硫酸盐腐蚀混凝土的热重分析结果与其中硫酸根离子含量之间的相关性研究  

Study on the correlation between the test results of thermogravimetry and sulfate content of sulfate attacked concrete

作　　者：陈记豪[1] 赵顺波[2] 姚继涛[1] 贺瑞春 

机构地区：[1]西安建筑科技大学土木工程学院,陕西西安710055 [2]华北水利水电学院土木与交通学院,河南郑州450011 [3]珠江水利委员会珠江水利科学研究院,广东广州510610

出　　处：《西安建筑科技大学学报（自然科学版）》2010年第6期835-840,共6页Journal of Xi'an University of Architecture & Technology(Natural Science Edition)

基　　金：河南省杰出青年科学基金(04120002300);河南省外国专家局引进国外技术;管理人才重点项目(20094100099)

摘　　要：硫酸盐腐蚀研究中对遭受腐蚀后的混凝土进行热重分析和硫酸盐含量检测是常用手段,二者同属于对腐蚀产物的研究方法,其结果间关系如何?尝试采用直线、S曲线及BP神经网络对混凝土受硫酸盐腐蚀后的硫酸盐含量值和热重损失值进行了回归分析,并得到了二者的关系式.结果表明:硫酸盐含量和热重损失量具有较好的相关性,相关系数达到了0.72,BP神经网络拟合效果最好,但是预测能力稍差,S曲线拟合及预测均较好,直线拟合与预测效果均最差.Thermogravimetric analysis technique and sulfate content determination are commonly used in the research on sulfate attack on concrete.They are used to study the corrosion products and the relationship between the measurement results in the paper,the linear regression,S curve fitting and BP neural network fitting are used to regressive analysis of the thermogravimetry and sulfate content of the sulfate-attacked concrete.The results showed that the sulfate content has good correlation（correlation coefficient is 0.72） with the thermogravimetry;and the BP neural network has the best fitting effect,but its predictability for outside samples is bad.The S curve fitting has good fitting effect and high prediction accuracy while the fitting and prediction effect of the linear regression are worse.

关 键 词：混凝土 硫酸盐腐蚀 硫酸根离子含量 热重损失 拟合 预报 

分 类 号：TU528.01[建筑科学—建筑技术科学]