Performance Prediction of Contact Separation Mode Triboelectric nanogenerators using Machine Learning Models

Abstract

The use of Artificial Intelligence (AI) algorithms for analyzing practical data has increased with the advent of AI models. Combining physics and engineering has garnered a lot of interest so much, so that the triboelectric Nano-generators (TENG) industry may also use AI technologies. In this work, the classifiers suitable for predicting the system accuracy for TENG are analyzed. The experimental data used for training and testing, and two of the Machine Learning (ML) classifiers provided promising results: K Nearest Neighbor (KNN) and Neural Network (NN). Different ML parameters are generated such as precision, recall and F1 score with the help of Confusion matrix for KNN and NN of the practical TENG energy data. Additionally, we assess the TENG’s output quality in CS mode under various load factors using ML models.