基于K-means算法的飞机下降过程油耗分类  

Fuel Consumption Classification of Aircraft in Descent Based on K-means Algorithm

作　　者：陈静杰[1] 李吕琪 

机构地区：[1]中国民航大学航空自动化学院,天津300300

出　　处：《测控技术》2015年第11期16-19,共4页Measurement & Control Technology

基　　金：国家重点基础研究发展计划(973计划)项目(2010CB955401);国家科技支撑计划项目(2012BAC20B03);民航局科技基金项目(MHRD201008;MHRD201121)

摘　　要：飞机下降过程受飞行环境和流量影响而运行情况复杂,相应油耗也不同。提出一种基于下降段实飞距离、平均下降率、下降点标准高度等多种飞行特征指标的下降过程分类方法。该方法先对QAR数据进行统计分析,找出影响下降过程油耗的飞行特征指标,应用Pearson相关系数进行油耗特征选择,进而应用K-me卸s算法针对相关性大的指标进行油耗聚类。结果将油耗分为明显的两类,可看出两类的飞行特征差异,这为进一步设计节油下降轨迹提供了新视角。Aircraft descent is affected by the flight environment and traffic.Fuel consumption caused by complex operation environment is also different.A classification in descent based on the different of real flying distance,the average decline rate,standard height of drop point,is presented.This classification analyzes QAR data to find out flight characteristic indexes which influences the fuel consumption in descent,and analyzes the correlation between the indexes and fuel consumption,uses Pearson correlation coefficient to select greater correlation indexes,then the fuel consumption based on selected indexes is clustered by using K-means algorithm.The fuel consumption is divided into two categories,and the differences between them are apparent.It offers a new vision to the further design of fuel-saving descent trajectory.

关 键 词：QAR数据 下降过程 特征选择 Pearson相关系数 K-MEANS算法 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]