Optimization of olive pomace dehydration process through the integration of computational fluid dynamics and deep learning

ABSTRACT

This study introduces an innovative approach to optimize the dehydration process of olive pomace by combining computational fluid dynamics (CFD) and deep learning. Through CFD, it identifies the optimal air inlet velocity in a prototype of a passive direct solar dehydrator for olive pomace, which allows for the reduction of its moisture content for subsequent use as biomass. The prototype was simulated in ANSYS software, and this simulation consisted of the following steps: prototype design, meshing, selection of physical models, material assignment, boundary condition simulation, and validation of results with data obtained from the prototype. Following this process, it was concluded that the optimal air inlet velocity to the dehydration chamber is 0.1 m/s. Concurrently, an artificial neural network model was used to analyze data from sensors in the physical prototype, revealing that solar radiation and ambient temperature are the most influential variables on the temperature of the dehydration chamber. This analysis resulted in a predictive model for the optimization of the dehydration process, with a correlation coefficient of 0.9699 for temp_up and 0.9710 for temp_down, and a Willmott coefficient of 0.9999, demonstrating a high concordance between the model’s predictions and the experimental data. The model’s input variables include solar radiation, ambient temperature, and both external and internal air humidity. This integration of CFD and deep learning offers a promising methodology for improving olive pomace dehydration systems and the industry in general.

KEYWORDS:

• Olive pomace
• CFD
• dryer
• solar
• deep learning

Nomenclature

CFD	=	Computacional fluid dynamics
ANN	=	Artificial neural network
t	=	Time [s]
u	=	x component of velocity [m/s]
v	=	y component of velocity [m/s]
w	=	z component of velocity [m/s]
P	=	Pressure [Pa]
U	=	Specific internal energy [J/kg]
T	=	Temperature [K]
K	=	Thermal conductivity [W/m.k]
I	=	Radiation intensity [W/m2]
a	=	Absorption coefficient [1/m]
n	=	Refractive index
$\overrightarrow{\mathit{r}}$	=	Position vector
$\overrightarrow{\mathit{s}}$	=	Director vector
$\overrightarrow{s^{\prime }}$	=	Scattering direction vector
$\mathit{s}$	=	Path length
DO	=	Discrete ordinates radiation model
V	=	Velocity [m/s]
Dh	=	Hydraulic diameter [m]
Nu	=	Nusselt number
L	=	Characteristic length [m]
h	=	Convective coefficient [W/m2.K]
g	=	Gravity [m/s2]
Gr	=	Grashof number
Re	=	Reynolds number
Pr	=	Número de Pranddtl
y	=	Temperature prediction [K]
W	=	Layer weight matrix
X	=	Input vector
b	=	Bias
MSE	=	Mean square error
R2	=	Determination coefficient
RMSE	=	Root mean square error
SV	=	Standard deviation
MLR	=	Multiple linear regression model
Temp_up	=	Temperature at the top of the prototype [K]
Temp_down	=	Temperature at the bottom of the prototype [K]
Wet_up	=	Humidity in the upper area of the prototype [%]
Wet_down	=	humidity in the lower area of the prototype [%]
Wet env	=	Ambient humidity [%]
Rad	=	Radiation [W/m2]
temp_env	=	Ambient temperature [K]
r	=	Pearson correlation coefficient
d	=	Willmott’s concordance index
i	=	Iteration number
F	=	All Feature Set
v(S)	=	Feature subset without feature i
$\left|\mathit{S}\right|$	=	Number of features in S
$\left|\mathit{F}\right|$	=	Total number of features
Letras griegas	=
ρ	=	Density [kg/m3]
$\mathit{\sigma }$	=	Stefan-Boltzmann constant [5.67×10−8 W/m2.K4]
${\sigma }_s$	=	Scattering coefficient [m−1]
$\mathrm{\Phi }$	=	Phase funtion
${\mathrm{\Omega }}^{\prime }$	=	Solid angle [Degrees]
µ	=	Dynamic viscocity [Pa.s]
β	=	Coefficient of thermal expansion [1/K]
α	=	Thermal diffusivity [m2/s]
ω	=	Kinematic viscosity [m2/s]
ϕi	=	SHAP value of characteristic i
Subíndices	=
s	=	Surface
∞	=	Ambient

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Notes on contributors

José Cabrera-Escobar

José Cabrera-Escobar is a research professor at the National University of Chimborazo, a mechanical engineer, a master’s in mathematical engineering and computing, a master in mechanical design and currently studying a doctorate in advances in materials engineering and sustainable energy.

David Vera

David Vera has a PhD in Industrial Engineering from the University of Jaén. He has worked for 3 years in the private sector (Energy lab, Vigo) on technology transfer projects to companies in the energy sector such as Naturgy, Repsol, Finsa and Inditex, among others. Regarding his research activity, he has published more than 40 indexed works and has participated as a researcher in more than 15 projects. Currently, he is Coordinator of the European Project REFFECT AFRICA funded by the European Commission with a total budget of €8.1 M. Its objective is to promote renewable resources in African countries through access to electricity and drinking water.

Francisco Jurado

Francisco Jurado (Senior Member, IEEE) was born in Linares, Jaén, Spain. He received the M.Sc. and Dr.Ing. degrees from the National University of Distance Education, Madrid, Spain, in 1995 and 1999, respectively. Since 1985, he has been a Professor with the Department of Electrical Engineering, University of Jaén, Jaén. His current research interests include power systems, modeling, and renewable energy. His research activities have been devoted to several topics, including power systems, and renewable energy. He is author of more than 380 papers in journals included in the Journal Citation Report (JCR), about 190 papers in the proceedings of International Conferences and 6 books. He has been involved in research projects funded by Spanish Ministries and European Commission. Researcher among the most influential in the world according to the classification made by Stanford University (USA).

Manolo Córdova-Suárez

Manolo Córdova-Suárez is a research professor at the National University of Chimborazo.

Gonzalo Santillán-Valdiviezo

Gonzalo Santillán-Valdiviezo is a research professor at the National University of Chimborazo

Antonio Rodríguez-Orta

Antonio Rodríguez- Orta is a doctoral student at the Universidad de Huelva, specializing in Electronic Engineering, Computer Systems and Automation.

Raúl Cabrera-Escobar

Raúl Cabrera-Escobar is a doctoral student at the University of Jaen, specializing in Renewable Energies.