A data-driven method of product quality monitoring for industrial furnace tube centrifugal casting process

Industrial furnace tube plays an important role in petrochemical process, such as ethylene cracking process and hydrogenation process. Therefore, the quality control on furnace is important, as if some undetected unqualified tubes were produced and installed in industrial furnace, production accident would occur. Traditional quality control methods mainly depend on individual experience of engineers and operation parameter tuning is conducted manually. Critical causes for quality problems are not easy to be identified automatically. Thus, a multivariate statistical model is constructed to monitor the quality of furnace tubes and ensure the stable quality.