摘要
为了模拟港口水域的浮游植物丰度变化及评估外来生物入侵风险,收集了天津港口水域本底生态环境监测数据,建立了港口生态环境数据仓库,以中肋骨条藻为例,运用聚类分析和关联分析等数据挖掘技术,挖掘影响中肋骨条藻丰度的港口水质环境指标,然后利用这些指标建立了基于人工神经网络模型中的反馈网络(BP神经网络)的中肋骨条藻丰度预测模型。研究发现,中肋骨条藻丰度的预测值与真实值的均方根误差为0.0715,相对误差绝对值的平均值为14.7%。该结果表明,BP神经网络对于中肋骨条藻丰度有较好的预测效果,可作为模拟港口水域浮游植物丰度变化的技术手段。数据挖掘技术和BP神经网络模型的综合应用,可以解决港口水域浮游植物丰度变化及外来生物入侵风险评估的关键技术难题。
In order to simulate the phytoplankton abundance and estimate the invasion risk of exotic organisms,the monitoring data of background ecological environment at Tianjin Port were collected,and the port′s ecological environment database was established.Taking Skeletonema costatum as an example,data mining techniques such as cluster analysis and association analysis were used to mine the port water quality environmental indicators that affected the abundance of Skeletonema costatum.Then,the prediction model of Skeletonema costatum abundance based on BP neural network was established with these indicators.The results show that the RMS error between the predicted value and the true value of the algal abundance of the middle rib is 0.0715 and the average value of the absolute value of the relative error is 14.7%,which indicates that BP neural network has a good prediction effect on the abundance of the algae in the middle rib,and can be used as a technical means to simulate the variation of phytoplankton abundance in the port water area.The integrated application of the data mining technology and BP neural network model can solve the key technical problems of phytoplankton abundance change and risk assessment of alien invasion in port water.
作者
李涛
郁百成
邹德武
任利利
LI Tao;YU Bai-cheng;ZOU De-wu;REN Li-li(China Waterborne Transport Research Institute,Beijing 100088,China;Zhejiang Province Port and Shipping Administration Centre,Hangzhou 310011,China;Quzhou Port and Shipping Administration,Quzhou 324000,China)
出处
《交通运输研究》
2021年第2期100-107,共8页
Transport Research
基金
国家重点研发计划项目(2016YFC0503602)
浙江省港航管理局科技项目(2019K001)。
关键词
压载水
生物入侵
数据挖掘
BP神经网络
中肋骨条藻
ballast water
biological invasion
data mining
BP neural network
Skeletonema costatum
