Numerical and Machine Learning‐Based Triboelectric Nanogenerator Simulators: Contact‐Separation Mode DOI
Abdülkerim Okbaz, Adem Yar, Geng‐Sheng Lin

и другие.

Energy Technology, Год журнала: 2025, Номер unknown

Опубликована: Апрель 1, 2025

Triboelectric nanogenerators (TENGs) hold great potential as portable, cost‐effective, and flexible energy sources. It is essential to understand in depth how the triboelectric properties of materials operating conditions change TENG performance improve their electrical outputs. In this study, effects various material parameters on voltage, current, power outputs TENGs are numerically investigated. The surface charge density improves at all load resistances, while dielectric thickness, constant, area, separation velocity effective medium low resistances. distance, unlike these, decreases However, high it has opposite effect performance. Furthermore, a broad range data obtained from numerical simulations used train machine learning‐based simulator. This simulator based multilayer perceptron (MLP) model with an input layer nine neurons, two hidden layers, one neurons other 55 output three for predicting power. MLP model, trained using TensorFlow, demonstrates accuracy R² values over 0.99 achieves remarkably mean absolute percentage error (MAPE) 4.22%, 3.35%, 7.57% predictions, respectively.

Язык: Английский

Numerical and Machine Learning‐Based Triboelectric Nanogenerator Simulators: Contact‐Separation Mode DOI
Abdülkerim Okbaz, Adem Yar, Geng‐Sheng Lin

и другие.

Energy Technology, Год журнала: 2025, Номер unknown

Опубликована: Апрель 1, 2025

Triboelectric nanogenerators (TENGs) hold great potential as portable, cost‐effective, and flexible energy sources. It is essential to understand in depth how the triboelectric properties of materials operating conditions change TENG performance improve their electrical outputs. In this study, effects various material parameters on voltage, current, power outputs TENGs are numerically investigated. The surface charge density improves at all load resistances, while dielectric thickness, constant, area, separation velocity effective medium low resistances. distance, unlike these, decreases However, high it has opposite effect performance. Furthermore, a broad range data obtained from numerical simulations used train machine learning‐based simulator. This simulator based multilayer perceptron (MLP) model with an input layer nine neurons, two hidden layers, one neurons other 55 output three for predicting power. MLP model, trained using TensorFlow, demonstrates accuracy R² values over 0.99 achieves remarkably mean absolute percentage error (MAPE) 4.22%, 3.35%, 7.57% predictions, respectively.

Язык: Английский

Процитировано

0