Abstract
Industrial big data has received much attention from both academia and industry in recent years. In this talk, distributed parallel computing frameworks are introduced for big data modeling in the process industry, based on which various applications can be carried out, such as process monitoring, fault diagnosis, key performance indices prediction and diagnosis, etc. After introduction of the research background of this talk, distributed parallel data-driven models are demonstrated for big data analytics under different application scenarios, with evaluation experiments in real industrial processes. To conclude this talk, several promising issues are highlighted for future work.
