Method and system for importing hypersonic flow field into BP neural network for refined processing

Abstract

The invention belongs to the technical field of combination of sound velocity physical calculation and artificial intelligence, and discloses a method and system for importing a sound velocity flow field into a BP neural network for fine processing. The method comprises the steps: employing an error reverse neural network algorithm to refine a sparse flow field data sample, and forming refractiveindex flow field distribution with higher data density; and refining the sound velocity refractive index six-field distribution at the window by adopting the convergence error and the sampling point number. According to the method, sparse data are used as samples to be imported into the neural network for training, a single hidden layer is adopted for propagation of the neural network, and when the number of output layers is calculated, and when the error of the output layers does not meet the specified requirement, the weight of connection between neurons is changed, and calculation is conducted again; after the sampling points are denser, the refractive index distribution near shock waves is smoother and clearer, the fidelity is obviously improved, and the method is closer to the actualsituation.

Description

technical field [0001] The invention belongs to the technical field of the combination of sound velocity physical calculation and artificial intelligence, and in particular relates to a method and system for importing a sound velocity flow field into a BP neural network for refined processing. Background technique [0002] At present, the research on space target attack and hypersonic aircraft has attracted more and more attention. When a hypersonic aircraft flies in space at high speed, it will compress the surrounding air to form a high-temperature shock wave around the flow field, and the surrounding air will ionize to form plasma. The body sheath, electromagnetic waves will be difficult to penetrate, which makes the coupling of optics and hypersonic flow field rise. The complex flow field interferes with the beam transmission and causes the wavefront distortion of the beam, which is called the aero-optical effect. Most of the light beam transmission in the hypersonic flo...

AI Technical Summary

Problems solved by technology

[0006] Through the above analysis, the existing problems and defects of the existing technology are as follows: (1) The cost of the existing hypersonic flow field experiment is relatively high, the data sampling process is relatively complicated, and it is difficult to meet the requirements when fine flow field data is required at the position of the flight window;

[0007] (2) For optical transmission, the fine refractive index field is often the decisive factor in determining the calculation of the aero-optical effect. If the flow field is not extremely fine, it is difficult to obtain accurate results;

[0008] (3) The ordinary interpolation method reconstructs the flow field, and the accuracy is not high. To achieve the required accuracy, it takes a lot of computing time and the efficiency is too low

Images