Parametric study and design of deep learning on leveling system for smart manufacturing

Sheet metal is widely used in the industry for metal forming purposes, such as metal stamping and metal cutting. It is often winded and storage in a coil form for transportation purposes. However, before any manufacturing process such as, cutting, or stamping, leveling is required as the residual stress inside coil is present which can cause distortion to the metal forming/cutting process. In conventional coil leveling machines, the machine parameters are often set by machine technicians with many years of experiences. In addition, the optimized machine parameter is achieved by trial and error method or based on experiences. However, the machine parameters are also not exactly trivial due to too many input factors which may cause changes to the outcome result. In the recent years, industry 4.0 and smart manufacturing has been a widely discussed topic in terms of industry manufacturing solutions in many different industrialized countries. In smart manufacturing, communication and interaction between machines have become an important role to improve manufacturing efficiency, flexibility and customization. As smart manufacturing focused on information process through real objects, it is required to digitize the experience through deep learning method. This paper is aimed to describe and study the deep learning application based on coil leveling system. Finally, through this study and experiment verification, analyzes on research directions and prospects of deep learning.