论文部分内容阅读
为准确评估桥梁的安全性能,根据钢筋混凝土桥梁的结构特征,首先构建桥梁安全性评估指标体系,确定评估指标的分级标准;然后使用径向基函数神经网络(RBF)替代传统的BP神经网络,优化学习速度和适用范围;其次结合自适应模糊推理,建立基于自适应RBF神经网络-模糊推理的桥梁安全性评估系统;最后用该系统评估某钢筋混凝土桥梁的安全性能。示例分析结果表明,大量专家评估意见调查数据,可为评估系统提供足够的输入数据,学习后的系统的输出结果与专家的评估意见误差减小,可用于评估桥梁的实时工作状态。
In order to accurately evaluate the safety performance of the bridge, according to the structural characteristics of the reinforced concrete bridge, firstly, the index system of the bridge safety assessment is established and the classification criteria of the assessment index is established. Then, RBF neural network (RBF) is used to replace the traditional BP neural network, Optimize the learning speed and scope of application; Secondly, combined with adaptive fuzzy inference, establish a bridge safety assessment system based on adaptive RBF neural network-fuzzy inference; finally use the system to evaluate the safety performance of a reinforced concrete bridge. The results of the example analysis show that a large number of expert opinion survey data can provide enough input data for the evaluation system. The error between the output result of the system and the expert’s assessment opinion can be reduced, which can be used to evaluate the real-time working status of the bridge.