Aero-engine fault prediction method based on Logistic regression and Xgboost model

Abstract

The invention discloses an aero-engine fault prediction method based on Logistic regression and an Xgboost model, and the method is characterized in that the method comprises the steps: obtaining a data set formed by a plurality of engine parameters of an airplane; constructing a time series data set according to the plurality of engine parameters; performing feature extraction on the time seriesdata set; manually marking the data corresponding to the engine fault in the time series data set; constructing an aero-engine fault prediction model based on an Xgboost model according to the plurality of marked time series data sets; obtaining a real data set of a known aero-engine fault, and enabling the real data set to pass through an aero-engine fault prediction model based on an Xgboost model to obtain a result value; and determining whether the aero-engine breaks down or not by substituting the result value into the Logistic regression function. The engine fault prediction is carried out on the aero-engine based on the Xgboost ensemble learning model and the Logistic regression model, so that the fault condition of the aero-engine can be effectively predicted.

Description

technical field [0001] The invention relates to the field of computer technology, in particular to an aeroengine fault prediction method based on Logistic regression and Xgboost model of a military flight big data maintenance field autonomous support information support system. Background technique [0002] From the 1990s to the present, aviation equipment technology has developed rapidly, especially in the context of the adjustment of military strategy and changes in the combat use of aviation equipment, the requirements for aircraft ground support are getting higher and higher, and the support of aero-engines the most fundamental factor. The rapid development of military technology has put forward higher requirements for the guarantee and failure prediction of aero-engines. But in the long-term development, the support technology of aero-engine always lags behind the technology of other aerospace equipment. The original aero-engine support system is facing huge challenge...

AI Technical Summary

Problems solved by technology

The original aero-engine support system is facing huge challenges under the new equipment conditions, and the lack of support for it will greatly reduce the combat readiness rate of military aircraft

[0003] The failure prediction of aero-engines lacks quantitative analysis, and the experience and data accumulated in the actual use and maintenance process have not been well combined with the design data for analysis, resulting in a separation between theory and practice

There is no early warning mechanism when an aero-engine fails, and it is difficult for the maintenance personnel of the outfield equipment to know the aero-engine equipped, lack of predictability, and the coexistence of excessive maintenance and insufficient maintenance, resulting in a decline in the integrity rate of the aero-engine

[0004] When an aero-engine fails, the fault data at this stage is not structured, and it is difficult for field maintenance personnel to conduct a clear fault diagnosis based on comprehensive analysis of fault phenomena, reliability data, index data, etc., making it difficult to find the most Optimal Failure Prediction Method for Aeroengine Replacement

This increases the maintenance cost of the aero-engine, and at the same time, the faulty engine cannot be well maintained, resulting in a waste of resources

Images