Experimental investigation on the seismic performance of FRP wrap confined FRP reinforced concrete columns

Design of test specimens

A total of six identical specimens were constructed for this experimental study and tested, with a total height of 2700 mm and a cross-sectional dimension of 300 mm × 300 mm. The specimens were constructed with column footing of dimensions 1500 mm × 1000 mm×700 mm to fulfill the fixed-end conditions which further helps to fix the specimens on the testing machine set up during the test. Furthermore, a column head of dimensions 600 mm × 400 mm×400 mm was constructed to the top of each column so that the MTS actuator (lateral loading actuator) could connect. Both the column footing and column head were heavily reinforced to avoid any premature failure during testing. The actual calculated height from the top of the stub to the loading point of the specimens was 1800 mm and the shear span ratio of the specimens was 6. The detailing of the specimen can be seen in Figure 1 and detailed specifications of the specimens have been listed in Table 1.

OPEN IN VIEWER

Figure 1.

Specimens and reinforce details (unit: mm).

OPEN IN VIEWER

Table 1.

Details of test specimens.

Specimen	GFRP bar	Hoop’s spacing	No. of confined layer
S65G0	12 ϕ 14	65	0
S85G0	12 ϕ 14	85	0
S65G3	12 ϕ 14	65	3
S85G3	12 ϕ 14	85	3
S85G4	12 ϕ 14	85	4
S85G5	12 ϕ 14	85	5

All six columns were designed with 12 longitudinal GFRP bars of 14 mm in diameter, resulting in a reinforcement ratio of 2.05% to prevent the premature failure of longitudinal bars. Additionally, referring to the previous studies ((PDF) Confined concrete columns with Stubs, no date; Priestley and Seible, 1995; Tirasit and Kawashima, 2007; Jiang et al., 2014; Youssf et al., 2015; Cai et al., 2016; Wang et al., 2018) on RC and FRP-confined reinforced concrete columns, the additional wrap seemed beneficial for seismic loading. Tirasit & Kawashima’s (Tirasit and Kawashima, 2007) studies showed that damage height of column was 0.5 to 1.5D. Similar outcomes were observed when others models applied (Jiang et al., 2014; Youssf et al., 2015). However, for convenient, it was chosen as 500 mm in this study because the finished GFRP fiber width was 500 mm. The specimens were confined by wrapping GFRP sheets with a specified number of layers along the height of the plastic hinge region starting from the top of the column stub. To facilitate the wrapping process, the GFRP wrap is applied using the wet layup method, which involves cleaning the concrete surface and then applying the adhesive mixture to the surface before sticking the GFRP wraps. The adhesive mixture was applied after every layer. This process should be done carefully to avoid wrinkles and bulges in the GFRP wraps during the wrapping process. During the formwork process, a quarter circle with a radius of 30 mm made of an iron wrap is fixed at the chamfered part of the specimen.

Preparation of closed CFRP hoop

This study combined on-site fabrication techniques with wet lay-up technology to manufacture closed CFRP hoops, the manufacturer steps were as follows (Figure 2):

(1) Firstly, the CFRP wraps made from commercial unidirectional carbon fiber UT70-30 supplied by the manufacturer were cut into the desired width (i.e., 20-30 mm).

OPEN IN VIEWER

Figure 2.

Preparation of reinforcement cage.

Material properties

The ready-mix concrete was ordered from a local concrete mixing plant to cast specimens. All the specimens were cast with a single batch of concrete and cured under the same conditions. To measure the compressive strength of concrete during the experiment three standard (150 mm × 150 mm × 150 mm) cubes were cast and tested. The equivalent compressive strength of the concrete was 41.1 MPa. The CFRP sheets were produced by the Japanese company Toray whereas GFRP sheets were produced from unidirectional fiber HICOMA®-HITEX supplied by Nanjing Haituo Composite Materials Co., Ltd. All the GFRP longitudinal reinforcement was supplied by Jiangsu Haichuan New Materials Technology Co., Ltd. It was manufactured by wrapping GFRP fiber yarns and epoxy resin with a volume content of 75% and 25%, respectively, into ribbed bars with an effective diameter of 14 mm. The longitudinal GFRP bars were tested according to ASTM, and the tensile strength of the GFRP reinforcement was measured. The mechanical properties of the materials are shown in Table 2.

