-
摘要: 通过隶属度函数确定的加权KNN-BP神经网络方法,建立PM2.5浓度动态实时预测模型,以PM2.5、PM10、NO2、CO、O3、SO2等6种污染物前1 h的浓度及天气现象、温度、气压、湿度、风速、风向等6种气象条件,以及预测时刻所在一周中天数和该时刻所在一天当中的小时数为KNN实例的维度,选取3个近邻,根据得到的欧氏距离确定每个近邻变量的隶属度权重,最终将所有近邻的维度作为BP神经网络的输入层数据,输出要预测的下1 h PM2.5浓度,该方法避免了传统BP神经网络方法不能体现历史时间窗内的数据对当前预测影响的问题。对北京市东城区监测站2014-05-01T00:00—2014-09-10T23:00的数据进行预测试验,结果表明,加权KNN-BP神经网络预测模型相较其他方法的预测误差最低,且稳定性效果最好,是PM2.5浓度实时预测的有效方法。Abstract: Through the weighted KNN-BP neural network method determined by membership function , the dynamic real-time prediction model of PM2.5 concentration was established. The concentration of six pollutants, i.e. PM2.5, PM10, NO2, CO, O3 and SO2, six meteorological data including weather condition, temperature, pressure, humidity, wind speed and wind direction in the first hour, as well as the days of a week and the hours of the days for projection were regarded as the dimensions of the KNN instance. Three nearest neighbors were selected and, according to the Euclidean distance obtained, the membership weight of each neighbor point variable determined. Finally, the dimension of all nearest neighbor points were taken as the input layer of BP neural network, and the next hour PM2.5 concentration to be predicted as the output layer data. The method avoided the problem that the traditional BP neural network method failed to reflect the influence of the data in the historical window on the current predicting. The data of 2014-05-01 from 00:00 to 23:00 2014-09-10 in Dongcheng District monitoring station in Beijing was tested. The results showed that the prediction model with weighted KNN-BP neural network had the lowest deviation compared with other methods, and the stability showed the best. Therefore, this model is an effective method for the PM2.5 real time prediction.
-
Key words:
- BP neural network /
- K-nearest neighbor /
- degree of membership function /
- haze forecast
-
[1] 王金南 . 控制PM2.5污染:中国路线图与政策机制[M]. 北京: 科学出版社, 2016. [2] 李岚淼, 李龙国, 李乃稳 . 城市雾霾成因及危害研究进展[J]. 环境工程, 2017,35(12):92-97.LI L M, LI L G, LI N W . Studies review on cause and damage of urban haze[J]. Environmental Engineering, 2017,35(12):92-97. [3] 杨继东, 章逸然 . 空气污染的定价:基于幸福感数据的分析[J]. 世界经济, 2014,37(12):162-188. [4] LEE K H, KIM Y J, KIM M J . Characteristics of aerosol observed during two severe haze events over Korea in June and October 2004[J]. Atmospheric Environment, 2006,40(27):5146-5155.
doi: 10.1016/j.atmosenv.2006.03.050[5] 彭斯俊, 沈加超, 朱雪 . 基于ARIMA模型的PM2.5预测[J]. 安全与环境工程, 2014,21(6):125-128.
doi: 10.3969/j.issn.1671-1556.2014.06.023PENG S J, SHEN J C, ZHU X . Forecast of PM2.5 based on the ARIMA model[J]. Safety and Environmental Engineering, 2014,21(6):125-128. doi: 10.3969/j.issn.1671-1556.2014.06.023[6] 黄思, 唐晓, 徐文帅 , 等. 利用多模式集合和多元线性回归改进北京PM10预报[J]. 环境科学学报, 2015,35(1):56-64.
doi: 10.13671/j.hjkxxb.2014.0728HUANG S, TANG X, XU W S , et al. Application of ensemble forecast and linear regression method in improving PM10 forecast over Beijing areas[J]. Acta Scientiae Circumstantiae, 2015,35(1):56-64. doi: 10.13671/j.hjkxxb.2014.0728[7] 艾洪福, 潘贺, 李迎斌 . 关于空气雾霾PM2.5含量预测优化研究[J]. 计算机仿真, 2017,34(1):392-395.
doi: 10.3969/j.issn.1006-9348.2017.01.085AI H F, PAN H, LI Y B . Research on optimization of PM2.5 content prediction in air haze[J]. Computer Simulation, 2017,34(1):392-395. doi: 10.3969/j.issn.1006-9348.2017.01.085[8] 孙宝磊, 孙暠, 张朝能 , 等. 基于BP神经网络的大气污染物浓度预测[J]. 环境科学学报, 2017,37(5):1864-1871.
