【作者】 陈向阳；

【导师】 夏元友；

【作者基本信息】 武汉理工大学 ， 桥梁与隧道工程， 2008， 博士

【摘要】 随着我国高速公路的大规模建设,工程中遇到的软土问题越来越突出,已成为影响工程质量、建设工期和工程造价的关键性因素之一。因此,加强高速公路软基处理技术的研究,已成为影响我国交通事业发展的一个重要课题。在高速公路软基处理的技术研究中,存在着大量的决策问题。在设计阶段,首先需要对是否进行处理做出决策,对需要处理的地段,则须对采取深层处理还是浅层处理做出决策;对于浅层处理或深层处理的具体方法也需要做出决策;对于某种具体的处理方法,需要对其技术参数做出决策。本文以系统工程、人工智能、模糊数学、灰色理论、神经网络、遗传算法和粒子群优化算法等为理论基础,结合岩土工程、土质学和土力学等学科的最新发展,对高速公路软基处理方案智能评价与优化方法进行了研究。按照研究的内容和论文的章节顺序,所取得的具体成果有:(1)简要地介绍了本课题研究的目的和意义,介绍了有关软基处理技术的国内外研究现状,系统地阐述了现有的软基处理方案评价方法,如评分优化法、模糊评判法、层次分析法以及地基处理智能辅助设计系统等,指出了它们存在的主要问题和不足,在此基础上,提出了本研究的工作思路和主要内容。(2)针对软基处理技术决策中是否需要进行处理,是深层处理还是浅层处理的问题,通过研究有关土工参数的相关性等,提出以软土厚度、软土压缩模量、路堤填土高度、地表硬壳层厚度作为决策参数,以人工智能中的范例推理为基础,并考虑到在不同的决策环境下相同的属性对决策输出会有不同的影响,提出了基于变权的高速公路软基处理方案范例推理评价方法。实例分析表明,该模型具有原理简单、直观,使用方便等特点;利用神经网络工具箱建立了BP神经网络评价模型,对所建的神经网络模型进行训练、回判和预测,得到了满意的结果,证明模型是有效的。(3)提出了基于AHP的模糊相似优先比的高速公路软基处理评价模型。在该模型中,定义了模糊相似比,构造了软基处理方案影响因素的模糊相似优先关系,利用层次分析法确定了各影响因素的权重,从整体上找出了与理想方案最为相似的软基处理方案。(4)鉴于软基处理方案评价中各指标的不相容性问题,以模糊物元分析理论为基础,为避免在权重分配问题上的主观性,用信息熵所反映数据本身的效用值计算指标的权重,将信息熵理论与模糊物元分析相结合,建立了基于信息熵的模糊物元分析决策模型。将该模型应用于高速公路软基处理方案的评价中,得到了合理的处理方案,为选择合理的软基处理方案提供了一条新的可行途径。(5)软基处理方案决策影响因素具有不完全性、未知性等特点,软基处理方案决策实质上是灰色系统问题。本文运用灰色系统理论,建立了一种新的软基处理方案评价方法,合理确定了方案评价指标及其权重,根据计算得出的灰色关联度来实现对软基处理方案的优选。实例分析表明,该模型使用方便、决策结果合理,具有实用价值。(6)将自适应共振理论用于高速公路软基处理的方案决策,建立了基于ART网络的高速公路软基处理方案评价方法。实例表明,模型具有易于实王见、效率高等特点。(7)论述了粉体搅拌桩的传统设计理论和方法,应用遗传算法、粒子群优化算法和非线性优化方法,建立了粉喷桩的三种优化设计模型,开发了粉喷桩设计与优化系统,最后运用工程实例验证了优化模型的可行性和有效性。更多还原

【Abstract】 With the large-scale construction of expressway, the problem of soft ground is becoming more and more protrusive and has been one of key factors influencing project quality, project cycle and project cost. So it has become a very important task to strengthen the research on the improvement of expressway soft ground, which affects the development of traffic project in China.There are much decision problems in the study of expressway soft ground improvement. During the design phase, decision whether to improve the ground or not should be made firstly, then it should make clearly whether deep improvement or shallow improvement in the segment which needs improvement. And the detailed method on improvement should be made distinctly too. Finally, the technical parameters of the method should be decided. Based on system engineering, artificial intelligence, comprehensive evaluation, neural network, fuzzy mathematics, genetic algorithm and particle swarm optimization, etc. and combing the latest developments of geo-technique and soil mechanics, the dissertation studies many aspects of problems to establish intelligent decision support system of expressway ground improvement.According to the research contents and sequences of chapters and sections, the main achievement of the dissertation are concisely mentioned as follows:1. It expresses briefly the purpose and significance of the article, and introduces comprehensively present research situation, existing problems and research prospect of ground improvement. Based on systems analysis of decision problems of expressway ground improvement, it puts forward research subject and chief contents of intelligent decision model of soft soil improvement methods.2. In view of the incompleteness and uncertainty of these effective factors, a new intelligent decision model of the soft ground improvement method of the highway using case-based reasoning is presented for the first time. Considering the sensitivity of attribute weights to the environment, the algorithm of attribute weights is set up on the basis of the concept of changeable weights. It is shown from example that results of the decision model is simple, visual, practical and convenient to use. 3. In this dissertation, the decision model of the soft soil improvement methods based on fuzzy analogy preferred ratio is put forward. In this model, fuzzy analogy preferred relationship between feasible methods and the ideal one is determined, and the similar series in them in every affecting factor is given. Determining their factor weights by using the analytic hierarchy process (AHP), the comprehensive similar series in them is carried out, and the most similar soft ground improvement method to ideal one is found out. It is discussed how to choose the input parameters of the decision model. Using the MATLAB-NNT, a decision model of the soft soil improvement of the expressway with 4 parameters is established. After trained by input data and target data, recollected by same data and simulated by new input data, the result of the model is satisfied, and the model is effective.4. In view of the uncertainty of index for evaluating soft ground improvement programs, combining with fuzzy matter-element method and the entropy value theory, the fuzzy matter-element model is established based on entropy value theory. The weights are calculated according to the method of information entropy indicating the availing value of data, and the problem of weight allocation is solved. The model is applied for programs evaluation of improving expressway soft ground and a reasonable program is got which provides a new way to select programs for improving soft ground.5. The factors that affect the decision of soft ground improvement method have the features of incompleteness and uncertainty, making a decision of the soft soil improvement methods is a Grey System Theory problem essentially. By means of Grey system theory, the new decision model is established. In this model, the evaluation indexes of the schemes and their weights are reasonably decided; the optimal scheme is selected by calculating the relational grade to attain the selection and order arrangement for the schemes. The results of the actual examples indicated that the decision model is reasonable, valuable and it is convenient to use.6. Adaptive resonance theory is used in technical decision of expressway ground improvement. Decision model based on adaptive resonance theory is set up for the first time. The result proves that the model is valid and easy to realize.7. The basic design computational process theory of dry jet mixing pile is studied. Three optimization models are set up based on the genetic algorithm, particle swarm and nonlinear theory for dry jet mixing pile, and the optimization design system of dry jet mixing pile is programmed. Finally, the results of the actual examples indicated the validity and feasibility of the two optimization methods.更多还原