Keywords

24.1 Introduction

Fatigue damage of reinforced concrete (RC) structure cumulatively increases with applied load cycles. Damage is related to the irreversible process [1] in a structure with respective of cyclic accumulation that causes the significant damage led to structural failure. Cumulative fatigue damage analysis is vital in the prediction of RC structures subjected to load histories [2]. Damage and failure process during load history require for the application of Non-Destructive Evaluation (NDE) such as Acoustic Emission (AE) technique that possible for real-time damage monitoring. Besides, the damage processes often generate a vast amount of AE hits or events from one damage zone [3].

The fatigue damage process divides into three stages, namely, the initiation, the steady propagation, and the coalescence of microcracks to macrocracks [4]. Md Nor et al. [5] identified the fatigue damage based on the bath-tub curve and three stages were found. The stages are burn-in, steady state, and burn-out. The burn-in is known as the infant mortality or initiation of damage. The steady state is the constant damage severity. The burn-out is wear-out, life end, or failure of a structure. Md Nor et al. [5] enhanced that the high damage severity is found to be at first point of the initiation stage of fatigue failure. Xie and Lai [6] stated that at an initiation stage, the rate of failure is high and affect to the production to be released. It is because high failure rate can be found at initiation stage on any system relates to fatigue. It gives an indication that thorough investigation needs to be carried out especially at initial stage prior to steady stage.

However, in fatigue study, most of the researchers focus on the fatigue damage process as a whole system [7,8,9] and lack on investigation at the initial stage. It is vital to take action the cracking process at initiation stage by accumulating the AE hits or events. AE hits or events are accumulating with crack growth [10], and it is reasonable to investigate the relationship between cumulative absolute energy and deflection. As stated by Md Nor et al. [5], the initiation stage of fatigue life of a structure is 7 and 0.8% if 0.8 and 0.7 of ultimate loads were applied to the beam.

The prediction of fatigue life of a structure can be done using AE technique. It has been extensively reported by Md Nor et al. [5] and found that AE characteristic for beam subjected to fatigue load with load ratio of 0.8 is higher than load ratio of 0.7. It is because the higher load ratio would generate the higher stress to the beam. Karimian et al. [11] inferred that the AE waveform is very reliable indicator of fatigue crack initiation. To give a better interpretation of fatigue life, the AE energy as well as signal strength is usually utilised [5, 11]. At the same time, the intensity analysis, historical index, and severity index can be exploited. From the intensity analysis, the classification of damage of the structure can be performed. Intensity analysis is a method of evaluating the structural significance of an event by a combination of historical index and severity index [12]. The intensity analysis has been applied for a structure subjected to static loading to failure for characterisation of damage for glass fibre-reinforced composites decks [8], and carbon fibre reinforced polymer reinforced concrete slabs [13]. It also has been used for damage detection of RC beam under alternating static and dynamic loads and health index evaluation of RC beam under stepwise-static loading [14].

One of the AE energy is signal strength. It is defined as “the measured area of the rectified AE signal with units proportional to volt-seconds” [15] and related to the relative energy of a specimen. The signal strength is defined in Eq. (24.1), where f+ is the positive signal envelope function, f-is the negative signal envelope function, t1 is time at first threshold crossing and t2 is the time at last threshold crossing. Meanwhile, the intensity analysis is a technique for evaluating intensity fatigue damage zones by calculating historic index (HI) and severity index (Sr) [13]. The equations for HI and Sr are presented in Eqs. (24.2) and (24.3).

$${S}_{0}=\frac{1}{2}\int\limits_{{t}_{1}}^{{t}_{2}}{f}_{+}\left(t\right)dt-\frac{1}{2}\int\limits_{{t}_{1}}^{{t}_{2}}{f}_{-}\left(t\right)dt$$
(24.1)
$$HI=\frac{N}{N-K}\frac{\sum_{i=K+1}^{N}{S}_{oi}}{\sum_{i=1}^{N}{S}_{oi}}$$
(24.2)
$${S}_{r}=\frac{1}{J}\sum\limits_{m=1}^{J}{S}_{om}$$
(24.3)

where: N is the number of hits up to time t, Soi is the signal strength of the ith event, K and J are empirical derived constant based on material [14], and Som is the signal strength of the mth hit.

However, damage evaluation using acoustic emission analysis under constant fatigue loading is still lack of attention. This paper presents the investigation of AE activities at an initiation stage of fatigue life of RC beam subjected to fatigue loading. In this study, the initiation stage was based on the Md Nor et al. [5], about 7% of the fatigue life of the beam.