OPEN IN VIEWER

Table 2.

Material properties of CFRP and GFRP sheet.

Material	E t (GPa)	E c (GPa)	σ t (MPa)	σ c (MPa)	ε t (%)	ε c (%)	d or t (mm)	e (%)	Wt/m2 (g)
CFRP sheet	240	-	4340	-	1.7	-	0.167	-	300
GFRP sheet	72	-	1500	-	2.1	-	0.169	-	400
GFRP bar	48.6	35.2	1020	431	2.1	1.2	12	2.1	-

Test setup

The experiment was conducted at the structural and seismic laboratory of the Harbin Institute of Technology where the specimens were fixed and tested under constant axial load and cyclic lateral displacement. The constant axial load was applied by a vertically mounted hydraulic jack, whereas cyclic lateral displacement was applied by a horizontally positioned MTS actuator of 1000 kN to the pier column as shown in Figure 3. A specially designed loading apparatus was employed to maintain a constant axial load and move it along with the upper part of the column during testing, as depicted in Figure 3, to include and simulate the P-delta effect during the test. The column was first compressed, and later this load was adjusted using a hydraulic system with a pressure-relief valve to preserve stability.

OPEN IN VIEWER

Figure 3.

Test setup.

Among different loading methods, a displacement control loading system was adopted in this study. The loading system used in the experiment was according to the 0.5% drift ratio at each stage. In the first stage, a specimen was loaded at a 0.25% drift ratio (i.e.4.5 mm) for one cycle, whereas the second stage was loaded at a 0.5% drift ratio (i.e.9 mm) for one cycle. After that, each stage was increased by 0.5% drift ratio (i.e.9 mm) for two cycles. After each stage of loading, before completing the second cycle, the specimens were observed, and images were captured. At the same time, crack development was also observed, and it was drawn along its developing directions. The experiment was terminated in three conditions: (i) when the load-bearing capacity of the specimen drops to 70%, (ii) when the displacement becomes too large and affects the loading, and (iii) when there is significant damage to the GFRP cloth, longitudinal reinforcement or concrete. The loading mode of the specimen is shown in Figure 4.

OPEN IN VIEWER

Figure 4.

Loading procedure.

The strain gauges were arranged on the longitudinal bars, stirrups, and GFRP sheet. Figure 1 shows the arrangement positions of the strain gauges on the longitudinal bars and stirrups, whereas Figure 5 displays the layout of the strain gauges on the GFRP sheets. The strain gauges were pasted at 150 mm from the bottom of the wrapped sheets. There was a transverse strain gauge on each of the four sides and the corner of the column, and there was a vertical strain gauge on the north and south push-pull sides of the column, Figure 5. A total of three linear variable displacement transducers (LVDT) were arranged for the detection and control of the lateral displacement of the column. The lower LVDT was arranged in the middle of the column stub to observe the small sliding of the column. The middle LVDT is arranged in the middle of the column and is mainly considered with strain and vertical displacement recording corresponding to the MTS system. The upper LVDT is arranged in the middle of the lateral loading head, and the value of the lower LVDT is subtracted from the upper LDVT to measure the magnitude of the lateral displacement of the column to ensure normal loading. The locations of the LVDTs are shown in Figure 3.

OPEN IN VIEWER

Figure 5.

Layout of strain gauge.

