摘要
钻芯和回弹是两种常用的混凝土测强方法。为了充分利用这两种测强手段的特点,本文将模糊神经网络应用到混凝土强度综合评定中。由于模糊神经网络具有很强的自学习、泛化和模糊逻辑推理功能,它可以有效地映射出钻芯、回弹数据间复杂的非线性关系。通过对试验数据的仿真计算,其强度预测精度高于常规的综合方法.
Drill and rebound are two common concrete strength testing methods. This paper presents an integrated evaluation approach for concrete strength using Fuzzy Neural Networks （FNN） to take full advantage of the two methods. FNN efficiently maps the complex non-linear relationship between data by drill and rebound methods for its automatic learning, generation and fuzzy logic inference. FNN simulation indicates that the predicted results are more accurate than that estimated with traditional method.
出处
《科技通报》
2006年第5期666-670,共5页
Bulletin of Science and Technology
基金
浙江大学城市学院教师科研基金资助课题
关键词
自适应模糊神经网络
混凝土
强度
adaptive neuro-fuzzy inference system（ANFIS）
concrete
strength
