554, 2006.
• [24] Mike Schuster and Kuldip K Paliwal. Bidirectional recurrent neural networks. IEEE Transactions on Signal Processing, 45(11):2673–2681, 1997.
• [25] Tomáš Mikolov, Martin Karafiát, Lukáš Burget, Jan Cernocky, and Sanjeev Khudanpur. Recurrent neural network based language model. In Eleventh annual conference of the international speech communication association, 2010.
• [26] Alex Graves, Abdel-rahman Mohamed, and Geoffrey Hinton. Speech recognition with deep recurrent neural networks. In 2013 IEEE international conference on acoustics, speech and signal processing, pages 6645–6649. IEEE, 2013.
• [27] Nal Kalchbrenner and Phil Blunsom. Recurrent continuous translation models. In Proceedings of the 2013 Conference on Empirical Methods in Natural Language Processing, pages 1700–1709, 2013.
• [28] Vladimir Vapnik. Statistical learning theory. 1998, volume 3. Wiley, New York, 1998.
• [29] Harris Drucker, Christopher JC Burges, Linda Kaufman, Alex J Smola, and Vladimir Vapnik. Support vector regression machines. In Advances in neural information processing systems, pages 155–161, 1997.
• [30] Richard Lippmann. An introduction to computing with neural nets. IEEE Assp magazine, 4(2):4–22, 1987.
• [31] Teuvo Kohonen. Self-organization and associative memory, volume 8. Springer Science & Business Media, 2012.
• [32] David E Rumelhart and James L McClelland. Parallel distributed processing: explorations in the microstructure of cognition. volume 1. foundations. 1986.
• [33] Teuvo Kohonen. An introduction to neural computing. Neural networks, 1(1):3–16, 1988.
• [34] Sepp Hochreiter and Jürgen Schmidhuber. Long short-term memory. Neural computation, 9(8):1735–1780, 1997.
• [35] Jiawei Han, Jian Pei, and Micheline Kamber. Data mining: concepts and techniques. Elsevier, 2011.
• [36] Ian H Witten, Eibe Frank, Mark A Hall, and Christopher J Pal. Data Mining: Practical machine learning tools and techniques. Morgan Kaufmann, 2016.