Liquid hydrogen filling system safety assessment method based on evolutionary clustering learning

Abstract

The invention discloses a liquid hydrogen filling system safety assessment method based on evolutionary clustering learning, and belongs to the technical field of space launch safety assessment. The method includes the following steps that 1, a training sample is selected from historical perception data; 2, segmentation processing is carried out on the training sample; 3, similarity measurement is carried out with the l2-norm distance measure; 4, the group sparsity coefficient is solved with the segment coordinate relaxation algorithm; 5, a support vector machine is trained, and an evolutionary clustering segment dictionary is updated in real time; 6, the liquid hydrogen filling system safety measurement standard is quantitatively analyzed in combination with the system and state motion rule; 7, liquid hydrogen filling system safety assessment indexes are constructed, and the system safety level is described; 8, in online running, the steps 2-7 are repeated to achieve real-time assessment. According to the method, evolutionary clustering learning is introduced into a safety assessment theoretical framework, a dictionary learning and updating mechanism under data driving is described, the safety assessment indexes are constructed, and liquid hydrogen filling system safety assessment is achieved.

Description

technical field [0001] The implementation of the present invention relates to safety assessment technology, in particular to a safety assessment method for a liquid hydrogen filling system based on evolutionary clustering learning. Background technique [0002] In the 21st century, mankind will move toward space, explore space, develop space and utilize space in an all-round way. Space launch, as an important symbol to judge a country's comprehensive national strength, has important strategic significance. Space launch is a typical large-scale and complex engineering system involving multiple people, multiple machines, and multiple environments. It is composed of spacecraft, launch vehicles, and launch facilities. It may lead to operational accidents or even disasters, and its exploratory, experimental, dangerous and social characteristics determine the safety of space launch has a very important position. [0003] With the continuous development of science and technology, ...

AI Technical Summary

Problems solved by technology

[0010] It can be seen that the existing safety assessment methods do not fully consider the large-scale system and the strong coupling between subsystems under the influence of risk factors such as system / equipment process parameter overrun, failure and misoperation, which leads to the occurrence of system operation accidents. Mechanisms and evolution laws present complexity and uncertainty over time series

Images