Probabilistic Based Automated Tool for Unbalance Calculation Forecasting

Abstract One of the most critical values to forecast in the production of a Gas Turbine rotor is the initial unbalance of the machined component. Depending on the value of initial unbalance, different types of balancing operations could be needed to target essential rotordynamic performances during engine operation. Therefore, the ability to forecast and optimize the initial unbalance of the disks, wheels and spacers, may bring several benefits for customers, as the decrease of the vibration, and producers, as properly designing a balancing feature or the reduction of several steps of an ordinary low speed balancing procedure. Typically, the definition of an initial unbalance limit is scaled up/down from existing engines where the values of unbalances were measured over time and statistically processed to design a smooth balancing process. The unbalance encountered in a rotor part depends on the asymmetry along its rotation axis which is determined from the accuracy in performing the required machining features. As the worst-case scenario represents an unrealistic approach to set the upper limit of the unbalance, here a probabilistic approach is suggested and compared with the measured data to obtain a good match. The developed tool is then able to determine the unbalance of rotor parts processing machining tolerances in input through a Montecarlo analysis. The article explains how the tool works along with a case study highlighting its application. The findings confirm that this innovative tool can simplify and speed up the definition of machining and unbalance tolerances.