PSI - Issue 64

Henrik Becks et al. / Procedia Structural Integrity 64 (2024) 1279–1286 Henrik Becks / Structural Integrity Procedia 00 (2019) 000 – 000

1281

3

For continuous two-dimensional strain measurement, the monitoring concept 2D-FOS was developed, which is based on an externally applied FOS network. A Python algorithm was devised for processing raw measurement data and visualizing two-dimensional strain profiles. This algorithm automates the filtering, segmentation, decomposition into direction-dependent strain components, and processing into strain profiles (Figure 1a, bottom). Node-dependent principal axes transformations are employed to determine the maximum tensile strains in the measurement field from individual strain components (Figure 1b, top). These tensile strains and their temporal evolution are utilized as input by a lifetime prediction model (stage I) to estimate the remaining service life until crack initiation (Figure 1c, top). Following crack initiation, aggregated measurement data are used to track crack opening and progression (Figure 1b, bottom). Identification of crack propagation is based on a simple peak detection algorithm (Becks et al., 2022; Richter et al., 2023), while calculation of the strain integral (considering the inherent fiber transmission length) provides the crack opening (Berrocal et al., 2021; Herbers et al., 2023; Janiak et al., 2023). The steel stress in the transverse reinforcement is approximated from the crack opening, and the remaining service life of the entire structure is estimated using another lifetime prediction model (stage II) (Figure 1c, bottom).

b) data interpretation & aggregation

a) measurement & preparation of raw strain data

c) forecast of remaining service life

lifetime prediction model (stage I)

Raw data acquisition using 2D FOS

calculation of principal strains

ε x

ε z ε 1

ε 1

ε 1

time-variant 1D strain

ε

ε ξ

Principal axes transformation

relevant

. ε

 t,max

ε ε x z +

( )²  xz

(ε ε x z - )²

ε x

+-

ε 1,2 =

+

t rest

2

4

4

ε 2

x

1

2

3

x 1

x 2

 t

t

ε x correction & filtering filtered

calculation of crack opening

lifetime prediction model (stage II)

1D to 2D

Δσ s s ( ) w

Δσ s s ( ) w S-N-curve

ε z

ε ξ

ε x

w s

ε load+constraint ε temp

crack

N( ) i s Δσ

n i

0

0.25 0.5 [%]

N

t

Figure 1. Schematic overview of the developed service life prediction concept: a) measurement and preparation of raw strain data, b) data interpretation/aggregation, and c) forecast of remaining service life. 3. Experimental campaign To implement the service life prediction concept described in Section 2, it is necessary to first evaluate the application limits of the monitoring system 2D-FOS. As a preliminary step, the fiber optic measurement technology utilized was thus examined for its capability to detect and track tensile stresses under monotonic and fatigue loading. Particularly critical for the success of the concept was the ability to capture both very small strains (prior to crack initiation) and very large strains (after crack initiation) using a single fiber-adhesive combination. The following section presents and discusses the preliminary experimental. 3.1. Test setup and material properties The experimental campaign was conducted using uniaxial tensile tests on specimens measuring 100 cm in length and 12 cm in width (Figure 2). Some of these tests were reinforced with two rebars (6 mm diameter). The height of the plain specimens was 2.5 cm, while that of the reinforced specimens was 4 cm. To predefine the location of crack initiation, the plain tests were notched 2 cm from both sides in the middle of the specimen (Figure 2a). The concrete strength was set to a target concrete class of C30/37. The concrete mix consisted of cement type CEM I 52.5 R, quartz aggregates with a maximum grain size of 8 mm, and a water-cement ratio of 0.6. To assess the concrete properties, various material samples in the form of cylinders (300 mm in height and 150 mm in diameter) were prepared and tested at the outset of the test campaign. The compressive strength, modulus of elasticity, and