Abstract
The presented dataset accurately records biomass generation and readings of environmental variables that directly affect the growth of Verrucodesmus verrucosus crops in bubble column photobioreactors. The information was obtained through a monitoring system based on sensors and manual quantification of dry weight, which allowed the acquisition of physicochemical parameters such as irradiance, temperature (°C), nitrate concentration (NO₃), dissolved oxygen (DO), oxygen gas (O₂ gas), carbon dioxide gas (CO₂ gas), and electrical conductivity. This dataset is essential for creating and validating kinetic models and machine learning algorithms crucial for biomass estimation. It contains 1080 observations, serving as a vital resource for examining growth kinetics and creating highly accurate predictive models that aid in optimizing microalgae culture systems. Advanced preprocessing methods were employed to maintain data integrity and quality, such as outlier detection, data interpolation, and key feature extraction. The combination of different inference models, including multiple linear regression, second and third-degree polynomial models, and Gaussian regression, demonstrates the feasibility of predictive strategies for biomass quantification. In addition to its relevance for developing prediction models, this dataset facilitates reproducibility in studies on microalgae biomass. It drives the advancement of bioprocess engineering and environmental monitoring by generating high-quality experimental information.