Esquema de detecção e diagnóstico de falhas baseado em dados para Benchmark de Turbina Eólica

Abstract

This paper investigates a new scheme for fault detection and isolation based on time series and data analysis. This scheme is applied in a wind turbine model and illustrates the power of the proposed approach in the context of renewable energy. The proposed scheme is performed in two steps and it is based on process data without using any kind of mathematical modeling. The first step, the fault detection, is based on an alternative method based on the Gibbs sampling algorithm in which the occurrence of a sensor fault is modeled as a change point detection in a time series. The second step, the fault isolation, is handled via a Fuzzy/Bayesian network scheme classifying the kind of fault. This approach presented a good performance for detection and diagnostics of sensor faults in a standard wind turbine benchmark. In addition, this work presents proposals for research extension with enhancements of the fault detection and isolation system and formulation of fault tolerant control system.