APP下载

基于经验模态分解的空气质量指数组合预测方法及应用

2019-07-08李婷婷

价值工程 2019年16期

李婷婷

摘要:空气质量发展趋势的预测对于空气污染问题的防治具有非常重要的意义。因此,本文提出了基于经验模态分解(EMD)的空气质量指数(AQI)的一种组合预测方法。我们首先运用经验模态分解(EMD)的方法对非平稳、非线性且呈剧烈波动的时间序列即AQI原始数据进行多尺度分解。其次,我们分别使用4种常用的单项预测方法:灰色预测(GM)、ARIMA、BP神经网络和支持向量回归(SVR),分别对于分解后的本征模态函数(IMF)序列和趋势序列进行预测,得到单项预测结果。为了提高预测的精度,我们选用平均相对误差(MRE)较小的前三种单项预测方法,并对它们的预测结果进行组合预测。最后,运用熵权法分别计算出IMF序列和趋势序列的组合预测值,并将所有预测值求和得到AQI的最终预测结果。为了评价模型的预测效果,我们选用四种常用误差评价指标,对各个模型的预测结果进行评价比较,而仿真实验的结果表明了本文提出的基于经验模态分解的空气质量指数组合预测方法具有较高的预测精度和良好的适用性。

Abstract: The prediction of air quality development trend is very important for the prevention and control of air pollution problems. Therefore, this paper proposes an air quality index (AQI) combination forecasting method based on empirical mode decomposition (EMD). First, the empirical modal decomposition method (EMD) is used to perform multi-scale decomposition of non-stationary, nonlinear and violently fluctuating time series AQI raw data. Secondly, four different single prediction methods, gray prediction (GM), ARIMA, BP neural network and support vector regression (SVR), are used to predict the decomposed intrinsic mode function (IMF) sequence and trend sequence respectively, and obtain the single prediction methods forecast results. Then, in order to improve the prediction accuracy, we use the first three single prediction methods with small mean relative error (MRE) for combined prediction. Finally, the combined prediction values of the IMF sequence and the trend sequence are calculated by the entropy weight method, and all the predicted values are summed to obtain the final prediction result of the AQI. At the same time, in order to evaluate the prediction effect of the model, four error evaluation indicators are used to evaluate the prediction results of each model. The simulation results show that the proposed method based on empirical mode decomposition has high prediction accuracy and good applicability.

关键词:空气质量指数;EMD分解;灰色预测;ARIMA;BP神经网络;SVR;组合预测

Key words: air quality index;EMD decomposition;grey prediction;ARIMA;BP neural network;SVR;combined prediction

中图分类号:O212                                        文献标识码:A                                  文章编号:1006-4311(2019)16-0134-05

0  引言

随着我国工业化和城市化进程的加快,空气污染逐渐成为环境科学中的重要问题,而能源需求和消耗量持续增加,同时也造成日趋严重的空气污染問题,更加严重影响了人们的日常生活和健康[1]。有研究表明在被污染的空气中存在着许多对人体健康有不利影响的空气污染物,如PM2.5、一氧化碳(CO)、SO2、NO2、SPM、臭氧(O3)等。……

登录APP查看全文