Dynamic Analysis of Bolted Joint Rotor System Subjected to Multiple Nonlinearities

Abstract

Bolted joints are widely used in aero-engine rotors and gas turbines. The dynamic analysis of such joints is complicated to model due to the inherent nonlinearities due to stick-slip friction forces. The nonlinear behaviour included in assembled structures mostly depends on the interface properties. In several systems, the stator-rotor rub impact forces also arise due to several unbalance effects. In this paper, dynamic analysis of rotor bearing system with bolted joint disk system is presented with multiple excitations including contact interface dynamics as well as rub-impact forces and disk unbalances, bearing forces. A finite element model of the rotor system with 2-node Timoshenko beam elements is used to discretize the rotor system and different forces are applied at appropriate nodes in the assembled model. The resulting system of differential equations is solved by fourth-order Runge-Kutta time integration method to obtain the dynamic response. In order to identify the interface and contact stiffness properties using dynamic response, a multilayer perceptron neural network model is trained. The results are presented at different speeds of operation.