摘要
为了动态优化高铁路基碾压参数,以提高高铁路基填料碾压效率,提出了一种基于振动能量最小原则的振动参数优化方法。首先,开展室内振动压实试验,以建立振动参数与干密度之间的定量关系;其次,采用改进神经网络方法建立干密度增量预测模型,构建基于能量最小原则的压实过程振动参数优化方法;最后,基于Modbus通讯协议,优化设计出振动参数可调的智能振动压实仪,并通过振动压实试验验证优化结果。研究结果表明:振动强度达到阈值后,进一步增大振动强度会导致压实仪发生“跳振”,严重影响振动设备使用寿命;改进神经网络模型能够较好用于预测干密度增量;提出适用于动态优化的遗算法(genetic algorithm,GA)算法的最佳参数,即种群数量为150,选择概率为0.9,交叉概率为0.6,变异概率为0.05;在此最佳参数下,能有效保证动态优化方法的准确性;采用优化后振动压实方案能够减少振动能量127.58 J,占优化前能量输出的25.61%,填料破碎率减小0.9%。
To achieve the dynamic optimization of high-speed railway subgrade compaction parameters,and improve the efficiency of filling compaction,a vibration parameter optimization method based on the principle of minimum vibration energy was proposed.Firstly,vibration compaction experiments were conducted to quantify the relationship between vibration parameters and dry density.Secondly,an improved neural network method was employed to establish a dry density increment prediction model,and a compaction process vibration parameter optimization method based on the principle of minimum energy was constructed.Finally,an intelligent vibration compaction instrument with adjustable vibration parameters was designed and optimized using the Modbus communication protocol.The optimization results were validated through vibration compaction experiments.The results show that increasing the vibration intensity beyond a certain threshold can lead to"jumping vibration"of the compaction instrument,significantly affecting the service life of the vibration equipment.The improved neural network model proves suitable for predicting dry density increments with high accuracy.The proposed genetic algorithm(GA)parameters for dynamic optimization ensure the accuracy of the dynamic optimization method.The optimal parameters of the genetic algorithm(GA)applicable to dynamic optimization are proposed,that is the number of populations is 150,the probability of selection is 0.9,the probability of crossover is 0.6 and the probability of variation is 0.05.At the optimization condition,the accuracy of the dynamic optimization method can be effectively ensured.The adoption of the optimized vibration compaction scheme reduces vibration energy by 127.58 J,which accounts for 25.61%of the energy output before optimization,and decreases the filling breakage rate by 0.9%.
作者
陈晓斌
谢康
尧俊凯
惠潇涵
王业顺
邓志兴
CHEN Xiaobin;XIE Kang;YAO Junkai;HUI Xiaohan;WANG Yeshun;DENG Zhixing(School of Civil Engineering,Central South University,Changsha 410083,China;Key laboratory of Engineering Structures of Heavy Haul Railway of Ministry of Education,Central South University,Changsha 410083,China;Railway Engineering Research Institute,China Academy of Railway Sciences Co.Ltd.,Beijing 100081,China;China Railway Guangzhou Bureau Co.Ltd.,Guangzhou 510440,China)
出处
《中南大学学报（自然科学版）》
EI
CAS
CSCD
北大核心
2023年第9期3731-3742,共12页
Journal of Central South University:Science and Technology
基金
国家自然科学基金资助项目(51978674)
中国铁道科学研究院科技研究开发计划重点课题(2023QT002)
中南大学研究生创业计划项目(2022ZZTS0622)。
关键词
高铁路基填料
智能压实
振动强度
振动参数
动态优化
神经网络
high-speed railway subgrade fillers
intelligent compaction
vibration intensity
vibration parameters
dynamic optimization
neural network
