ABSTRACT
Rotating systems are the main components in turbomachinery, and are presents in several industrial applications, such as oil and gas or energy. The efficiency of these machines is directly dependent on the mechanical characteristics and their dynamic behaviours. Thus it is important to develop specific numerical methods and algorithms that can help to predict and to optimize the dynamic behaviour of these machines. The goal of this work is to implement a mathematical model to simulate and to optimize the dynamic behaviour of rotating machines and to develop an evolutionary optimization method in order to improve their response for a large frequency band, placing the nominal rotational speed as far as possible from resonance regions. A modal analysis was made considering the effects of mass, stiffness, damping and gyroscopic effects, in order to have an optimized reparation of the shaft diameters by using a finite element analysis, considering its components (shaft, disks and bearings). The optimized distribution of the shaft diameters was obtained to maximize the frequency span of two natural frequencies around the operating speed. Constraints related to the total shaft volume and to limit values for the diameters were considered. The optimization problem is solved by using the Bidirectional Evolutionary Structural Optimization (BESO) method. A MATLAB code was developed to simulate and to optimize the rotating systems. The Campbell diagram and the unbalance response of the rotor bearing system has been used to assess the performance of the optimized configuration.
