A Logarithmic Penalty Function Approach for Enhancing the Alkylation Process in Kaduna and Warri Refineries

Abstract

Petroleum industries play a vital role in the global economy, and research institutions need to focus on optimizing their chemical processes. Achieving the ideal operating conditions for these processes is crucial for effective management decision-making. In the alkylation reaction, isobutane reacts with light olefins, like butylenes, in the presence of a strong acid catalyst to produce high-octane compounds. To optimize the profit function without relying on complex computer programming codes, we utilized a novel logarithmic penalty function (LPF) approach. This approach was specifically designed to handle nonlinear programming problems with irregular features. The approach was implemented using the fminunc routine function, employing quasi-Newton methods, and compared it with existing conventional methods (using nonlinear programming system optimization laboratory (NPSOL) code and Hock-Schittkowsky (HS)). The results demonstrated higher efficiency compared to NPSOL and HS approaches as the LPF method yielded a higher objective value, implying a more favorable outcome in terms of maximizing the desired outcome (e.g. profit).