Structural Performance of GFRP-Wrapped Concrete Elements: Sustainable Solution for Coastal Protection

Abstract

Protecting coastal regions is crucial due to high population density and significant economic value. While numerous strategies have been proposed to mitigate scouring and protect coastal structures, existing techniques have limitations. This paper introduces a novel approach, SEAHIVE^®, which enhances the performance of engineered structures by utilizing hexagonal, hollow, and perforated concrete elements externally reinforced with glass fiber-reinforced polymer (GFRP). Unlike conventional steel bars, GFRP offers superior durability and requires less maintenance, making it a sustainable solution for any riverine and coastal environment. SEAHIVE^® aims to provide robust structural capacity, effective energy dissipation, and preservation of natural habitats. Although some research has addressed energy dissipation and performance in riverine and coastal contexts, the structural performance of SEAHIVE^® elements has not been extensively studied. This paper evaluates SEAHIVE^® elements reinforced with externally bonded GFRP longitudinal strips and pretensioned GFRP transverse wraps. Testing full-size specimens under compression and flexure revealed that failure occurred when the pretensioned GFRP wraps failed in compression tests and when longitudinal GFRP strips slipped in flexure tests. Strength capacity was notably improved by anchoring the GFRP strips at both ends. These findings underscore the potential of the SEAHIVE^® system to significantly enhance the durability and performance of coastal and riverine protection structures. FEM simulations provided critical insights into the failure mechanism and validated the experimental findings. In fact, by comparing FEM model results for cases before and after applying GFRP wraps under the same compression load, it was found that maximum stresses at crack locations were significantly reduced due to compression forces resulting from the presence of pretensioned GFRP wraps. Similarly, FEM model analysis on flexure samples showed that the most vulnerable regions corresponded to the locations where cracks started during testing.

1. Introduction

1.1. The Context

With the advent of technologies in producing concrete elements, it is feasible to fabricate concrete elements with unique structural shapes. These new structural shapes not only provide a way to implement them in special cases but also help to design structures in optimum ways based on imposed loads, such as flexure and compression. This paper aims to evaluate the structural performance of hollow hexagonal concrete elements with perforation on their sides known as SEAHIVE^®. This geometry has the potential to be used in many marine and riverine applications, such as coastal protection and scour protection of piers, as a sustainable solution.

About 40% of the population in the USA lives near the coast, with the southeast states hosting over 70 million people and covering 29,000 miles of coastline [1]. Coastal areas are vulnerable to natural disasters such as hurricanes and tropical cyclones, which lead to increased wave heights and storm surges. In addition to physical and psychological disruptions, natural hazards are responsible for substantial economic losses [2,3,4]. Catastrophic occurrences like Hurricanes Ian and Michael emphasize the pressing need to explore efficient and cost-effective measures aimed at reducing the impact and risk to coastal states. According to a recent report by the National Centers for Environmental Information, weather and climate disasters in the USA have resulted in costs of about USD 2 trillion since 1980 [5]. Furthermore, over the past decade, tropical cyclones and hurricanes have caused more than 50% of the total damage and over 6500 deaths.

Given that 29% of the total U.S. population resides near the coast, the identification of sustainable solutions to mitigate hazards like flooding and wave overtopping is a critical societal imperative. When it comes to protecting such coastal regions, conventional solutions, such as seawalls or breakwaters [6,7], are rather ineffective in terms of wave energy dissipation. For example, Kraus [8] discussed the challenges related to erosion and seawalls in the Mississippi Sound. The findings indicated that this protection technology is ineffective when concerning wall flanking and backfill loss. Additionally, reflected waves have the potential to stir up suspended sediments, rendering shorelines more vulnerable to erosion. The potential of seawalls to exacerbate wave energy is of concern [9]. Moreover, such measures fail to foster biodiversity; seawalls typically sustain 23% lower biodiversity and 45% fewer organisms compared to natural shorelines [10].

