Bu-Ali Sina University Journal of Stress Analysis 2588-2597 7 1 2022 09 01 Parametric Investigation of Carbon Nanotube-Based Nanomechanical Mass Sensors using Structural Mechanics and an Artificial Neural Network Approach 5211 10.22084/jrstan.2023.26558.1216 EN Z. Naghibi Department of Computer Engineering, Hamedan University of Technology, Hamedan, Iran. J. Payandehpeyman Department of Mechanical Engineering, Hamedan University of Technology, Hamedan, Iran. K. Moradi Department of Mechanical Engineering, Hamedan University of Technology, Hamedan, Iran. Journal Article 2022 07 07  The use of single-walled carbon nanotubes (CNTs) as mechanical sensors to<br />detect tiny objects has dramatically expanded in the last decade. In this article, the parameters affecting the efficiency of sensors, including the diameter of<br />single-walled carbon nanotubes (SWCNTs), the length of SWCNTs, SWCNT<br />chirality, applied strain, and added mass, were investigated. At first, the<br />effects of the desired parameters were investigated using structural mechanics.<br />Then, an artificial neural network (ANN) was trained to predict the sensor<br />behavior in other design points. After the training phase, the ANN-based<br />model provided an accurate macro-model of a sensor. The results showed<br />that the nanotube-based sensor could detect a mass of even 10 zeptograms<br />(1zg=10<em>−</em>21g) and that the applied axial strain significantly increased the<br />efficiency of the sensor. According to the results, the ANN-based model can<br />model the dynamic behavior of this type of sensor with significant accuracy.<br />Moreover, the ANN-based model is 104 orders of magnitude faster than the<br />existing models in structural mechanics.