Experimental validation of a robust model-based balancing approach

Balancing procedures are periodically applied to rotating machines to reduce vibration amplitudes, thus keeping the system operating within acceptable safety limits. Various methods have been developed to balance rotating machines, such as the so-called signal-based balancing techniques. This paper presents the experimental validation of a balancing approach dedicated to rotating machines. This approach is based on a representative mathematical model of the system and it first performs the model updating of the machine. The unbalance condition is obtained through the solution of an inverse problem by taking into account the uncertain parameters of the rotor that can affect the balancing result. This robust balancing methodology is performed by considering a mono-objective optimization method in which the uncertain parameters are represented by random variables. In the present contribution, only the unbalance distribution along the rotor is considered as uncertain information. For this aim, uncertainty was modeled as a Gaussian field and was represented by Monte Carlo simulations. The corresponding experimental investigation considers a rotor system composed of a horizontal steel shaft, three discs, and two self-alignment ball bearings. The obtained results demonstrate that the proposed approach is an effective alternative for the balancing of rotating machines.