PSI - Issue 14

6

K Lakshmi et al. / Procedia Structural Integrity 14 (2019) 282–289 Lakshmi and Rama Mohan Rao/ Structural Integrity Procedia 00 (2018) 000–000

287

(b)

-0.1 0.0 0.1 -0.1 0.0 0.1 -0.1 0.0 0.1 -0.1 0.0 0.1 -0.1 0.0 0.1 -0.1 0.0 0.1 -0.1 0.0 0.1 -0.1 0.0 0.1

( a )

0.4 0.8 0.0 0.1 0.2 0.0 0.1 0.2 0.3 0 100 200 300 400 500 600 700 800

0.5 0.0 0.4 0.8 0.0 0.4 0.8 0.0 0.1 00 . . 20

Amplitude

Amplitude

0 100 200 300 400 500 600 700 800 0.0 0.2 0.4 00 .. 06 00 . . 10

0.00 0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00 -- 21 01 2 Time(s )

Frequency(Hz)

Fig. 2. EMD of the response at node-6 of simply supported beam-Test Case-1: (a) IMFs of healthy data; (b) FFT spectrum of IMFs of healthy data.

The critical IMFs are isolated from the extracted IMFs, using the correlation coefficient presented in the earlier section. All the isolated critical IMFs are added up to reconstruct the new current time series. Similarly, the new baseline time series is reconstructed using the selected critical IMFs. The typical reconstructed signals of sensor node 15 (i.e., node closer to crack location) using the critical IMFs of the current and healthy signals based on the presence of damage rich features of the current data are shown in Fig. 3(a). Once the current data subset and the baseline subset are reconstructed using the critical IMFs, distances of ARMAX models in terms of the subspace angles are computed for each of the sensor node signals to locate the damage as shown in Fig. 3(b). The normalized values presented in Fig. 3(b), clearly indicate that the proposed method based on EMD-ARMAX is effective in detecting as well as locating the spatial damage present in the structure in the form of minor crack. Using the count of the current dataset being analyzed, we can arrive at the exact time instant of damage. In this example, the time instant of damage is found to be 4s. 4.2 Experimental verification: The second example to validate the proposed technique to detect minor/incipient damage is the laboratory experimental studies, conducted on a simply supported reinforced cement concrete (RCC) beam. The test structure considered is a simply supported RCC beam with dimensions: 3000mm x165mm x200mm. The bottom longitudinal reinforcement was 2# 16 and the upper was 2# 12 with 25# 6 stirrups. The beam is instrumented with 16 micro electromechanical systems (MEMS) accelerometers, placed equidistantly along the beam to record the acceleration time history data. The beam is excited using a modal shaker of sine peak force capacity of 200N and the tests were carried out in the frequency range of 0-1000Hz, for harmonic and random excitations. The loading frequencies, as well as amplitude, is varied during each set of measurements to simulate operational variability. Initially, acceleration responses of the beam are measured at all the 16 sensor nodes for the undamaged state of the beam. This scenario is named as ‘Healthy’. Each signal is measured for 19 sec and is sampled at 3000 Hz so that the frequencies of the beam can be uncovered, in 0–1500 Hz frequency range.