Exhaust temperature prediction method of aero-engine starting process

Abstract

The invention belongs to the technical field of aircraft engine prediction, and provides an exhaust temperature prediction method of an aero-engine starting process. The method needs to solve the technical problem of filling a vacancy of exhaust temperature prediction of aero-engines in starting processes. According to the provided exhaust temperature prediction method of the aero-engine startingprocess, aero-engine ground test running data are utilized, a method of machine learning is employed to obtain an exhaust temperature prediction model of the engine starting process, the model is highin prediction precision and good in generalization performance, a prediction result can be further used for an engine control aspect and the like, and occurrence possibility of engine overheating isdecreased. Compared with traditional single-parameter prediction, the method employs fusion prediction, contains more information, and thus enables prediction errors to be reduced; and compared with single prediction algorithms, the method employs an AdaBoost.RT integration algorithm, integrates weak learning machines, and is thus lower in the prediction errors.

Description

technical field [0001] The invention belongs to the technical field of aero-engine prediction, and in particular relates to a method for predicting exhaust gas temperature during the start-up process of an aero-engine. Background technique [0002] When the aircraft is started, the aeroengine is in a state of high temperature, high load and high speed, and the possibility of overheating of the engine is high, which increases the flight risk. Therefore, it is necessary to predict the exhaust temperature and control the aeroengine in time to prevent An over temperature has occurred. There are three main methods for aeroengine exhaust temperature prediction: model-based methods, regression-based methods, and machine learning-based methods. Among them, the model-based method is computationally complex, and real-time calculations may have problems such as iterative non-convergence; the regression-based method sometimes does not necessarily have obvious linear or other functional...

AI Technical Summary

Problems solved by technology

The above methods all use the exhaust temperature of multiple flight cycles of the aero-engine as data to predict the performance status and degradation of the aero-engine, but they do not involve the exhaust temperature prediction of the entire start-up process, so they cannot Control the engine before the phenomenon

Images