Reliability prognostic and life assessment of wind turbine blade

Algolfat, Amna (2023) Reliability prognostic and life assessment of wind turbine blade. Doctoral thesis (PhD), Manchester Metropolitan University.

File will be available on: 9 August 2025. Available under License Creative Commons Attribution Non-commercial No Derivatives. Download (2MB)

Abstract

Wind is one of the leading renewable energy sources, especially with the increase in wind turbine efficiency through larger rotors and economies of scale. The world needs to reach net-zero emissions by 2050. However, the wind energy business has not yet reached commercial maturity, and to achieve that it is necessary to reduce maintenance costs, and the cost of energy production. The blade is considered one of the most critical components in a wind turbine, being a major contributor to downtime. Offshore wind turbine blades differ significantly from their onshore counterparts and there is a greater need to protect them from failure and breakdown. This study aims to reduce breakdowns of wind turbines by enhancing the reliability of offshore Horizontal Axis Wind Turbine (HAWT) blades by employing vibration-based methods. There are differences in the free and forced vibration results of previous literature studies. The examined studies have used different techniques and adopted different commercial software to find the dynamic properties and dynamic deflections of wind turbine blades. This study adopted most of the previous techniques to find the most comprehensive model. In addition, the code is written by the author to ensure applying the same numerical procedure of all the adopted models. The approach includes the computational-structure model (CSM) to determine relevant structural properties, the aerodynamic-loading model (ALM) to determine the exciting aerodynamic loads, the dynamic response model (DRM) to calculate the dynamic response in the flap-wise and edge-wise directions, the health-prediction model (HPM) to generate a structural health monitoring paradigm of the offshore wind turbine blade, the detection-prediction model (DPM) to detect any discrepancy of the blades’ parameters with those of the healthy structure via an energy technique, the fault diagnosis model (FDM) to identify where the mechanical damage occurred and a prognosis-prediction model (PPM) to decide the location of any damage and its severity. The different damage states are used to create a structural monitoring paradigm that can accurately predict the behaviour of the blade under operational conditions. The results show the Rayleigh theory is the most accurate where the dynamic deflection decreased by 0.327m compared with Bernoulli. Adding the angular speed directly, it affects the modal features, which impacts the dynamic response to decrease 1.13m in the flap-wise direction and 0.05m in the edge-wise direction. The results are used to demonstrate the effectiveness of using the change in the modal characteristics, modal strain energy and dynamic deflections as indices to identify the damage, its location, and its severity.