24.2 Experimental Programme

24.2.1 Sample Preparation

The concrete beam was prepared for concrete strength of C40. The concrete was made up of cement, water, fine aggregate and coarse aggregate with a proportion of 1: 0.43: 2.16: 2.60, respectively. Then, 1% of retarder of cement weight was added to the concrete mix to improve the workability of the fresh concrete. The maximum coarse aggregate of 20 mm was used. The beams were designed as a singly RC beam with two high yield steel bars of 16 mm to strengthen the tension part and two mild steel bars of 8 mm as hanger bars. In the stirrups, 12 diameter of mild steel with a spacing of 100 mm centre to centre was used. In the preparation of each RC beam, the reinforcement was submerged in the standard mould of 150 mm × 150 mm × 750 mm with a cover of 20 mm, then the concrete mix was cast. All beams were demoulded after +24 h and submerged in curing tank for 28 days. The compressive strength for 7 days was found to be 32.35 and 44.65 N/mm2 at the age of 28 days. All samples were tested at the age exceeded 80 days. The reason for this was to avoid any strength increase that would influence the results.

24.2.2 Fatigue Test

Three-point fatigue tests were performed under load control with a clear span of 630 mm. A servo-hydraulic testing machine was employed for subjecting prismatic RC beams of 150 mm × 150 mm × 750 mm to the required fatigue loading. The specimen setup is represented in Fig. 24.1. The fatigue test was performed at a low frequency of 1.0 Hz, with the applied load ratio of 0.8 of the ultimate load of the beam. The ultimate load was identified from the static loading carried out on five beams. The ultimate load was 158.85 kN. Hence, the maximum fatigue loading, Pmax was 127 kN. The minimum fatigue loading, Pmin was taken about 20% of the ultimate load. Therefore, the Pmin is 32 kN. The selection of the Pmin and Pmax was based on Md Nor et al. [5]. Then, a sine wave load cycle is used to induce the fatigue load to the beams until failed. An AE system monitored the test for all beams and the data recorded by data acquisition software.

Fig. 24.1
figure 1

Setup of the third point bending test for the RC beam

Full size image

24.2.3 Acoustic Emission Monitoring

A total of four AE sensors (R6I—resonant 60 kHz Integral sensor) were applied on the beam, as shown in Fig. 24.1. After installation of the sensors, the calibration test was performed to check the sensitivity of both sensor and beam surfaces. Both surfaces were connected using a thin layer of high vacuum grease. The sensitivity was performed using pencil lead fracture close to the sensor on the beam surface and the results were monitored on the AE visual. The amplitude that appeared on the AE visual for each lead fracture was observed, in which the required value is above ±97 dB in accordance with [15]. In the AE hardware, the threshold level of 45 dB and wave velocity were applied throughout the test. The rearm time of 1.62 ms, duration discrimination time of 400 µs, pre-trigger samples of 200 and the sample rate of 10 MHz were set in the AE system.

In this study, only data collected on Channel 1 (CH1), as shown in Fig. 24.1, would be analysed and discussed. From the data acquisition, the signal strength corresponding to the normalised cycle at an early stage was extracted for analysis. Then, the cumulative signal strength, historical index, severity index and intensity analysis were analysed to investigate the AE activities at an earlier stage of fatigue load.

24.3 Results and Discussion on Acoustic Emission

24.3.1 Signal Strength

Figures 24.2 and 24.3 present the AE parameter of signal strength with respect to the normalized cycle. The signal strength is used to observe the damage activities in the RC beams under fatigue load at an earlier stage. The normalised cycle is the present cycle divided by the fatigue life, N/Nf [8]. From the fatigue test, the number of cycles at failure, Nf of RC beam subjected to a load ratio of 0.8 is 848461 cycles on Channel 1 (CH1). In this analysis, the AE hits, or events in the first 3 s of loading are not considered for analysis. It is due to the signal collected at the beginning of fatigue loading, which may originate from complex sources such as the contact of supports and the specimen [7]. Otherwise, Gostautas et al. [10] were discarded the first 100 hits in the analysis. It is because within the first 100 hits usually are associated with cracking of excessive matrix material under initial loading. In Fig. 24.2, high AE signal strength is produced at the early stage of cracking due to hasty of crack propagation of concrete. The highest signal strength at an early stage is 3.6 × 109 nVs, as presented in Figs. 24.2 and 24.3. Therefore, the cumulative signal strength becomes higher in the early stage of the cycle, as shown in Fig. 24.4. A similar finding has been found by El Batanouny et al. [16] that the presence of high signal strength is closely related to the formation of an initial crack in the specimen as it can be seen through the data collected by the fixed sensor on specimen’s surface. This trend of the curve of cumulative signal strength—normalised cycles—is quite similar to the curve represented by Roberts and Talebzadeh [8] on absolute cumulative energy—normalised cycles.

