2019 CSCE Annual Conference - Laval (Greater Montreal)

2019 CSCE Annual Conference - Laval (Greater Montreal) Conference


Title
Optimal maintenance plan for a vibrating-grate biomass boiler: availability and cost saving approach

Return to Session

Author(s)
Mr. Mohammad Hosseinirahdar, Concordia University (Presenter)
Dr. Fuzhan Nasiri, Concordia University
Dr. Bruno Lee, Concordia University
Abstract

In recent years, political and economic issues motivated Canada government to make northern Canada more crowded. To hit this aim, it is vital to make these regions more energy secure since they are off-grid and consequently the long harsh winter worsens this situation. Currently, diesel generators mostly provide demanded energy through the remote areas in Canada while high capital and operational cost of system burden government and people. By brainstorming through available energy technologies, Biomass boilers can be pondered as a promising technology for such these areas particularly through areas far from source of diesel fuel. However, having the system available during the long winter plays a key role to ensure us about energy security. A mathematical optimization of the maintenance plan regarding a vibrating-grate biomass boiler, which is the case study in this paper, is carried out based on the system availability so that the maintenance cost is minimized while the system availability remains higher than initial availability. A modified failure density function is utilized to bring the maintenance effect to component reliability following the maintenance. A regression analysis is performed on the constant hazard rate to turn them into Weibull distribution parameters (?, ?) to form a time-dependent hazard functions. To examine the system availability mathematically, Fault Tree (FT) method is employed and since only series and parallel arrangement exist through the system, this method is most proper one in this case study. A flowchart of maintenance optimization is designed to optimize the system availability function with 33 unique components and results shows 32% decay in the number of maintenance tasks over 10 years of operation. The flowchart performs fast and efficient somehow that it promises to stay efficient even for more complex system.