Mimo nonlinear process control using neuro-fuzzy system | |
| Author | Nguyen Quoc Dinh |
| Call Number | AIT Diss. no.ISE-07-06 |
| Subject(s) | Tiles Kilns |
| Note | A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Engineering in Industrial System Engineering, School of Engineering and Technology |
| Publisher | Asian Institute of Technology |
| Series Statement | Dissertation ; no. ISE-07-06 |
| Abstract | In ceramic tile manufacturing line, roller kiln plays a very important role because performance of firing process in roller kiln decides quality as well as price of final products. If the temperature values at various areas in roller kiln do not track the reference values very well, many defects will occur in finished wall or floor ceramic tiles. Therefore, it is essential to improve the control for temperature process in roller kiln. This requires good controller design to make temperature process control in roller kiln as accurate as possible. Actually, this study comprises of two separate phases: modeling and control. In the first one, we conducted a comparative study using Artificial Neural Networks (ANNs) and Co-Active Neuro-Fuzzy Inference System (CANFIS) in modeling a real, complicated Multi¬Input-Multi-Output (MIMO) nonlinear temperature process of roller kiln. The input-output technical data used for training models was collected from a real industrial roller kiln. Using this study, we proved that CANFIS is better suited for modeling this temperature process. In control phase, a neural network (NN) controller has been constructed to control the above-mentioned temperature process using a feedback control diagram. The designed controller performance is then tested by a Visual C++ project and the resulting numerical data showed that error between real temperature values and set-point ones at various points in roller kiln, for three cases of set-point temperature set, is reduced and acceptable when compared with the typical error in literature. Therefore, this controller can work accurately and reliably when the roller kiln set-point temperature set changes. Because of the good performance of the designed NN controller, many defects occurring in the finished tiles such as cracks and fractures, black heart, tonality variations of the glaze, size variations, planarity defects will be minimized. |
| Year | 2007 |
| Corresponding Series Added Entry | Asian Institute of Technology. Dissertation ; no. ISE-07-06 |
| Type | Dissertation |
| School | School of Engineering and Technology (SET) |
| Department | Department of Industrial Systems Engineering (DISE) |
| Academic Program/FoS | Industrial Systems Engineering (ISE) |
| Chairperson(s) | Afzulpurkar, Nitin V.; |
| Examination Committee(s) | Manukid Parnichkun ;Phan Minh Dung; |
| Scholarship Donor(s) | Ministry of Education and Training (MOET) Vietnam; |
| Degree | Thesis (Ph.D.) - Asian Institute of Technology, 2007 |