Engineering strength of fiber-reinforced soil estimated by swarm intelligence optimized regression system

摘要: Fiber-reinforced soil (FRS) has been used as a promising alternative material for civil and construction engineering. Shear strength of FRS is influenced complexly by many factors including fiber properties, stress conditions. This inherent complexity limits the ability designers to assess shear parameters made it difficult establish a mathematical model accurately predicting strength. Accurately estimating vital engineers in designing geotechnical structures management. Thus, this work proposed novel computational method, namely swarm intelligence optimized regression (SIOR) system estimate peak randomly distributed FRS. The SIOR integrates machine learning technique with an enhanced algorithm obtain reliable good performance prediction process. real-world dataset collected over past 30 years was validate system. To demonstrate capability system, modeling results were compared those obtained using numeric predictive models. analytical confirm that superior baseline empirical methods via cross-fold validation hypothesis test accuracy improvement from 44.7 99.7% mean absolute percentage error. Therefore, can significantly improve facilitate