Failure modes

The failure modes of each specimen were evaluated based on lateral displacement level or drift ratio δ (i.e. the ratio of lateral displacement to the effective height of the specimen) and depicted in Figure 6. For the unconfined specimens S85G0 and S65G0, both exhibited similar failure modes typical of flexural behavior. Flexural cracks were observed at an early stage of loading (4.5 mm displacement). As loading increased, these cracks widened and extended, with additional cracks forming, particularly in the plastic hinge region. Over time, the cracks propagated toward the web of the columns, eventually developing into a combination of flexural and shear-flexural cracks. At a displacement level of 18 mm (δ = 1%), the crushing of concrete was observed in S85G0 and with further loading, the crushing zone expanded and began to connect across the section. In S65G0, concrete detachment was observed at the bottom corner of the column at a drift ratio of 1.5%, followed by progressive concrete fracture in that region. At a loading displacement of 45 mm (δ = 2.5%), S85G0 showed no new crack development, although the crushing zones on both sides had connected. At this point, the load-bearing capacity had dropped to less than 70% of the peak value in both directions and the test was terminated. The concrete almost detached whereas the longitudinal bars were exposed on both sides and fractured due to compression in S65G0. Additionally, the load bearing capacity dropped below 70% at 45 mm on one side and 54 mm on the other side after which the test was terminated. During post-test observations, a void in S85G0 was found at the column base, with both tension and compression longitudinal bars fractured after the removal of the surrounding concrete, whereas concrete fragment and detachment, and fracture of bar due to compression was found in S65G0, resulting both in flexural failure, as shown in Figure 6(a) & (b). The fracture of bars is highlighted with a red circle in Figure 6(a) and (b). A similar kind of failure pattern was studied by (Cai et al., 2017).

OPEN IN VIEWER

Figure 6.

Failure modes.

In the case of specimens confined with GFRP wraps, a cracking sound attributed to adhesive debonding of the GFRP was heard during the initial stage of loading displacement at 4.5 mm, except for S85G3 which occurred at a loading of 9 mm. The development of minor flexure cracks in concrete and fracture in GFRP wraps began to appear at 9 mm of loading for S65G3, Figure 6(d) while for for S85G3 and S85G5, it was observed at a loading of 18 mm, Figure 6(c) and (e). These cracks typically developed just above the top edge of the GFRP wrap, consistent with observations reported in previous studies (Cai et al., 2017; Wang et al., 2018; Zeng et al., 2022; Zeng et al., 2022), suggesting that the wraps effectively delayed crack formation. With the increasing displacement level, additional cracks developed and existing ones widened in both the concrete and the GFRP wrap. Three distinct cracks along the upper, middle, and lower portion of the GFRP wrap were appeared in specimen S85G3 at a drift ratio of 2.5%, Figure 6(c). There was no further cracks were noticed in S85G3 and S65G3 till at drift ratios of 3% and 3.5%, respectively. However, at 63 mm of displacement, a loud bang sound was heard in S85G3, coinciding with the formation of two through-cracks in the GFRP wrap at this stage. These cracks widened significantly, up to approximately 4-5 mm, Figure 6(c), and continued to expand as the drift ratio approached 4%, leading to a noticeable decline in load-bearing capacity and eventual termination of the test. Additionally, the maximum GFRP crack width of 5 mm was measured at a loading of 72 mm followed by another bang sound heard at 81 mm of loading for S85G5, Figure 6(e). These cracks in the GFRP wrap widened and the load bearing capacity of the specimen decreased to below 70%, leading to the termination of the experiment. Likewise, in S65G3, the concrete near the GFRP wrap started to fragment and approach detachment at a drift ratio of 4% and the test was terminated at a drift ratio of 5% due to load load-bearing capacity reaching to limit. During post observation, after removing GFRP sheet, it was found that the concrete portion of all three confined specimens in the plastic hinge zone mostly remained intact without any signs of crushing, and only minor damage to the longitudinal bars either slight buckling or fracture was observed Figure 6(c)–(e) which is highlighted by red circle. Importantly, no damage or failure of the CFRP hoops was noted, which is likely due to their higher tensile strength compared to the GFRP bars.

Load-displacement hysteresis curve

The load-displacement hysteresis responses of all the test specimens are presented in Figure 7. In these graphs, the positive and negative values on both axes correspond to the push and pull directions, respectively. Lateral load data were recorded using an MTS actuator, while displacement measurements were determined by calculating the differential readings between LVDT-1 and LVDT-3. The hysteresis behavior of specimens was analyzed based on their hoop spacing and the application of GFRP confinement layers. For example, unconfined specimens S65G0 and S85G0 were compared based on hoop spacing alone, whereas confined specimens S65G3, S85G3, and S85G5 were assessed based on both hoop spacing and the number of GFRP wrap layers.

