摘要
为提高桥梁病害诊断分析的智能化水平,提出了基于神经网络诊断的桥梁病害智能诊断分析方法。利用产生式规则知识表示法,将大量的病害案例、指南规范、专家经验等知识构造成为可用于神经网络学习的规则知识,实现了桥梁病害智能知识库的构建。在此基础上,对BP神经网络智能学习算法进行训练,构建了智能化桥梁病害诊断分析模型,开发了桥梁病害成因分析模块。桥梁病害知识库可以涵盖各类桥梁的常见典型病害成因与处治措施,并且在实践中可以不断更新完善。模块可自动实现桥梁病害的诊断分析,智能诊断准确率达80%以上。所提出的方法将传统的桥梁病害检测分析与人工智能进行跨学科融合,为桥梁病害成因分析提供了创新性的手段和工具。
In order to improve the intelligent level of diagnosis and analysis of bridge diseases,an intelligent diagnosis method of bridge diseases based on neural network diagnosis is proposed.Utilizing the production rule knowledge representation method,a large number of disease cases,guidelines,expert experience and other knowledge are constructed into rule knowledge that can be used for neural network learning,as a result,the intelligent knowledge base of bridge disease is constructed.On this basis,the BP neural network intelligent learning algorithm is trained,the model of intelligent bridge disease diagnosis analysis is constructed,and the analysis module of bridge disease cause is developed.The bridge disease knowledge base can cover the common typical disease causes and treatment measures of various bridges,and can be continuously updated and improved in practice.The module can automatically realize the diagnosis and analysis of bridge diseases,and the accuracy of intelligent diagnosis is more than 80%.The proposed method integrates traditional bridge disease detection and analysis with artificial intelligence,which provides innovative means and tools for bridge disease cause analysis.
作者
徐强
宋建永
赵安
XU Qiang;SONG Jian-yong;ZHAO An(Research Institute of Highway,Ministry of Transport,Beijing 100088,China)
出处
《公路交通科技》
CAS
CSCD
北大核心
2022年第S01期53-58,共6页
Journal of Highway and Transportation Research and Development
基金
交通强国试点项目(QG2021-3-14-1)
关键词
桥梁工程
病害成因
智能诊断
处治措施
知识库
神经网络
bridge engineering
disease cause
intelligent diagnosis
treatment measure
knowledge base
neural network
