摘要
为实现PM_(2.5)、PM_(10)和TSP粉尘质量浓度预测,利用哈尔乌素露天煤矿环境监测数据,以随机森林算法为基础,建立环境因素影响下的粉尘质量浓度预测模型,采用网络搜索的方法对随机森林模型进行参数调整.研究结果表明:采用加权算术平均值法对粉尘质量浓度数据进行降噪处理,能有效提高预测模型的准确性.特征重要性分析表明,环境影响因素中相对湿度对预测效果影响最大,其次是温度、噪声,风速、风力、风向影响较小.
To achieve the prediction of PM_(2.5),PM_(10)and TSP dust mass concentrations,a random forest algorithm was used to establish a dust mass concentration prediction model under the influence of environmental factors based on the environmental monitoring data of Halwusu open-pit coal mine,and a network search method was used to reconcile the random forest model.The results of the study show that the noise reduction of dust mass concentration data by the weighted arithmetic mean method can effectively improve accuracy of the prediction model.The feature importance analysis shows that the relative humidity has the greatest influence on the prediction effect,followed by temperature and noise,while wind speed,wind force and wind direction have less influence.
作者
霍文
栾博钰
周伟
陆翔
王春丽
周永利
赵彬宇
HUO Wen;LUAN Boyu;ZHOU Wei;LU Xiang;WANG Chunli;ZHOU Yongli;ZHAO Binyu(Information Center,Shenhua Group Zhungeer Energy Company Limited,Ordos 010300,China;Shool of Mines,China University of Mining and Technology,Xuzhou 221116,China;Research Institute of Science and Technology,Shenhua Group Zhungeer Energy Company Limited,Ordos 010300,China)
出处
《辽宁工程技术大学学报(自然科学版)》
CAS
北大核心
2021年第5期409-414,共6页
Journal of Liaoning Technical University (Natural Science)
基金
国家自然科学基金(52174131,51804299)