OPEN IN VIEWER

Figure 7.

Load-displacement hysteresis curve.

A comparison of hysteresis behavior of unconfined shows that decreasing the hoop’s spacing enhances the restraint effect of hoops on the core concrete, which is beneficial for improving the lateral load bearing capacity for unconfined specimens and ductility for confined specimens. The maximum lateral load resistance capacity of S65G0 is higher than S85G0 by 19.19% in the push direction. It also leads to a slight increase in the pull direction, although the magnitude is insignificant as depicted in Figure 7(a). Additonally, it can be observed from Figure 7(a) that the hysteresis loop seems similar in pull direction however the hoop spacing is different. It is due to the fracture of longitudinal bars of S65G0 in that direction as the surrounding concrete spalled off which resulted in the exposure of bars. Furthermore, the specimen lost the bonding between concrete and bars which resulted in dropping the load bearing capacity below 70%. In contrast, Figure 7(b) shows that the confined specimen S65G3, which had smaller hoop spacing, exhibited a broader hysteresis loop and more gradual stiffness degradation beyond the peak load, indicating that the improved ductility due to GFRP wrapping, without a significant compromise in stiffness. Notably, S85G3 also demonstrated favorable performance up to a drift ratio of approximately 3.5% in the push and 3.0% in the pull direction, showing a response comparable to S65G3 within this range. However, beyond these points, it can be seen that the lateral resistance force decreased rapidly before its termination, Figure 7(b).

The use of external GFRP wraps to the columns proved to be very effective in protecting the concrete, especially in the plastic hinge areas where damage typically starts. These wraps helped prevent the concrete from crushing or spalling and also supported the longitudinal GFRP bars, reducing the risk of breaking. As a result, the wrapped (confined) columns showed stronger and more stable performance, with wider hysteresis loops and higher peak loads. Similar conclusions was drawn from previous studies on the aaplication FRP wraps (Cai et al., 2016; Wang et al., 2017, 2018). For example, specimen S65G3 demonstrated a substantial increase in lateral displacement 68.2% in the push direction and 100.8% in the pull direction compared to S65G0. While the peak load in the push direction didn’t change much, in the pull direction it increased by 52.9%, as seen in Figure 7(c). Figure 7(d) compares unconfined and confined specimens with the same hoop spacing but different numbers of GFRP wrap layers. As mentioned earlier, the inclusion of GFRP wraps in the plastic hinge region not only enhanced the hysteresis loop shape but also contributed to higher lateral load capacities. For instance, specimen S85G3 exhibited a much wider hysteresis loop with a more gradual degradation in load bearing capacity compared to S65G0. Specifically, S85G3 exhibited 33% and 51% higher peak lateral load compare to S85G0 in push and pull directions, respectively, and 61% and 65% increases in lateral displacement. However, increasing the number of GFRP confinement layers seemed limited benefit. Increasing the GFRP wraps from 3 to 5 layers led to a 24.5% gain in lateral displacement and only about a 6% gain in load capacity. Still, specimen S85G5 performed superior among all the test specimens exhibiting a much wider hysteresis curve, reaching to drift ratio of 5.02% before failure and recorded the highest lateral load capacity of 82.86 kN.

Load-displacement skeleton curve

The “envelope curve” of each specimen obtained from the quasi-static hysteresis test which represents the relationship between the load and displacement of the specimen as shown in Figure 8.

OPEN IN VIEWER

Figure 8.

Envelope curve.

The “yielding point” method proposed by Feng et al. 2015) was used in current study to define the yield point, which identifies the yield point as the farthest point from the origin on the skeleton curve that shares the same slope as the line connecting the origin and the peak point. This method has a clear physical interpretation and can be easily calculated using computer programming and this approach has been applied in previous research (Ren et al., 2022) to determine the yield point. Yielding point method was adopted to measure various performance points for all specimens in this study. The specific calculation formula is shown in equation (1).

