中文摘要：

由于CO2浓度、大气温度、碳化腐蚀电流密度和混凝土强度等影响碳化锈胀开裂的随机参数分布具有时间和空间变异性，在基于可靠度的耐久性分析中如何考虑这一特性是目前研究热点和难点。基于随机场理论，使用中点法，通过合理选取单元相关长度和波动系数，发展空间时变可靠度模型来预测RC结构在不同环境条件下将来100年内的碳化腐蚀开始比例和顺筋锈胀开裂比例。研究表明：①碳化腐蚀过程考虑累积效应和温度效应更接近实际退化过程，预测结果更合理；②考虑腐蚀参数空间变异性比完全相关和完全不相关条件下的顺筋锈胀开裂比例分别高22．9％和低32．8％，基于可靠度的耐久性分析需计人材料、几何尺寸以及腐蚀过程的空间变异性的影响；③做好结构防水是减少碳化锈胀损伤的有效措施之一。

英文摘要：

Due to the spatial variability in the distributions of CO2 concentration, temperature, humidity, carbonation- induced corrosion current density, compressive strength of concrete, etc. , nowadays, how to take these parameters into accounts for reliability-based durability analysis remains a hot and difficult research topic. Based on the random field theory, in the present study a midpoint method, which selects the correlation length and scale of fluctuation （0） appropriately, and a spatial time-dependent reliability model are employed to predict the proportion of carbonation- induced cracking under various atmospheric environments over next 100 years. Results indicate that： （1） If the carbonation cumulative effect and temperature influences arc considered, carbonation-induced corrosion process is close to the actual deterioration situation, and the prediction results should be more reasonable; （2） Proportion of corrosion- induced cracking with spatial variability of corrosive parameters considered is 22.9% higher than that predicted from totally correlated cracking or 32. 8% lower than that obtained from perfectly uncorrelated cracking, respectively, indicating that the spatial variability of materials, geometrical sizes and corrosion process should be considered in the reliability-based durability analysis; and （3） To decrease the carbonation-induced corrosion, waterproof is one of the best measurements.