Component prediction method combining machine learning and CVD modeling

Abstract

The invention provides a component prediction method combining machine learning and CVD modeling. The method comprises the following steps: determining the size of a reactor and the shape and size ofa substrate; establishing a corresponding reactor geometric model; adding material attributes to the geometric region and the boundary; carrying out model mesh generation; selecting to establish a multi-physical field model, and carrying out multi-physical field coupling; establishing a fluid heat transfer and laminar flow model; establishing a concentrated substance transfer model; calculating aphysical field of laminar flow and fluid heat transfer coupling, and calculating a physical field interface of chemical and concentrated substance transfer by taking the obtained solution as an initial value to obtain concentration distribution of various intermediate substances obtained by reaction of the boron-carbon system precursor gas; respectively obtaining different results for contrastiveanalysis, and finally obtaining substance concentration distribution results under various deposition process conditions; a machine learning algorithm is utilized to associate the deposited boron-carbon ratio with the boron-carbon ratio, the boron-carbon ratio deposited under different deposition conditions is predicted, and the error is analyzed. The method can accurately predict the component ratio of the deposition product.

Description

technical field [0001] The invention relates to the technical field of material component analysis, in particular to a component prediction method combining machine learning and CVD modeling. Background technique [0002] Ceramic matrix composites are a type of composite material that is based on ceramics and combined with other fibers. It has good properties such as high strength, high modulus, low density, high temperature resistance, wear resistance and corrosion resistance. In particular, the high temperature resistance of ceramic matrix composites has drawn attention to their application research in high temperature environments. However, the biggest disadvantage of ceramic materials is that they are brittle and easily oxidized and corroded in high-temperature water-oxygen environments. Therefore, introducing a coating on the surface of ceramic matrix composites or designing a multi-component ceramic matrix is ​​an effective way to improve its high temperature perform...

AI Technical Summary

Problems solved by technology

[0005] Although the experiment has intuitive observation results, due to the guarantee of deposition quality, deposition usually occurs at lower temperature and pressure, resulting in long production cycle and high production cost of boron carbide deposition

Moreover, due to the special reaction system and harsh reaction conditions of the chemical vapor phase method, it is difficult to determine the depth and accuracy of the experimental determination.

like figure 2 As shown, two machine learning methods of error backpropagation neural network (BP) and support vector machine (SVM) were used to establish macroscopic process parameters (temperature, mole fraction of inlet components, pressure, flow rate) and the ratio of boron to carbon of deposition products ( The correlation between the molar ratio of B / C) is not satisfactory, indicating that the macroscopic process parameter correlation product component ratio is not the best way to describe the deposition mechanism (function)

Images