Reliability evaluation of generation system with multi-state generating units using Monte Carlo simulation technique

Abstract

The reliability assessment of generation systems nowadays has drawn prime interest in the area of power system planning, designing and evaluation of reliability. As the demand of the modem society is to have electrical energy as economically as possi15le with a specified degree of reliability, therefore, there is a significant need to assess the reliability of generation systems. There are two basic techniques available in the literature for reliability evaluation of generation systems: the analytical approach and the simulation approach. Most of the techniques available in the literature for assessing the reliability of generation systems are based on analytical methods. The simulation approach estimates reliability indices by simulating the actual behaviour of the system. A new methodology has been developed in this thesis for inclusion of multi-state failures of generating units in sequential Monte Carlo simulation procedures. The developed approach for accommodating multi-state failures is first applied to solve the state space diagram of such failures and the results are compared with those obtained by using analytical technique. The results show very close comparison. Finally the new algorithm is included in sequential Monte Carlo simulation procedures and applied to evaluate the IEEE-RTS with multi-state generating units. The results are also compared with that obtained by using analytical techniques. They compare favourably indicating the effectiveness of the new approach.