为解决参数分类和构建最小误差目标函数中瞬态温度时域过程的表达问题,分别定义了瞬态温度时均量和波动量对修正参数敏感度和分析值与试验值间瞬态温度均方差.利用蒙特卡洛混合方法对敏感参数进行分层修正的结果表明,修正前后瞬态温度分析值与试验值间误差由8％～16％降低至5％以下,分析值的时均量及波动量均有明显改善,且该结果被修正中未使用的试验数据所验证.这说明了修正中瞬态温度描述方式对改善瞬态热分析精度有效,蒙特卡洛混合法分层修正合理可行.%The susceptibilities of the time-average and fluctuation quantity of revising parameters was defined,and the mean square deviation of transient temperature between analysis and experiment values were built,so that the expression about time domain process of transient temperature was solved in parameter classification and the building of the minimum error target function.The sensitive parameters were corrected by the means of Monte-Carlo mixed method according to the classification layer by layer.The correction result shows that the transient temperature error is reduced from 8％ ～16％ to less than 5％,and the timeaverage and fluctuation quantities are improved obviously.Moreover,the conclusion is proved by the experiment data which is not utilized in the correction process.It indicates the proposed method about transient temperature and Monte-Carlo mixed correction method according to the classification layer by layer in correction process is rational with improved transient thermal analysis accuracy.
展开▼
