Establishment of regression model for estimating shape parameters for vacuum-sealed glass panel using design of experiments

Presently, the technique of edges sealing of vacuum glass panels is being utilized in a variety of industrial applications, such as in displays and home appliances. The sealing conditions of a vacuum glass panel strongly affect its key performance parameters, such as its insulation and strength. In the present study, edges of two glass panels were melted and sealed using a hydrogen mixed gas torch. The edges were sealed after appropriately setting process parameters that affected the shape of the sealed edges. Regression models were established for estimating the edge thickness, deflection, and maximum radius of the sealed part, which were considered as shape parameters. The effects of the process parameters on the shape parameters, as well as the interactions among the process parameters, were analyzed, and a polynomial regression model that considered these interactions was established. The feasibility of all the regression models was verified through analysis of variance.