Prediction method of aeroengine transition state acceleration process parameters based on spatial reconstruction

Abstract

The invention belongs to the technical field of prediction of aero-engine performance parameters, and provides a method for predicting key performance parameters in a transient-state acceleration process of an aero-engine based on space reconstruction. According to the method provided by the invention, a training data set and a test data set are established by utilization of test data in the transient-state acceleration process of the aero-engine provided by a certain research institute; dimension rising of the data set is carried out based on data space reconstruction of an automatic coder; model parameters are optimized by adoption of a population optimization algorithm represented by a particle swarm optimization; and finally, transient-state performance parameters are regressed by utilization of a random forest regression algorithm having great performance to high-dimensional data; and thus, starting from the engineering application angle, effective real-time prediction is realized.

Description

technical field [0001] The invention belongs to the technical field of aero-engine performance parameter prediction, and in particular relates to a method for predicting parameters of an aero-engine transition state acceleration process based on space reconstruction. Background technique [0002] Aeroengines operate in a complex environment of high temperature, high pressure, and high speed for a long time, and the possibility of failure increases with time. The performance of the transition state acceleration process is directly related to the progress of the aircraft take-off and acceleration flight process. Due to the extremely complex mechanism of the aero-engine, it is very difficult to model the parameters of the transition state process. Therefore, the use of a data-driven advanced prediction method for aeroengine performance parameters can avoid model building for complex engine mechanisms, and at the same time predict the parameter state of the engine transition st...

AI Technical Summary

Problems solved by technology

However, the traditional forecasting algorithm has high requirements on model parameters and input features. Usually, an optimization algorithm is used to adjust the parameters before modeling; the influence of each parameter feature of different types of engines on the prediction accuracy of the model is also different, and it needs to be re-calculated. Feature selection; the change rule of the transition state acceleration process parameters is complex, and the model parameters have a greater impact on the prediction accuracy; the traditional regression prediction algorithm performs poorly on high-dimensional samples, and the transition state acceleration process data sample space dimension is far from enough Describe the performance of aero-engines; the generalization ability of the model is poor. When the engine changes, it is necessary to re-select the model parameters and input features, which increases the consumption of manpower and financial resources to a certain extent.

Images