铝合金腐蚀损伤统计研究

运用LY12CZ的腐蚀实验数据,根据高强铝合金的失效模式:点蚀—晶间腐蚀—剥蚀,建立了对最大腐蚀深度分类的概率神经网络模型,输出结果与实验数据比较吻合。在此基础上,对分类后的最大腐蚀深度的统计研究表明:高强铝合金点蚀的最大腐蚀深度服从Gumbel第一型极值分布,晶间腐蚀阶段服从正态分布和三参数威布尔分布,发生剥蚀后的最大腐蚀深度服从三参数威布尔分布。对高强铝合金的最大腐蚀深度,应用三种或两种分布类型,而非单一分布来表征其分布规律。

A Statistical Study on Corrosion Damage of Aluminum Alloy

Making use of the available experimental data of LY12CZ alloy, a probabilistic neural network is developed to classify the maximum corrosion depth ranges based on the material failure model. The output is in good agreement with experiment result .Statistical study on classified corrosion damage is carried out. The results show that maximum pitting corrosion depth for aluminum alloy is in conform to Gumbel distribution. The normal distribution and three parameters Weibull distribution fit well with period of inter-granular corrosion, and the maximum exfoliation corrosion depth is consistent with the three parameters Weibull distribution law. It may be necessary to use several distribution functions rather than a single distribution to represent corrosion damage characteristics due to the large distribution of maximum corrosion depth in aircraft materials.