Hypersonic aircraft fuel tank temperature field prediction method based on point cloud network

Abstract

The invention provides a hypersonic aircraft fuel tank temperature field prediction method based on a point cloud network. The hypersonic aircraft fuel tank temperature field prediction method comprises the following steps of S1, obtaining shape features, material composition and working condition variables of a hypersonic aircraft fuel tank; s2, according to the heat transfer layering condition of the fuel tank, grid division is carried out, and boundary conditions are set; establishing a numerical simulation model for three-dimensional heat transfer of the fuel tank and verifying the numerical simulation model; s3, temperature field point cloud data on the longitudinal section of the middle position of the numerical simulation model are selected, and a point cloud sample data set of the fuel tank temperature field is established; s4, dividing a training data set and a test data set based on the point cloud sample data set; s5, establishing a point cloud network model, performing training verification on the point cloud network model by using the training data set and the test data set, and obtaining a temperature field prediction model based on the point cloud network; and S6, predicting the temperature field by using the temperature field prediction model.

Description

technical field [0001] The invention relates to the technical field of temperature prediction of a fuel tank of a hypersonic aircraft, in particular to a method for predicting the temperature field of a fuel tank of a hypersonic aircraft based on a point cloud network. Background technique [0002] Hypersonic vehicles have strong penetration capability and survival probability, and are of great significance to national defense security and scientific and technological progress. As the core power equipment of a hypersonic aircraft, the stable operation of the fuel tank directly affects the performance and safety of the aircraft. However, hypersonic aircraft mostly fly at high speed in the atmospheric region prone to high temperature effects, resulting in a sharp rise in the temperature of the outer surface of the aircraft, causing uneven changes in fuel temperature, resulting in thermal stratification in the fuel tank due to changes in fuel density. Greatly affect the reliab...

AI Technical Summary

Problems solved by technology

[0004] However, the existing temperature field prediction method based on the data-driven deep learning model cannot effectively use the grid point structure characteristics of the fuel tank, and can only give a color temperature field composed of pixel values ​​​​after the given operating conditions variables The image does not give a sparse temperature field including the temperature values ​​of all grid points in the shape space of the fuel tank, so it is difficult to obtain a high-precision temperature field, and it is impossible to effectively analyze the temperature field of the entire fuel tank space

At the same time, the temperature field prediction method based on the data-driven deep learning model can only model a single fuel tank structure alone, but the actual demand is that the temperature field of different fuel tank structures can be applied

If the existing deep learning model is used to model multiple fuel tank structures and working condition variables separately, the correlation between the fuel tank structural features, working condition variables and temperature fields will be artificially split, which may lead to some useful The loss of information makes the model less accurate

Images