Monitoring and path optimization of catalytic reformer in a refinery: Principal component analysis and A* algorithm application

Abstract

Catalytic naphtha reforming converts naphtha into gasoline or aromatics and is a vital petrochemical process in refineries. This paper developed a hybrid framework combining principal component analysis (PCA) and A* algorithm to find optimal operating conditions through process monitoring and path optimization. PCA is employed to develop an efficiency monitoring model using the process history data. The operating conditions with profit information are projected onto a two-dimensional plane using the first and second principal components. High- and low-profit intervals can be clearly distinguished on this map. The A* algorithm is then employed in finding an optimal trajectory from low-index to high-index on the projection map. Finally, the inverse transformation is used to obtain optimum variables for optimal operating conditions. The effectiveness of the proposed framework is verified by applying it to a Sinopec refinery.