Application of machine learning in optimization of thermochemical conversion of waste biomass

Abstract

Renewable energy has attained a global focus due to its less carbon newlineemission and sustainable energy nature. Energy sources such as wind energy, newlinesolar energy, geothermal energy, hydrogen gas, tidal energy, biomass energy, newlineand biofuels were considered non-depleting sustainable energy sources. newlineVarious studies have been carried out on these energy sources to find the newlineefficient extraction of energy. Among these energy resources, biomass and newlinebiofuels are the instantaneous sources of energy compared with other newlinementioned sources. Biomass feedstock comprises purpose-grown crops, newlineresidues of crops, wood, algae, fatty acids, edible plant oils, and wastes from newlinesewage, and food. Choosing appropriate input feedstock and process newlineconditions from the wide range of sources of biomass claims a strong newlineknowledge. newlineArtificial intelligence and machine learning approaches were used newlinefor analysing a large collection of data irrespective of the fields. Classification newlineand regression were the two major techniques used in machine learning for newlineprocessing a large volume of data. Some of the machine learning approaches newlineshall be applied over the available biomass dataset to find the correlation newlineamong the data and help to predict the optimal yield. newlineThis study involved the development of a machine learning newlinealgorithm based Tunable Decision Support System (TDSS) and Tunable newlineRecommendation System (TRS) for optimizing the process conditions and newlineproduct yield values of hydrothermal liquefaction (HTL) and hydrothermal newlinegasification (HTG) processes individually. newline