Scholarlink

<p>Wind turbine drivetrains are exposed to fluctuating torque, turbulent inflow, cyclic loading, contamination, and thermal and alignment disturbances, making them vulnerable to interacting gear, bearing, lubrication, and interface failures. This study develops an uncertainty-aware Failure Mode and Effects Analysis–multi-criteria decision-making framework for prioritizing 20 drivetrain failure modes. Three domain experts assessed Severity, Occurrence, Detectability, and Cost/Criticality through a two-round elicitation process. Triangular fuzzy numbers represent linguistic uncertainty; Z-numbers add assessment reliability; and a proposed consensus-adjusted Z-number representation, abbreviated as a ZE-number in this study, attenuates reliability when expert agreement is weak. Criteria weights were estimated using Fuzzy SWARA, Z-SWARA, and ZE-SWARA, and failure modes were ranked using an extended Risk Priority Number baseline, Fuzzy ARAS, Z-ARAS, and ZE-ARAS. Lubricant particle contamination (<InlineEquation ID="IEq1"><EquationSource Format="TEX">\(\:{\text{F}\text{M}}_{17}\)</EquationSource></InlineEquation>) ranked first in every method. Catastrophic gear-tooth fracture (<InlineEquation ID="IEq2"><EquationSource Format="TEX">\(\:{\text{F}\text{M}}_{4}\)</EquationSource></InlineEquation>), tapered-roller-bearing spalling (<InlineEquation ID="IEq3"><EquationSource Format="TEX">\(\:{\text{F}\text{M}}_{9}\)</EquationSource></InlineEquation> and <InlineEquation ID="IEq4"><EquationSource Format="TEX">\(\:{\text{F}\text{M}}_{10}\)</EquationSource></InlineEquation>), misalignment (<InlineEquation ID="IEq5"><EquationSource Format="TEX">\(\:{\text{F}\text{M}}_{6}\)</EquationSource></InlineEquation>), and conical-interface slip (<InlineEquation ID="IEq6"><EquationSource Format="TEX">\(\:{\text{F}\text{M}}_{19}\)</EquationSource></InlineEquation>) formed the dominant upper-risk group. Pairwise Spearman rank correlations ranged from <InlineEquation ID="IEq7"><EquationSource Format="TEX">\(\:0.838-0.955\)</EquationSource></InlineEquation>, and Kendall’s coefficient of concordance was <InlineEquation ID="IEq8"><EquationSource Format="TEX">\(\:0.936\)</EquationSource></InlineEquation>, indicating strong internal agreement. A <InlineEquation ID="IEq9"><EquationSource Format="TEX">\(\:\pm\:20\%\)</EquationSource></InlineEquation> one-at-a-time weight sensitivity analysis retained <InlineEquation ID="IEq10"><EquationSource Format="TEX">\(\:{\text{F}\text{M}}_{17}\)</EquationSource></InlineEquation> as the highest-priority mode in all scenarios and limited the maximum rank shift to three positions. The framework supports reliability-centered maintenance, condition-based monitoring, renewable-energy asset management, and risk-informed inspection. Its practical and socio-economic value lies in focusing maintenance resources on failure mechanisms that drive downtime, repair logistics, energy losses, and safety exposure.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Uncertainty-aware FMEA-MCDM prioritization of wind turbine drivetrain failure modes

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年7月29日

查看全文
  • Ali Paşaoğlu,
  • Ali Ashkani,
  • Elaheh Yaghoubi,
  • Elnaz Yaghoubi,
  • Raheleh Ghadami,
  • Saeid Jafarzadeh Ghoushchi,
  • Mohamed Mazlan

摘要

Wind turbine drivetrains are exposed to fluctuating torque, turbulent inflow, cyclic loading, contamination, and thermal and alignment disturbances, making them vulnerable to interacting gear, bearing, lubrication, and interface failures. This study develops an uncertainty-aware Failure Mode and Effects Analysis–multi-criteria decision-making framework for prioritizing 20 drivetrain failure modes. Three domain experts assessed Severity, Occurrence, Detectability, and Cost/Criticality through a two-round elicitation process. Triangular fuzzy numbers represent linguistic uncertainty; Z-numbers add assessment reliability; and a proposed consensus-adjusted Z-number representation, abbreviated as a ZE-number in this study, attenuates reliability when expert agreement is weak. Criteria weights were estimated using Fuzzy SWARA, Z-SWARA, and ZE-SWARA, and failure modes were ranked using an extended Risk Priority Number baseline, Fuzzy ARAS, Z-ARAS, and ZE-ARAS. Lubricant particle contamination (\(\:{\text{F}\text{M}}_{17}\)) ranked first in every method. Catastrophic gear-tooth fracture (\(\:{\text{F}\text{M}}_{4}\)), tapered-roller-bearing spalling (\(\:{\text{F}\text{M}}_{9}\) and \(\:{\text{F}\text{M}}_{10}\)), misalignment (\(\:{\text{F}\text{M}}_{6}\)), and conical-interface slip (\(\:{\text{F}\text{M}}_{19}\)) formed the dominant upper-risk group. Pairwise Spearman rank correlations ranged from \(\:0.838-0.955\), and Kendall’s coefficient of concordance was \(\:0.936\), indicating strong internal agreement. A \(\:\pm\:20\%\) one-at-a-time weight sensitivity analysis retained \(\:{\text{F}\text{M}}_{17}\) as the highest-priority mode in all scenarios and limited the maximum rank shift to three positions. The framework supports reliability-centered maintenance, condition-based monitoring, renewable-energy asset management, and risk-informed inspection. Its practical and socio-economic value lies in focusing maintenance resources on failure mechanisms that drive downtime, repair logistics, energy losses, and safety exposure.