GASOLINE BLENDING SYSTEM MODELLING VIA STATIC AND DYNAMIC NEURAL NETWORKS

摘要

Gasoline blending is an important unit operation in the gasoline industry. A good model for the blending system is beneficial for supervision operation, prediction of gasoline qualities, and realizing of model-based optimal control. The gasoline blending process involves two types of properties: a static blending property and a dynamic property of blending tanks. As the blending cannot follow the ideal mixing rule in practice, we propose static and dynamic neural networks to approximate the two types of blending properties. Input-to-state stability approach is applied to access robust learning algorithms of the two neural networks. Numerical simulations are provided to illustrate the neuro modelling approaches.