Abstract: Severe temperature changes during the rapid melting and solidification of 3D printing metal powder inevitably produce residual deformations. The accumulated excessive residual deformation may lead to deformation, distortion, even cracking and failure while printing workpiece. By considering this “manufacturing constraints” of metal 3D printing in the structural design stage, the residual deformation in the 3D printing process can be controlled effectively upon the topology optimization method. Through the elastoplastic thermo-mechanical coupling finite element analysis of the metal powder melting process in two adjacent printing layers, the equivalent inherent strain load in printing process is extracted, and then the 3D printing process simulation and residual deformation prediction can be quickly realized. A followed 3D printing experiment is carried out to validate the prediction method. Through introducing the P-norm of global residual deformation and deducing its sensitivity, a multi-load case multi-objective optimization algorithm is proposed to improve the design and manufacturing performances, namely minimize structural compliance and 3D printing residual deformation simultaneously based on the floating projection topology optimization (FPTO) method. At last, several 2D and 3D examples are conducted to demonstrate the effectiveness of the method proposed.