No abstract

Failure mode effective analysis (FMEA) is a quality instrument that is being used to recognize potential failure modes and its related consequences on different complex sub-structures in a system. The tool is effectively used for continuous improvement of efficiency, reliability, and quality. However, the traditional FMEA methods using risk priority number (RPN) values have been criticized for having many immanent limitations, affecting its effectiveness in real-world applications. Many risk priority models are emerging in the field of safety and reliability engineering. Among these models, Multi-criteria decision making (MCDM) methods are amongst the most widespread approaches employed to rank failure modes. The main aim of the present work is to implement FMEA analysis using fuzzy MULTIMOORA method. The proposed model has the competence of representing the imprecise knowledge and uncertainties of FMEA team members. FMEA analysis has been performed on offshore wind turbines consisting of four sub-assemblies for nine potential failure modes. The failure modes and its risk factors have been well-defined and appropriate linguistic variables are assigned to the failure modes according to the knowledge of team members. The responses of the team members have been aggregated and operated by the fuzzy model which prioritizes failure modes. The comparison of traditional FMEA rankings with the proposed model displays that a more precise and rational ranking can be attained by the incorporation of MULTIMOORA method using fuzzy set theory to FMEA.

No abstract

Majority of the previous research investigations on fault diagnostics in a wind turbine gearbox are limited to binary classification, i.e., either detecting the type of defect or severities of defect. However, wind turbine gearbox consists of multiple speed stages and components, therefore performing the binary classification is not adequate. In the present study, a multi-level classification scheme which is capable of classifying the defects by stage, component, type of defect and severity level is proposed. Experiments are performed and the response is recorded through vibration, acoustic signal and lubrication oil analysis. Later, an integrated multi-variable feature set is achieved by combining the statistical features of the above mentioned individual condition monitoring strategies. Further, the obtained integrated multi-variable feature set is subjected to multi-level classification using various machine learning models and the learning model that best suits for carrying the multi-level classification is investigated. Finally, the hyperparameters of the learning models are optimized by an iterative process of reducing the objective function. It is observed that, optimized support vector machine model has yielded favorable results when compared to other machine learning models with the overall classification accuracy of 82.52 % for the four-level classification.