Heavy landing analysis method and device based on a multi-branch time convolutional network

Abstract

The invention provides a heavy landing analysis method based on a multi-branch time convolutional network. The method comprises the following steps: acquiring original parameter data and a dynamic time point; performing convolution operation on the original parameter data by using the improved time convolution network to generate a feature map of each parameter; performing feature extraction on the feature map to generate overall feature representation; learning a preset category by using the overall feature representation to obtain a parameter level of the preset category and a weight occupied by a feature map of each parameter; and according to the parameter level and the weight occupied by the feature map of each parameter, carrying out linear combination on the feature maps in the overall feature representation to obtain a final class activation mapping map, and according to the class activation mapping map, carrying out analysis on airplane heavy landing. According to the method, a new thought is provided for safety accidents or overrun events in the aviation field, reference is provided for interpretability work of the time sequence classification problem, technical reference is provided for flight safety, and the method has good theoretical and application values.

Description

technical field [0001] This application relates to the technical field of heavy landing, and in particular to an analysis method and device for heavy landing based on multi-branch temporal convolution network. Background technique [0002] Flight safety is the core concern of the civil aviation industry. According to Boeing's statistics on commercial aircraft accidents from 1959 to 2019, among all flight phases, the approach and landing phases are the most dangerous and prone to major safety accidents. Such as figure 2 As shown, the flight time in the landing phase only accounts for 1% of the entire flight process, but the accident rate is as high as 24% in this phase. Therefore, it is particularly important to ensure flight safety during the landing phase. Aiming at the safety issues in the landing phase, most studies mainly focus on the overrun events with a high occurrence rate, such as running off the runway, heavy landing, tail rubbing, etc.; Impact. Among them, Ha...

AI Technical Summary

Problems solved by technology

[0020] For this reason, the first purpose of this application is to propose a heavy landing analysis method based on multi-branch time convolutional network, which solves the problems of difficult acquisition of QAR data, numerous parameters, high professional barriers and incomplete compiled data. Problems; At the same time, it also solves the technical problems of poor interpretability and low prediction / classification accuracy of existing methods, and realizes the purpose of fully learning the time dimension characteristics of each parameter and improving the prediction accuracy of heavy landings. The idea of ​​the importance of image visualization features refines the interpretation of heavy landing events to the changes of each parameter at each moment, providing strong interpretability for heavy landings; at the same time providing It provides a new idea, provides a reference for the interpretability of time series classification problems, and also provides a new technical reference for flight safety, which has good theoretical and application value

Images