Optimization of condition-based maintenance in wind energy systems using numerical approach

• Author / Creator

  Zuo, Yunqiu

• Effective maintenance strategies play a critical role in ensuring the operational integrity of wind farms while concurrently reducing costs. Among a variety of maintenance strategies, condition-based maintenance (CBM) stands out as an effective approach nowadays. This is an approach to maintenance that prioritizes actions based on the observed condition of an asset or equipment, rather than following fixed schedules or time-based intervals. It relies on continuous monitoring, data analysis, and predictive algorithms to identify potential issues or failures before they occur. Through optimization of CBM policies, operating and maintenance (O&M) costs can be minimized, while enhancing the reliability of wind turbines and components. This in turn brings significant benefits for the wind industry, fostering its sustainability and competitiveness.
  In existing studies, simulation methods were commonly used for CBM policy cost evaluations. The simulation methods demonstrated flexibility in modeling diverse scenarios and factors. However, due to its reliance on sampling, variations occur in the evaluation of CBM costs, leading to non-smooth surfaces in the cost function. This sampling characteristic may lead to issues such as local minima and convergence, thereby complicating the optimization process. A numerical method was originally proposed in a previous study to evaluate the maintenance cost rate of wind farm systems, aiming to address limitations identified in the simulation methods mentioned above. However, it had issues in obtaining reasonable cost evaluation and optimization results.
  
  This thesis develops a modified numerical method based on previous studies for more accurate wind farm CBM cost evaluation. The CBM policy used in this study is the same as those used in the referred studies in the literature. Compared to the existing numerical method, the proposed modified numerical method addresses the issues and achieves more accurate results mainly by introducing a new 4-dimensional structure for the component age combination probability matrix, and integrating a previously separate age combination probability transition matrix into the main cost evaluation loop. With the modified numerical method, we are able to find the optimal CBM policy, which were not possible before. The results are verified using multiple examples. The cost surface generated by of the numerical method exhibits greater smoothness and cost estimation outcomes are stable, which supporting the optimization process. Thus, the optimal CBM policy corresponding to the minimum maintenance cost rate can be estimated more accurately compared to the simulation method. Sensitivity analysis and comparative studies with the simulation method were conducted to emphasize the advantages of the numerical method. This thesis provides basis for further advancing numerical methods for wind farm CBM policy evaluation and optimization.

• Subjects / Keywords

  • Condition based maintenance optimization
  • Operating and maintenance
  • Policy evaluation
  • Maintenance strategies

• Graduation date

  Fall 2024

• Type of Item

  Thesis

• Degree

  Master of Science

• DOI

  https://doi.org/10.7939/r3-9j8a-hs37

• License

  This thesis is made available by the University of Alberta Library with permission of the copyright owner solely for non-commercial purposes. This thesis, or any portion thereof, may not otherwise be copied or reproduced without the written consent of the copyright owner, except to the extent permitted by Canadian copyright law.