基于两步分解法和SARIMA的非饱和机场能耗预测  

ENERGY CONSUMPTION FORECAST FOR UNSATURATED AIRPORT BASED ON TWO-STEP DECOMPOSITION AND SARIMA

作　　者：陈静杰 孟琦 Chen Jingjie;Meng Qi(College of Electronic Information and Automation, Civil Aviation University of China, Tianjin 300300, China)

机构地区：[1]中国民航大学电子信息与自动化学院,天津300300

出　　处：《计算机应用与软件》2019年第4期46-50,78,共6页Computer Applications and Software

基　　金：国家科技支撑计划项目(2012BAC20B0304);中美绿色航线项目(GH201661279)

摘　　要：针对非饱和机场能耗时间序列的非线性和非平稳性特点,提出一种基于两步分解法和季节差分自回归滑动平均模型相结合的组合预测法。利用自适应噪声完整集成经验模态分解法和样本熵,将原始能耗时间序列分解为从高频至低频且复杂度不同的分量。再利用变分模态分解法对高频复杂分量再次分解,得到一系列呈现弱非线性且相对平稳的子序列。采用季节差分自回归滑动平均(SARIMA)模型对各子序列进行建模预测,将各子序列预测结果叠加得机场能耗预测值。实验结果表明,该方法可以有效提高非饱和机场能耗的预测精度。In view of the non-linear and non-stationary characteristics of the energy consumption time series in unsaturated airports, we proposed a combined prediction method based on two-step decomposition and SARIMA model. The complete integrated empirical mode decomposition with adaptive noise and sample entropy were adopted to decompose the original energy consumption time series into components with different complexity from high frequency to low frequency. The high-complexity components were decomposed again by the variational mode decomposition. We obtained a series of subsequence which presented weak nonlinearity and were relatively stable. The SARIMA model was used to model and predict each subsequence, and the prediction results of each subsequence were superimposed to obtain the prediction value of airport energy consumption. Experimental results show that this method can effectively improve the prediction accuracy of unsaturated airport energy consumption.

关 键 词：机场能耗预测 自适应噪声完整集成经验模态分解 样本熵 变分模态分解 季节差分自回归滑动 平均模型 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]