Artificial neural network approach for predicting tunneling-induced and frequency-dependent electrical impedances of conductive polymeric composites

Jang, Daeik; Kil, Taegeon; Yoon, H.N.; Seo, Joonho; Khalid, Hammad R.

doi:10.1016/j.matlet.2021.130420

Cited by 17 publications

(9 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Similar findings have been reported in the literature that the CNT-composites showed frequency-dependent behavior at lower CNT dosages, which became independent of frequency at higher CNTs content [ 33 ]. Thus, the findings derived from this study show that as the conductive network becomes denser, the variation of electrical characteristics under the application of DC and AC signals decreases, which are in close agreement with previous studies [ 32 , 33 ].…”

Section: Resultssupporting

confidence: 92%

“…For C2 and C3 sensors, the electrical impedance was found to be frequency-dependent, i.e., the electrical impedance decreased as the input frequency of the signal increased. This result can be explained by the property of capacitance in the CNT-embedded composites [ 30 , 32 ]. It is the fact that the capacitance is inversely proportional to the input frequency, and thus, the results are in close agreement with the property of capacitance reported in the previous studies [ 30 , 32 ].…”

Section: Resultsmentioning

confidence: 99%

“…This result can be explained by the property of capacitance in the CNT-embedded composites [ 30 , 32 ]. It is the fact that the capacitance is inversely proportional to the input frequency, and thus, the results are in close agreement with the property of capacitance reported in the previous studies [ 30 , 32 ]. Interestingly, the extent of the decrease in electrical impedance decreased as the CNTs content increased and became negligible for the C4 sensor.…”

Section: Resultsmentioning

confidence: 99%

See 2 more Smart Citations

Electrical Stability and Piezoresistive Sensing Performance of High Strain-Range Ultra-Stretchable CNT-Embedded Sensors

et al. 2022

Self Cite

View full text Add to dashboard Cite

Highly flexible and stretchable sensors are becoming increasingly widespread due to their versatile applicability in human/robot monitoring sensors. Conductive polymeric composites have been regarded as potential candidates for such sensors, and carbon nanotubes (CNTs) are widely used to fabricate such composites. In the present study, CNT-embedded high flexible sensors were fabricated using a facile three-roll milling method, which mitigates the drawbacks of the conventional fabrication methods. CNTs content varied between 0.5 and 4.0 wt.%, and the percolation threshold range was obtained via conductivity/resistivity values of the fabricated sensors. Following this, the electrical stability of the sensors was examined against the various DC and AC signals. Furthermore, the fabricated sensors were stretched up to 500% strain, and their sensitivity against varying strain amplitudes was investigated in terms of the change in resistance and gauge factors. Lastly, the fabricated sensors were applied to human fingers for monitoring finger bending and releasing motions to validate their potential applications. The experimental results indicated that these sensors have a percolation threshold of around 2% CNTs content, and the sensors fabricated with 2 to 4% CNTs content showed measurable resistance changes against the applied strain amplitudes of 50–500%. Among these sensors, the sensor with 2% CNTs content showed the highest sensitivity in the studied strain range, exhibiting a resistance change and gauge factor of about 90% and 1.79 against 50% strain amplitude and about 18,500% and 37.07 against 500% strain amplitude, respectively. All these sensors also showed high sensitivity for finger motion detection, showing a resistance change of between 22 and 69%.

show abstract

Section: Resultssupporting

confidence: 92%

Section: Resultsmentioning

confidence: 99%

Section: Resultsmentioning

confidence: 99%

See 1 more Smart Citation

Electrical Stability and Piezoresistive Sensing Performance of High Strain-Range Ultra-Stretchable CNT-Embedded Sensors

et al. 2022

Self Cite

View full text Add to dashboard Cite

show abstract

“…The Levenberg−Marquardt method was used in the optimization algorithm with the learning rate set to 0.01. In the ANN optimization process, the mean square error (MSE) was employed as the loss function; it is described by eqn (1): 66 where n is the number of samples, and y p i and y t i are the predicted and true value of ANN, respectively. The average relative error (ARE) was calculated by eqn (2) and used as the criterion to evaluate the model's performance in predicting AA.…”

Section: Methodsmentioning

confidence: 99%

Machine-learning-assisted molecular design of phenylnaphthylamine-type antioxidants

Wang

et al. 2022

Phys. Chem. Chem. Phys.

View full text Add to dashboard Cite

show abstract

“…Subsequently, experimental synthesis can be performed based on these predictions, thus improving the efficiency of research and development to a large extent [25,26]. Consequently, material scientists have actively applied these methods to predict various material properties, such as formation energy [27], band gap [28,29], electrical impedance [30], and fatigue life [31]. Moreover, machine learning approaches have been widely employed for predicting relative permittivity [32].…”

Section: Introductionmentioning

confidence: 99%

Machine learning enhanced prediction of permittivity of spinel microwave dielectric ceramics compared to traditional C-M calculation

Liu,

Su,

Huang

et al. 2024

Modelling Simul. Mater. Sci. Eng.

View full text Add to dashboard Cite

Microwave dielectric ceramic (MWDC) is crucial in advancing the development of 5G technology and the communication field. The prediction or calculation of its properties is of great significance for accelerating the design and development of MWDCs. Therefore, the prediction of permittivity of spinel MWDCs based on machine learning was investigated in this work. Firstly, we collected 280 single-phase spinel MWDC entries and constructed feature engineering, which includes feature generation and feature selection (five dominant features, including Mpo, Dar, Mmbe, Aose and Dgnve, were selected from 208 generated features). Next, seven commonly used algorithms were utilized during the training process of machine learning models. The eXtreme Gradient Boosting (XGBoost) model shows the best performance with R-squared (R2) of 0.9095, Mean Absolute Error (MAE) of 1.02 and Root Mean Square Error (RMSE) of 1.96. Furthermore, all the machine learning models show enhanced prediction (calculation accuracy) of the permittivity of spinel MWDCs compared to the traditional Clausius-Mossotti (C-M) equation, which can provide a guide for the design and development of spinel MWDCs applied for wireless communication.

show abstract

Artificial neural network approach for predicting tunneling-induced and frequency-dependent electrical impedances of conductive polymeric composites

Cited by 17 publications

References 15 publications

Electrical Stability and Piezoresistive Sensing Performance of High Strain-Range Ultra-Stretchable CNT-Embedded Sensors

Electrical Stability and Piezoresistive Sensing Performance of High Strain-Range Ultra-Stretchable CNT-Embedded Sensors

Machine-learning-assisted molecular design of phenylnaphthylamine-type antioxidants

Machine learning enhanced prediction of permittivity of spinel microwave dielectric ceramics compared to traditional C-M calculation

Contact Info

Product

Resources

About