Date of Publication
7-2025
Document Type
Master's Thesis
Degree Name
Master of Science in Mechanical Engineering
Subject Categories
Engineering
College
Gokongwei College of Engineering
Department/Unit
Mechanical Engineering
Thesis Advisor
Aristotle T. Ubando
Kathleen B. Aviso
Defense Panel Chair
Ivan Henderson V. Gue
Defense Panel Member
Robby B. Manrique
Raymond Girard R. Tan
Abstract/Summary
Microalgae are considered third-generation biomass feedstock that have the potential for sustainable biofuel production. With the use of the torrefaction process, the microalgal biomass can be upgraded to biochar, which is a solid biofuel. In converting the microalgal biomass to biochar as an end-product, there are significant factors considered in order to ensure that good-quality biochar is generated. These factors may range from the various microalgal species to the compositional material of the biomass, and to the operational factors of the torrefaction process. Out of these factors, it is important to classify the quality of biochar generated from microalgal biomass. To achieve this, the study proposes the development of an enhanced binary hyperbox model for the classification of microalgae biochar quality from the various factors, including ultimate composition, proximate composition, and process parameters that translate to 11 attributes. The enhanced binary hyperbox model is considered a machine learning toolbox for classifying good-quality microalgal biochar. The performance of the model based on the rules generated is evaluated based on common machine learning performance metrics such as F1-score. The best performing model is that with training hyperparameters of Δ = 0.1 and ε = 0.2 corresponding to a user-defined error margin and maximum allowable type II error value, which resulted in an F1-score of 0.8889. The optimal model found only temperature and time as relevant attributes, implying that the microalgae attributes are irrelevant and only the torrefaction process matters in determining the quality of the biochar for solid biofuel use. The rule generated can aid engineers and entrepreneurs in generating consistent quality microalgae biochar.
Abstract Format
html
Language
English
Format
Electronic
Keywords
Biomass energy; Microalgae
Recommended Citation
Millo, S. M. (2025). A hyperbox classifier model approach for the classification of microalgae for biofuel use. Retrieved from https://animorepository.dlsu.edu.ph/etdm_mecheng/32
Upload Full Text
wf_yes
Embargo Period
7-2028