摘要
为对港口货运量进行科学精准预测,结合天牛须搜索(beetle antennae search, BAS)算法和蒙特卡洛准则,提出一种改进BAS的Elman神经网络预测模型。收集上海港1989—2018年内的货运量以及当地各项经济数据,建立港口货运量预测评估体系,对各项影响因子进行预处理,消除数据冗余信息对预测的影响,对预处理后的数据进行仿真测试。实验结果表明,该模型预测准确率可达95%以上,有效地提高了港口货运量的预测精度。
To predict a port freight volume scientifically and accurately,a modified Elman neural network prediction model was proposed based on the beetle antennae search(BAS)and Monte Carlo criteria.The freight volume and local economic data of the Shanghai Port from 1989 to 2018 were collected,the prediction-and-evaluation system of port freight volume was established,various influencing factors were pre-processed,the influence of redundant data on the prediction was eliminated,and simulation test on the pre-processed data was conducted.Results show that the prediction accuracy is>95%,which effectively enhances the prediction accuracy of port freight volume.
作者
廖列法
欧阳宗英
LIAO Lie-fa;OUYANG Zong-ying(School of Information Engineering,Jiangxi University of Science and Technology,Ganzhou 341000,China)
出处
《科学技术与工程》
北大核心
2021年第7期2937-2944,共8页
Science Technology and Engineering
基金
国家自然科学基金(71761018,71462018)。
关键词
天牛须搜索
ELMAN神经网络
吞吐量预测
蒙特卡洛准则
主成分分析
beetle antennae search(BAS)
Elman neural network
throughput prediction
Monte Carlo criteria
principal component analysis
