Markov State Model for Optimization of Maintenance and Renewal of Hydro Power Components

Abstract

In this paper a reliability model is presented which can be used for scheduling and optimization of maintenance and renewal. The deterioration process of technical equipment is modeled by a Markov chain. A framework is proposed how the parameters in the Markov process can be estimated based on a description of the technical condition of components and systems in hydro power plants according to the Norwegian Electricity Industry Association. A time dependent solution of the Markov model is presented. Imperfect periodic inspection can be modeled by the proposed approach. The length of the inspection interval depends on the system condition revealed by the previous inspection. The model can be used to compute performance measures and operational costs over a finite time horizon. Finally, simulation results for a dataset for a Norwegian hydro power plant are presented