Flight quality adaptive evaluation method

Abstract

The invention discloses a flight quality adaptive evaluation method, which comprises the following steps of: firstly arranging flight actions to be evaluated, then selecting flight action entries of related standards from historical flight data, and establishing a standard action reference library; constructing a depth variational self-encoding network, then calculating distance measurement by adopting a dynamic planning mode on the sorted flight action to be evaluated and corresponding standard action compression characteristics obtained after processing of the depth variational self-encoding network, and then calculating similarity measurement between the flight action to be evaluated and the standard action; and then, according to similarity measurement between a flight action to be evaluated and a standard action, global information and local information are considered at the same time, a multi-view fusion scoring rule is formed, so that a scoring model is constructed, namely, the completion condition of a single action is considered, and scoring is performed in combination with performance in a complete flight sortie. Therefore, the multi-view automatic comprehensive evaluation of the flight action quality is realized.

Description

technical field [0001] The invention relates to the technical field of data mining, in particular to a method for self-adaptive evaluation of flight quality. Background technique [0002] The evaluation of flight quality is mainly divided into subjective method and objective method. The subjective method is generally used to score the pilots after the experts observe the flight process or the three-dimensional trajectory reproduction process and the parameter transformation rules. It relies on the expert experience of the flight instructor and has a strong Therefore, the degree of scoring in practical applications is not high, which not only consumes a lot of manpower and material resources, but also makes it difficult to evaluate the quality of pilot training objectively and fairly; the objective method is to select the parameters in the flight data to At present, there are few literatures on the evaluation of flight quality by objective method. In 2018, Xiao Yanping et al....

AI Technical Summary

Problems solved by technology

[0002] The evaluation of flight quality is mainly divided into subjective method and objective method. The subjective method is generally used to score the pilots after the experts observe the flight process or the three-dimensional trajectory reproduction process and the parameter transformation rules. It relies on the expert experience of the flight instructor and has a strong Therefore, the degree of scoring in practical applications is not high, which not only consumes a lot of manpower and material resources, but also makes it difficult to evaluate the quality of pilot training objectively and fairly; the objective method is to select the parameters in the flight data to At present, there are few literatures on the evaluation of flight quality by objective method. In 2018, Xiao Yanping et al. conducted a comprehensive analysis of the aircraft’s lateral flight quality evaluation indicators from the perspective of system dynamics model. In 2019 In 2010, Wang Yuwei and others built a five-level comprehensive evaluation structure system of flight quality, and used the comprehensive weighting method to carry out weighted evaluation of the basic action scores. Therefore, for different flight actions, it is necessary to extract the standard flight quality of the action. It is time-consuming and labor-intensive to calculate the parameter indicators again, and does not consider the impact of other flight parameters on quality evaluation

Affected by various factors such as aircraft performance, pilot's operating habits, and flight environment, and complex maneuvers are highly random and fuzzy, the above algorithms use hard division of precise thresholds, knowledge expression and The reasoning process does not fully reflect the randomness and ambiguity of complex maneuvers, and the promotion value in practical applications is not high

Images