Choosing the optimal sequences of switching for a repairable system with passive redundancy by Petri networks coupled with the Monte-Carlo simulation

Abstract Working on this research, we suggest studying a system composed by: three generator sets, a control unit and an inverter system that ensures switching, in case that the generator in use failed. Devices of this system have a potential for faults. After a malfunction, a restorative process is immediately starting. When a failure occurs on a machine, this last one just stopped. Therefore, the control unit is activated and trigs automatically the other operational generator set, through the inverter system which provides the changeover. This solution is considered to increase the system's productivity. The purpose of this research is to submit an ultimate performance method. This approach will provide us to determine the switching sequences of a specific configuration, thus guaranteeing a cost and availability optimization. Models shown here all differ by assumptions considered: by systems where machines have different costs, and by the switching command of a given configuration. By applying Petri networks coupled with the Monte-Carlo simulation on this system of diversified technical characteristics, we are going to determine optimal configurations. Depending on the switching sequences command of subsystems. But also, main-tenance actions on the performances of an operational heterogeneous system with passive redundancy. A special thanks to the company TOTAL who put at our disposal its GRIF software platform developed, which enabled us to determine the fundamental indicators of operational safety