Artificial intelligence model for monitoring biomass growth in semi-batch Chlorella vulgaris cultivation

Abstract

There is a great demand for a clean, economical, and long-term energy source, due to the depletion of fossil fuels. Large-scale production of microalgae biomass for biofuel production is likely attributable to several challenges, including the high cost of photobioreactors, the need for a sustainable medium for optimum development, and time-consuming algal growth monitoring techniques. Firstly, the research novelty aims at improving the strategy of recycling culture media for semi-batch cultivation of Chlorella vulgaris. Two cycles were performed with varying amounts of recycled medium replacement to evaluate algal growth and biochemical content. As compared to all other culture ratio combinations, the mixing ratio of recycled medium to fresh medium is at 40 % (40RB) combination yielded the greatest biomass growth (4.52 g/L), lipid (317.40 mg/g), protein (280.57 mg/g), and carbohydrate (451.37 mg/g) content. Next, custom vision was applied to Chlorella vulgaris maturing stages, and a unique digital architecture framework was developed. The iteration model delivers result interpretation with an accuracy of more than 92 % of every data set based on the trained Model Performance.