Experimental study for artificial neural network (ANN) based prediction of electric energy production of diesel engine based cogeneration power plant

Abstract

In this study artificial neural network (ANN) has been developed in order to estimate the electricity production of cogeneration power plant, which produces a total of 11.52 MW electric power, consisting of two V type and 12 cylinders each of which is 5.760 kW diesel engines running with heavy fuel oil no 6. In the ANN which was developed for the estimation of electric power generation of cogeneration, power plant(W), Time period (t), working hours (h), fuel consumption (m) and internal power consumption (Wp) values were used as input variables. After evaluating the performance of different ANNs, an ANN, consisting of one hidden layer and 10 neurons, was considered to be the most ideal one. As a result of the comparison with experimental data, it is concluded that this model estimates the electricity generation values of the cogeneration power plant with an R-value of 0,99073 and mean square error 4.734e-8 Bu çalışmada iki adet V tipi 12 silindirli dizel motordan oluşan ve her biri 5.760 kW olmak üzere toplam 11.52 MW elektrik enerjisi üreten kojenerasyon enerji santralinin elektrik üretiminin tahmin edilmesi için yapay sinir ağı (YSA) geliştirilmiştir. 6 no'lu fuel oil ile çalışan kojenerasyon enerji santralinin elektrik enerjisi üretiminin (W) tahmini için geliştirilen YSA'da, zaman (t), çalışma saatleri (h), yakıt tüketimi (m) ve iç tüketim (Wp) değerleri giriş değişkenleri olarak kullanılmıştır. Farklı YSA'ların performansı değerlendirildikten sonra, bir gizli katman ve 10 nörondan oluşan YSA en ideal model olarak değerlendirilmiştir. Deneysel verilerle yapılan karşılaştırma sonucunda, bu modelin kojenerasyon enerji santralinin elektrik üretim değerlerini 0,99073 R değeri ve 4.734e-8 MSE ile tahmin edebileceği sonucuna varılmıştır.

Determination of the heat capacity of cellulosic biosamples employing diverse machine learning approaches

“…Recent progress in the data analysis and computer science areas resulted in implementing statistical 40 and machine learning 41,42 techniques in various fields, mainly in vision‐based learning, 43 energy storage 44 and management, 45 and material design 46 and characterization. Moreover, it is possible to develop machine learning methods utilizing extracted/selected features to enhance the prediction accuracy 47 .…”

A comparative analysis of maximum likelihood estimation and artificial neural network modeling to assess electrical component reliability

“…ANN models, which have higher performance compared to traditional simulation methods, are often preferred 22 . It can be shown that the ANN model is effective and also applied in scientific studies conducted in different fields such as energy, 23 electronic, 24 nanotechnology, 25 construction 26 and medicine 27 …”

“…21 ANN models, which have higher performance compared to traditional simulation methods, are often preferred. 22 It can be shown that the ANN model is effective and also applied in scientific studies conducted in different fields such as energy, 23 electronic, 24 nanotechnology, 25 construction 26 and medicine. 27 Liu et al 28 suggested an ANN model for mean function and reliability estimation in the Wiener process for failure modeling and reliability estimation.…”