Markovian jump system approach for the estimation and adaptive diagnosis of decreased power generation in wind farms

Abstract

In this work, we propose a Markovian jump model of the power generation system of a wind turbine and we present a closed-loop model-based observer to estimate the faults related to energy losses. The observer is designed through an H∞-based optimization problem that optimally fixes the trade-off between the observer fault sensitivity and robustness. The fault estimates are then used in data-based decision mechanisms for achieving fault detection and isolation. The performance of the strategy is then ameliorated in a wind farm level scheme that uses a bank of the aforementioned observers and decision mechanisms. Finally, the proposed approach is tested using a well-known benchmark in the context of wind farm fault diagnosis.

Identification of discrete time systems with random jump parameters and incomplete information

“…Estimation and identification problems are relevant for different systems. As an example of such systems, one can consider, for example, economic systems [1,2], energy systems [3,4], flight systems [5], communication systems [6,7]. Such problems occupy a special place in the problem of fault detection [8][9][10].…”