Prediction of Moisture Content of Chlorella vulgaris Microalgae Using Hybrid Evolutionary Computing and Neural Network Variants for Biofuel Production

Aquino, Heinrick; Concepcion, Ronnie; Mayol, Andres Philip; Culaba, Alvin B.; Cuello, Joel L.; Dadios, Elmer P.; Ubando, Aristotle T.; Juan, Jayne Lois G. San

doi:10.1109/hnicem54116.2021.9731926

Cited by 5 publications

(1 citation statement)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The processes of the data-driven model in the form of symbols can be recapitulated as follows [44]: GPTIPS initiates with a population of MGGP from randomly generated vector trees. The fitness of various solutions is evaluated, and then regular tournament selection is carried out based on a probabilistic Pareto tournament [45]. The regular tournament develops by repeating mutation and crossover up to a pre-set truncation criterion.…”

Section: Data Collection and Machine Learning Algorithmmentioning

confidence: 99%

Predicting the Degree of Reaction of Supplementary Cementitious Materials in Hydrated Portland Cement

Degefa,

Park,

Yang

et al. 2023

Sustainability

View full text Add to dashboard Cite

Determination and prediction of degree of reaction (DOR) of supplementary cementitious materials (SCMs) in hydrated Portland cement are important for designing concrete with lower levels of embodied carbon dioxide. Herein, a model for predicting the DOR of SCMs in hydrated cement was developed using a set of collected data and a machine learning algorithm based on genetic programming toolbox for the identification of physical systems. The results suggest that the model reliably predicts the DOR of slag, fly ash, metakaolin, and silica fume with a coefficient of determination (R2) value of 0.89. The predicted DOR of SCMs is found to be directly proportional to water-to-cement ratio and curing time, while it is highly reliant on the oxide composition and differs amongst SCMs. For instance, the DOR of slag substantially increased with a higher alumina content, while the DOR of metakaolin remained more stable, primarily influenced by the silica-to-alumina ratio. The proposed model is particularly useful for predicting phase assemblages of SCMs-blended Portland cement when experimental data and information on SCMs are limited and properties of SCMs are highly variable. The insights gained from this study offer a pathway towards more sustainable and efficient concrete design, aligning with contemporary environmental objectives.

show abstract

Section: Data Collection and Machine Learning Algorithmmentioning

confidence: 99%