| Paper Details: | Downloads: 310 |
| Serial Number: | P1111140861
|
| Title: | Optimization of the Pulping Process: Polynomial vs. Neural Network Models
|
| Authors: | Wan Rosli Wan Daud, Zarita Zainuddin, Ong Pauline, Amran Shafie
|
| Abstract: | Optimization of the experimental conditions for cost efficiency and time saving purposes is highly sought for an economically viable pulping process. In this paper, the integration of the polynomial model and wavelet neural networks (WNNs) in exploring the influence of the pulping variables (viz. cooking temperature and time, ethanol and sodium hydroxide concentration) on the resulting pulp and paper properties, i.e., screened yield, kappa number, tensile index and tear index during the pulping of the oil palm fronds was examined. Performance assessment signified that the WNNs demonstrated superior predictive capability than the polynomial model, where the former reproduce the experimental results with errors less than 6%, and preserve satisfactory determination coefficient.
|
| Keywords: | Organosolv, palm fronds, pulping, response surface methodology, wavelet neural networks
|
| Journal/Conference: | ICGST Conference on Computer Science and Engineering, CSE-11
|
| Volume: |
|
| Issue: |
|
| Submission Date: | 10/4/2011 12:00:00 AM
|
| Review Date: | 11/30/2011 1:30:04 AM
|
| Publishing Date: | 12/19/2011 12:00:00 AM
|
| Article Downloads: | 310
|
| Download: |
|