Journal of Soft Computing in Civil Engineering (Jul 2020)

Prediction of Mechanical Strength Attributes of Coir/Sisal Polyester Natural Composites by ANN

  • B.S. Keerthi Gowda,
  • G.L. Easwara Prasad,
  • R. Velmurugan

DOI
https://doi.org/10.22115/scce.2020.226219.1200
Journal volume & issue
Vol. 4, no. 3
pp. 79 – 105

Abstract

Read online

Coir and Sisal are agriculture wastes that are effectively and financially accessible in the distinctive piece of Karnataka and other various states of republic India. These are generally treated as bio-compostable material by the customary horticulture/agriculture professionals. Aftereffects of past research related to fabrication, testing, analysis and design of conventional (synthetic fiber reinforced) composite materials portray that, strength to weight proportion is the basic criteria for a tailored design of composite materials. Viable utilizations of low-density reinforcing materials as the constituent materials of composites demonstrate great strength to weight ratio. Hence, 2 mm, 3 mm, 4 mm, 5 mm and 6 mm thick composite panels made up of 10 mm long coir/sisal fiber fortified in a polyester matrix of coupons are utilized for the experimentation process. The present study exhibits that the feed-forward Artificial Neural Network (ANN) model developed to predict the mechanical properties of coir/sisal polyester composite could be the acceptable mathematical tool for the prediction of mechanical properties of treated and untreated, arbitrarily oriented coir/sisal fiber strengthened polyester composite instead of the complicated experimental procedure. It exhibits that where traditional technique feels hard to estimate mechanical properties of coir/sisal fiber fortified polyester composite materials, the ANN model supports to foresee it. ANN approach avoids remembrance of equations and generalizes the problem domain and reduces the human error.

Keywords