کاربرد مدلهای شبکه عصبی مصنوعی (MLP و RBF) و ماشین بردار پشتیبان (SVM) به منظور تخمین میزان سایه در جمع‌کننده‌های خورشیدی صفحه تخت در ایران

نوع مقاله : مقاله پژوهشی

نویسندگان

1 گروه مهندسی ماشینهای کشاورزی و مکانیزاسیون-دانشگاه علوم کشاورزی و منابع طبیعی خوزستان-ملاثانی

2 گروه مهندسی ماشین‌های کشاورزی و مکانیزاسیون، دانشکده مهندسی زراعی و عمران روستایی، دانشگاه علوم کشاورزی و منابع طبیعی خوزستان، ملاثانی

چکیده

در این تحقیق از مدل‌ شبکه عصبی مصنوعی و رگرسیون بردار پشتیبان به‌منظور تخمین میزان سایه در جمع کننده‌ صفحه تخت خورشیدی با توجه به شرایط جغرافیایی ایران استفاده شد. دو نوع الگوریتم آموزش LM و BR همراه با تابع انتقال تانژانت سیگموئیدی و تعداد متنوعی از نورون­ها در لایه پنهان همراه با مدل اعتبارسنجی تقاطعی به منظور ایجاد مجموعه داده‌های تصادفی مورد استفاده قرار گرفت. نتایج نشان داد که مدل MLP با الگوریتم آموزشی BR و ساختار (1-23-5) می‌تواند داده‌هایی با دقت بالا و شبیه به مقادیر واقعی ایجاد کند. میانگین آماره­های MAPE و R2 برای مدل فوق به ترتیب 10/0±42/0 درصد و 01/0±99/0، برآورد شد و نتایج آماری مقایسه میانگین، واریانس و توزیع آماری در سطح احتمال 95% بین داده‌های واقعی و مقادیر پیش‌بینی شده، معنی‌دار نبودند. نتایج آنالیز حساسیت نشان داد که فاصله صفحه جاذب تا پوشش شیشه‌ای مهم‌ترین فاکتور تاثیرگذار بر ایجاد سایه است.

کلیدواژه‌ها


عنوان مقاله [English]

Application of Artificial Neural Network Models (MLP and RBF) and Support Vector Machine (SVM) to Estimate the Shadow in Flat-plate Solar Collectors in Iran

نویسندگان [English]

  • Morteza Taki 1
  • Rouhollah Farhadi 2
1 Department of agricultural machinery and mechanization- Agricultural Sciences and Natural Resources University of Khuzestan-Mollasani
2 Department of Agricultural Machinery and Mechanization, Agricultural Sciences and Natural Resources University of Khuzestan, Mollasani, Iran
چکیده [English]

In this study, the amount of shadow in different types of flat-plate solar collectors according to the geographical conditions of Iran was estimated by using artificial neural network models (MLP and RBF) and Support Vector Machine (SVM). In this study, two types of LM and BR training algorithms with sigmoid tangent transfer function (TanSig) and different number of neurons in a hidden layer with k-fold cross validation method were used to create random datasets at each stage of modeling. The results showed that the MLP model with BR training algorithm and (5-23-1) structure, can create high-precision data similar to real values. The average MAPE and R2 statistics for the above model were estimated to be 0.42 ± 0.10 and 0.99±0.01, respectively. Also, there was no significant difference between the actual data and the predicted values (95% probability) at mean, variance and distribution. The results of sensitivity analysis showed that the distance of the absorber plate and the glass cover is the most important factor influencing the formation of shadows.

کلیدواژه‌ها [English]

