Loading...

Development of an Intelligent Decision Support System to Operate Gas Transmission Network

Khadem, Ahmad | 2015

512 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 46815 (06)
  4. University: Sharif University of Technology
  5. Department: Chemical and Petroleum Engineering
  6. Advisor(s): Bozorgmehry Boozarjomehry, Ramin
  7. Abstract:
  8. In plant-wide systems, faults can easily propagate due to the interconnection of streams. The main objective of this paper is to develop a systematic framework for online decision support system (DSS) in order to make the right decision to get the process out of critical conditions (which cannot be handled by the plant controllers) smoothly. One of the key features of the proposed approach is its lack of dependence on prior knowledge of the fault signals (e.g. number of faults, and their origin). In the proposed scheme, the plant is modeled by a fuzzy directed graph (FDG). The proposed scheme utilizes a backward reasoning based on the deviations that exists in the process variables (attributes) of target nodes, in order to detect the most crucial node (equipment) whose manipulating variables can be changed appropriately (i.e., upward or downward) to get the process out of the abnormal condition gracefully. Quantitative decisions are made based on qualitative decisions. Thereafter, forward reasoning is used to determine the impact of each quantitative decision on the attributes of target nodes. The attributes of the affected nodes are obtained based on their estimations by Fuzzy Inference System (FIS) through traversing from affecting nodes to other nodes in a sequential manner. To keep the generality and flexibility of the proposed scheme, FIS uses preliminary intuitive rules between two connected nodes rather than empirical rules for the entire plant. Based on this procedure, all tentative feasible decisions are built. These feasible decisions get ranked and prioritized through Analytic Hierarchy Process (AHP) method. The decision set with highest priority is then implemented whose effect on the process operation is assessed and if it seems the process still requires some changes to get out of abnormal situation, the proposed scheme is repeated based on the newly established condition and the appropriate set of decision is obtained accordingly The performance of the proposed scheme is shown based on its application for a gas transmission network as a benchmark. The proposed scheme is not only capable of offering decisions to alleviate abnormal condition, but also capable of leading the plant to a new desired state from a normal condition
  9. Keywords:
  10. Decision Making Support System ; Analytical Hierarchy Process (AHP) ; Gas Transmission Network ; Forward Reasoning ; Fuzzy Directed Graph (FDG)

 Digital Object List

 Bookmark

No TOC