Updating boring bar's dynamic model using particle swarm optimization

The present article deals with analytical modeling of boring bar dynamics as well as identification of unknown parameters for the dynamic model. Experimental modal analysis is utilized to measure the Frequency Response Functions (FRFs) of cutting tool. Using the analytical methods of modal analysis theory¡ dynamic parameters of boring bar (i.e. natural frequencies¡ damping ratios and modeshapes) are extracted from curve fitting of experimental FRFs. A new physical configuration is proposed¡ in order to accurately estimate the dynamic response of boring bar in time/frequency domains. In the proposed dynamic model¡ boring bar is modeled as an Euler-Bernoulli beam with flexible support and tip mass. The mechanical properties (i.e. modulus of elasticity and density) are considered to be constant along beam length. The flexibility of boring bar''s clamping interface is modeled by linear translational/torsional spring elements. Particle Swarm Optimization (PSO) is utilized to identify the unknown parameters of dynamic model. The parameters include translational/rotational clamping stiffness and dimensionless correction factors for boring bar''s diameter/tip mass. These parameters directly control the mass/stiffness distribution of proposed dynamic model. The FRFs obtained from updated model of boring bar are compared with experimental FRFs. It is shown that¡ by optimal selection of unknown parameters¡ boring bar FRFs can be accurately calculated at any point along its length. Hence¡ by incorporating the dynamic model of passive/active actuator into the proposed dynamic model¡ the stability lobes of dampened boring bars can be predicted.