V y , D y = max d V y , D y = V , D = | V p · D − D p · V | V p 2 + D p 2

(1)

where D and V represent displacement and load-bearing capacity respectively, and D _y , D_p, V_y, and V_p represent the corresponding yield and peak displacement, yield, and peak load-bearing capacity, respectively. The values of D _y , D_p, V_y, and V_p of each specimen was calculated from the envelope curve and summarized in Table 3.

OPEN IN VIEWER

Table 3.

Main stiffness and ductility index of each specimen.

Specimens	Dy (mm)	Vy (kN)	Dp (mm)	Vp (kN)	Du (mm)	K1 (kN/mm)	K2 (kN/mm)	De (%)	xu (%)
S85G0	9.03	47.16	22.52	54.52	36.46	5.22	0.587	-	2
S65G0	10.68	54.34	25.52	62.92	41.9	5.09	0.576	-	2.3
S85G3	12.6	66.36	35.78	78.18	73.83	5.25	0.385	28.72	3.7
S65G3	13.53	67.26	36.07	79.54	84.64	5.12	0.313	22.20	4.1
S85G5	14.03	69.37	40.53	82.86	92.13	5.27	0.295	11.40	4.6

As seen in Figure 8, there is slight differences in the positive and negative load bearing capacity and curve trend of the specimen. It likely due to minor experimental inconsistencies such as loading head misalignment, concrete casting imperfections, fabrication errors, or slip between the column base and the foundation. To address this, the performance values of each specimen reported in Table 3 represent the average of the positive and negative values. The ability to bear loads and the capacity to deform the columns were evaluated by analyzing the load-displacement envelope curves. Figure 8(a)–(b) depicts the effect of hoop spacing whereas Figure 8(c)–(d) illustrates the effect of confinement and the number of layers of confinement. The specimens with smaller hoop show slightly higher peak load and gradual deformation capacity, whereas the specimens confined with a GFRP sheet in its plastic hinge region demonstrated a significant improvements in strength, ductility, and post-peak behavior. A comparative analysis revealed that, relative to unconfined specimens S65G0 and S85G0, the yield displacements of S65G3 and S85G3 increased by approximately 18% and 7%, respectively, while their corresponding yield strengths rose by 15% and 1%. The related peak displacements and loads of the unconfined specimens also increased by 13% and 15%, respectively but remained nearly unchanged for the confined specimens. However, the benefits of GFRP confinement became more evident when comparing the confined and unconfined cases directly. Specimens S65G3 and S85G3 exhibited 27% and 40% higher yielding displacement, and 23% and 41% greater yield strengths, respectively, than their unconfined counterparts. In terms of peak values, the displacements of confined specimens S65G3 and S85G3 raised by 41% and 26%, respectively and the corresponding peak load increased by 59% and 43% in comparison to unconfined specimens. On further increasing the confinement layer from 3 to 5 (i.e., from S85G3 to S85G5), the yield displacement and load uplifted by 11% and 4%, respectively. The peak displacement and load also saw gains of 13% and 6%, respectively. While the rise in peak load was relatively modest, the improvement in ultimate displacement was significant. Compared to the unconfined specimen S85G0, the ultimate displacement increased by 82% in S85G3 and by a notable 127% in S85G5. These results clearly show that GFRP wraps play an essential role in enhancing both the strength and deformation capacity of the columns.

Yield stiffness K₁ represents the secant stiffness of the yield point and the origin which can be calculated by the expression as shown in equation (2). The yield stiffness of confined and uncofined specimens are very close around 5 kN/mm. The application of FRP wrap has neglible influence on secant stiffeness.

The load degradation rate K₂ represents the limit and peak points of the secant stiffness and is a parameter widely used to characterize the ductility performance of test pieces, as presented in equation (3).

Table 3 shows that higher levels of confinement helped reduce the rate at which load capacity dropped after reaching the peak. In comparison to S85G0, the load degradation rate of S85G3 and S85G5 decreased by 34% and 50% respectively. Based on the envelop curve, yield stiffness K₁, load degradation rate K₂, bearing capacity degradation coefficient D_e, ductility factor x_u, and other stiffness and ductility data were calculated and expressed in Table 3. Together, these values offer a well-rounded understanding of how increased confinement improves the structural performance of the specimens under lateral loading.

