一种基于图卷积网络的电弧炉状态预测方法及系统
By using a graph convolutional network-based method, combined with Bernstein polynomial spectral filtering and neural controlled differential equations, we have achieved modeling of the complex nonlinear coupling relationship and continuous-time dynamic evolution of multi-source operating parameters of electric arc furnaces. This solves the problem of accurately predicting furnace conditions during the electric arc furnace smelting process and improves the operational stability and production efficiency of the electric arc furnace.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- 西冶科技集团股份有限公司
- Filing Date
- 2026-05-20
- Publication Date
- 2026-07-17
AI Technical Summary
Existing technologies struggle to accurately predict furnace conditions during electric arc furnace smelting, especially during the switching between different smelting stages such as melting, oxidation, and reduction. They are unable to effectively characterize the high-order nonlinear relationships and continuous dynamic evolution characteristics among multiple operating parameters, leading to difficulties in electric arc furnace process optimization and energy consumption control.
A graph convolutional network-based approach, combined with Bernstein polynomial spectral filtering and neural controlled differential equations, is used to construct an electric arc furnace operating parameter graph. By combining the continuous-time multivariate state observation sequence with the graph-coupled state features, the complex nonlinear coupling relationship and continuous-time dynamic evolution of the multi-source operating parameters of the electric arc furnace are modeled.
It improves the accuracy and stability of electric arc furnace condition prediction, can adapt to inconsistent sampling frequencies and data gaps in industrial sites, enhances the ability to identify different smelting stages, and improves the operational stability and production efficiency of the electric arc furnace smelting process.
Smart Images
Figure CN122245526B_ABST