SSA-BP评价共振碎石化沥青加铺路面施工质量  

SSA-BP Evaluation of Construction Quality for Resonant Rubblization Asphalt Overlay Pavement

作　　者：肖秋明[1] 程志伟 吕茂丰 XIAO Qiuming;CHENG Zhiwei;LYU Maofeng(School of Traffic and Transportation Engineering,Changsha University of Science&Technology,Changsha 410114,China.;不详)

机构地区：[1]长沙理工大学交通运输工程学院,湖南长沙410114 [2]中铁五局集团机械化工程有限责任公司,湖南长沙410100

出　　处：《武汉理工大学学报（信息与管理工程版）》2023年第1期103-108,116,共7页Journal of Wuhan University of Technology：Information & Management Engineering

基　　金：国家自然科学基金项目(51878077);中铁五局集团有限公司科技项目(2020[12]).

摘　　要：目前,利用共振碎石化技术将旧水泥混凝土路面板破碎后再加铺沥青面层的改造方式在工程中应用广泛。为合理评价旧水泥混凝土共振碎石化沥青加铺路面的施工质量,采用麻雀搜索算法(SSA)优化BP神经网络的权值和阈值,建立了基于SSA-BP的施工质量评价模型,并对模型的性能进行了评估。研究结果表明:SSA-BP模型预测结果的相对误差绝对值可控制在1.0%以内,相比于标准BP模型和PSO-BP模型,SSA-BP模型的均方误差、均方根误差及平均绝对误差指标值均小于BP和PSO-BP模型,说明了麻雀搜索算法对BP神经网络起到了优化作用,且模型具有较高的精度和稳定性,可供同类工程应用参考。Nowadays,retrofit method of asphalt overlay after crushing the old cement concrete pavement by resonant rubblization technology is widely used in engineering.In order to reasonably evaluate the construction quality of old cement concrete resonant rubblization asphalt overlay pavement,the sparrow search algorithm(SSA)was used to optimize the weights and thresholds of BP neural network,and the construction quality evaluation model based on SSA-BP was established,and the performance of the model was evaluated.The results show that the absolute value of relative error of SSA-BP model prediction results can be controlled within 1.0%.Compared with the standard BP model and PSO-BP(particle swarm optimization,PSO)model,the mean square error,root mean square error and mean absolute error of SSA-BP model are less than those of BP and PSO-BP models,indicating that the sparrow search algorithm has played an optimization role in BP neural network.The model has high accuracy and stability,which can be used for reference in similar engineering applications.

关 键 词：共振碎石化 沥青加铺路面 麻雀搜索算法 BP神经网络 施工质量评价 

分 类 号：U415.11[交通运输工程—道路与铁道工程]