基于神经网络冷再生层最大剪应力预测

发布时间：2018-01-09 09:04

本文关键词：基于神经网络冷再生层最大剪应力预测　出处：《沈阳建筑大学学报(自然科学版)》2017年03期 　论文类型：期刊论文

【摘要】：目的针对乳化沥青冷再生路面内部剪应力过大易导致路面产生车辙等路面破坏问题,对其内部剪应力进行预测,减少此类病害,更好地选择路面结构参数,提高冷再生层内部抗剪强度.方法以乳化沥青冷再生层的厚度、模量,水泥稳定碎石的厚度、模量以及土基模量为输入参数,以冷再生层最大剪应力为输出参数,运用遗传算法对初始参数进行优化,运用灰色神经网络理论构建冷再生层最大剪应力预估模型;构建多元线性回归模型预测最大剪应力,对二者的预测能力进行分析.结果笔者建立的神经网络模型预测值与实测值拟合效果良好,最大误差仅为4.119 2%,能够进行准确预测.多元线性回归和灰色神经网络预测模型,都可用于冷再生层最大剪应力的预测,但灰色神经网络模型对冷再生层最大剪应力数据的预测结果较优.结论把灰色神经网络预测模型与沥青路面结构的设计联系起来,可以更好地控制乳化沥青冷再生路面的剪切破坏.
[Abstract]:This paper aims to predict the internal shear stress of emulsified asphalt cold recycled pavement due to the excessive internal shear stress, which can lead to rutting and other pavement damage, to reduce such diseases, and to select better pavement structure parameters. The thickness, modulus, thickness, modulus of cement stabilized macadam, modulus of soil foundation and thickness of cold regenerated layer of emulsified asphalt were used as input parameters to improve the internal shear strength of cold regenerated layer. Taking the maximum shear stress of the cold regenerated layer as the output parameter, the initial parameters are optimized by genetic algorithm, and the prediction model of the maximum shear stress of the cold regenerated layer is constructed by using the grey neural network theory. A multivariate linear regression model was constructed to predict the maximum shear stress. The prediction ability of the two models was analyzed. Results the predicted value of the neural network model and the measured value were fitted well. The maximum error is only 4.1192.The multivariate linear regression model and grey neural network model can be used to predict the maximum shear stress of cold regenerated layer. But the grey neural network model is better than other models in predicting the maximum shear stress data of cold regenerated layer. Conclusion the grey neural network prediction model is related to the design of asphalt pavement structure. The shear failure of cold recycled asphalt pavement can be better controlled.
【作者单位】：沈阳建筑大学交通工程学院;
【基金】：国家自然科学基金项目(51178278) 辽宁省自然科学基金项目(201602631)
【分类号】：U418.6
【正文快照】： 随着我国经济的快速增长,早期修筑的沥青路面已慢慢步入大修、中修时期,我国公路行业的发展已逐渐进入建养并重的阶段,处理翻修铣刨后的沥青路面旧料已然是当下亟待解决的问题[1-3].乳化沥青冷再生技术是将旧料(RAP)按比例与乳化沥青、水泥、新集料和水进行常温拌和并铺筑,继

【相似文献】