首页> 中文期刊> 《中国机械工程》 >基于盲数理论的萤火虫神经网络失效征候结构可靠度预测

基于盲数理论的萤火虫神经网络失效征候结构可靠度预测

         

摘要

Under any feasibility conditions,the reliability of large arm structure with failure symp-toms was difficult to be determined due to various uncertain factors of load effects and resistances in notch-crack propagation.The firefly neural network prediction method of the reliability of structure with failure symptom was presented based on the blind number theory by building a blind number fail-ure symptom structure reliability model (BNFSSRM),serial type firefly neural network prediction model consisting of far field stress prediction sub-model (STFNN-FFSPS)and reliability prediction sub-model (STFNN-RPS)of damaged structure.On the basis of the stress-strength interference mod-el,the firefly neural network being as a forecasting method,the idea converting the uncertainty prob-lems to blind numbers was introduced into the interference model of intensity factors of notch-crack and fracture toughness according to the rule of notch-crack instability propagation,and the reliability of notch-crack damaged structure was obtained by the experiments and finite element simulation com-bined with STFNN-FFSPS under typical working conditions.The reliability being taken as the target output and the notch-crack size being as the input data of expanded samples,the real-time prediction of reliability of damaged structure may be achieved by BNFSSRM and STFNN-RPS,thus realizing the evaluation of the capacity of the structure resistant to failure and determining the orders of different conditions out of the feasible condition regions.QY130 remanufacturing arm structure was taken as an example and the effectiveness of the proposed method were verified.%在任意可行性工况下,针对大型臂架失效征候结构的载荷效应与抗力的多种不确定因素导致结构可靠度难以确定的问题,通过构建盲数失效征候结构可靠度模型(BNFSSRM)、串行式萤火虫神经网络预测模型——缺陷结构远场应力预测子模型(STFNN-DSSSPSM)与缺陷结构可靠度预测子模型(STFNN-DSRPSM)串联,提出了基于盲数理论的萤火虫神经网络失效征候结构可靠度预测方法.以应力强度干涉模型为基础,萤火虫神经网络为预测方法,根据裂纹失稳扩展准则,将不确定性问题盲数化的思想引入裂纹强度因子与断裂韧性的干涉模型,通过试验仿真与STFNN-DSSSPSM,得到典型工况下裂纹缺陷结构的可靠度;再以此为扩展样本的目标输出、裂纹扩展尺寸为扩展样本的输入,通过BNFSSRM、STFNN-DSRPSM,实时预测不同工况下缺陷结构的可靠度,评估抵抗失效的能力以及不同工况退出可行性工况域的先后顺序.以QY130流动式起重机再制造臂架结构为例,验证了该方法的有效性,为结构再制造准入期实时判断以及再制造方案的选择提供了理论指导.

著录项

相似文献

  • 中文文献
  • 外文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号