Abstract
This paper presents the comparative studies between Response Surface
methodology (RSM) and Artificial Neural Network (ANN) in phase change nanofiber diameter optimization. In this study, ultra fine fibers of Polyethylene glycol/Cellulose acetate (PEG/CA) composite (1:1, w/w) ,in which PEG acts as the PCM and CA acts as the polymer matrix, were successfully prepared and characterized by scanning electron microscopy (SEM). Two methods were compared for their modeling and optimization abilities. The average percentage error for ANN and RSM models were 4.1 and 0.91 and the coefficient of determination was 0.98 and 0.97, respectively. The results indicate that both of this two optimization models are in good agreement with experimental data. Moreover, the RSM model shows much lower average percentage error than the ANN model. Therefore, in this study, the efficiency of RSM was better than ANN in phase change nanofiber diameter optimization.
methodology (RSM) and Artificial Neural Network (ANN) in phase change nanofiber diameter optimization. In this study, ultra fine fibers of Polyethylene glycol/Cellulose acetate (PEG/CA) composite (1:1, w/w) ,in which PEG acts as the PCM and CA acts as the polymer matrix, were successfully prepared and characterized by scanning electron microscopy (SEM). Two methods were compared for their modeling and optimization abilities. The average percentage error for ANN and RSM models were 4.1 and 0.91 and the coefficient of determination was 0.98 and 0.97, respectively. The results indicate that both of this two optimization models are in good agreement with experimental data. Moreover, the RSM model shows much lower average percentage error than the ANN model. Therefore, in this study, the efficiency of RSM was better than ANN in phase change nanofiber diameter optimization.
| Original language | English |
|---|---|
| Publication date | 2017 |
| Number of pages | 10 |
| Publication status | Published - 2017 |
| Externally published | Yes |
| Event | 5th Texteh International Conference - Bucharest, Romania Duration: 18 Oct 2012 → 19 Oct 2012 |
Conference
| Conference | 5th Texteh International Conference |
|---|---|
| Country/Territory | Romania |
| City | Bucharest |
| Period | 18/10/2012 → 19/10/2012 |
Keywords
- Artificial neural network (ANN)
- Electrospinning
- Optimization
- Phase change materials (PCMs)
- Response surface methodology (RSM)