自适应神经模糊推理系统在桥梁状态评估中的应用

以钢筋混凝土中小跨径梁式桥为对象,根据结构特征和《公路养护技术规范》(JTJ 073 96)建立了主梁耐久性评估指标体系,确定了评估指标的分级标准,开发了基于自适应神经 模糊推理系统（ANFIS）的主梁耐久性评估系统.提出了根据统计学原理模拟桥梁检查数据和专家意见调查数据的方法,比较了教师数据数量对系统训练结果的影响.模拟教师数据验证结果表明,系统具有良好的学习能力和实际应用能力.应用某市三环路上的13座桥梁110根主梁的实测数据,验证了学习后的系统能够快速有效地模仿专家非线性模糊推理能力.

Application of adaptive neuro-fuzzy inference system in bridge condition state evaluation

An evaluation index system for girder's durability of existing reinforced concrete bridges with medium and short spans was established based on the structure properties and "technical specification of maintenance for highway"(JTJ 073-96).The rating standard of the evaluation index was defined,and a durability evaluation system for girders of existing bridges based on an adaptive neuro-fuzzy inference system(ANFIS) was constructed.A method was proposed based on statistics method to simulate the bridge detection records and opinions of bridge experts.The effect of training data on the system's training result was investigated.The system's training ability and applicability were proved to be good according to the simulated training data.At last,a case study was taken using the experimental results of 13 bridges and 110 girders in the third round road of a city.The system can imitate the nonlinear fuzzy inference ability of experts quickly and effectively,thus the feasibility of this method is validated through the existing bridge.