The Research Of Monitoring Aircraft Fuel Consumption Supported By Large Data

The improved fuel-efficiency is of great significance for our country to promote the construction of "resource saving" and "environment friendly" society. In the field of civil aviation, to improve the fuel utilization efficiency is one of the important goals of airlines especially in the management of aircraft. It is very difficult for airlines to evaluate fuel efficiency by monitoring a single data to measure, since the complex flight course diverse in domestic and international. Therefore, it’s necessary to take a configuration of multipoint data monitoring. With a large number of fuel consumption data, it’s not easy to determine the accurate evaluation index. We lays emphasis on monitoring index of fuel consumption based on a large amount of operation data, building a reasonable fuel consumption performance model, designing an appropriate baseline and boundary line, establishing airplane fuel consumption monitoring system. The methodology of airline fuel consumption monitoring is also be discussed in the paper.Firstly, the background of paper are introduced in detail, including the concept of airplane fuel consumption monitoring system, domestic and foreign airlines in fuel consumption monitoring methods and the application and research of the evaluation method.Secondly, a suited airplane fuel consumption monitoring method based on a large amount data is proposed. In the method, we refine the analysis theory. By means of index selection, we build an exponential decline model based on flight distance. The fuel consumption model could classified the patterns of fuel performance and summarize the parameter of distance.Thirdly, with the methods of neural network, signal separation, which is mainly introduced based on RELAX algorithm and NHD generalized inner product, and least-squares principle, we calculate the baseline and boundary in three ways. By comparing the advantages and disadvantages of three kinds of methods, we put forward the method of optimization for different fuel consumption model, it turned to be correct with the results have certain reference.Finally, according to the actual Boeing-737 and Airbus-A330 airplane data from airlines, the fuel consumption model and methods is being validated. The test focus on the noise performance, fitting performance of the algorithm. And it is proved effective.