Application of Grey System Model and Particle Swarm Optimization on CNC Machine Thermal Error Modeling

Based on the complexity of mechanism and variation rules of thermal error on CNC machine, a new grey system model GM(1,1,¿) combined with particle swarm optimization (PSO) was proposed and applied in thermal error modeling on CNC machine. Particle swarm optimization has very strong capacity to solve complicated multi-peaked optimum question. In the new model, the old information was deleted and the new information was entered in time to enhance the precision of modeling, and PSO was used for finding the global optimal point of ¿. Then, one thermal error series is researched using traditional GM(1,1) method, metabolic GM(1,1) model and metabolic GM(1,1,¿)model respectively for modeling and prediction. It can be seen form the analysis results that metabolic GM(1,1,¿) model was proved the most optimal model.