Parameter Sensitivity Analysis of Mounting Pedestals and Multi-Objective Optimization for a Multi-Support Rigid Body System

Abstract

In this paper, a parameter sensitivity analysis of mounting pedestals and a multi-objective optimization design for vibration reduction in a multi-support rigid body system, taking an aeroengine-lubricating oil tank supported by multiple mounting pedestals as an example, are conducted based on the third version of non-dominated sorting genetic algorithm (NSGA-Ⅲ) combined with Sobol’s sensitivity analysis method (SSAM). An aeroengine-lubricating oil tank with three mounting pedestals is simplified as a three-support dynamic system, and its dynamics model is established. Several structural parameters of mounting pedestals are taken as the design variables, and the system vibration response and the reaction force of the front and rear mounting pedestals are considered as the objective functions. The first-order results and total sensitivity index of different design parameters for each objective function are obtained via SSAM, and the five most sensitive parameters are selected. Based on the above five design parameters, multi-objective optimization designing for vibration reduction in a simplified lubricating oil tank system is conducted based on NSGA-Ⅲ, and the results of the above triple-objective optimization are obtained as a Pareto-front surface with an obvious frontier. It can be observed from the simulation results that the oil tank vibration of the optimized system is effectively suppressed under the unbalanced excitation of two typical engine speeds. The established method and the main results can provide guidance for designers of aeroengine external structural systems, which can help to achieve superior system dynamic performances in engineering applications.

Keywords

• multi-support rigid body system
• structural optimization
• sensitivity analysis
• Sobol’s method
• multi-objective optimization
• NSGA-Ⅲ