基于自适应Kriging的不确定可靠性功能优化算法研究
投稿时间:2022-06-16  修订日期:2022-09-10  点此下载全文
引用本文:
摘要点击次数: 17
全文下载次数: 0
作者单位邮编
钟维宇 湖南三一工业职业技术学院 410129
马娇 湖南三一工业职业技术学院 
蔡敢为 广西大学机械工程学院 
柳林燕 南京理工大学机械工程学院 
中文摘要:产品研发中功能失效是一个复杂的系统性工程,失效过程包含大量不确定性因素。为此,本文构建自适应Kriging的不确定可靠性功能优化算法,进行产品总体功能失效分析、认知集合可信任度、样本点的生成、自适应Kriging计算及优选功能组合,获取在指定的概率约束下的最优解。以大数定律及极限定理为基础,保证了样本点在重要区域及Kriging模型的收敛条件。以工程机械储能系统为例,说明算法的迭代性、收敛性、准确性及稳定性。结果表明,该算法能够得出准确的敏感度,节省计算时间、提高计算效率。
中文关键词:Kriging模型  认知不确定  功能分析法  可靠性优化  优化算法
 
Research on Uncertain Reliability Function Optimization Algorithm based on Adaptive Kriging Model
Abstract:Function failure in product development is a complex systematic engineering, and the failure process contains a lot of uncertain factors. To this end, this paper constructs an adaptive Kriging algorithm for uncertain reliability function optimization, carries out product overall function failure analysis, cognitive set reliability, sample point generation, adaptive Kriging calculation and optimal function combination, and obtains the optimal solution under the specified probability constraint. Based on the law of large numbers and the limit theorem, the convergence conditions of the sample points in important regions and Kriging model are guaranteed. Taking the energy storage system of construction machinery as an example, the iteration, convergence, accuracy and stability of the algorithm are illustrated. The results show that the algorithm can get accurate sensitivity, save computing time and improve computing efficiency.
keywords:Kriging model  cognitive uncertainty  Functional analysis  Reliability optimization  optimization algorithm
查看全文   查看/发表评论   下载pdf阅读器