Virtual Sensors for Biodiesel Production in a Batch Reactor

Fossil fuel combustion produces around 98% of coal emissions. Therefore, liquid and gaseous biofuels have become more attractive due to their environmental benefits. The biodiesel production process requires measurements that help to control and supervise the variables involved in the process. The m...

Full description

Saved in:
Bibliographic Details
Published in:Sustainability 2017-03, Vol.9 (3), p.455
Main Authors: López-Zapata, Betty Y., Adam-Medina, Manuel, Álvarez-Gutiérrez, Peggy E., Castillo-González, Juan P., León, Héctor R. Hernández-de, Vela-Valdés, Luis G.
Format: Article
Language:English
Subjects:
Batch reactors
Biodiesel fuels
Biofuels
Diglycerides
Extended Kalman filter
Fourier transforms
Fuel combustion
Glycerol
Kalman filters
Monoglycerides
NMR
Noise
Nuclear fuels
Nuclear magnetic resonance
Sensors
Sustainability
Triglycerides
Variables
Virtual sensors
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Fossil fuel combustion produces around 98% of coal emissions. Therefore, liquid and gaseous biofuels have become more attractive due to their environmental benefits. The biodiesel production process requires measurements that help to control and supervise the variables involved in the process. The measurements provide valuable information about the operation conditions and give estimations about the critical variables of the process. The information from measurements is essential for monitoring the state of a process and verifying if it has an optimal performance. The objective of this study was the conception of a virtual sensor based on the Extended Kalman Filter (EKF) and the model of a batch biodiesel reactor for estimating concentrations of triglycerides (TG), diglycerides (DG), monoglycerides (MG), methyl ester (E), alcohol (A), and glycerol (GL) in real-time through measurement of the temperature and pH. Estimation of the TG, DG, MG, E, A, and Gl through this method eliminates the need for additional sensors and allows the use of different types of control. For the performance analysis of the virtual sensor, the data obtained from the EKF are compared with experimental data reported in the literature, with the mean square error of the estimate then being calculated. In addition, the results of this approach can be implemented in a real system, since it only uses measurements available in a reactor such as temperature and pH.
ISSN:2071-1050
2071-1050
DOI:10.3390/su9030455