A 2008 joint report from the National Oceanic and Atmospheric Administration (NOAA) and the U.S. Fish and Wildlife Service revealed that the Eastern United States lost over 361,000 acres of coastal wetlands between 1998 and 2004 [11]. The Pew Charitable Trusts reports that 14% of the U.S. coastline is currently armored. In 2015, NOAA projected that if development continues at the current rate, nearly one-third of the shoreline will eventually be protected by hard structures, such as seawalls and bulkheads.

In this context, living shorelines present a transformative solution by integrating erosion control with the preservation of natural shorelines and habitats. The term “Living Shoreline” is used here to refer specifically to the stabilization of estuarine, embayed, or sheltered coastlines using native vegetation [12,13]. This nature-based approach helps in combating shoreline erosion while enhancing coastal resilience through the restoration and maintenance of natural functions [11]. Examples of living shoreline protection approaches include the use of native vegetation, reefs, coir logs, and marsh rock sills.

To address the mentioned issues, efforts have been undertaken to study and develop man-made solutions that can offer effective, efficient, and sustainable protection for shorelines while maintaining a hospitable environment for marine creatures. Among these solutions, Ghiasian et al. [6]. investigated the dissipation of wave energy of new hollow hexagon-reinforced concrete (RC) shapes, known as SEAHIVE^®. This new RC system, with its perforations, effectively dissipates wave energy and is found to be more eco-friendly for fostering natural habitats in protected coastlines [9].

In addition to its application in coastal protection and wave energy dissipation, SEAHIVE^® technology has the potential to serve as a countermeasure against scouring in bridge piers, retaining walls, and riverbed bulkheads. The predominant factors for bridge collapse are attributed to hydraulic causes, such as scour, floods, stream instability, lateral migration, and floating debris [14,15]. Among the mentioned factors, scour is to blame as the most frequent culprit, accounting for about 66% of bridge collapses in North America and Europe [14]. The available data for the United States showed that 20% of bridge collapses are attributed to scour [1,10,16,17].

Different types of scours influence bridges in the erosion process. They are divided into three main categories, namely local scour, general scour, and contraction scour, acting independently or in combination with other hydraulic agents to cause a bridge collapse [18]. The mechanism of local scouring is because the bridge piers impede the flow stream, and this causes large-scale turbulence structures. This turbulence not only exacerbates the turbulence in the flow downward towards the bed (i.e., horseshoe vortex) but also increases the turbulence behind the bridge pier (i.e., wake vortex) [19]. It can be concluded that the main root of local scouring is attributed to the hydraulic structures’ interference by obstructing the natural flow field [20]. When it comes to general scour, it can be categorized as either prolonged or short-term erosion. The root of this mechanism involves removing the sediments from the bed river and bank river from the width of the channel [21]. As opposed to the local and contraction scour, the bridge pier is not a factor in causing this mechanism. As for the contraction scour, this is attributed to the acceleration of the flow due to the obstruction and contraction, such as a bridge. This type of scouring is narrowed down to the length of contraction [21,22].

Among scour categories, local scour is typically regarded as the primary type of action that impacts bridge stability. It occurs around the piers and abutments of the bridge, where water flow accelerates, resulting in significant local erosion. Local scour can severely compromise the bridge foundation, making it a key concern in the design and upkeep of bridges. While general and contraction scours are also relevant, local scour represents the most immediate threat to the integrity of a bridge [23].

1.2. Proposed Solution—SEAHIVE^®

To protect the bridge piers from scouring, incorporating SEAHIVE^® technology is proposed. The presence of perforations on the surfaces of hollow hexagons holds the potential to alter hydraulic flow dynamics in areas surrounding a pile cap, as well as within the interstices between piles, where contraction and local scour may occur [6].

