Implementasi Support Vector Machine pada Sistem Fermentasi Produksi Bioetanol

Abstract

The biomass fermentation system is used to produce bioethanol from organic basic materials with the help of microorganisms. The bioethanol production process produced ethanol concentration with Saccharomyces Cerevisiae in this study using experimental methods by implementing a monitoring system using ESP32, cloud service data logger, pH sensor, MQ-2 sensor and temperature sensor. The test dataset is stored in the next data logger cloud service used for the process of classifying fermentation results.

The real-time data logger from the monitoring system has not been able to detect the parameters that categorize the concentration of fermentation products have been declared successful, so the user must analyze the data and carry out periodic checks to ensure the success of the fermentation process. The results of the SVM test show that the level of accuracy in the classification of bioethanol fermentation production shows satisfactory results, namely 99.79% using a linear kernel with training data of 60% and testing data of 40%, while the lowest accuracy using the RBF kernel is 95.85 with 80% training data and 20% testing data.