Research On Flight Planning Optimization Based On Multidimensional Data In Terminal Area

With the increasing number of flights across the country in recent years,flight delays have become more serious,and the proportion of flight delays caused by unreasonable flight planning is also increasing.Reasonable optimization of flight plans has important practical value for alleviating flight delays and improving flight operation efficiency and passenger satisfaction.In order to optimize the flight plan more reasonably,this paper starts from one of the flight schedule elements,selects the scheduled departure time in the terminal area as the optimization target,and optimizes the flight plan by predicting and analyzing flight delays in the terminal area.First,through the analysis of domestic and foreign literature on flight plan optimization,the flight plan optimization is divided into three research directions,namely,flight plan optimization based on flight plan preparation link,flight plan optimization based on flight delay prediction,and coordinated release based on airport.Decision-making flight planning optimization,this paper chooses an optimization method based on delay prediction.Secondly,the flight plan,flight delay prediction and Chengdu terminal area are introduced,the general process of flight delay prediction is given,and the relevant data required for flight delay prediction is summarized.Subsequently,the required data was collected and sorted out,and the flight data,weather data,and departure data from January to August 2018 were collected,and the data was processed,including data cleaning,conversion,and fusion.According to the analysis of characteristics,it is found that the proportion of flight delays in Chengdu terminal area from July to August is relatively large,and the delay rate of daily and hourly departure flights is about 25%.This is closely related to the weather at the same time.Meteorological factors are an important factor affecting flight regularity.Then,a flight delay prediction model based on random forest was constructed.The planned take-off time,wind speed,settlement and other characteristics in the data were selected as the input items of the model,and the flight delay grade was used as the output item to predict the delay of departing flights in the Chengdu terminal area.The accuracy is over 80%,which is in line with expectations.Finally,through the analysis of flight delay levels,a flight plan optimization model was constructed based on the flight delay prediction model,and the flight schedule was modified based on the principle of strong meteorological correlation to verify whether the optimized delay situation is better than the original one.The optimization model is analyzed for robustness,meteorological data is selected to carry out the probabilistic perturbation of meteorological factors on the model,and the robustness of the model is analyzed.Finally,it is concluded that the flight planning optimization model of this paper has better robustness.