Although the capabilities of SEAHIVE^® have been well-documented in terms of wave energy dissipation, further evaluation is needed to understand its structural performance, particularly in relation to the impact of various fabrication methods used for these units. Drawing from the existing literature, insights can be gathered from research on the flexural behavior of solid cross-section beams that have openings on their sides. Design guidelines have been proposed [24,25,26] to help determine the optimal sizes and placements of web openings. Furthermore, Tan and Mansur [27] addressed both ultimate and serviceability limit states in their guidelines for designing RC beams with large web openings. Daniel and Revathy [28] found that beams with rectangular openings experience a notable decrease in both ultimate flexural resistance and stiffness when compared to solid beams. Al-Sheikh [29] investigated the flexural behavior of RC beams with circular, square, and rectangular openings, concluding that beams with a single circular opening performed best, with ultimate capacity reductions of about 1.5% for small openings and 10% for large openings. Aykac and Yilmaz [30] determined that circular openings are more effective than triangular openings in terms of ductile behavior. Moreover, it was found that introducing large openings without adequate internal reinforcement could significantly reduce ultimate capacity [19]. However, providing sufficient diagonal reinforcement around the openings prevented the shear failure of the web posts and premature failure of the beam [31]. Despite the mentioned research on the flexural performance of perforated solid cross-section beams, there is a lack of data on the performance of the perforated hollow units with a hexagonal cross-section.

When it comes to the fabrication method of SEAHIVE^® units, they can currently be carried out by at least three different methods, namely wet-cast, dry-cast, and 3D printed methods. The main focus of this research is on the potential of the dry-cast method. Dry-cast concrete consists of using zero slump concrete (See Figure 1a), which is formed with various methods and immediately removed from its forming setup and let to cure. The very first dry-cast product was concrete blocks and concrete pipes at the very beginning of the 1900s (See Figure 1b), with the method extending to structural applications like T-joists (See Figure 1c) in the 1930s and prestressed concrete hollow core slabs in the 1950s (See Figure 1d).

The dry-cast method is normally used for high-volume production since it requires a high capital investment, but it permits much higher productivity compared with traditional wet-cast processes (wet-cast products are manufactured using fluid concrete that is poured into forms for shaping and curing). Another significant advantage of the dry cast with an immediate stripping process is the product strength. This happens for two main reasons: the first is because the water-to-binder (w/c) ratio of the concrete never exceeds 0.4 (most common is 0.3) for process reasons, and the second is because the packing action is normally one order of magnitude higher than what is obtainable with traditional wet-cast concrete.

When it comes to reinforcing SEAHIVE^®, using conventional materials such as black steel bars is not a sustainable solution due to the vulnerability of steel to corrosion. For example, the portion of the seawalls exposed to tides and waves becomes a weaker link, reducing their service life. As an alternative, SEAHIVE^® can be reinforced by either external GFRP wraps or internal GFRP bars [9,32,33,34,35,36,37,38]. Many research projects have investigated the effect of external wrapping of the concrete elements by FRP materials. The results showed that the effect of external reinforcement increases the compressive strength of the concrete column [39,40].

Because of the expenses related to providing equipment for fabricating concrete elements with the dry-cast method, this research is initially focused on evaluating the structural performance of wet-cast SEAHIVE^® reinforced externally with GFRP strips and wraps. The outcomes of this study are, however, applicable to concrete units produced by the dry-cast method since structural performance is almost solely affected by the methodology of reinforcement consisting of externally bonded longitudinal GFRP strips and pretensioned transverse GFRP wraps (See Figure 2).

2. Experimental Work

2.1. Test Specimens

To assess the capability of SEAHIVE^®, four specimens underwent structural testing: two subjected to pure compression and two to flexure, respectively. The configuration and preparation details of each specimen are outlined in

Table 1. Configuration of SEAHIVE^® samples.

