Method and system for predicting in-orbit deformation of remote sensing satellite

Abstract

The invention provides a remote sensing satellite in-orbit deformation prediction method and system, and the method comprises the steps: S1, collecting test result data under a reference working condition before and after each test working condition starts; s2, establishing a multi-layer neural network by using the test working conditions and the test result data, and identifying the neural network; s3, after identification of reference working conditions before and after all test working conditions is completed, deviation evaluation is carried out on identification results and test result data of the reference working conditions, and whether the established neural network is acquired or not is selected according to deviation evaluation conditions; and S4, on the basis of the step S3, substituting the thermal load corresponding to the thermal environment of the orbit into the acquired neural network, and predicting the in-orbit deformation condition of the satellite. According to the method, the state characteristics of the satellite real object can be represented more directly and more closely; and singularity and divergence of the established multi-layer neural network can be effectively avoided, so that the multi-layer neural network has relatively high accuracy and feasibility.

Description

Technical field [0001] The present invention relates to the field of spacecraft design and testing technology, relates to a remote sensing satellite deformation prediction method based on deep learning of thermal deformation test data, specifically, relates to a method of using superimposed working conditions to establish a multilayer neural network and using reference working conditions to identify, and finally substituted into orbital thermal environmental conditions to predict the deformation of satellites in orbit, in particular to a remote sensing satellite in orbit deformation prediction method and system. Background [0002] The orbital thermal environment is a typical external payload of a satellite during its orbital operation, which causes a significant deformation of the satellite during its orbital operation, which in turn adversely affects the accuracy retention ability of satellites, especially high-precision remote sensing satellites and spaceborne high-precision ...

AI Technical Summary

Problems solved by technology

In this invention, although the reverse correction of the simulation model based on the ground test data can maintain a good agreement between the analysis results of the simulation model and the test data to a certain extent, it is only based on the "maximum influencing factor of the thermal deformation index". On the basis of identification, the simplification and deviation between the simulation model and the actual state cannot be eliminated, so the limitation of this method is relatively large

Images