33 Sub-components of Wind Turbine Blades: Proof of a Novel Trailing Edge Testing Concept 271 Fig. 33.6 Speckle pattern for the digital image correlation (DIC) system shown on the suction side surface (a) and virtual strain gauge positions (b). Three reference markers are positioned at the trailing edge and spar cap edge of the specimen to define the common coordinate system for suction and pressure side. Post-processing lines are defined along the target cross-section (TCS-line) and along the middle of the trailing edge cell (TEC-line). Additionally, the trailing edge electrical strain gauge line (TE-line) is highlighted. The color scale in (b) shows the longitudinal strains εz Two pairs of 2.8 megapixel cameras were installed to record the movement of each panel separately - two cameras on the SS and two on the PS. The use of two stereo DIC setups and a global coordinate system provided the ability to measure out-of-plane displacements of both panels relative to the other. Virtual strain gauges placed next to the electrical strain gauges compared the TCS-line (Fig. 33.6b). 33.3 Experimental Results and Model Validation 33.3.1 Geometry Comparison The comparison of the finite element model (FEM) with the measured geometry of the specimen shows that the FEM is thicker compared to the actual geometry (Fig. 33.7). 33.3.2 Strain Response Along the Trailing Edge The trailing edge specimen was loaded in a configuration aiming to reach the 100 % strain level of the leading-towardstrailing edge (LTT) full-scale blade test load case [21]. The trend of the longitudinal strains measured with strain gauges (SG) at the trailing edge pressure (PS) and suction side (SS) (TE-line at s ≈50 mm, see Fig. 33.6a) are in a good agreement with the measured values of the LTT full-scale test (Fig. 33.8a). The slope of the strain distribution along the trailing edge was successfully replicated. The discrepancy close to the load frames is caused from the stiffening of the support wooden blocks. Moreover, the back to back values at both sides of the TE compare very well to each other. At the target cross-section of the specimen they move apart indicating a bending deformation of an S-shape. The analytical (AM) and finite element models (FEM) are validated against the experimental, longitudinal strains along the specimen length. The strain data were derived from SGs. The longitudinal strain along the trailing edge bond-line is benchmarked (Fig. 33.8a). In general, it can be seen that the experimental strain level in both full-scale test (FST) and balljoint sub-component test concept (BJ-SCT) is lower compared to the FEM, but for both experiments the strain is at the same level. The relative strain trend between suction (SS) and pressure side (PS), the pathway of the FEM and experiment BJ-SCT is similar: Along the length the strain level on the PS is getting lower than on the SS, finding its peak value at the target cross-section at ≈24 m blade length. Strain peaks were recorded next to both load frames as also prognosed by the FE model The peaks are caused by the load frame constraints on to the local Poisson’s ratio of the structure.
RkJQdWJsaXNoZXIy MTMzNzEzMQ==