基于人工神经网络的混凝土实时强度影响因素敏感性分析

发布时间：2018-03-14 10:41

本文选题：混凝土　切入点：实时强度　出处：《混凝土》2014年11期 　论文类型：期刊论文

【摘要】：准确地预测出混凝土材料在使用过程的实时强度对于正确评估结构安全性有着重要的意义。影响混凝土材料实时强度的主要因素包括龄期、环境类别、水灰比、胶凝材料用量等等。采用人工神经网络进行混凝土实时强度影响因素敏感性分析。首先,对影响混凝土实时强度的各类因素进行分析,确定敏感性因素。其次,针对龄期敏感性因素,建立两个神经网络,一个神经网络的输入变量包含龄期,另一个不包含龄期,将训练好的两个神经网络用同组数据进行测试,比较两组测试结果,以此来确定龄期因素对混凝土强度的敏感性。采用上述方法分别对环境类别、水灰比、胶凝材料用量等因素进行敏感性分析。最后,通过比较确定混凝土龄期、环境类别、水灰比为影响混凝土实时强度的敏感性因素。
[Abstract]:It is important to accurately predict the real time strength of concrete materials in the process of use for the correct evaluation of structural safety. The main factors affecting the real time strength of concrete materials include age, environmental category, water / cement ratio, etc. Using artificial neural network to analyze the sensitivity of concrete real time strength factors. Firstly, the factors affecting concrete real time strength are analyzed to determine the sensitivity factors. According to the age sensitivity factors, two neural networks were established. The input variables of one neural network included the age and the other did not include the age. The two trained neural networks were tested with the same set of data, and the results of the two groups of tests were compared. In order to determine the sensitivity of age factors to concrete strength, the above methods were used to analyze the sensitivity of environmental category, water-cement ratio, cement content and so on. Finally, the concrete age and environmental category were determined by comparison. The water-cement ratio is a sensitive factor affecting the real time strength of concrete.
【作者单位】：辽宁工程技术大学建筑工程学院;
【基金】：国家自然科学基金资助(51008148) 辽宁省教育厅年杰出青年学者成长计划资助(LJQ2013037)
【分类号】：TU528

【参考文献】