Nano Boron Nitride and Carbon Nanotube in Sensors: Comparative Investigation with ANN

Authors : Kübra Keser, Zafer Kaya

Pages : 248-262

Doi:10.52795/mateca.1683835

View : 97 | Download : 133

Publication Date : 2025-12-30

Article Type : Research Paper

Abstract :Carbon nanotube (CNT) and nanosized boron nitride (BN) materials are widely used support elements in sensor development due to their compatibility and structural versatility. Studies examining physical parameters in such a detailed comparison with ANN are rarely encountered in the literature. The presented study prepared sensors with different amounts of nanomaterials, water, glycerol and gelatin content. The created sensors\\\' glossiness (20°-60°), colour change (ΔE) and hardness parameters were measured. The measurement data were evaluated using statistical methods and the Multilayer Perceptron (MLP) artificial neural network model. Tansig was used as the transfer function, and the mean square error was used as the performance function. The best classification for training and test success was made in the hardness parameter with 99.9% accuracy, followed by ΔE with 94.9% accuracy. ΔE and glossiness were the most effective common structural materials for the experiments with 6 ml of water at 60° test conditions and 9 ml of water for glossiness at 20° and hardness. For ΔE, 0.12 g BN and 0.06 g CNT and for 60° gloss, 0.06 g for both BN and CNT provided the most significant effect on the changes. The experimental conditions of 0.06 g for BN and CNT were also the most influential parameters for the hardness change.
Keywords : Sensör, BN, CNT, MLP, ΔE, parlaklık

ORIGINAL ARTICLE URL