Development of machine learning-based models for describing processes in a continuous solar-driven biomass gasifier

Tasneem, Shadma; Ageeli, Abeer Ali; Alamier, Waleed M.; Hasan, Nazim; Goodarzi, Marjan

doi:10.1016/j.ijhydene.2023.08.043

Cited by 5 publications

(1 citation statement)

References 58 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The use of machine learning methods is justified in predictive modeling initiatives because these approaches provide benefits that are unmatched in their ability to handle the intricate and everchanging interactions that are inherent in PTC systems. In contrast to more conventional approaches to analysis, ML algorithms are particularly effective at processing vast amounts of heterogeneous data that include a wide range of characteristics, including solar radiation, temperature outside, collector orientation, and different fluid properties [18], [19], [20], [21]. Through the use of this data, machine learning models are able to discover nuanced patterns and connections that may be beyond the comprehension of human intuition or standard statistical methods.…”

Section: Introductionmentioning

confidence: 99%

Model-prediction of Efficiency of a Parabolic Trough Collector Using Data-Driven Soft Computing

Nguyen,

Paramasivam

2024

COMIEN

View full text Add to dashboard Cite

The study highlights the need to use suitable modeling techniques to accurately predict the efficiency of parabolic trough collectors in light of their significance in renewable energy. An examination of prediction models using the Linear Regression (LR), Support Vector Regression (SVR), and Decision Tree (DT) algorithms for the efficiency of parabolic trough collectors provides insightful information about how well they work. In terms of prediction accuracy and precision, the Decision Tree model regularly performs better than its rivals throughout the training and testing phases. What sets it apart from SVR and LR models is its ability to identify minute relationships within the data. SVR performs better than LR, although it is not as exact or accurate as the DT model. Among the three models, Linear Regression has the lowest performance, underscoring its limitations in terms of capturing non-linear relationships. Given its exceptional performance, the Decision Tree model may prove to be a crucial instrument in encouraging the design and construction of solar energy systems, hence advancing the growth of sustainable development projects and renewable energy technology.

show abstract