University of Tasmania, Australia

A stochastic model of the operation and maintenance of power generation systems.

The aim of the project is to study the effect of various operational modes, including maintenance, on the lifetime and profitability of a hydro-power generation system and other similar power systems. In order to achieve this, we will construct and analyze a suitable stochastic Markovian model. Our choice of stochastic over deterministic models is due to the element of uncertainty present in such systems, which requires methods more complex than those that deliver only answers to predetermined scenarios. There is little literature in this area and a probabilistic model on the effect of maintenance on reliability of power systems have been recently proposed by Bean, O’Reilly and Sargison. The model is based on continuous-time Markov Chains, a class of stochastic models with a structure rich enough to model complex real-life scenarios, while remaining numerically tractable at the same time. In the modeling approach, the states of the Markov Chain are used to model various modes of the system, such as inspection, maintenance, and operation. The deterioration process is modeled by introducing the deterioration level, a continuous variable which takes real values from 0 to some b>0, where 0 describes a brand new system, and b describes a system that needs to be replaced. We assume that the deterioration rate at a particular time depends on the mode of operation of the system at that time. The project will involve collaboration with the industry in order to construct a suitable model, analysis of the model, collection of data from the industry, and programming.