Specimen ID	D/t*	Loading Type	Specimen Length (mm)	Characteristics	Objectives
CS-1	0.25	Monotonic quasi-static pure compression	910	Hollow unit with circular perforations (203-mm dia.) reinforced with externally bonded GFRP	Study the effect of pure compression and bending on the structural performance of SEAHIVE®
CS-2
FS-1		Monotonic quasi-static flexure	1830
FS-2

Notes D/t* = side perforation diameter to total height of the unit. CS and FS in the table mention compression and flexure samples, respectively.

and illustrated in Figure 3.

The total span length used for the flexural test was 1.83 m, as depicted in Figure 3a, a dimension aligned with production specifications. As for the compression test, specimens were cut in to two parts by saw-cutting, resulting in two elements, each with a length of 0.91 m. The diameter of the holes on the hexagon was 203 mm, spaced at a center-to-center distance of 346 mm, as illustrated in Figure 3a. Each leg of the GFRP-RC hexagon measured 440 mm in length and 127 mm in thickness. The overall depth of the cross-section was 792 mm, as depicted in Figure 3b. The SEAHIVE^® elements were transversally pretensioned with resin-impregnated fiberglass, as illustrated in Figure 3c. The width of prestressing wraps and the distance between them equals 50 mm and 40 mm, respectively (See Figure 3a).

For the SEAHIVE^®, GFRP strips made of resin-impregnated fiberglass roving were applied to each edge using the wet layup method, with 10 roving applied to each edge (about 10 mm² cross-section area of GFRP pack). Subsequently, GFRP wraps made of resin-impregnated fiberglass roving were applied around the hexagonal section, with 250 N of tension applied to each roving during the wrapping process. Each wrap was positioned as close as possible to the holes to compensate for the lower shear capacity at these locations. In addition to compressing the concrete, these wraps have the function of anchoring the longitudinal strips by enhancing the bond between strip and concrete. This is because the pretensioned wrap creates a compressive force at each corner between strips and concrete, thus significantly increasing the shear force capacity between them (See Figure 3d). For each wrap, 80 turns of 2400 Tex E-glass roving were applied (about 80 mm² cross-section area of GFRP pack), with each wrap tensioned to a total of 20 kN. The 250 N tension applied to each 2400 Tex roving is approximately 20% of its ultimate tensile capacity (refer to

Table 2. Material properties of fiberglass and concrete.

Designation	Specification	Density (g/cm3)	Elastic Modulus (MPa)	$\mathbf{Ultimate} \mathbf{Tensile} \mathbf{Strength} ({\mathit{f}}_{\mathit{f}\mathit{u}}$) (MPa)	Ultimate Tension Force (N)	Ultimate Strain	Concrete Strength (MPa)
Fiberglass	2400 Tex	2.54	81,200	1280	1250	0.034	N/A
Concrete	C30/37	2.5	32,837	N/A	N/A	0.003	30.0

), which is below the limit of creep rupture for GFRP estimated at 30% of ultimate [41].

2.2. Materials

2.2.1. GFRP Characterization

The fiberglass used for wrapping is 2400 Tex E-glass, which was paired with a high-modulus epoxy Elan-tech^® EC 152/W 152 MR. The specifications from the manufacturer are based on ASTM D1475-13 [42] for resin and ASTM D2343-17 [43] for fiberglass. The minimum guaranteed tensile capacity is 0.4 N/Tex, with a typical strength of 1151 N for a 2400 Tex epoxy-impregnated roving. The typical Young’s modulus for the impregnated roving is 81.2 GPa, based on tests conducted on a single roving.

To obtain accurate data on the mechanical performance of the GFRP strips and wraps, experimental testing was conducted in collaboration with the University of Bologna. The tensile properties of the GFRP samples were determined following the provisions of ASTM D2343 [43]. For this characterization, three cylindrical samples were wrapped with 40 turns of fiberglass roving (see Figure 4a), each subjected to a pretension of 250 N. During the test, the cylinders, separated into halves, were tensioned till GFRP failure. After three repetitions, the average ultimate tensile strength was found to be approximately 100 kN. The standard deviation of the samples from these tests was found to be 29.3 kN. Given that there were 40 turns, this equates to approximately 1250 N per roving (refer to Figure 4b).

