Stochastic modeling approaches based on neural network and linear–nonlinear regression techniques for the determination of single droplet collection efficiency of countercurrent spray towers

Abstract

This paper presents a new mathematical model and a two-layer neural network approach to predict the single droplet collection efficiency (SDCE), h d , of countercurrent spray towers. SDCE values were calculated using MATLAB \ algorithm for 205 different artificial scenarios given in a large range of operating conditions. Theoretical results were compared with outputs obtained from a two-layer neural network and DataFit \ scientific software. The predicted model developed from linear-nonlinear regression analysis and network outputs agreed with the theoretical data, and all predictions proved to be satisfactory with a correlation coefficient of about 0.921 and 0.99, respectively. By using the proposed model, iterations between Reynolds number (Re), drag coefficient (C D ) and terminal velocity values (v T ) were neglected for a large range of operating conditions. SDCE values were also obtained speedily and practically for five main operating inputs used in the model.

A Prognostic Approach Based on Fuzzy-Logic Methodology to Forecast PM10 Levels in Khaldiya Residential Area, Kuwait

“…For comparative purpose, the measured data were evaluated by a multiple regression software package (DataFit® V8.1.69, Copyright© 1995, Oakdale Engineering, PA, RC167), containing 298 two-dimensional (2D) and 242 three-dimensional (3D) non-linear regression models. The regression analysis was performed based on the Levenberg-Marquardt method with double precision, as similarly done in several studies of the first author (Yetilmezsoy, 2007;Yetilmezsoy and Saral, 2007;Yetilmezsoy and Sapci-Zengin, 2009;Yetilmezsoy and Sakar, 2008;Turkdogan-Aydinol and Yetilmezsoy, 2010;Yetilmezsoy, 2011;Yetilmezsoy, 2012).…”

“…Because of their non-parametric regression capabilities, generalization properties and easiness of working with high-dimensional data, several artificial intelligence-based methods, such as artificial neural networks (Abdul-Wahab and Al-Alawi, 2002;Yetilmezsoy, 2006;Yetilmezsoy and Saral, 2007;Akkoyunlu et al, 2010) and fuzzy-logic/neurofuzzy (Nunnari et al, 2004;Yildirim and Bayramoglu, 2006;Carnevale et al, 2009;Noori et al, 2010) methodology, have recently been utilized in the modeling of various reallife problems in air pollution field. There have also been other specific studies reporting the advantages and adaptability properties of artificial intelligence-based models for the prediction of daily and/or hourly particulate matter (PM 2.5 and PM 10 ) emissions in many urban and residential areas (Chaloulakou et al, 2003;Chelani, 2005;Grivas and Chaloulakou, 2006;Karaca et al, 2009).…”

Performance evaluation and kinetic modeling of the start-up of a UASB reactor treating municipal wastewater at low temperature

“…The start-up experimental data was evaluated by DataFit Ò scientific software (version 8.1.69, Copyright Ó 1995-2005 Oakdale Engineering, RC167) containing 298 two-dimensional (2D) and 242 three-dimensional (3D) nonlinear regression models. As done in our previous studies [22,25,36,37], in this study, the non-linear regression analysis was conducted based on the Levenberg-Marquardt method with double precision.…”

Stochastic modeling applications for the prediction of COD removal efficiency of UASB reactors treating diluted real cotton textile wastewater

“…It will depend on many factors, including the complexity of the problem and the number of data points in the training set (Yetilmezsoy and Saral 2007). In general, on networks, which contain up to a few hundred weights the Levenberg-Marquardt algorithm (LMA) will have the fastest convergence.…”