Abstract: In order to determine cutting force coefficients of a ball-end milling cutter， in which cutting edges have differences， a new method combining the average milling force method with particle swarm optimization(PSO)was proposed. Firstly， the milling force models and the identification models based on the average milling force method are constructed ignoring the differences of cutting edges. Then， the algorithm for searching more accurate cutting force coefficients based on PSO was given. The algorithm uses the coefficients obtained by the average milling force method as initial values， correction factors of cutting force coefficients as design parameters， and sum of squared deviations between the minimum cutting force of simulation and experiment as objective function. Finally， a series of simulations and experiments are performed. The results show that the predicted milling forces with corrected cutting force coefficients have a better consistency with measured forces.

Key words: ball-end milling cutter, cutting force coefficients, milling forces, coefficient identification, particle swarm optimization(PSO)