Loading...

Continues Risk Assessment by Exploiting Artificial Intelligence

Molla Mohammadi Sadafi, Amir | 2022

269 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 54863 (46)
  4. University: Sharif University of Technology
  5. Department: Energy Engineering
  6. Advisor(s): Rajabi, Abbas; Rashtchian, Davoud
  7. Abstract:
  8. The Fourth Industrial Revolution brought about a fundamental change in the system of management and production, especially in the field of industry, which was accompanied by the development of technology in the fields of artificial intelligence, robotics, nanotechnology and biotechnology. In line with these changes, the industry needs a new method to evaluate the reliability of operational equipment using artificial intelligence and intelligent systems, which aims to optimize the balance between a variety of repair methods with the possibility of timely replacement of parts. In this new method, the life of the parts is increased and the unplanned maintenance costs and labor are reduced. One of the most important achievements of these studies is to increase the reliability of continuous operation time in an operational unit. The purpose of implementing artificial intelligence (Ai) in this case study is to examine the health status of process equipment in the future.This model has been investigated on the information received from IPSORB technology of isomerization unit of Tehran Oil Refinery. The input information to the network includes the outlet temperature of the furnace before the filled beds, the operating pressure of the inlet feed pipeline to the filled beds, the vibration of the pipeline before the four main control valves, the flow rate entering the equipment, the relevant alarms and the failure list of sensitive control valves. This model should report the probability of the failure of each control valves. this model has reached 99.97% accuracy, 98.74% in F-1 score with a run time of 58 seconds per training of each epoch
  9. Keywords:
  10. Artificial Intelligence ; Artificial Neural Network ; Reliability Evaluation ; Risk Assessment ; Failure Rate ; Operational Risk ; Prognostics Failure ; Oil Refinery

 Digital Object List

 Bookmark

No TOC