Spacecraft energy system design method

Abstract

The invention relates to a spacecraft energy system design method. The method comprises the following steps: S1, constructing a general spacecraft power supply and distribution system model; s2, inputting prior on-orbit data into a recurrent neural network (RNN) for data generalization, and realizing refined mining of prior empirical data; s3, constructing an actual spacecraft power supply and distribution system model; s4, in a ground design stage, classifying energy balance according to a spacecraft task sequence, establishing sub-samples, and generating a ground test case; s5, correcting the modelable state of the actual spacecraft power supply and distribution system model by adopting a data mining method according to the actual ground test data of the spacecraft, and performing data fitting on the non-modeled state of the actual spacecraft power supply and distribution system model; and S6, in an in-orbit operation stage, performing in-orbit high-precision correction on the actual spacecraft power supply and distribution system model by adopting a data mining method according to the in-orbit measured data. The method has the beneficial effects that the spacecraft energy subsystem is designed according to the operating environment, the task sequence, the working state of each single machine and the previous on-orbit data.

Description

【Technical field】 [0001] The invention relates to the field of aerospace technology, in particular to a design method for a spacecraft energy system. 【Background technique】 [0002] With the increasing diversification and complexity of space missions, it is necessary to carry out detailed verification of each subsystem in the process of initial design and ground verification of spacecraft. The energy subsystem is an important support and guarantee for the effective work of the spacecraft in orbit. The reasonable supply and distribution of the energy subsystem is directly related to the performance of the spacecraft to complete the mission. The calculation is based on the envelope, and the charging performance of the sailboard under different working modes and the switching timing of each stand-alone machine are not fully considered. [0003] Recurrent Neural Network (RNN) is a kind of neural network, which is very effective for data with sequential characteristics, and can ...

AI Technical Summary

Problems solved by technology

The energy subsystem is an important support and guarantee for the effective work of the spacecraft in orbit. The reasonable supply and distribution of the energy subsystem is directly related to the performance of the spacecraft to complete the mission. Envelope calculations do not fully consider the charging performance of sailboards in different working modes and the switching timing of each stand-alone machine, etc.

Images