2.2.2. Concrete Characterization

The manufacturer of the concrete used in constructing the SEAHIVE^® specimens declared, based on their internal testing and quality control procedures, that the concrete used is of type C30/37. This classification aligns with the typical standards for construction-grade concrete. According to EN 1992-1-1 [35,45], the characteristic compressive strength ($f_{ck}$) of this type of concrete is 30 MPa, with a tensile strength ($f_{ct}$) of 2.9 MPa for analysis purposes (refer to Table 2).

2.3. Test Set-Up and Instrumentation

2.3.1. Half Unit Under Pure Compression

The experimental configuration for the compressive test is illustrated in Figure 5a,b, for specimens measuring 910 mm in length. To ensure an even distribution of the applied load from the hydraulic jack across the surface of the element, two masonite sheets measuring 914 × 457 × 5 mm were placed on the element’s surfaces. Following this, two steel plates measuring 900 × 400 × 25 mm were positioned, one on top and one on bottom of the specimens. Subsequently, load cells were situated over the sample, and the load was applied using a hydraulic jack.

2.3.2. Unit Under Flexure

Two full-size units were tested under four-point bending to investigate the flexural behavior of the specimens. As shown in Figure 6a–c, the distance between a support and a loading point (i.e., shear arm = a) is 627 mm, while the total height of the sample (i.e., h) is approximately 792 mm. The a/h ratio is about 0.8 (See Figure 6c), indicating that the unit being tested can be characterized as a deep beam. Two steel plates (25 × 25 × 700 mm) were used as the loading knives positioned over the center of the middle holes.

2.4. Loading Protocol

Four specimens underwent testing: two were subjected to pure compression, and two to flexural testing. Quasi-static load protocols were applied for both compression and flexural tests, with each specimen loaded monotonically until failure occurred.

3. Results and Discussion

This section provides information on experimental results regarding crack patterns and load–displacement curves for both pure compression and bending tests. The first cracking load, ultimate loads, and maximum displacement or mid-span deflection were recorded.

3.1. Compression Test Results

3.1.1. Crack Pattern and Failure Mode

In Figure 7a, the CS-1 specimen is shown at the point of failure. Cracks are evident along the mid-length of the horizontal section and at the corners of the inclined legs. Initially, horizontal cracks developed, followed by inclined cracks likely attributed to shear. The failure occurred as the horizontal cracks extended, ultimately connecting with the holes in the hexagon. Similarly, the crack pattern observed in the CS-2 specimen (depicted in Figure 7b) closely resembled that of the CS-1 specimen.

The presence of external reinforcement effectively prevented the sudden failure of the samples after the first cracks were initiated. It seems the compressive pressure induced by the external reinforcement significantly increased the strength of the specimens under pure compressive load. The pretensioned wraps created a compressive force along the edges of the SEAHIVE^®’s hexagonal structure. This mitigated the effect of tensile stresses on the specimen. As a result, the ultimate strength of the SEAHIVE^® became dependent on the GFRP’s ultimate strength.

During the test, the SEAHIVE^® successfully resisted the applied force until some of the glass filaments reached their limit and broke with concrete cracks that began to propagate. The first cracks appeared at the outer points of the SEAHIVE^®, where the applied load generated maximum moments. Finally, two additional cracks appeared in the middle of the SEAHIVE^® top and bottom legs due to the deformation of the specimen around the center.

3.1.2. Load–Displacement Curves

Figure 8 shows the load–displacement diagram of CS-1 and CS-2 tested under pure compression. The horizontal axis indicates the average displacement, and the vertical axis shows the applied load. The first crack occurrence is shown in the diagram by a load drop in both CS-1 and CS-2 around 145 and 172 kN, respectively. After that, the specimens continued to carry the load until GFRP failure (about 358 kN for both).

