*2.4. Flow Analyses*

In wind turbines, flow energy is converted into mechanical energy, and mechanical energy is converted into electrical energy in the generator. During this transformation, the effect of air as a fluid on the blades and the aerodynamic forces resulting from this action are basically based on pressure distributions. Flow analysis is required to obtain these pressure distributions, and these analyses can be done with CFD software [17,42,43]. In this study, MSC CRADLE, one of the MSC One software products, was used to obtain pressure distributions. In the numerical flow analysis, the flow area was firstly created according to the blade length. The same mesh structure was applied to the blade and the created flow area as seen in Figure 6 in each analysis. The hexagonal network structure formed in the blade and flow area was determined as a total of 922,673 octant (number of elements). In Figure 7, it is seen that while the mesh structure is denser in the blade, it is evenly distributed over the flow area in order to measure the changes in the blade more precisely. The angles of attack ( α) were selected as 5◦ and 10◦ in the analyses. Pressure distributions were obtained by giving 5 m/s, 10 m/s, 15 m/s and 20 m/s fluid velocities at each angle of attack. The pressure distribution data obtained from the CRADLE program were applied to the blade beam models in order to determine the behavior of the blades under different wind speeds and different angles of attack and also to simulate the structurefluid interaction.

**Figure 6.** The mesh structure (**a**) the flow domain (**b**) the meshed blade and (**c**) length scale of the meshed blades.

**Figure 7.** Pressure distribution on varying wind speeds and angles of attack: (**a**) α = 5◦ v = 5 m/s (**b**) α = 10◦ v = 5 m/s (**c**) α = 5◦ v = 10 m/s (**d**) α = 10◦ v = 10 m/s (**e**) α = 5◦ v = 15 m/s (**f**) α = 10◦ v = 15 m/s (**g**) α = 5◦ v = 20 m/s (**h**) α = 10◦ v = 20 m/s.

#### **3. Results and Discussion**

The results obtained in this study include (1) the data results of the experimental tests of the blades with the DIC system, (2) the results of the flow analyses, and (3) the behavior of the blades modeled in the finite element method.

#### *3.1. Experimental Results Obtained from the DIC System*

The strengths of both commercial and new generation blades under static load (see Figure 5) were compared using the DIC system. Table 1 has been created to better understand this comparison. In Table 1, the displacement and surface strain values of both the commercial and new generation blades are shown corresponding to each test case. When these displacement and strain values are compared, it is revealed that the new generation blade has higher stiffness than a commercial one. For example, after applying a single load

to the tip region of both blades, the displacement values along the longitudinal axis of commercial and new generation blades vary between 5 mm–60 mm and 0.8 mm–30.3 mm, respectively. Since this example shows that the commercial blade makes about two-times more displacement than the new generation blade under the same load, it turns out that the stiffness value (stiffness = force/displacement) of the commercial blade is approximately equal to half of the stiffness of the other blade. In addition, if the strain values formed on the surfaces of the blades are to be compared with each other, the average strain values occurring in the root region of commercial and new generation blades are shown as approximately 0.0070 and 0.0029, respectively (see Table 1). The average strain values in the middle region of commercial and new generation blades are also shown as approximately 0.0147 and 0.0059, respectively. Finally, the average strain values in the tip region of commercial and new generation blades are approximately 0.0247 and 0.0129, respectively.

Table 1 also includes the displacement data for commercial and new generation blades after applying a single load to the middle and root regions. When the single load is applied to the middle region of the blades, the maximum displacements of the commercial and new generation blades are calculated as approximately 14 mm and 11.57 mm, respectively. This shows that the new generation blade again exhibits higher endurance performance. However, this is not the case when a single load is applied to the root zone. In other words, the maximum displacements of commercial and new generation blades were calculated as approximately 0.41 mm and 0.55 mm when the single load was applied to the root regions of the blades. In addition, considering that the displacement values calculated at such low values will be within the margin of error of the DIC system, it would be difficult to claim anything through comparison of the new generation and commercial blades with these data.

When the data in Table 1 are examined, the surface strain distributions that occur when a single load is applied to the middle regions of both blades show that the new generation blade is 77% stiffer than the commercial blade. In addition, the surface strain distributions formed when a single load is applied to the root regions of both blades show that the new generation blade is 43% more rigid than the commercial blade. However, as stated above, the new generation blade displaces more when a single load is applied to the root region of both blades. In this context, it can be thought that there is a mismatch between the displacement data from the DIC system and the stain data. However, it is thought that the main reason for this incompatibility is not related to the DIC system or commercial software, but that a possible fixation problem may have occurred during the testing of the new generation blade under the single load at its root region. In other words, since the strain data are independent of the perfection of the fixation of the blades, it is thought that the displacement data may have been greatly affected by a problem with the fixation of the blade root.

