Numerical model and optimization strategy for the annealing process of 3D coil cores. (September 2020)

• Record Type: Journal Article
• Authors: Dou, Ruifeng; Zhao, Haoxiang; Zhao, Pengfei; Wen, Zhi; Li, Xianhao; Zhou, Liang; Zhang, Rongzhao
• Is Part Of: Applied thermal engineering. Volume 178(2020)
• Publication Date: 2020-09
• Subjects: 3D coil core -- Grain-oriented electrical steel -- Anisotropic conduction -- Monte Carlo method -- Optimization strategy; Heat engineering -- Periodicals; Heating -- Equipment and supplies -- Periodicals; Periodicals; 621.40205
• DOI: 10.1016/j.applthermaleng.2020.115517
• Languages: English
• ISSNs: 1359-4311

Abstract

Highlights: Digital twin model gives a near real-time prediction. 3D features of heat transfer inside annealing furnace are reserved. Typical error of annealing model for 3D coil cores is less than 2.5%. Optimization strategy reduces 15% of annealing time. Abstract: Three-dimensional (3D) coil core is one kind of transformer iron core that is made of grain-oriented electrical steel. This material should be annealed before use because of the residual stress generated inside the silicon steel strip during production. The temperature evaluation of the 3D coil core is important to its annealing quality, but this temperature cannot be measured easily and the prediction method is seldom mentioned in the open literature. A numerical model, which considers the anisotropic heat conduction inside the 3D coil core and convection and radiation heat transfers inside the furnace, is built to link the temperature of furnace with the temperature of 3D coil core. Monte Carlo method is used to calculate the view factors between surfaces in the furnace. Thermal radiation resistance network method is used to obtain radiant heat flux. A computational fluid dynamics model is established to predict the convection heat transfer coefficient. The finite volume method is used to solve the anisotropic conduction inside the 3D coil core, wherein radiant heat flux and convection heat transfer coefficient serve as boundary conditions. The model considers the effects of size, position, location, orientation, and number of 3D coil cores on the temperature evolution. Onsite experiments verify that this digital model is precise in practical application, and the typical simulation error is less than ±2.5%. A straight forward optimization strategy is proposed on the basis of the digital model. The onsite experiments also show that the optimization method can effectively shorten the annealing time and increase the efficiency of the annealing process. Making a near real-time prediction and keeping 3D heat transfer features are the advantages of the present method.