Prediction Model of Fuel Consumption and Flight Time Based on AdaBoost
Traditional operation control systems use performance and other disciplinary calculation formulas to calculate aircraft fuel consumption and flight time,the calculation accuracy is low and the work processes need to be optimized.A fuel consumption and flight time prediction model based on AdaBoost is proposed to address the issue of low accuracy of traditional flight plan performance calculation formulas in calculating aircraft fuel consumption and flight time.This model uses grey correlation analysis to screen correlation indicators,optimizes decision tree base learners using genetic algorithms,and introduces the concept of ensemble learning.Experimental verification is conducted using historical QAR data from the ARJ routes of China Southern Airlines.The root mean square error of the combined model is reduced by 23.68％and 36.21％respectively compared to the flight planning system.The model provides algorithmic support for the study on fuel saving strategies and flight normalcy of airlines.
NETL
NSTL
万方数据