According to the Canadian standard CSA/CAN S806-12 (CSA S806-12 (R2021)) for FRP-reinforced concrete, the ductility drift ratio of concrete columns reinforced with FRP must not be less than 4%. To evaluate the ability of the test specimens to meet this requirement, the degradation of load-bearing capacity at a 4% drift ratio was assessed using the parameter D_e, which is defined in equation (4), where V_e represents the load-bearing capacity at a 4% drift ratio, which is the average load-bearing capacity at a drift ratio of 4% in positive and negative directions. The essential aim of calculating the degradation load-bearing capacity is to ensure that FRP reinforced concrete structures have sufficient ductility to withstand seismic events and other types of loads. From Figure 8, it is evident that all the confined specimen failed after 4% drift ratio and meets CSA/CAN S806-12 (CSA S806-12 (R2021)). In contrast, the unconfined specimens failed before reaching this limit, as their load-bearing capacity dropped below 85% of the peak, leading to premature termination of the test. Among the confined specimens, the load-bearing degradation of specimen S85G3 is approximately 2.5 times lower compare to S85G5, showing the most stable performance. It implies that increasing the number of GFRP confiment layer helps reduce the rate of load degradation, resulting in more stable and gradual failure behavior.

The drift ratio (x_u) functions as an immediate indicator of the capacity of a structure or element to experience plastic deformation subjected to earthquake load. It is directly related to the deformation capacity of the structure and used in previous studies to describe the ductility of the structure (Vu et al., 2016; Peng et al., 2023). The drift ratio of the column should be considered by the ratio of ultimate displacement to the height of the column, and the calculation expression is shown in equation (5). A higher factor of ductility signifies an increased capability for plastic deformation and the dispersion of energy. It can be observed from Table 3 and Figure 8, the drift ratio of specimens increased as the confinement effect increased to the core concrete. The drift ratio of confined specimens increased in the ranges from 104% to 155% compared to their unconfined counterparts. For instant, the drift ratio specimens S65G3 and S85G3 is 104% and 105% higher than its counterparts (i.e., S65G0 and S85G0), respectively. When compared the specimens with different hoop spacings (i.e., S65G0 and S85G0, S65G3 and S85G3), this ratio was was about 15% higher for smaller hoop spacing. Additionally, the drift ratio increased by 24% when the confinement layer increased from 3 to 5. It clearly demonstrate that the addition of external GFRP wrap played a crucial role in improving the ductility of FRP-reinforced columns, allowing them to meet seismic design standards and improving their ability to sustain inelastic deformation during earthquakes.

Residual drift ratio

The residual drift ratio is generally defined as the ratio of the average of positive and negative residual displacement to the calculated height of the specimen when the horizontal hysteresis force is unloaded to 0 at each level of displacement. Through the residual drift ratio, the self-centering performance of the specimen can be understood, and then the recoverability and functionality of the specimen after earthquake action can be evaluated. It can be calculated using equation (6).

Figure 9 illustrates the relationships between the residual drift ratio and the loading drift ratio of the tested specimens. From Figure 9, it can be seen that with the increase in the loading drift ratio, the residual drift ratio increased. Furthermore, the specimens in Figure 9(a)–(b) were compared and evaluated based on hoop spacing. It is evident that the residual drift ratio of specimens with small hoop spacing at a certain drift level was decreased. For example, the residual drift ratio of specimen S65G0 at a loading drift level of 2.5% (δ = 2.5%) decreased by 39% compared to S85G0, as shown in Figure 9(a). Additionally, it can be seen from Figure 9(b) that the residual drift ratio of S65G3 was slightly higher compared to S85G3 till 2.5% of the loading drift ratio after which the degradation in residual drift ratio was observed. It may be due to the damage experienced by S65G3 at the plastic hinge zone at the initial stage of loading. This specified that decreasing hoop spacing improved the self-centering capacity of the specimens. It is worth mentioning that according to Chinese seismic design standards for buildings (GB 50011-2010 (2016)), the inter-story drift ratio cannot exceed 2%.

