Real Time monitoring Glycerol Esterification Process with Mid IR Sensors using Support Vector Machine Classification

Abstract

The commercial synthesis of fatty acid esters of glycerol is important as it plays role in other derivative production varieties. This research aims to construct the monitoring system for faster esterification condition identification and increase efficiency of the glycerol production. The monitoring systems are based on the measurement parameters from two sources LED mid IR 3,4 µm and 5,5 µm and two detectors that connected using the data acquisition tool with computer database via USB 2.0 and classifying the esterification condition using the Support Vector Machine (SVM) classifier. The purpose of SVM method is to classify the variations of parameter inputs from the LED mid IR sensors in real time monitoring. In this research, three conditions according to esterification stages divided to monitor condition of esterification. The construction of classification based on SVM deployed in orange system software. In the application of esterification monitoring, the influence of various parameters such as temperature set in the reactor has relation to the process time needed. By monitoring this classifying system, we obtained one of the experiment in optimum process conditions in temperature 210oC is 2 hours.