A three-dimensional (3D) geometric component-based explicit method that is focused on solving both single-scale and multiscale hollow structural optimization problems is proposed. The 3D hollow geometric component, which enables the hollow feature during optimization, is generated from the subtraction between the solid geometric components. The geometric parameters, instead of densities at each element, are regarded as design variables. The group of components will be projected onto a fixed grid so that the density field can be constructed. The combination of different components at different scales proves the potential of solving complex structural optimization problems, such as the concurrent optimization of hollow lattice structures. Numerical examples for both single-scale and multiscale problems are delivered to validate the proposed method. The results show that the proposed method can ensure the hollow features of the optimized structure. The method provides a novel direction in 3D multiscale hollow structural optimization.