Çok Katmanlı Yapay Sinir Ağı Modeli ve Kültürel Algoritma Modeli Kullanılarak Geliştirilen Melez Yöntem ile Kısa Vadeli Fotovoltaik Enerji Santrali Çıkış Gücü Tahmini

Öz

Güneş enerjisinden elde edilen sürdürülebilir enerji, konut, ticari ve endüstriyel uygulamalarda enerji talebini karşılamada önemli kaynaklardan biri haline gelmiştir. Ancak güneşten elektrik enerjisi üretimindeki temel zorluk, fotovoltaik enerji santrallerinde hava koşullarından kaynaklanan güçteki anlık değişimlerdir. Büyük ölçekli güneş FV enerji santralleri için, güç dengesizlikleri elektriksel olarak sistem verimliliği ve kararlılığını olumsuz yönde etkilemektedir. Bu nedenle, FV enerji santrallerinin çıkış gücünün kısa vadede doğru tahmin edilmesi, elektrik şebekesi üretim, dağıtım ve depolamanın günlük/saatlik verimli yönetimi ve enerji piyasasında karar verme için büyük önem taşımaktadır. Bu makalede, FV enerji santralinin güç üretimini tahmin etmek için kültürel geçiş hedefi temelinde popülasyon tabanlı bir algoritma geliştirmeyi amaçlamaktadır. Aynı zamanda, her yinelemede tüm değişkenleri göz önünde bulundurarak daha hızlı yakınsamaya olanak sağlaması özelliği ile Parçacık Sürü Optimizasyon (PSO) yöntemi ile kısa vadeli tahmin yapılmaktadır. Kısa vadeli FV panel çıkış güç tahminin sonuçlarını en az hata oranı ile elde etmek için çok katmanlı yapay sinir ağı modeli PSO ve Kültürel Algoritma (KA) ile kullanılarak melez yöntem oluşturulmuştur. KA iterasyon sırasında toplanan bilgileri depolama ve daha sonra kullanma özelliği ile evrimsel algoritmalardan daha hızlı yakınsama sağladığı için FV enerji çıkış gücü kısa vadeli tahmininden etkin sonuçlar elde edilmiştir.

Anahtar Kelimeler

• Fotovoltaik enerji
• Kültürel algoritma
• Parçacık sürü optimizasyonu
• Kısa vadeli güç tahmini
• Melez yöntem

Short Term Photovoltaic Power Plant Output Power Forecasting with Hybrid Method Developed Using Multilayer Artificial Neural Network Model and Cultural Algorithm Model

Abstract

Sustainable energy obtained from solar energy has become one of the important sources in meeting the energy demand in residential, commercial and industrial applications. However, the main difficulty in generating electricity from solar energy is the instantaneous changes in power caused by weather conditions in photovoltaic (PV) power plants. For large-scale solar PV power plants, power imbalances electrically negatively affect system efficiency and stability. Therefore, accurate forecasting of the output power of PV power plants in the short term is of great importance for efficient daily/hourly management of electricity grid generation, distribution and storage and for decision making in the energy market. This paper aims to develop a population-based algorithm estimating the power generation of a PV power plant based on the cultural transition target. Also, short-term forecasting is carried out based on the Particle Swarm Optimization (PSO) method, with the feature of allowing faster convergence by considering all the variables in each iteration. In order to obtain the results of short-term PV panel output power estimation with the least error rate, a hybrid method was created by using the multilayer artificial neural network model with PSO and Cultural Algorithm (CA). Since the CA provides faster convergence than evolutionary algorithms with the ability to store and use the information collected during iteration, effective results are obtained from the short-term forecasting of the PV energy output power.

