Loading...

Review of Network Reduction Techniques and Their Applications in Transmission Network Fault Diagnosis

Lotfi, Mehdi | 2019

537 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 51650 (06)
  4. University: Sharif University of Technology
  5. Department: Chemical and Petroleum Engineering
  6. Advisor(s): Bozorgmehry Boozarjomehry, Ramin
  7. Abstract:
  8. Monitoring and control of gas transmission networks are very challenging and important, this is due to the fact that interactions between various elements in the transmission network severely affect network’s overall performance. This makes it crucial for the monitoring and control systems to take into account the topology of the network, in their decision making and control scenario’s. On the other hand, in the event of any abnormality in the network behavior, , in order to accurately and quickly detect the cause of the abnormal conditions, there is a need for rapid and effective analysis of the network based on the measurements made therein and to identify the key and effective situations throughout the network. The rapid and online analysis of the network based according to the rigorous model of the network that contains all pieces of equipment and their corresponding details is extremely time consuming and computationally demanding (if not impossible). The purpose of this project is to reduce the whole transmission network and come up with an equivalent reduced network that has similar input-output behavior and can mimic the necessary abnormal situations of the original transmission network. This approach makes it possible to have a reduced order model by which the prediction of the network performance in a short time span right after sensing the abnormalities would be possible. If this prediction is accurate enough, due to its fast (yet accurate enough) computations, one can use its results to detect the fault in the transmission network. In this project, first, an algorithm is proposed to remove the less significant (those which does not have significant effect on the input-output behavior of the network), and then approximates the remaining pipes with a set of tanks with input and output valves and converting the pipeline distributed system to a limited number of lumped systems. By doing so, the dynamical behavior of the system can be rebuilt more quickly, and the reduced order model can be used for control, fault detection and other applications which requires a fast and yet similar input-output behavior of the network. Simulation results for various benchmarks show that the proposed approach is quite capable of the prediction of input-output behavior of the transmission network enough accuracy with a computational times atleast 10 times less than it corresponding value with the original model
  9. Keywords:
  10. Order Reduction ; Model Reduction ; Fault Diagnosis ; Gas Transmission Network ; Interconnected System ; Model Order Reduction

 Digital Object List

 Bookmark

No TOC