STUDY OF REINFORCEMENT DESIGN H-PROFILE OF AIRCRAFT WING SPAR

In today's world, with ever-increasing safety requirements, there is a growing demand to maintain or reduce production costs. In aviation, in addition to factors like weight and related variables such as resistance to vibration, corrosion, temperature and other are also considered. The task of this paper is to analyse unconventional designs of wing beams with respect to the current requirements of the aviation industry. In the article, the authors analyse the possibilities of design modification either by adding ribs to the profile, or by changing the cross-section of the profile itself. In practice, such design changes would increase weight, production time and finances, but also increase strength and thus safety. All proposed changes were subjected to strength analyses by FEM (Finite element method) computer simulations. The article output is the selection of suitable designs for further observation and experimental verification to ensure comprehensive results for the possibility of implementation in practice. Despite the non-traditional shapes of the proposed wing beam cross-sections, the authors assume that traditional beam shapes will be gradually modified more efficiently.


Introduction
Historically, the first flight dates back more than 100 years. Even then, aircraft designers were looking for ways to build solid wings. However, in terms of beam strength, the problem is much older. Beams are the basic element of the longitudinal system of the wing. In the case of purely girder wings, they ensure the full extent of the transmission of bending loads; in the case of girder half-shells, they play a significant role in the bending transmission (Kindmann, 1993). Their weight is in the range of 25 -50% of the whole wing weight, while higher values relate to beam wings, lower values to half-shells wings (Chen, C.-C, 2019). According to the structural arrangement, we divide the beams into full-walled and, rarely occurring, lattice beams. Like in aviation, as well as in the civil engineering industry beam with I-or H-profile occur. The area of building constructions inspired authors with the ability to drop the collapse risk by adding supporting elements to the beams, for example in the form of ribbing (Figure 1-right) (Abdel-Ghaffar, 2003). The ribbing allows a more efficient distribution of internal material stresses and prevents the beam from collapsing (Figure 1-left). In both fields, civil and aerospace engineering, the main effort is focused on prevention of bursts, collapse, or any other damage. To ensure the best mechanical properties, it is necessary to know the position of a beam neutral axis. The neutral axis is a line (more precisely in the spatial concept of the area), which changes the sense of its internal axial effective countervailing forces under bending stress ( Figure 3). That means that the best use of the mechanical properties of the beam flange material is in the places furthest from its neutral axis (Nardin, 2009). Assuming the use of the same material and of the same crosssection for both the upper and the lower flange, the neutral axis will lie in the middle of the beam height. For most of the wing beams is, indeed, the material of the two flanges is the same, but usually with different cross-sections. The neutral axis then passes through the common gravity centre of these two flanges and it tends to be closer to the thicker, predominantly of the upper flange (Starší, 2020).
The figure 3 shows a metal embedded beam of a Bf-109 fighter aircraft. The beam has, for the most part of the span, a profile I, near the end of the wing it passes into a profile C. The flanges (1) are made of duralumin L-profiles (3), the cross-section of which decreases towards the wing end. The web (2) is made of duralumin sheet, in the root part thicker, at the end of the wing the sheet is thinner. The rear flange (pictured from the visible side of the beam) degrades from its L-profile to its end in the area between sections A and C only into a vertical shear plate riveted to the web. The front L-profile is terminated even earlier than the rear and passes into the sheet metal flange of the wing end, which is formed by bending the end part of the web into a C-shape. The hinge fittings (5) are made of highstrength steel. The lower hinge is equipped with a ball joint both to reduce the parasitic load on the hinges by operational deformations of the spar, and to facilitate its assembly and disassembly. A number of spacers (6) are arranged along the length of the beam to reinforce the web. These are also used to connect the ribs to the beam. It is worth mentioning the reinforcement (9) of the web in the place of the large hole, which was forced by the cannon installation in the wing. The view of the entire skeleton of the wing is also shown in the Figure 3. The front L-profile is terminated even earlier than the rear and passes into the sheet metal flange at the wing end, which is formed by bending the end portion of the web into the C-shape (Starší, 2020). Along with the size and shape of the flange cross-section, the mechanical properties of the used material, namely its tensile modulus E, contribute to the final beam stiffness (Chu AYT, 2002). In the following chapters we will go through an analysis of proposed new beam types and their properties with respect to the requirements of the smallest possible stress under load and the smallest possible weight.

