Reliability evaluation of power systems incorporating maintenance policy with partial information

This paper proposes a partially observable semi-Markov decision process (POSMDP) based preventive maintenance model for power system reliability assessment. In this paper, we utilize a model that considers the components can fail due to both random and aging failure. Based on this model, optimal maintenance and inspection policy are performed to prevent aging failure. Moreover, to account for condition monitoring uncertainties, the inspection result gives a probabilistic assessment of component's conditions. The proposed method is applied to a modified IEEE RTS-79. Comparing with the traditional power system reliability assessment model, this one enables a more realistic power system.