Multi-bolt pretightening process optimization method considering structural deformation and its application in computer numerical control machine tools

In precision equipment assembly, the quality of bolted joints is crucial for structural performance and geometric accuracy, yet it often relies on error-prone manual adjustment. This study proposes a deformation-aware optimization method for the multi-bolt pretightening process to simultaneously control accuracy and ensure stiffness. A geometric accuracy prediction model is established to link the dynamically changing bolt preloads (due to elastic interactions) with structural deformation. Concurrently, a layout-independent pretightening sequence constraint model (PSCM) is formulated to generate feasible tightening sequences that guarantee sufficient interfacial contact. These two models are integrated within a hybrid Particle Swarm Optimization-Genetic Algorithm (PSO-GA) framework to optimize the tightening sequence and force distribution. Experimental validation on a large CNC machine bed assembly shows that the proposed method directly achieves the required guideway levelness and reduces straightness and pitch errors by up to 60% and 40%, respectively, compared to the conventional process, demonstrating its effectiveness for precision assembly.