ROBUST OPTIMIZATION OF FLEXIBLE ROTOR SYSTEMS WITH UNCERTAIN PARAMETERS VIA INTERVAL ANALYSIS METHOD

Uncertainties in the input parameters are inevitable in any design process. Along with the demands for higher rotational speed and higher efficiency of rotating machinery, parameter uncertainties (e.g. support stiffness, the effective bending stiffness of connecting structures) resulted from the increasing load on rotor systems lead to significant scatter of its dynamic performance. These parameters are "uncertain but bounded" which means the distributions are unknown, but the intervals are always got easier. This paper presents a method to robustly optimize the dynamic performance of flexible rotor systems taking into account parameter uncertainties via interval analysis method. Interval analysis methods for modal properties and dynamic response behavior of rotor systems are developed with the interval variables introduced into the equation of motion. The aim of the robust design method is to optimize the critical speed margins and dynamic load on bearings, in the meanwhile, minimizing the variability of the objective items by the means of reducing their sensitivity to parameter uncertainties. A numerical example is presented, results show that, for the highspeed flexible rotor systems, the optimal choices of design variables could reduce of sensitivity to rotor parameter uncertainties, thus optimizing the variability of dynamic performance, which has important practical significance in engineering.