利用系统可靠度约束下的优化设计求解公差分配问题

一个较好的公差分配方案可以有效地提高产品的性价比.然而目前所采用的公差分配方法或者并不能真正反映出产品的性能状况,或者计算的效率较低.文中研究系统可靠度约束下优化设计的序列近似规划求解方法及其在机械公差分配问题中的应用.分析文献中经常出现的算例,最终结果用Monte Carlo模拟证实,基于系统可靠度的优化方法比文献常用的简化方法在公差的分配方面更经济合理,不同算法的比较结果表明计算方法的有效性.结果分析为机械公差分配方法的工程应用提供建议.

TOLERANCE ASSIGNMENT USING ALGORITHM FOR SYSTEM RELIABILITY BASED OPTIMAL DESIGN

An optimal tolerance allotment can increase the perfonnance/eost ratio of a design. Current tolerance allotment methods either cannot really reflect the performnce of a design considering assembly stack-up conditions, or they are time consuming. In order to improve their computational efficiency and accuracy, a sequential approximate programming method for system reliability based optimal design is proposed and applied to the tolerance assignment problems. A series of approximate programming fomulations, where the approximate objective function and reliability constraints instead of the accurate ones are utilized, is established and conducted.The sequential optimums of the approximate programroing lead to the solution of the original problem. Two tolerance allotment examples mostly quoted in the literatures are reanalyzed and the results are checked using Monte Carlo simulation. The comparisons show that the new method for system reliability based optimal design is more efficient and appropriate than its simplified method currently utilized. A suggestion about tolerance synthesis is suggested after the comparisons.