基于混合遗传神经网络的高速公路沥青路面使用性能评价方法研究

发布时间：2018-04-20 01:16

本文选题：高速公路 + 沥青路面　；参考：《武汉理工大学》2003年硕士论文

【摘要】： 高速公路把国民经济各领域和各个地区紧密地联系起来，对繁荣经济发展、增进文化交流、促进社会进步都起着重要的推动作用。路面使用性能是高速公路最重要的功能之一，要保证高速公路安全、经济、高效的运营，必须对它实行科学的管理与养护。路面使用性能的正确评价是合理制定养护维护计划、进行投资决策的重要依据之一，是路面养护管理系统中重要的一环。因此，对路面使用性能评价方法的研究在国内外受到极大的重视，是当前道路工程领域研究的热点问题之一。目前国内外常用的四类路面使用性能评价方法中回归法评价结果与实测数据相关性不太理想，系统分析法客观性不强，灰色理论法存在一定的主观经验性，其它一些方法还处于研究阶段；而且这些方法都是基于普通公路建立，高速公路具有交通量大、汽车轴载重、交通渠化和行车速度高等特点，，路面使用性能的特点和发展变化规律与普通公路存在较大差异。所以，对高速公路路面使用性能评价新方法的研究，对确保道路运输的畅通和国民经济顺利的发展有着极其重要的理论和现实意义。针对这些问题，本文在分析现有沥青路面使用性能评价方法的基础上，对高速公路沥青路面使用性能评价问题进行了比较深入的研究，本文的具体研究成果如下： 1、对国内外路面使用性能评价方法的现状进行了评述，详细论述了各种评价方法的基本思路及其优缺点，并深入探讨了路面使用性能评价的发展趋势，为建立新的评价方法提供了基础与参考； 2、分析了高速公路沥青路面的特点，对常用的路面使用性能评价指标进行了研究，并针对高速公路的特点对各指标进行改进，探讨了一些能充分反映高速公路沥青路面使用性能特点的评价指标； 3、针对遗传算法与神经网络优缺点互补的特点，探讨了GA与NN结合的必要性与可行性，将其结合与完善，建立了能充分发挥遗传算法与神经网络优势的混合GANN模型；选择简单、高效、开放的Matlab语言实现了混合GANN模型； 4、首次提出了高速公路沥青路面使用性能混合GANN评价方法；该方武汉理工大学硕士学位论文法运用混合GANN模型，以路面最大裂缝率CR、最大车辙深度RD、平整度 o、强度系数551、抗滑能力SFC实测数据为基础进行模型建立，实现了对高速公路沥青路面使用性能的评价;通过对京珠高速公路湖北段路面的评价，结果表明该方法准确、科学、高效、实用，评价过程方便、快捷，预测结果科学、准确; 5、首次提出了沥青路面路面状况指数PCI的混合GANN预测方法;该方法运用混合 GANN模型，以路面平整度。、强度系数551与抗滑能力SFC 实测数据为基础进行模型建立，实现了对路面状况指标PCI的预测;依据该方法可以对路面破损进行有针对性的调查，减少调查范围，节约人力物力，提高工作效率。本研究项目得到湖北省科技攻关项目(高速公路路面结构优化方法研究 98 1 P 1201)资助。
[Abstract]:The expressway links all areas of the national economy with each area closely. It plays an important role in promoting economic development, promoting cultural exchange and promoting social progress. The performance of pavement is one of the most important functions of the highway. It is necessary to ensure that the expressway is safe, economical and efficient. Management and maintenance. The correct evaluation of pavement performance is one of the important bases for making maintenance and maintenance plan and making investment decision. It is an important part of pavement maintenance management system. Therefore, the research on pavement performance evaluation method is greatly emphasized at home and abroad, and it is a hot topic in the field of road engineering. One of the questions.
At present, the correlation of four kinds of pavement performance evaluation methods commonly used at home and abroad is not very ideal, the objectivity of the system analysis method is not strong, the grey theory method has certain subjective empiricism, and the other methods are still in the study stage, and these methods are based on the common highway and high speed public. The road has the characteristics of heavy traffic, truck axle load, traffic channelization and high speed, and the characteristics of pavement performance and the law of development and change are quite different from that of ordinary highway. Therefore, the research on the new method of evaluating the performance of highway pavement performance is very important to ensure the smooth transportation of roads and the smooth development of the national economy. Important theoretical and practical significance.
In view of these problems, based on the analysis of the existing evaluation methods of the existing asphalt pavement performance, this paper makes a thorough research on the performance evaluation of the expressway asphalt pavement performance. The concrete results of this paper are as follows:
1, the current situation of pavement performance evaluation methods at home and abroad is reviewed, the basic ideas and advantages and disadvantages of various evaluation methods are discussed in detail, and the development trend of pavement performance evaluation is discussed in depth, which provides the basis and reference for the establishment of new evaluation methods.
2, the characteristics of expressway asphalt pavement are analyzed, and the commonly used evaluation indexes of pavement performance are studied. According to the characteristics of the expressway, the indexes are improved, and some evaluation indexes which can fully reflect the performance characteristics of expressway asphalt pavement are discussed.
3, in view of the advantages and disadvantages of genetic algorithm and neural network, the necessity and feasibility of combining GA with NN is discussed, and a hybrid GANN model which can give full play to the advantages of genetic algorithm and neural network is established, and a hybrid GANN model is realized by selecting simple, efficient and open Matlab language.
4, a hybrid GANN evaluation method for expressway asphalt pavement performance is put forward for the first time.
Master's degree thesis of Wuhan University of Technology
The mixed GANN model was applied to the maximum pavement crack rate CR, the maximum rutting depth RD, and evenness.
O, strength coefficient 551, skid resistance SFC measured data based on the establishment of the model, the realization of the right.
Evaluation of expressway asphalt pavement performance; evaluation of Hubei section of Beijing Zhuhai expressway.
The results show that the method is accurate, scientific, efficient and practical, and the evaluation process is convenient, fast and forecast.
Scientific and accurate;
5, a hybrid GANN prediction method for asphalt pavement condition index PCI is put forward for the first time.
Methods the mixed GANN model was applied to pavement roughness, strength factor 551 and skid resistance SFC.
Based on the measured data, the model is established, and the prediction of pavement condition index PCI is realized.
The method can conduct a targeted investigation of the road surface damage, reduce the scope of investigation, and save manpower and material resources.
Improve work efficiency.
This research project has been studied in Hubei province.
981 P 1201) funded.

【学位授予单位】：武汉理工大学
【学位级别】：硕士
【学位授予年份】：2003
【分类号】：U416.217

【引证文献】