OPEN IN VIEWER

Figure 9.

Residual drift ratio.

In addition, Figure 9(c)–(d) compared the performance of specimens based on confinement and the number of layers of confinement applied. From Figure 9(c)–(d) it can be observed that the application of exterior GFRP wraps successfully reduced the residual drift ratio of the specimens. In comparison to S65G0, the residual drift ratio of S65G3 at drift ratios 2% and 2.5% decreased by 51% and 55% respectively. Similarly, the residual drift ratio of specimens S85G3 and S85G5 compared to S85G0 decreased by 76% and 82 % respectively. Furthermore, it can be observed from Figure 9(d) that the δ_r decreased with the increasing number of layers of GFRP wraps. The residual drift ratio of S85G5 at drift ratio 2.5% and 4% decreased by 25% and 43% respectively. It may be due to the following two main factors: (1) the GFRP sheets effectively confine concrete from cracking, ensuring the bond between GFRP bars and concrete and preventing the GFRP bars from compressed and fractured due to concrete crushing; (2) the constitutive action of internal concrete may alter due to GFRP sheets confinement, increasing the peak and ultimate strains, improving ductility of concrete, and reducing the residual drift ratio.

Hysteresis energy dissipation

Hysteretic energy dissipation is an important parameter that determines the performance of structural components under seismic events. Inadequate energy dissipation in a structure can result in an excessive drift ratio required during seismic excitations (Cai et al., 2017). Previous experimental research (Hyung et al., 2008; Saiidi et al., 2009; Bu et al., 2015) has demonstrated that enhancing the self-centering capability of a reinforced concrete (RC) column typically leads to a reduction in its ability to dissipate hysteretic energy. However, the test results from the current study suggested a different conclusion. Figure 10(a) depicts the comparison of the hysteresis loop of specimens S85G0 and S85G5 at a loading displacement level of 45 mm (δ = 2.5%). From Figure 10(a), it can be noticed that the residual displacement of confined specimen S85G5 was much smaller compared to unconfined specimen S85G0 however the area enclosed by those two hysteresis loops was almost the same. Similar results were observed by Wang et al. (Cai et al., 2017). The hysteresis energy of S85G0 and S85G5 dissipated during those two particular cycles were 2.98 kNm and 3.07 kNm, respectively. This outcome may be due to the enhanced load bearing capacity and improved post-yielding stiffness of S85G5 compared to S85G0.

OPEN IN VIEWER

Figure 10.

Energy dissipation capacity.

To further study the hysteresis energy dissipation capacity of tested specimens, energy dissipation to displacement diagram was plotted and presented in Figure 10(b). The cumulative hysteresis energy dissipation capacity represented the summation of the area enclosed by each loop of the hysteresis curve. It is worth mentioning that during the calculation of energy dissipation (E_d) for the displacement level at which two cycles were executed only the area associated with the first cycle was concluded. From Figure 10(b), it can be clearly seen that at the initial stages of loading displacement, the E_d of specimens was generally the same. It can also be observed that at a particular drift ratio level the E_d of the specimens were almost the same or differed by a very small margin because the area enclosed by hysteresis loop at certain level is almost same. In the case of unconfined specimens S65G0 and S85G0 which were compared based on their hoop spacing was observed that the E_d of S85G0 was only less by 6% compared to S65G0. However, comparing the E_d of FCFRCs S65G3 and S85G3, it was found that the E_d of S85G3 was identical with S65G3 till the drift ratio of 4% at which specimen load bearing capacity dropped and the test was terminated but E_d of S65G3 was continued until drift ratio of 5%. Likewise comparing unconfined specimens with confined ones, it was seen that the E_d of confined specimens was 2.5 to 3.6 times the unconfined specimens. This implied that the external GFRP wraps effectively enhanced the energy dissipation capacity even though the residual drift ratio decreased. Similar conclusions were drawn in a previous study (Cai et al., 2017).