Indexed In Scopus
  Scopus ID: 21100926589

Modeling for Predicting the Mechanical Properties of Calcium Carbonate Filled Coir-silk Squash Hybrid Composites Using Response Surface Methodology and Artificial Neural Network

Mohana Krishnudu Doni, Venkateshwar Reddy Pathapalli and Venkata Saikumar Reddy Ravipati

Abstract

Presently, there is a need to switch for more sustainable and renewable materials because of its low weight, economic, eco-friendly, biodegradable and easily recyclable. The samples of coir, silk squash hybrid composites filled with calcium carbonate (CaCO3) powder have been prepared using hand lay-up technique and the mechanical properties of the prepared hybrid composite material are determined. The aim of the present work is to evolve the knowledge-based system to estimate the mechanical properties (tensile strength, flexural modulus, and impact strength) of these hybrid composites. Mechanical properties like tensile strength, flexural modulus, and impact strength have been evaluated by varying the input parameters. The morphological properties of the polymer composites were studied in the present study. Feed forward neural networks were successfully trained and the model is validated by testing it with experimental data. Also, regression equations are developed for tensile, flexural and impact strengths in terms of the input variables using Response Surface Methodology (RSM). The prediction accuracy of both RSM and Artificial Neural Network (ANN) models is good and the three statistics namely R2 , average absolute deviation error, and mean absolute error calculated for both the models suggest that ANN modelling has better predictability.

Published on: November 27, 2023
doi: 10.17756/nwj.2023-s4-028
Citation: Doni MK, Pathapalli VR, Ravipati VSR. 2023. Modeling for Predicting the Mechanical Properties of Calcium Carbonate Filled Coir-silk Squash Hybrid Composites Using Response Surface Methodology and Artificial Neural Network. NanoWorld J 9(S4): S161-S166.

Download Citation (XML)

26 Downloads
Bitnami