Analysing and predicting dielectric properties of bamboo fiber filled polypropylene composites using artificial neural network

This article investigates the efficacy of machine learning algorithms and proposes Artificial Neural Network (ANN) based model to predict the dielectric parameters of bamboo/polypropylene blends. Bamboo fibers with filler concentration of 20, 30, 40 and 50% were reinforced with polypropylene to prepare composites as per ASTM standards. Dielectric parameters such as dielectric constant, tan δ and AC conductivity were measured using impedance analyser. Extensive analysis of three activation functions such as ReLU, sigmoid and tanh is performed here. Robust regression metrics Mean Absolute Error (MAE), Mean Squared Error (MSE), Root Mean Squared Error (RMSE) and R² have been used for detailed model evaluation. Output components were predicted using fiber concentration, polymer content, frequency and temperature as input elements. Empirical results demonstrate that ANN model has potential to precisely and accurately predict dielectric properties of bamboo fiber filled Polypropylene composites that would be useful in future research.