SPRINGBACK PREDICTION OF COVER PANELS BASED

ON SIMULATED ANNEALING ALGORITHM

LI Fuzhu¹  GUO Yuqin¹  Wang Xiaochun²  SonG Wei³

(1. Technology Institute, Xuzhou Normal University, Xuzhou 221011;

2. School of Mechanical Engineering, Xi’an Jiaotong University, Xi’an 710049;

3. Department of Mechanical Engineering, Naval Aeronautical Engineering Academy, Yantai 264001)

Abstract: From the perspective of the energy being decreased gradually in order to make the substance system transfer to a new state of balance, it is suggested that the simulated annealing (SA) is applied to handle the springback prediction problem of cover panels after being formed according to the intense resemblance of this problem with that of the metal annealing. Therefore, a BFGS-SA hybrid algorithm is designed to solve the problem of the springback prediction of cover panels with many design variables. It combines the Broyden fletcher goldfarb shann (BFGS) algorithm with the SA so that take full advantages of each of them, such as the powerful ability of searching the local optimization solution, the higher efficiency, the simple programming for the former and the general applica bility, the great ability of climbing out the local optima to reach a higher accurate global optimal solution for the latter. And the corresponding optimization model and the cooling schedule are established for the BFGS algorithm. By comparing the results of the simulated calculation with those of the practical measurement at the specified points, the rationality of the proposed schedule and the validity of the BFGS－SA algorithm for the springback prediction are verified.

Key words: Cover panels  Springback prediction  Key words: Simulated annealing  Anneal control strategy

CLC No: TG386.41

Received 20050525,received in revised form 20051102