The invention discloses a bird risk prediction and evaluation method in an aero-engine bird-attracting environment. The method comprises steps of obtaining the probability of an aero-engine bird-attracting event according to historical data, writing an M file program through employing
MATLAB language, achieving a BP neural network
algorithm, predicting the economic loss caused by the aero-engine bird-attracting event through employing a BP neural
network model, and obtaining the bird-attracting risk of an aero-engine;
risk assessment being carried out in combination with the aircraft engine bird suction event probability and the
risk assessment matrix; and finally, designing a GUI (
Graphical User Interface), selecting eight factor data such as aircraft
mass, engine installation number, engine type, month, weather, bird type, bird number and bird size in the statistical data according to the correlation degree of influence factors in the aircraft engine bird attraction event, and calling a BP (Back Propagation) neural network
algorithm through an M file to obtain a result. The method is advantaged in that the model can be used for predicting the probability value, the severity degree and the
risk assessment level of a certain bird inhaled by a certain engine of a certain flight aircraft in a specified time period and a specified height in the future.