Keywords

• Photovoltaic energy
• Cultural algorithm
• Particle swarm optimization

Kaynakça

1. Abedinia O., Amjady N., Ghadimi N. Solar energy forecasting based on hybrid neural network and improved metaheuristic algorithm. Computational Intelligence 2018;34:241–60.
2. AlHakeem D., Mandal P., Haque AU., Yona A., Senjyu T., Tseng T-L. A new strategy to quantify uncertainties of wavelet-GRNN-PSO based solar PV power forecasts using bootstrap confidence intervals. 2015 IEEE Power Energy Society General Meeting, 2015; 1–5.
3. Almonacid F., Pérez-Higueras PJ., Fernández EF., Hontoria L. A methodology based on dynamic artificial neural network for short-term forecasting of the power output of a PV generator. Energy Conversion and Management 2014;85:389–98. Barrera JM., Reina A., Maté A., Trujillo JC. Solar Energy Prediction Model Based on Artificial Neural Networks and Open Data. Sustainability 2020;12:6915.
4. Cervone G., Clemente-Harding L., Alessandrini S., Delle Monache L. Short-term photovoltaic power forecasting using Artificial Neural Networks and an Analog Ensemble. Renewable Energy 2017;108:274–86.
5. Eberhart R., Kennedy J. A new optimizer using particle swarm theory. MHS’95. Proceedings of the Sixth International Symposium on Micro Machine and Human Science, IEEE; 1995, p. 39–43. Hazem Mohammed O., Amirat Y., Benbouzid M. Economical Evaluation and Optimal Energy Management of a Stand- Alone Hybrid Energy System Handling in Genetic Algorithm Strategies. Electronics 2018;7:233.
6. Jang HS., Bae KY., Park HS., Sung DK. Solar Power Prediction Based on Satellite Images and Support Vector Machine. IEEE Trans Sustain Energy 2016;7:1255–1263.
7. Kennedy J., Eberhart R. Particle swarm optimization. Proceedings of ICNN’95-international conference on neural networks, IEEE; 1995;4:1942–1948.
8. Li Z., Rahman SM., Vega R., Dong B. A Hierarchical Approach Using Machine Learning Methods in Solar Photovoltaic Energy Production Forecasting. Energies 2016;9:55. Ma H., Wang Y. Cultural Algorithm Based on Particle Swarm Optimization for Function Optimization, Fifth International Conference on Natural Computation, 14-16 Ağustos 2009, sayfa no:224-228, Tianjian, Çin.

9. Martin R., Aler R., Valls JM., Galvan IM. Machine learning techniques for daily solar energy prediction and interpolation using numerical weather models: Machine learning techniques for daily solar energy prediction and interpolation using numerical weather models. Concurrency Computat: Pract Exper 2016;28:1261–74. Mishra M., Byomakesha Dash P., Nayak J., Naik B., Kumar Swain S. Deep learning and wavelet transform integrated approach for short-term solar PV power prediction. Measurement 2020;166:108250.
10. Reynolds R.G. An introduction to cultural algorithms. In Proceedings of the third annual conference on evolutionary programming, River Edge, NJ: World Scientific 1994;24:131-139. Reynolds RG., Peng B. Cultural algorithms: modeling of how cultures learn to solve problems. 16th IEEE International Conference on Tools with Artificial Intelligence, 15-17 Kasım 2004, sayfa no:166–172, Boca Raton, FL, USA.
11. Saberian A., Hizam H., Radzi MAM., Ab Kadir MZA., Mirzaei M. Modelling and Prediction of Photovoltaic Power Output Using Artificial Neural Networks. International Journal of Photoenergy 2014;2014:1–10.
12. Shuang H.,Qiao YH., Yan J., Liu Y., Li L., Wangb Z. Mid-to-long term wind and photovoltaic power generation prediction based on copula function and long short term memory network. 2019; 239: 181-191. Talbi E. Population-based metaheuristics, Metaheuristics from Des. to implementation. John Wiley Sons, Inc., Hoboken, New Jersey 2009; 190–200.
13. Theocharides S., Makrides G., Livera A., Theristis M., Kaimakis P., Georghiou GE. Day-ahead photovoltaic power production forecasting methodology based on machine learning and statistical post-processing. Applied Energy 2020;268:115023.
14. VanDeventer W., Jamei E., Thirunavukkarasu GS., Seyedmahmoudian M., Soon TK., Horan B, et al. Short-term PV power forecasting using hybrid GASVM technique. Renewable Energy 2019;140:367–79.
15. Zeng J., Qiao W. Short-term solar power prediction using a support vector machine. Renewable Energy 2013;52:118–27.