Research Of Flight Conflict Resolution Based On Rough Set Theory And Neural Network Method

As an important part of the transportation industry,the civil aviation is in a rapid developing period.It has also brought many problems such as increased aviation traffic flow and frequent flight conflicts.Therefore,it is important to ensure the safe operation of civil aviation aircraft and resolve conflict situations quickly and efficiently.There are many regulations and research models for aircraft separation and conflict relief.Responds to the needs of the development of the information age,the research starts with aircraft conflict historical data,and uses data mining to analyze the selection of conflict resolution programs.The method combining the rough set theory with the neural network model was introduced into the study of aircraft conflict resolution.Based on the empirical data information,the aircraft conflict resolution program was selected efficiently and objectively.Analyzing several factors affect the selection of aircraft conflict resolution methods,it integrates information sources for the topic,and selects data warehouses that can support decision-making as data storage.In accordance with requirements analysis,conceptual model design,logical model design,and physical model design,build a complete aircraft conflict resolution program data warehouse system to provide data samples for the entire algorithm model.The attribute reduction of rough set is used to reduce the attributes of the decision table generated by the aircraft conflict resolution method data warehouse.The redundancy attribute is removed.Through the calculation of the importance degree,identified the key attributes and assigned the weights.And the reduced minimum attribute set is generated.Then,established a classification model of aircraft conflict resolution method combined with rough set and neural network.Designing the model structure,the neural network was constructed by using MATLAB software to classify the resolution objects when the aircrafts were in conflict.Finally,the typical case base was introduced to match the specific conflict resolution methods.The classification information is further combined with case-based reasoning,computed the similarity of typical cases.The nearest neighbors algorithm is used to find the most similar cases to provide guidance for actual conflict liberation.The research is aimed at getting the information from the data to guide the selection of the aircraft conflict resolution program.Using the rough set to reduce the input data of the neural network when dealing with the fuzzy problem can simplify the network structure and improve the efficiency of network training.The combination of network and case-based reasoning techniques for the selection of aircraft conflict resolution methods can effectively combine the network model with expert experience,which is conducive to obtaining more specific and accurate resolution programs.Overall,this research on data mining and aircraft conflict resolution have certain theoretical references and practical operating guidance values.