Abstract

The computer-aided design optimization and vibration analysis, especially avoiding resonance are fundamental challenge to piping system design of complex products. Conventional pipe design methods commonly focus on either automatic geometry layout or vibration analysis, which usually leads to repeated modifications between both disciplines. In this paper, a multi-objective pipe routing algorithm considering vibration performance for aero-engine is presented based on Kriging model and non-dominated sorting genetic algorithm-II (NSGA-II), where pipe length, smoothness, and natural frequency are taken into account. Pipe node positions are optimized by using NSGA-II subject to several constraints. In order to improve computational efficiency, a Kriging model reflecting the relationship between the pipe node position and natural frequency is constructed, and then is applied instead of computer-aided engineering analysis program to evaluate the individual fitness during evolutionary computations. Several numerical computations are performed on a simplified piping model, and the results show that the constructed Kriging model can guarantee high modeling accuracy and the presented routing algorithm is able to obtain a set of non-dominated solutions of pipe layouts, in particular, the natural frequency of pipes are far from engine excitation frequency, which guarantees to avoid possible resonance.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call