Fig. 24.2
figure 2

Relationship between signal strength and normalised cycle

Full size image
Fig. 24.3
figure 3

Relationship between signal strength and normalised cycle at an early stage

Full size image
Fig. 24.4
figure 4

Cumulative signal strength at an early stage

Full size image

24.3.2 Intensity Analysis

The presence of a crack in the beam specimen at an early stage of load cycle is pronounced by the relationship between the severity index and historical index, which is known as AE intensity analysis. The intensity analysis provides a quantitative assessment of damage classifications [10]. Five damage zones, namely, Zone A (insignificant emission), Zone B (minor), Zone C (intermediate), Zone D (follow-up) and Zone E (major) were adopted to classify the damage. Figure 24.5 shows the intensity chart of RC beam subjected to cyclic loading at an early stage. In the chart, the plots divide into three parts as depicted in Table 24.1: P1, P2 and P3. P1 is the AE hits at an early stage of 1000 cycles. P2 is the AE hits data collected from cycles 1001–2000. Meanwhile, P3 is the AE hits from cycles 2001–3000. The three parts identified from the investigation were carried out by Md Nor et al. [5] on the early stage of fatigue failure.

Fig. 24.5
figure 5

Intensity analysis of the RC beam under fatigue

Full size image
Table 24.1 The cycles for each part
Full size table

From the chart, most of the plots for data collected at P1 indicate the beam in damage Zone E. It indicates that the beam is suffering to significant major damage. Gostautas et al. [10] stated that, as the structure suffering to major damage, the structure is required an immediate shut-down and follow-up inspection. The major defect (Zone E) is produced by the first 1000 events of AE signals, and it is generated by the first few loading cycles under high load ratio. It is due to the occurrence of crack initiation in the concrete at an early stage of the fatigue load. As the load cycles increase, from load cycles 1001–2000, the plots are in the middle of Zones E and D. Meanwhile, as the load continuously cycles, the plots turn to Zone D. Zone D means to follow up where the significant defect is requiring to carry out further inspection. The inspection is carried out by analysing the additional data collected from the testing. Gostautas et al. [10] recommended a non-destructive examination for further data analysis. Usually, the inspection is carried out periodically to inspect the condition of the structure.

The plots in the intensity chart are reflected in the high signal strength at an early stage, as presented in Fig. 24.2. It was also pronounced by the robust increase in the accumulated signal strength, which becomes a hint on the occurrence of the second stage and the third stage of fatigue failure. At the same time, due to rapid crack initiation, it produces high AE activities. According to Roberts and Talebzadeh [8], an early stage of fatigue damage has occurred in the first 10% of Nf before the curve stabilises. Meanwhile, Md Nor et al. [5] found that 7% of Nf was the endpoint of the early stage of fatigue damage under the same load ratio of 0.8 of the ultimate loads.

24.3.3 Intensity Analysis

Figures 24.6 and 24.7 show the historical index and severity index with respect to normalised cycle for beam subjected to fatigue loading, which are representing the overall performance of the beam condition. From the historical index, as shown in Fig. 24.6, high historical index can be observed at initiation stage as shown in small graph in Fig. 24.6. This high historical index is due to the presence of crack at the concrete beam. However, as the loading continuously cycles at the same range of 32 and 127 kN, several peaks of historical index were noticed until the beam failed. It is because as generally known that the historic index has been found to be a sensitive parameter to quantify changes in the slope of the cumulative signal strength by weighing the average signal strength of the last 20% or 200 hits. From Fig. 24.7, high severity index is found at the beginning of load cycle. The severity reduces as the cycle increases. A significant high in severity at the beginning of fatigue loading indicates the presence of damage in the structure [17, 18].

Fig. 24.6
figure 6

Relationship between historical index and normalised cycle

Full size image
Fig. 24.7
figure 7

Relationship between severity index and normalised cycle for the whole testing

Full size image

At an early stage of fatigue damage, the high historical index is found in Fig. 24.8 with a value of 6.15. The similar pattern can be found through the severity index as presented in Fig. 24.9 with the value of more than 4200. According to Xie and Lai [6], at the beginning of fatigue failure, the initial failure rate is too high for the product to be released directly after production. It gives an indication to any system that relates to fatigue that would have a high failure rate or damage at the initial load cycle. When turning back to this study, it gives a clear picture of the development of an initial crack in the concrete, which would reflect on the production of high historical index and severity index.

Fig. 24.8
figure 8

High historical index at an early stage

Full size image
Fig. 24.9
figure 9

Severity index at an early stage

Full size image

24.4 Conclusions

The acoustic emission activities at an initiation stage of fatigue damage for reinforced concrete beam were investigated. It is found that high signal strength was noticed at an earlier fatigue damage of the beam with the value of 3.6 × 109 nVs. This high value is closely related to the formation of initial crack in the concrete. A rapid increase in the cumulative signal strength at this stage was presented. The results were verified by the intensity analysis with the plots at first 1000 load cycles were on Zone E. The Zone E means that major damage and rigorous inspection of the beam needs to be carried out. As the load cycles accumulated, the plots were reduced to Zone D, which requires follow-up inspection of the beam.

Similar pattern was found on the severity index and historical index with respect to normalised cycle, where high value was found at first load cycle. All data from the acoustic emission activities can be used as a hint to the formation of the next stage in the fatigue damage process. From the investigation of acoustic emission activities at an initiation stage of fatigue damage, a prediction of the beam performance can be identified earlier.