3.2. Flexure Test Results

3.2.1. Crack Pattern and Failure Mode

Figure 9a,b show the failed specimens tested under bending. The cracks started to propagate in the middle of the SEAHIVE^® at the holes and extended directly upwards. This crack pattern suggests failure due to the bending moment in the area where the moment of inertia is minimal due to the presence of holes. When the longitudinal GFRP strips placed on the bottom (first) and mid-height (second) corners could no longer withstand the tensile stress, they failed and caused the collapse of the specimen.

3.2.2. Load–Displacement Curves

Figure 10 shows the load–displacement diagram for FS-1 and FS-2, where the horizontal axis represents the average displacement, and the vertical axis indicates the applied load. According to the experimental data, the first drop in load for specimens FS-1 and FS-2 occurred at approximately 81 and 86 kN, respectively. Additionally, the final loads for FS-1 and FS-2 were 227 and 315 kN, respectively. The difference between the two results is due to the slipping of the longitudinal strips. In the FS-1 test, the GFRP strips held together until 231 kN but then suddenly slipped due to inadequate anchorage provided by the transverse wraps. Moreover, in this test, the longitudinal strips were adhered to using the layup method without any anchors at both ends.

In the FS-2 specimen, to overcome slippage, the longitudinal strips were anchored at both ends of the SEAHIVE^®. A slot was created at each end, and during the placement of the longitudinal fibers using the layup method, each fiber filament was wrapped around the slot. As a result, the pack of fibers became securely anchored at each end (See Figure 11). During the test of the FS-2 specimen, at 3.9 mm displacement, some of the longitudinal strip partially broke, resulting in a loss of load capacity. However, the remaining strip portions continued to hold, allowing the specimen to sustain a higher load. After a peak load was reached, fibers started to break gradually until sudden failure. Overall, the SEAHIVE^® in FS-2 withstood up to 315 kN before complete failure, which is significantly higher than the load in the FS-1 test.

Table 3. Results of the experimental test on specimens.

Load or Deflection	CS-1	CS-2	FS-1	FS-2
First cracking load (kN)	145	172	73	75
Ultimate load (kN)	360	354	227	315
Maximum deflection (mm)	17	15	8.3	14.68

summarizes the results of the experiments on four tested specimens.

3.3. Analysis

3.3.1. Specimens Under Compression

It was observed in Figure 12 that the first cracks appeared at the end of the inclined leg. The failure at this location resulted from the combined effect of axial and flexural stresses resulting from applying pure compression and prestressing load resulting from the GFRP wraps (See Figure 12).

The effective moment of inertia of the leg cross-section was calculated by subtracting the moment of inertia of two holes from the gross cross-section moment of inertia of the solid leg resulting in $\mathrm{I}=\mathrm{155,335,710} {\mathrm{m}\mathrm{m}}^4$. The corresponding cross-sectional area equals $A=\mathrm{115,570} {\mathrm{m}\mathrm{m}}^2$. The first visible cracks on the specimen appeared at a load between 145 and 172 kN, as per Table 3. Therefore, the load on each leg was assumed to be equal to about 72.5 kN. The moment at the mid-end of the leg was calculated by multiplying the applied load with its eccentricity (e) assumed to be e = 144 mm (See Figure 12). The stress at cracking was calculated as ${\sigma }_b=-8 \mathrm{M}\mathrm{P}\mathrm{a}$.

