Fiber reinforced composites have been increasingly used in the aerospace and automotive industry, due to their potential advantages for designing flexible, strong and lightweight structures. More recently they are also being considered for the manufacturing of marine propellers. Since they have the potential to lower the weight, lower the maintenance costs, increase the efficiency at off-design conditions, improve cavitation inception speed and minimize acoustic signatures. Exploiting the full potential of composite propellers however requires that they need to be cost-effective and to do so it is required that the lifetime of the blade is sufficient. To determine the lifetime of the blade it is
crucial to determine how the composite blade will respond to a wide variety of environmental and loading conditions that it will experience over its life. Basically one needs to determine what effects fatigue will have on the material properties of the blade. The main objective of this thesis is to estimate the fatigue lifetime of composite marine propellers subjected to a pressure distribution determined by a Fluid-Structure Interaction
(FSI)model with the use of a progressive damage model integrated into a Finite
Element Model (FEM). The fluid structure interaction model to determine the pressure distribution has already been created Maljaars (2019). Currently the mostly used and most reliable approach is to use an experimental approach in order to estimate the fatigue life. However, for custom designs such as marine propellers this is a long and costly process. Modelling fatigue the fatigue performance in an earlier stage of the design cycle will result in significant cost and time savings.

The expected fatigue lifetime of the composite marine propeller is concluded to be at least 1012 cycles under the considered loading condition. If the propeller would rotate at 600 rpm for 24 hours per day this would translate to over 3000 years. The reason behind this large number is that the combination of the applied pressure load and the strength of the carbon fiber propeller is such that the stresses in each ply are very low. These low stresses create almost no damage throughout the propeller blade even for ultra high cycles.

Fiber-reinforced composite (FRC) marine propellers potentially outperform metallic propellers in terms of efficiency and underwater radiated noise (URN) by hydro-elastic tailoring of the blades. Several methods can assess the extent of these potentials. Research shows that embedded sensing methods can be used in dynamic measurements of composites. This thesis studies a full-scale application of a network of embedded piezoelectric sensors in an FRC marine propeller blade. The study prefers using piezoelectric sensors because of their ability to operate in a relatively wide frequency range. The focus of the thesis starts with designing the full-scale network of embedded piezoelectric sensors. Since no literature includes this application on FRC blades, this study holds a pioneering role in embedding piezoelectric sensors in an FRC marine propeller blade. Detailed analysis of material dimensions - including sensors, wiring, and fiber plies - leads to a successful sensor network design. Considerations regarding the location of 24 sensors included both the in-plane and the in-depth position within the FRC laminate. Fabrication of an FRC blade has been done using a resin transfer moulding (RTM) process. For the first time, an FRC marine propeller blade is embedded with piezoelectric sensors. Demoulding of the blade caused damage to some of the sensor wires. An amount of 54% of the embedded sensors survived the process with full connectivity. The performance of the intact sensors after fabrication is assessed. These sensors are exposed to free vibration tests of the FRC blade. An excitation is imposed on the blade with an impact hammer. A data acquisition (DAQ) system is used to capture the responses of the embedded piezo-sensors. The frequency response functions (FRFs) of multiple locations on the blade are computed. These FRFs provide more insight into the dynamic behavior of the blade. A frequency range of 1-1000Hz is used in the modal analysis. The first five natural frequencies are found between 240Hz and 840Hz. Natural frequencies measured by the embedded piezo-sensors and surface-mounted strain gauges differ up until 25% from natural frequencies computed by a finite element model (FEM) of the blade. The mode shape of the blade at the natural frequencies is computed for by the FEM and embedded piezo-sensors. Some difference in mode shapes is demonstrated between measurements computed by FEM and those measured by the embedded piezo-sensors and surface-mounted strain gauges. The piezo-sensors and strain gauges are in agreement regarding the measured natural frequencies. Therefore, it is expected that discrepancies exist between the physical blade and FEM. Several points for improvement of the results have been found. The study provides the first-time feasibility of dynamic measurements from embedded piezo-sensors in an FRC marine propeller blade. Additionally, a framework for reconstructing the full-field vibration response is provided, which can provide more accurate results when the agreement between piezo-sensor and FEM measurements has improved.