According to Table 1, it can be seen that there is more variation in the different color distributions on the blade surface for the commercial blade. It is thought that the excess in this change is due to the heterogeneity of the internal structure of the new generation blade. In order to further reduce the standard deviation of the strain values on the surface of the new generation blade and also to standardize its production, it is thought that an optimization and design study should be carried out on the number of knotted carbon reinforcements and the distance between these reinforcements.

#### *3.2. Results of Flow Analyses*

Figure 7 shows the pressure distributions on the blade as the wind speed increases from 5 m/s to 20 m/s at different angles of attack (5◦ and 10◦). In these distributions, it is seen that the flow separations increase with the increase of the angle of attack. In addition, with the wind speed increase, the total pressure values increased approximately 13.6 times for the angle of attack of 5◦; this value is approximately 16.4 for the angle of attack of 10◦. While the flow separation leaves the blade surface late in Figure 7a, it is seen in Figure 7e that the flow separation decreases with the effect of the inertia force. It is seen that the flow separation in Figure 7a is later than Figure 7b at the same wind speed. It can be concluded that, with the increase of the angle of attack, the flow is not able to cope with the reverse pressure gradient and viscous forces, and it cannot hold on to the blade surface. The separation of the flow from the blade surface reduces the aerodynamic performance of the blade.

According to the CRADLE analysis given in Table 2, Force-Y expresses the drag force (FD) on the blade, while Force-Z expresses the lift force (FL). When the wind speed value increases from 5 m/s to 20 m/s, it was observed that the lift force increased 15.23 and 15.09 times at the 50◦ and 10◦ angles of attack, respectively. It is also seen that the lift force is 0.91% more effective at the angle of attack of 5◦ than 10◦.


**Table 2.** Lift and drag forces at varying speeds and angles of attack.

#### *3.3. Blade Modeling Results*

To recall, the curvatures of the blades were calculated according to Equation (1) by using the blade surface strain values obtained from the DIC tests. Then, these curvature data together with the moment diagram of the blades formed the moment–curvature inputs for the blades in modelling. These DIC tests could be verified by using these inputs in the material definition section in the blade modeling. For this verification, the blade displacement outputs from the models can be compared with the displacement data obtained from the experimental DIC tests (see Table 3). The errors in Table 3 show that the model results generally show a good agreemen<sup>t</sup> with the test results. In Table 3, it can be accepted that one of these errors is high. As mentioned before, it is thought that this high error may stem from a problem with the fixation of the new generation blade during one of the DIC tests.

After validating the models with experimental data, the pressure values obtained from the flow analyses at different wind speeds and angles of attack were applied to the root, middle and tip sections of the blades. With these new simulation models, the mechanical behavior of both commercial and new generation blades at different angles of attack and different wind speeds were investigated. The displacement values in the root, middle and tip regions of the wings, which are dependent on these behaviors, are shared in Table 4. In addition, to facilitate the comparison of these displacement values with each other, these values are also discussed in Figures 8–10. Figures 8–10 show the displacement values in the

root, middle and tip regions of both blades, respectively. The "brown" and "blue colors" in these figures represent commercial and new generation blades, respectively.

**Table 3.** Displacements of both blades at the tip region corresponding to DIC tests and finite element modellings.


**Table 4.** Modeling results for commercial and new generation blades for varying wind speeds and different angles of attack.


**Figure 8.** Modelling displacements of both blades at the root region. Red and blue colors are for commercial and new generation blades, respectively.

**Figure 9.** Modelling displacements of both blades at the middle region. Red and blue colors are for commercial and new generation blades, respectively.

**Figure 10.** Modelling displacements of both blades at the tip region. Red and blue colors are for commercial and new generation blades, respectively.

According to the data in Figure 8, the displacement values of both wings increase at similar rates for both angles of attack (5◦ and 10◦) with increasing wind speed from 5 m/s to 15 m/s. However, the effect of the angle of attack comes to the fore when the wind speed is 20 m/s. Therefore, when the wind speed is 15 m/s or 20 m/s at a 10◦ angle of attack, the variation of the displacement values of both blades remains small. In addition, when the blue and brown colors in Figure 8 are compared with each other, it is revealed that the displacement values of the root zone of the new generation blade are always calculated

lower than the commercial one for each speed and each angle of attack. It is seen that the difference between the displacement values of the new generation and the commercial blades increases with the increase in wind speed.

The data in Figures 9 and 10 also show other displacement distributions compatible with the data in Figure 8. Therefore, according to the data in these figures, it is revealed that the displacement values of the middle and tip regions of the new generation blade are always lower than the commercial blade at different wind speeds and different angles of attack. This also indicates that the stiffness of the new generation blade is higher than the commercial one. Finally, in these figures, as in Figure 8, the difference between the displacement values of the new generation blade and the commercial blade increases with increasing wind speed.
