基于FGM(1,1)的航段油耗预测模型  

Flight segment fuel consumption prediction model based on FGM(1,1)model

作　　者：陈静杰 车洁 CHEN Jing-jie;CHE Jie(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：《计算机工程与设计》2018年第2期522-526,共5页Computer Engineering and Design

基　　金：国家重点基础研究发展计划基金项目(2010CB955401);民航局科技基金项目(MHRD201121);科技支撑基金项目(2012BAC20B03);民航局节能减排专项计划基金项目(DPDSR0010)

摘　　要：为精确预测给定外界条件下固定机型航段飞机燃油消耗,基于足够规模真实航班QAR(quick access recorder)数据,提出一种基于单调函数弱化缓冲算子和偏差调节的FGM(first-entry GM)灰色预测模型。该模型可以有效抑制传统FGM(1,1)模型背景值构造不精确以及原始数据序列波动较大对预测精度带来的不利影响。实验采用考虑部分航班数据缺失情况下的给定巡航高度和起飞重量的固定机型航段油耗面板数据,实验结果表明,该模型预测精度优于FGM模型。To accurately predict the fuel consumption in a given condition with fixed plane model and flight segment,an improved FGM(first-entry GM)model was put forward based on the weakening buffer operator applying strictly monotone function and deviation adjustment for enough scale of real flight QAR(quick access recorder)data.The proposed model can effectively restrain the negative impact on prediction precision brought by the inaccurate background value and volatile original data sequence of traditional FGM.The fuel consumption panel data were utilized in the experiment.The panel data were constructed using the fixed plane model and flight segment QAR data in given cruise altitude and gross weight.Experimental results suggest that the proposed grey model performs better than FGM model.

关 键 词：单调函数弱化缓冲算子 航班数据缺失处理 航段油耗面板数据 灰色预测 偏差调节 first-entry GM模型 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程]