针对网级评估的区域桥梁退化建模与演绎应用

来源期刊:中南大学学报(自然科学版)2021年第3期

论文作者:孙利民 夏烨 雷晓鸣 王鹏

文章页码:828 - 839

关键词:桥梁工程;网级评估;神经网络;区域桥梁;退化模型;遗传算法

Key words:bridge engineering; network-level assessment; neural network; regional bridges; degradation model; genetic algorithm

摘    要:针对区域桥梁网级评估和管养策略优化的问题,系统地提出网级状态评估与管养策略优化框架,建立区域桥梁状态退化预测方法,实现桥梁状态在不同维修策略下的动态预测,解决路网桥梁管养策略的多目标优化问题。首先,从路网数据库中提炼得到用于桥梁状态预测的特征子集,选用合适的二次转码技术以适应人工神经网络的训练;然后,引入代价敏感型训练误差,通过模型选择得到最优桥梁退化模型,并用实测数据进行了效果验证;最后,通过控制各部位维修记录的输入,借助已有退化模型对单体桥梁在不同维修策略下的状态变化进行动态预测,进而拓展至整个桥梁路网。为统筹考虑区域路网的安全指标和费用指标,采用带精英策略的非支配排序遗传算法对网级评估结果进行分析和提炼,研究结果表明:该方法可建立基于神经网络的退化模型,可实现考虑维修的桥梁状态评估及预测。

Abstract: To effectively assess the bridge condition, predict the degradation and optimize the management and maintenance strategy at the regional level, the framework of network-level condition assessment and optimization of management and maintenance strategy was proposed. The regional bridge condition degradation prediction methods and network-level management and maintenance optimization strategy were established, and multi-objective management strategy was optimized. Firstly, the selected features were extracted from road network database, and the appropriate transcoding technology was selected to adapt to artificial neural network. Then, the cost-sensitive training error was introduced, and the optimal bridge degradation model was obtained by model selection. After that, the condition of each bridge in the road network using different maintenance strategies were predicted with the proposed degradation models. In order to consider the structural safety and total cost of the road network, the genetic algorithm to analyze and refine the road network management and maintenance strategy was selected.

相关论文

  • 暂无!

相关知识点

  • 暂无!

有色金属在线官网  |   会议  |   在线投稿  |   购买纸书  |   科技图书馆

中南大学出版社 技术支持 版权声明   电话:0731-88830515 88830516   传真:0731-88710482   Email:administrator@cnnmol.com

互联网出版许可证:(署)网出证(京)字第342号   京ICP备17050991号-6      京公网安备11010802042557号