【作者】 许允之； 王舒萍； 边宁； 李鹏程；

【Author】 Xu Yunzhi;Wang Shuping;Bian Ning;Li Pengcheng;Laboratory of Electrical and Automation Engineering for Coal Mines,School of Electrical and Power Engineering,China University of Mining and Technology;School of Electronic Information and Electrical Engineering,Shanghai Jiaotong University;

【机构】 中国矿业大学电气与动力工程学院江苏省煤矿电气与自动化工程实验室； 上海交通大学电子信息与电气工程学院；

【摘要】 采用随机森林算法,对雾霾物质在沿面放电作用下的雾霾成分分解特征进行了分类预测,并选择了支持向量机SVM、K最近邻算法KNN和朴素贝叶斯3种算法同样进行分类来比较分类效果。结果表明:相对于其他3种算法来说,随机森林算法对于雾霾分类具有更高的预测准确率。最后,对雾霾成分在沿面放电实验结束后玻璃板表面的剩余溶液进行了离子色谱分析,分析了离子浓度的变化和可能存在的离子反应。将溶液加电后离子组分与浓度发生变化的现象,引申到电场作用对雾霾组分的影响,指出电场去除雾霾也是一个很有潜力的研究方向。更多还原

【Abstract】 In this paper,we classified and predicted the fog-haze component decomposition characteristics of fog-haze material under surface discharge by using random forest algorithm. And the three algorithms of support vector machine,K nearest neighbor algorithm and naive Bayes were selected to classify the type of fog-haze pollution and compare the classification effect.The results showed that compared with the other three algorithms,the random forest algorithm had higher prediction accuracy for the fog-haze classification. Finally,we performed ion chromatographic analysis of the remaining solution on the surface of the glass plate after the end of the surface discharge experiment for fog-haze components. We analyzed changes in ion concentration and possible ion reactions. We linked the phenomenon of changes in ion composition and concentration after the solution was electrified to the influence of electric field on fog-haze components. And we pointed out that the removal of foghaze by electric fields was also a promising research direction.更多还原