MOISTURE ABSORPTION CHARACTERISTICS AND ADAPTIVE NEURO FUZZY MODELLING OF AMPELOCISSUS CAVICAULIS FIBER REINFORCED EPOXY COMPOSITE

A. J. Adeyi; O. Adeyi; A. D. Ogunsola; M. O. Fajobi; O. K. Ajayi; S. Oyelami; J. A. Otolorin

A. J. Adeyi Department of Mechanical Engineering, Ladoke Akintola University of Technology, P.M.B. 4000,Ogbomoso, Oyo State.Nigeria.
O. Adeyi Department of Chemical Engineering, Michael Okpara University of Agriculture, P.M.B 7267, Umudike Abia State Nigeria.
A. D. Ogunsola Department of Mechanical Engineering, Ladoke Akintola University of Technology, P.M.B. 4000,Ogbomoso, Oyo State.Nigeria.
M. O. Fajobi Department of Mechanical Engineering, Ladoke Akintola University of Technology, P.M.B. 4000,Ogbomoso, Oyo State.Nigeria.
O. K. Ajayi Department of Mechanical Engineering, Obafemi Awolowo Ile Ife, Osun State Nigeria
S. Oyelami Department of Mechanical Engineering, Osun State University, Oshogbo Nigeria
J. A. Otolorin Department of Chemical Engineering, Michael Okpara University of Agriculture, P.M.B 7267, Umudike Abia State Nigeria.

Abstract

Natural fibre reinforced composite is fast becoming an important class of engineering materials due to its low cost, light weight and good mechanical properties; therefore increased natural fibre composite development is desirable. In this study, the effect of water soaking time and Ampelocissus cavicaulis natural fiber (ACNF) size factors on the water absorption characteristics of ACNF reinforced epoxy composite was investigated. The optimum membership function in Adaptive Neuro Fuzzy Inference System (ANFIS) structure that modelled and predicted the observed water absorption characteristics of the developed composite was investigated by giving consideration to minimum training error. ANFIS was also utilized to evaluate the sensitivity of ACNF reinforced epoxy composite’s water absorption characteristics to water soaking time and ACNF size factors. Results showed that developed composites’ water absorption increased as water soaking time and ACNF size increased. While optimising the ANFIS structure, the training error associated with ANFIS gauss, tri, gbell, gauss2, pi and gsig memmbership functions were 0.5171, 0.08997, 0.6706, 0.08803, 1.3770 and 0.6167, respectively. The optimum ANFIS model structured (trimf) had a coefficient of determination (R2) value of 0.99947. The water absorption characteristics of ACNF reinforced epoxy composite was most dependent or sensitive to water soaking time with a root mean squared error (RMSE) of 1.577 followed by ACNF size with RMSE of 1.753. It is concluded that ACNF reinforced epoxy composite are best applied in non-moist or dry environments.