Fault diagnosis in the ethylene production process using an object-oriented expert system | |
| Author | Zhuang, Cheng |
| Call Number | AIT Thesis no.IE-93-18 |
| Subject(s) | Expert systems (Computer science) Ethylene plant |
| Note | A thesis submitted in partial fulfillment of the requirement for the degree of Master of Engineering, School of Engineering and Technology |
| Publisher | Asian Institute of Technology |
| Series Statement | Thesis ; no. IE-93-18 |
| Abstract | The increasing complexity of chemical production processes have caused the chemical industry to look towards automated and sn·uctured approaches for identifying and diagnosing process abnormalities during the normal course of the daily operation of a production process. One such approach is to make use of a knowledge-based expert system which can perform fault diagnosis. In this thesis, a fault diagnosis expert system, called FDES, for locating the basic (root) cause(s) of a set of abno1malities in the ethylene production process. A methodology that is used to build FDES is proposed. This methodology relates to the problem solving methodology, the diagnostic knowledge representation and the control strategy. The problem solving methodology contains diagnostic knowledge acquisition and the problem solving structure. Diagnostic knowledge is derived from shallow knowledge which relates symptoms to hypotheses, and deep knowledge which is based on the qualitative reasoning principles to capture the function and structure of the process and equipments used in the process. For the ethylene production process, the process decomposition method based on the functional and structural considerations is presented. The hierarchical problem solving structure is constructed by the method, the structure leads to the structured knowledge base. The two-level knowledge base is constructed for FDES. The diagnostic knowledge at the top-level is integrates shallow knowledge and deep knowledge while the diagnostic knowledge at the low-level is derived from deep knowledge. The object-oriented representation, with which the production rules are combined, is used for the knowledge representation. The reasoning strategy takes a hypothesis-test strategy. The control strategy which the diagnostic direction is decided at top-level is adopted. FDES has a friendly interface and is written in CLIPS. Three test cases are discussed in detail. |
| Year | 1993 |
| Corresponding Series Added Entry | Asian Institute of Technology. Thesis ; no. IE-93-18 |
| Type | Thesis |
| School | School of Engineering and Technology (SET) |
| Department | Department of Industrial Systems Engineering (DISE) |
| Academic Program/FoS | Industrial Engineering (IE) |
| Chairperson(s) | Batanov, D. N.; |
| Examination Committee(s) | Nagarnr, N. N. ;Yu-lu, Qi; |
| Scholarship Donor(s) | Asian Institute of Technology; |
| Degree | Thesis (M.Eng.) - Asian Institute of Technology |