There has indeed been quite a lot of research in this area, as is readily uncovered by searching for "improving the strength of graphene" in your search engine of choice; for example, this 2017 review https://www.sciencedirect.com/science/article/pii/S0079642517300968.
Graphene is often highlighted as having high mechanical strength, but this usually refers to its tensile strength and is not always the critical factor in mechanical applications. Do you only need the material to support tension, or also compression? What about shear stress? Some relevant considerations:
Graphene is a 2D material, only a single atom thick, so in order for it to be useful for macroscopic applications requiring high tensile strength, many sheets would need to be used.
Graphene sheets only interact very weakly, via van der Waals interactions. The sheets can slide very easily (which is why graphite is so good for pencils, lubrication etc) and cannot support even moderate shear stresses (in the out-of-plane direction). If you need high shear strength, then you need to do something to the graphene sheets, for example induce corrugation: https://www.sciencedirect.com/science/article/pii/S0008622320306230
Real graphene sheets are not infinite, they have edges. These edges are chemically reactive and need to be passivated or they will reconstruct. One alternative is to fold the sheet so that, for example, the left-hand edge wraps over and binds to the right-hand edge. This can maintain the pristine structure and high tensile strength of the graphene. The resultant structure is a carbon nanotube, which is one reason why these are of interest in mechanical applications.
Graphene sheets are strong under tension, but buckle easily under compression (nanotubes are stronger in this respect).
Pristine, single-crystal graphene is difficult to create in large quantities, e.g. using CVD, and polycrystalline graphene has a much wider spread of mechanical strengths. Not all of the defects and grains are detrimental, but understanding the nanostructure and implications is challenging for modelling due to the long-range interactions between point defects and the extremely soft flexural phonon modes which lead to large thermal effects (see e.g. https://www.nature.com/articles/nphys3183)
There are many, many more studies in this area, both experimental and modelling, but I hope this has given you a flavour of it at least.