(核工业理化工程研究院 四所，天津 300180)
摘 要: 针对高强铝合金蠕变数据存在幂律失效而难于拟合的问题，采用Monkman-Grant关系式，在经典时间温度参数模型的基础上，推导4种不同形式的速率温度参数模型。根据高强铝合金7075在34.4~148.9 ℃下的蠕变数据，利用多元线性回归方法，计算得出基于不同形式速率温度参数的材料稳态蠕变方程。结合统计检验方法，评价不同模型下蠕变方程的拟合效果。结果表明：在显著性水平5%的条件下，OSD’模型下稳态蠕变速率实测数据与拟合结果的对数差值不满足正态分布。OSD’、LM’模型下拟合数据的对数值无法完全落入理想拟合线双侧96.5%的置信区间内。RMB’、MH’模型下实测与拟合对数值之间的线性相关系数均大于0.994，对应的残差平方和分别为0.2797、0.2800。基于RMB’、MH’模型的蠕变本构方程可以较好地描述7075铝合金在实验条件范围内的稳态蠕变行为。
关键字: 速率温度参数；高强铝合金；幂律失效；蠕变本构方程；统计检验
（Institute of Physical and Chemical Engineering of Nuclear Industry, Tianjin 300180, China）
Abstract:Based on Monkman-Grant equation and time temperature parameter, four different rate temperature parameter models were proposed to resolve the difficulty of creep data fitting when power-law breakdown appears in high strength aluminum alloy. The 7075 creep data were fitted into different rate temperature parameter models to obtain the steady creep rate equations by multiple linear regression. The effectiveness of model prediction were discussed in comparison of different models by statistical test methods. The results demonstrate that when significance level is 5%, the logarithmic difference distribution of measured and fitted data by OSD’ model do not satisfy normal distribution. The fitted logarithmic creep rate by OSD’ and LM’ models could not completely drop into 96.5% confidence intervals of ideal fitted line. The regression curves are calibrated by logarithmic values of measured and fitted data by RMB’ and MH’ models, which linear correlation coefficients are greater than 0.994. The sum of squared residuals of RMB’ and MH’ models are 0.2797 and 0.2800, respectively. The steady state creep behavior of 7075 aluminum alloy in experimental conditions can be well described by creep constitutive equations of RMB’ and MH’ models.
Key words: rate temperature parameter; high strength aluminum alloy; power-law breakdown; creep constitutive equation; statistical test