doi: 10.13671/j.hjkxxb.2016.0391SUN B L, SUN H, ZHANG C N , et al. Forecast of air pollutant concentrations by BP neural network[J]. Acta Scientiae Circumstantiae, 2017,37(5):1864-1871. doi: 10.13671/j.hjkxxb.2016.0391[9] DÍAZ-ROBLES L A, ORTEGA J C, FU J S , et al. A hybrid ARIMA and artificial neural networksmodel to forecast particulate matter in urban areas:the case of Temuco,Chile[J]. Atmospheric Environment, 2008,42:8331-8340.
doi: 10.1016/j.atmosenv.2008.07.020[10] 艾洪福, 石莹 . 基于BP人工神经网络的雾霾天气预测研究[J]. 计算机仿真, 2015,32(1):402-405.
doi: 10.3969/j.issn.1006-9348.2015.01.085AI H F, SHI Y . Study on prediction of haze based on BP neural network[J]. Computer Simulation, 2015,32(1):402-405. doi: 10.3969/j.issn.1006-9348.2015.01.085[11] 赵晨曦, 王云琦, 王玉杰 , 等. 北京地区冬春PM2.5和PM10污染水平时空分布及其与气象条件的关系[J]. 环境科学, 2014,35(2):418-427.ZHAO C X, WANG Y Q, WANG Y J , et al. Temporal and spatial distribution of PM2.5 and PM10 pollution status and the correlation of particulate matters and meteorological factors during winter and spring in Beijing[J]. Environmental Science, 2014,35(2):418-427. [12] 刘金培, 汪官镇, 陈华友 , 等. 基于VAR模型的PM2.5与其影响因素动态关系研究:以西安市为例[J]. 干旱区资源与环境, 2016,30(5):78-84.
doi: 10.13448/j.cnki.jalre.2016.151LIU J P, WANG G Z, CHEN H Y , et al. Dynamic relationship between PM2.5 and its influence factors in Xi’an City based on the VAR model[J]. Journal of Arid Land Resources and Environment, 2016,30(5):78-84. doi: 10.13448/j.cnki.jalre.2016.151[13] 迈尔·舍恩伯格, 库克耶 . 大数据时代[M]. 杭州: 浙江人民出版社, 2013. [14] 潘竟虎, 张文, 王春娟 . 2011—2013年中国雾霾易发生期间API的分布格局[J]. 环境工程学报, 2016,10(3):1340-1348.PAN J H, ZHANG W, WANG C J . Spatial pattern of urban air pollution during large-scale range haze based on fractal net evolution approach and GIS in China[J]. Chinese Journal of Environmental Engineering, 2016,10(3):1340-1348. [15] 杨健, 宋冰, 谭帅 , 等. 时序约束NPE算法在化工过程故障检测中的应用[J]. 化工学报, 2016,67(12):5131-5139.
doi: 10.11949/j.issn.0438-1157.20160543YANG J, SONG B, TAN S , et al. Time constrained NPE for fault detection in chemical processes[J]. CIESC Journal, 2016,67(12):5131-5139. doi: 10.11949/j.issn.0438-1157.20160543[16] 王钰, 郭其一, 李维刚 . 基于改进BP神经网络的预测模型及其应用[J].计算机测量与控制, 2005(1):39-42.
doi: 10.3321/j.issn:1671-4598.2005.01.013WANG Y, GUO Q Y, LI W G . Predictive model based on improved BP neural networks and it’s application[J].Computer Measurement & Control,2005(1):39-42. doi: 10.3321/j.issn:1671-4598.2005.01.013[17] ZHENG Y, YI X W, LI M , et al. Forecasting fine-grained air quality based on big data[M]. Sydney:Association for Computing Machinery, 2015: 1231-1239. [18] ZHENG Y, CAPRA L, WOLFSON O , et al. Urban computing:concepts,methodologies and applications[J]. ACM Transaction on Intelligent Systems and Technology, 2014,5(3):1-55.
doi: 10.1145/2629592[19] ZHENG Y, LIU F R, HSIEH H P .U-air:when urban air quality inference meets big data[M].Chicago:Association for Computing Machinery, 2013: 1436-1444.
点击查看大图
计量
- 文章访问数: 653
- HTML全文浏览量: 170
- PDF下载量: 75
- 被引次数: 0