The tensile modulus of rupture in concrete, inclusive of the effect of prestressing, was calculated to be equal to 4.08 MPa, showing that capacity is lower than applied stress (i.e., 8 and 4.08 MPa), thus causing the first crack that eventually led the failure. Even though not strictly applicable to this case, ACI 440.11-22 [46] provisions were used to calculate the moment capacity of the section ($M_n$) compared to the cracking moment ($M_{cr}$). Moment capacity equals 5361 kN.mm, and cracking moment equals 5152 kN.mm, indicating that a brittle failure would be expected. To evaluate the performance of the element under pure compression, a Finite Element Method (FEM) analysis was performed using the full SEAHIVE^® model, excluding the influence of GFRP strips and wraps. In Figure 13a, the mesh distribution on the SEAHIVE^® is illustrated, with mesh sizing based on the blended curvature-based method, varying between 11.5 mm and 152.9 mm.

As shown in Figure 13b, the SEAHIVE^® subjected to a compressive load of 700 kN, applied at the top surface, demonstrated significant stress. This load is double the ultimate load endured by the SEAHIVE^® in the experimental setup (Table 3), as the FEM model encompasses the full-length SEAHIVE^®, while the experimental tests utilized a half-length specimen. The model was fully constrained at the bottom.

Since this analysis excluded the pre-tensioned fibers, the SEAHIVE^® experienced peak stress along the outer edge of the hexagonal inclined legs, with values reaching approximately 10 MPa. These stress concentrations align with the regions where cracks were observed to initiate in physical specimens during testing.

Figure 13. (a) Mesh and (b) stress distribution on the SEAHIVE^® under compression load without the wrapped fibers around. When incorporating the effect of the GFRP wrapping around the element (consisting of 80 turns of glass fibers tensioned at 250 N each, resulting in 20 kN of pre-tensioning), these forces were applied at the corners and along each inclined leg. As shown in Figure 14, the SEAHIVE^® experienced maximum stress along the outer edge of the inclined legs. The blue regions in Figure 14 indicate the maximum stress zones, which align with the locations where cracks initiated during the experimental tests. In these areas, the P1 stress (normal stress in the first principal direction) reached 3.9 MPa. Given the characteristics of C30/37 concrete, the specimen failed in these regions. Additionally, as illustrated in Figure 15, the shear stress reached its peak value of 5 MPa.

Figure 13. (a) Mesh and (b) stress distribution on the SEAHIVE^® under compression load without the wrapped fibers around. When incorporating the effect of the GFRP wrapping around the element (consisting of 80 turns of glass fibers tensioned at 250 N each, resulting in 20 kN of pre-tensioning), these forces were applied at the corners and along each inclined leg. As shown in Figure 14, the SEAHIVE^® experienced maximum stress along the outer edge of the inclined legs. The blue regions in Figure 14 indicate the maximum stress zones, which align with the locations where cracks initiated during the experimental tests. In these areas, the P1 stress (normal stress in the first principal direction) reached 3.9 MPa. Given the characteristics of C30/37 concrete, the specimen failed in these regions. Additionally, as illustrated in Figure 15, the shear stress reached its peak value of 5 MPa.

Figure 14. Normal stress distribution on the SEAHIVE^® under compression load with interaction of the GFRP wrapping.

Figure 14. Normal stress distribution on the SEAHIVE^® under compression load with interaction of the GFRP wrapping.

Figure 15. Shear stress distribution on the SEAHIVE^® under compression load with interaction of the GFRP wrapping.

Figure 15. Shear stress distribution on the SEAHIVE^® under compression load with interaction of the GFRP wrapping.

By comparing the FEM analysis results for the two cases—before and after the application of GFRP-wrapped fibers under the same compressive load—it is clear that the maximum stress at the crack points in the SEAHIVE^® was significantly reduced due to the compression forces applied by the prestressed GFRP fibers at each corner. Although the crack locations remained the same, the stress values were significantly lower. This reduction in stress is a key reason why, in the experimental tests, the wrapped specimen under pure compression was able to resist a load of up to 360 kN.

