Abdul Rahim, Adzrinna (2006) Analisis Data Untuk Rekabentuk Sistem Pintar Bagi Pengelas Corak Aliran Minyak-Gas. Project Report. Universiti Sains Malaysia, Pusat Pengajian Kejuruteraan Elektrik dan Elektronik. (Submitted)
|
PDF
Download (185kB) | Preview |
Abstract
The crude oil component fractions consists of oil, gas and water. The oil and gas industry requires an efficient technology that ensure that oil produced meet the standard requirement and the market needs. Therefore, it is vital to classify the oil according to its flow regime. Crude oil is transferred from offshore to onshore using pipes. ECT system is applied to measure the capacitance value inside the pipe. The capacitance value measurement represents the permittivity distribution of the oil flow. The capacitance data generated from the ECT system will be classified according to the flow regime. In order to enable the simulated ECT data to be classified, Artificial Neural Network (ANN) is implemented. MLP Neural network and the Levenberg Marquardt algorithm is implemented to create a desirable network. The simulated ECT data is divided into three groups namely, training data, validation data and testing data. These data will be used to train the MLP in order to get an optimum network. The best trained MLP is chosen based on its “intelligence” in classifying unseen data correctly. Two factors investigated in choosing the best network are the number of hidden neurons used and the size of training data. These two factors will determine whether the network has reached its optimum “intelligence” and has the potential to classify the oil according to its flow regime.
| Item Type: | Monograph (Project Report) |
|---|---|
| Subjects: | T Technology T Technology > TK Electrical Engineering. Electronics. Nuclear Engineering |
| Divisions: | Kampus Kejuruteraan (Engineering Campus) > Pusat Pengajian Kejuruteraaan Elektrik & Elektronik (School of Electrical & Electronic Engineering) > Monograph |
| Depositing User: | Mr Engku Shahidil Engku Ab Rahman |
| Date Deposited: | 16 May 2023 09:33 |
| Last Modified: | 16 May 2023 09:33 |
| URI: | http://eprints.usm.my/id/eprint/58551 |
Actions (login required)
 |
View Item |
