Please use this identifier to cite or link to this item:
https://ptsldigital.ukm.my/jspui/handle/123456789/457626Full metadata record
| DC Field | Value | Language |
|---|---|---|
| dc.contributor.author | Alvani Vida (P48326) | - |
| dc.date.accessioned | 2023-09-12T09:11:34Z | - |
| dc.date.available | 2023-09-12T09:11:34Z | - |
| dc.date.issued | 2013-10 | - |
| dc.identifier.other | ukmvital:81893 | - |
| dc.identifier.uri | https://ptsldigital.ukm.my/jspui/handle/123456789/457626 | - |
| dc.description | Biological carbon removal is a complicated process that is affected by variety of parameters, e.g. pH, Temperature, nutrient, COD/N/P ratio, organic loading rate (OLR) and many more. Tedious experiment is needed and long time is required to complete in order to get the maximum capacity of the system. Artificial neural network (ANN) is a method that can be applied for prediction of experimental results in such process. Data from an experimental operation of Biological Aerated Filters (BAF) were used to model the removal of Total Organic Carbon (TOC) at different OLR in a partially and fully packed BAF using ANN. Two identical reactors with 14 cm in diameter and 100 cm in height were used in the experiment and operated at step increased loadings of OLR 0.5 kg COD m 3 d -1 to 7.95 ± 0.40 kg COD m -3 d -1 . The influent and effluent TOC were analyzed using Matlab software (R2009b) and artificial neural network. Two neural network topologies, Feed Forward Neural Network (FFNN) and also Radial Basis Neural Network (RBFNN) were applied for the prediction of TOC removal concentrations in the effluent. A total number of 187 data, obtained from the experimental work, were used for training and confirming the model. For the first method of FFNN, a goal was set, and training of the network was stopped when the root mean square error (RMSE) on the training samples reaches to this goal. For each reactor the network was trained and tested for 1 to 10 neurons in a hidden layer. The best numbers of hidden neuron for the full bed and partial bed were 3 and 9 respectively. The best RMSE on the test set of these reactors was also 0.026 and 0.029 respectively. For the second method of FFNN a validation set was considered. The best RMSE on the test sets of these reactors were also 0.027 and 0.029 respectively. The second model is a RBFNN. The best value of spread obtained was 0.1 for the full bed and 0.5 for the partial bed. The ANN-based simulation models demonstrated accurate proximity for TOC removal and provided an efficient tool in estimating biological reactor performance.,Master | - |
| dc.language.iso | eng | - |
| dc.publisher | UKM, Bangi | - |
| dc.relation | Faculty of Engineering and Built Environment / Fakulti Kejuruteraan dan Alam Bina | - |
| dc.rights | UKM | - |
| dc.subject | Dissertations, Academic -- Malaysia | - |
| dc.subject | Biological Aerated Filters | - |
| dc.subject | Total Organic Carbon | - |
| dc.title | Modeling of carbon removal performance in partial bed and full bed configuration of the biological aerated filter utilizing artificial neural network | - |
| dc.type | theses | - |
| dc.format.pages | 116 | - |
| dc.identifier.callno | TA455.C3 A458 2013 3 | - |
| dc.identifier.barcode | 001545 | - |
| Appears in Collections: | Faculty of Engineering and Built Environment / Fakulti Kejuruteraan dan Alam Bina | |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| ukmvital_81893+SOURCE1+SOURCE1.0.PDF Restricted Access | 1.48 MB | Adobe PDF |  View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.