PSI - Issue 5

Alejandro Carvajal-Castrillón et al. / Procedia Structural Integrity 5 (2017) 729–736 Alejandro Carvajal-Castrillón/ Structural Integrity Procedia 00 (2017) 000 – 000

734

6

strings. Very high final acquisition rates can be obtained when package size is over 80 strings. For a 160-string package size it is possible to have almost the 6-KHz acquisition capability provided by the FBG interrogator with an 18-ms delay due to the wireless transmission. The testing exhibited a requirement of nearly 200-Kbps bandwidth to send the data in the most critical condition. Results also showed variability of data transmission rates due to network and processor utilization.

Table 1. Data transmission test results Strings per package sent

Received strings

Received packages

10 20 40 80

1540 2820 4120 4560 5920

154 141 103

57 37

160

3.2. Operational tests

During operational ground tests, data from 14 FBGs were acquired during a minute for each test, accomplishing five tests for each of the 12 different wing angles of attack. The strain fields obtained from these experiments were processed by the OBS and D2SL-SOM methodology, having a final quantization error of 0.032 and a final topographic error of 0.028, which means that proper clustering identification was achieved. The process resulted on five data clusters being found, three of them being major clusters and two being smaller clusters. This indicates that the wing supports and the load application on the test performed corresponded to five different operational conditions. The mean strain fields of the operational condition clusters are displayed on Figure 2. The data acquired shows that the highest strains were presented at sensor number three, which was located at the tension side, being at the half of the beam’s length. This behavior was expected as it corresponds to a typical doubly supported beam case, and in this experiment the wing acted like one. Sensors No. one to No. five (tension) and No. six to No. 10 (compression) showed the highest strains while No. 11 to No. 14 sensors (torsion) presented the lowest strains, this happens because the structure was tested by applying loads almost orthogonally to the wing’s chord on most conditions, hence torsion strains were not significant on the lateral sides of the instrumented beam.

Fig. 2. Mean strain fields.