Research On Optimization Of Fuel Policy Based On QAR Data

According to the promulgation of CCAR-121-R5 fuel policy and the full implementation at the end of 2019,after the implementation of the new version of R5 fuel policy,the unpredictable fuel has changed a lot compared with the previous R4,and with the increase of the voyage,the gap becomes more obvious,which has a great impact on the fuel consumption of airlines operating international long-distance routes.Therefore,this paper mainly studies how to reduce the performance-based Contingency Fuel(PBCF)of flights in operation by data validation,and more actively manage the amount of fuel carried in flight operation,so as to improve operation safety and efficiency.The three-year effective QAR historical data of the Chengdu-Auckland-Chengdu route is extracted by an aviation division A330.Firstly,the QAR effective data is cleaned,the historical fuel deviation data interpolation method,the data distribution fitting,the seasonal interweaving mechanism,and the fuel consumption.Multi-dimensional verification of process control theory and the like.By analyzing the deviation of the cruise segment fuel consumption by more than 95% of the voyage,the 10 eigenvalues affecting the fuel consumption deviation are extracted.Based on the grey correlation theory and correlation coefficient method,the weights of the comprehensive eigenvalues affecting the fuel deviation are calculated.BP(Back Propagation)is constructed.Network and genetic algorithm GA(Genetic Algorithms)improved BP neural network fuel consumption prediction model,based on the determined correlation feature values,through the cleaning QAR data respectively,the model training and testing,the two network iteration speed and prediction accuracy are compared and analyzed.Use the recent flight data to verify the improved forecasting network model;conduct risk management and control of fuel policy optimization,and use the flight planning system to alert,dispatch monitoring,and aircraft performance monitoring to effectively prevent the risk of fuel consumption deviation that may occur during operation.