Data-driven decision making and production optimization of higher mathematics in industrial science

Abstract

Since complex industrial chemical systems contain huge process data, these process data will present the operation characteristics and laws, and the use of appropriate methods to analyze the data is one of the feasible directions for fault diagnosis. In this paper, we analyze data from industrial chemical production processes using the support vector machine algorithm as our decision-making approach. Considering the large amount of data generated in industrial chemical systems and its nonlinear characteristics, this study applies Gaussian and non-Gaussian space to obtain the high-dimensional characteristics of the data before putting it into an SVM model. The rotary drying kiln production simulation experimental process is optimized by applying the fault diagnosis method, with the thermal efficiency shown by several tests being close to each other. The method proposed in this paper has obvious advantages in terms of search efficiency for a feasible global optimal solution.

Keywords

• gaussian space
• support vector machine
• dssvm integration algorithm
• production optimization
• industrial chemistry
• 03b70