摘要
采用目前方法预测土壤重金属元素含量时,在不同距路基垂直距离下无法准确地得出重金属元素含量,预测结果的均方根误差大以及错误比例高。提出基于Krigin插值法的土壤重金属元素含量预测方法,依据Pearson相关系数获取变量之间的关联程度,实现对相关数据的分析,同时,对土壤重金属元素进行主成分分析。根据数据统计分析结果,利用Krigin插值法原理进行重金属元素含量预测,并根据半变异函数对预测结果进行优化,最后设定阈值对重金属元素含量的等级进行划分,实现土壤重金属含量的预测。实验结果表明,所提方法在不同距路基垂直距离下能够准确得出重金属元素含量,预测结果的均方根误差小,并且错误比例较低。
When the current method is used to predict the content of heavy metals in soil,the content of heavy metals cannot be accurately obtained under different vertical distances from the subgrade,and the root mean square error of the prediction results is large and the error proportion is high.In this regard,a prediction method of soil heavy metal content based on Krigin interpolation method was put forward in this paper.For realizing the correlation data analysis,the correlation degree between variables was obtained via Pearson correlation coefficient.Meanwhile,the principal components of soil heavy metal elements were analyzed in detail.Based on the analysis results of data statistics,Krigin interpolation principle was adopted to predict and optimize the content of heavy metals by semi-variogram.At last,the threshold was set to classify the content of heavy metals,thus predicting the content of heavy metals in soil.The experimental results show that this method can accurately obtain the content of heavy metals at different vertical distances from the subgrade,and has small root mean square error and low error ratio.
作者
倪碧珩
陆胤
施维林
NI Bi-heng;LU Yin;SHI Wei-lin(College of Environmental Science and Engineering,Suzhou University of Science and Technology,Suzhou Jiangsu 215009,China;College of Biological and Environmental Engineering,Zhejiang Shuren University,Hangzhou Zhejiang 310015,China)
出处
《计算机仿真》
北大核心
2022年第5期234-237,392,共5页
Computer Simulation
基金
国家自然科学基金项目(NSFC31570515)
苏州市科技局项目(SS201721,SS201724,SS201728)。
关键词
插值法
土壤重金属
含量预测
概率阈值
半变异函数
Interpolation method
Heavy metals in soil
Content prediction
Probability threshold
Semi-variogram
