Multiple Linear Regression Based Silica Melting Process via Partial Differential Equations

Iron tailings are the waste after beneficiation and the main component of industrial solid waste. This paper addresses the issue of the change of centroid position, movement trajectory, edge contour characteristic index and actual melting rate during the melting process of silica. Discrete digital image is converted into a continuous model via the partial differential equation. With the thermal model derivation, multiple linear regression, energy function models, differential equations and other models; the centroid position, motion trajectory and characteristic index of silica during high temperature melting change and melting rate are obtained; then the process of solving the center’s mass position is optimized through the artificial neural network regression method. Image processing is based on partial differential equations, a center of gravity calculation model based on multiple linear regression methods, and energy function models for solving characteristic indicators.