3.3.2. Specimens Under Flexure

Gohari’s findings [36] indicated that when the diameter of holes in concrete beams subjected to bending exceeds 0.25 times the total depth of the beam, traditional beam theory becomes inapplicable. In this study, with a hole diameter of 203 mm and eight holes on each side of the unit (four stacked vertically), the ratio of hole diameter to the total depth of the concrete exceeds 0.25, surpassing the limit set by the referenced study. The presence of holes on all surfaces of the hexagonal, hollow specimen further complicates the analysis. Moreover, the shear span-to-depth ratio of the SEAHIVE^® was 0.8, indicating that conventional beam theory is unsuitable for this analysis.

To evaluate the element’s performance under flexural load and estimate the crack pattern, an FEM analysis was performed using the full-length SEAHIVE^® with GFRP reinforcement. The aim was to assess the maximum stress on the specimen and gain insight into the stress distribution on concrete. The load applied to the FEM model was 250 kN in total, with 125 kN applied at the center of each of the two middle holes. According to Figure 10, this is where the concrete in the experimental specimen under flexural load began to crack and fail. After this point, the concrete lost its resistance due to cracking, and the GFRP became responsible for holding the structure together. The experiment shows that the specimen continued to resist the load until the GFRP strips at the bottom of the SEAHIVE^® eventually failed, which occurred at around 310 kN (Table 3). The model was fixed at both ends with spherical joints, simulating the actual test conditions, where the specimen was supported on two cylindrical supports. The mesh distribution and seeding on the SEAHIVE^® under flexural loading are identical to those used for compression loading (Figure 16).

Figure 16. Mesh distribution on the SEAHIVE^® under flexure.

Figure 16. Mesh distribution on the SEAHIVE^® under flexure.

Figure 17a,b show the P1 (first principal) stress and shear stress distributions on the specimen. As depicted in these figures, near the holes, particularly in the direction toward the center of the applied load, the P1 stress reaches approximately 30 MPa, which is close to the breaking point of the concrete and marks the location where cracking initiates. Additionally, the tensile stress at the bottom surface of the unit, where the bending moment is largest, experiences the same stress magnitude. In the same region, the shear stress reaches approximately 3 MPa.

By generating a detailed contour of the stress levels, the FEM highlighted the most vulnerable regions where cracks were likely to start. This method provided critical insights into the failure mechanism and validated the experimental findings.

Figure 17. Stress distribution on the SEAHIVE^® under flexure (a) P1 stress, and (b) shear stress of specimen with wrapped GRRP.

Figure 17. Stress distribution on the SEAHIVE^® under flexure (a) P1 stress, and (b) shear stress of specimen with wrapped GRRP.

4. Conclusions

The study was conducted to analyze the structural performance of SEAHIVE^® reinforced with externally bonded GFRP longitudinal strips and pretensioned transverse GFRP wraps. This reinforcement methodology applied to dry-cast concrete appears to be effective. The study examined the structural behavior of hexagonal, hollow, and perforated units under both pure compression and flexure. Based on the outcomes of this study, the following conclusions are drawn:

• This method of casting and reinforcing presents several potential benefits, including the use of relatively inexpensive raw materials, easy automation, high productivity, and reduced need for reinforcing material due to the tensioned application of GFRP wraps on the product surface.
• The experimental results closely matched the predictions from the Finite Element Method (FEM) analysis, validating the accuracy and reliability of the FEM as a design tool for optimizing the reinforcing process in terms of amount as well as pre-tensioning.
• The pretension wrapping method enhances the bond between the longitudinal GFRP strips and the concrete, delaying the onset of catastrophic failure.
• Comparing the maximum tensile stress generated in samples during testing and the tensile modulus of rupture in concrete shows that causing the first crack in an element results in failure.
• Comparing the nominal moment capacity of the section ($M_n$) and the cracking moment ($M_{cr}$) shows that the brittle failure of the elements is expected.

Overall, these findings underscore the effectiveness and potential advantages of employing the externally wrapped composite method to enhance product strength and efficiency in diverse applications.