Materials & Methods
The study focused on the strength analysis of different wing beam is divided into two basic groups. In both cases the beams are the H-profile, but the changes are in additional elements, ribs or in the cross-section shape of the beams. All designs are based on the basic proposal, which shape and dimensions in millimetres are below in the figure 4. Total length of beams is 1500 mm. The material of the simulated beams is the Aluminium 6061 "Dural" with common use in the aviation industry. During the simulations the beams were fixed at one of the ends and loaded with a steady force with different values on the upper beam flange. In the first group of beams were those that were supplemented by ribbing. The following figure 5 shows the first group. From top to bottom goes first the ordinary beam with the dimensions according to the figure 4, followed by the beam with transverse ribbing, the beam with ribs at an angle of 45 ° to each other and the last beam with parallel ribbing at an angle of 45 °. Taking into account, the properties of the neutral axis and the danger of deformation of the beam wall, the authors investigated the modifications by additional elements with creating different cross-sections. The following Figure 6 shows the types of the beams that were examined. The figure 6 Cross-section patterns used for numerical simulations (from left to right -diamond, full dual "o", full "o", hollow "o", rectangular, hexagonal, dual beam The numerical analysis was divided into a total of 16 parts and 4 different loads were applied for each beam type, which were graduated from 200 N, 400 N, 600 N to 800 N. The load was applied perpendicular to the top of the beam as shown in the figure 7.

Results
The results of the simulations are shown in Table 1. From this table it is clear that when comparing the weight, the first group of beams was lighter than the second group. From the point of view of displacement evaluation and internal stresses, however, the results were opposite. The computational simulations show (Table 1) that the best rated beams include "dual beam" and "full dual "o"". For this reason, the team of authors focused on further monitoring the properties of these beams by their combination or dimensional modification, while maintaining a weight of approximately 5 kilograms. Through gradual analysis, we came to the fact that the direct displacement of the elements closer to the flanges affects the overall rate of stresses and displacements under load. The final proposal (Figure 9(b)) is designed so that the "o" elements are shifted to the closest possible distance to the flanges, therefore 43 mm from the centre of the beam. This distance was chosen because of the space for the placement of the structure welds. In addition to this design, another model was investigated with the ribs addition perpendicular to the beam wall (Figure 9(a)). The following Table 2 evaluates the numerical simulations of the additional designs. The final designs from the figure 9 are in the penultimate and last place in Table 2. It is obvious from the design that the additional ribbing and elements influenced the stress distribution in the materials. From the figure 10 (top) it can be determined that the design "Dual Beam (half thickness) + middle ribs 43 mm from centre" has a more even distribution of the stresses across the crosssection. In the proposal "Dual Beam (half thickness) + dual full" O "43 mm from centre" are the stress and forces carried by extra "O" elements in the cross-section. Authors also created combinations of the designs, but numerical simulations did not show significant improvements over the original designs The overview of selected design results is shown in the Graph 1, which contains proposals from Tables 1 and 2. The designs in the Graph 1 are numbered in ascending order from Table 1 (1-11) and Table 2 (12)(13)(14)(15)(16). From Graph 1 it is clear that the best strength characteristics has the design no. 5, i.e. "Dual Beam". Nevertheless, this beam has 16% more weight than the rest. For this reason, this design is inappropriate. From this point of view, however, it is necessary to consider the overall behaviour and susceptibility to fatigue and cracks (Čerňan, 2019 From the comparison of the other beams, the final properties from the figure 9 had the best properties. Therefore, these proposals were evaluated for further research and experimental validation. In addition to the designs, it is necessary to take into account the possibility of corrosion (Janovec, 2020) and fatigue wear. (Čerňan, 2018) In particular, wall thickness is prone to dimensional changes due to corrosion and cracks. However, the use of modern NDT diagnostic methods solves this problem (Janovec, 2019).

Conclusion
The study of the beam numerical simulations was focused on the design possibilities of increasing the wing structure strength. By performed simulations, it was possible to determine the beams with the best results, and subsequently to modify and optimize these designs. The study result are two designs of the H-profile with the double wall and additional elements as shown in the Table 2 and the Graph 1. Current designs are still the subject of research and appropriate optimization for further research. The given results can be experimentally verified and thus prove the correctness of numerical simulations. During the design process, the authors also tried to take into account the production technology of the given beams and the proposed designs to meet sufficient criteria for fast and efficient construction. In the case of the successful experimental verification, these optimized beams would be suitable in the wing construction use and could be considered to be used also in civil engineering, where the inspiration came from.