  • Artificial intelligence
  • k-fold cross validation model
  • Energy efficiency
Al-Waeli, A.H., Sopian, A. & Yousif, K. (2019). Artificial neural network modeling and analysis of photovoltaic/thermal system based on the experimental study. Energy Conversion and Management, 186, 368–379.
Amini, S., Taki, M. & Rohani, A. (2020). Applied improved RBF neural network model for predicting the broiler output energies. Applied Soft Computing Journal, 87, 106006.
Batzelis, E., Georgilakis, I. & Papathanassiou, S.A. (2015). Energy models for photovoltaic systems under partial shading conditions: a comprehensive review. IET Renewable Power Generation, 9 (4), 340-349
Blaga, R., Sabadus, A., Stefu, N., Dughir, C., Paulescu, M. & Badescu, V. (2019). A current perspective on the accuracy of incoming solar energy forecasting. Progress in Energy Combustion Sciences, 70, 119–44.
Çakmak, G. & Yıldız, C. (2011). The prediction of seedy grape drying rate using a neural network method. Computer and Electronic in Agriculture, 75 (1), 132–138.
Dimri, N., Tiwari, A. & Tiwari, G.N. (2019). An overall exergy analysis of glass-tedlar photovoltaic thermal air collector incorporating thermoelectric cooler: A comparative study using artificial neural networks. Energy Conversion and Management, 195, 1350–1358.
Duffie, J.A. & Beckman, W.A. (2013). Solar Engineering of Thermal Processes, Hoboken, New Jersey, John Wiley & Sons, Inc
Elsheikh, AH., Sharshir, SW., Abd Elaziz, M., Kabeel, AE., Guilan, W. & Haiou, Z. (2019). Modeling of solar energy systems using artificial neural network: a comprehensive review. Solar Energy, 180, 622–39.
Esen, H., Esen, M. & Ozsolak, O. (2017). Modelling and experimental performance analysis of solar-assisted ground source heat pump system. Journal of experimental and theotorical artificial intelligence, 29 (1), 1–17.
Farhadi, R. & Taki, M. (2020). The energy gain reduction due to shadow inside a flat-plate solar collector. Renewable Energy 147, 730-740
Ghritlahre, H.K. & Prasad, R.K. (2018). Prediction of heat transfer of two different types of roughened solar air heater using Artificial Neural Network technique. Thermal science and engineering progress. 8, 145–153.
Hamdan, M.A., Abdelhafez, E.A., Hamdan, A.M. & Khalil, R.A.H. (2016). Heat transfer analysis of a flat-plate solar air collector by using an artificial neural network. Journal of infrastructure systems, 22 (4), A4014004.
Heng, S.Y., Asako, Y., Suwa, T. & Nagasaka, K. (2019). Transient thermal prediction methodology for parabolic trough solar collector tube using artificial neural network. Renewable Energy, 131, 168–179.
Hu, F., Wei, E. & Wang, ZJ. (2013). Average Daily Shading Factor Variations with Aspect Ratios for Different Flat-Plate Collector Arrays. Applied Mechanics and Materials, 368, 949-952
Jia, Y., Alva, G. & Fang, G. (2019). Development and applications of photovoltaic–thermal systems: a review. Renewable and sustainable energy review, 102, 249–65.
Kalani, H., Sardarabadi, M. & Passandideh-Fard, M. (2017). Using artificial neural network models and particle swarm optimization for manner prediction of a photovoltaic thermal nanofluid based collector. Applied Thermal Engineering, 113, 1170–1177.
Lalot, S. & Lecoeuche, S. (2003). Neural models of solar collectors for prediction of daily
performance. International journal of sustainable energy, 23 (1–2), 39–49.
Loni, R., Asli-Ardeh, E.A., Ghobadian, B., Ahmadi, M.H. & Bellos, E. (2018). GMDH modeling and experimental investigation of thermal performance enhancement of hemispherical cavity receiver using MWCNT/oil nanofluid. Solar Energy, 171, 790–803.
Nahar, NM. & Gar, HP. (1980). Free convection and shading due to gap spacing between an absorber plate and the cover glazing in solar energy flat-plate collectors. Applied Energy, 7 (1), 129-145
Rohani, A., Taki, M. & Aodollahpour M. (2018). A novel soft computing model (Gaussian process regression with K-fold cross validation) for daily and monthly solar radiation forecasting (Part: I).  Renewable Energy, 115, 411-422
Roy, K., Mandal, K.K. & Mandal, AC. (2019). Ant-Lion Optimizer algorithm and recurrent neural
network for energy management of micro grid connected system. Energy, 167, 402–16.
Taki, M., Abdanan Mehdizade, S., Rohani, A., Rahnama, M. & Rahmati-Joneidabad, M. (2018a). Applied machine learning in greenhouse simulation; new application and analysis. Information Processing in Agriculture, 252-268.
Taki, M., Rohani, A., Soheilifard, F. & Abdeshahi, A. (2018b). Assessment of energy consumption and modeling of output energy for wheat production by neural network (MLP and RBF) and Gaussian process regression (GPR) models. Journal of Cleaner Production, 172, 3028-3041.
Tang, RS. & Liu, NY. (2012). Shading Effect and Optimal Tilt-Angle of Collectors in a Collector Array. Advanced Materials Research, 588, 2078-2082.