@article{LIU2026, 
author = {Yingjie LIU and Zeen HUANG and Ping LIU and Haiqun CHEN and Junfeng QIAN and Wenjie LIU and Fuan SUN},
title = {An integrated virtual simulation platform with a modeling-optimization-evaluation framework based on multi-objective optimization: A case study of fractionation and absorption-stabilization system for fluid catalytic cracking},
year = {2026},
journal = {Experimental Technology and Management},
volume = {43},
number = {8},
pages = {235-241},
keywords = {virtual simulation, process simulation, fluid catalytic cracking, multi-objective optimization, experimental teaching reform},
url = {https://www.sciopen.com/article/10.16791/j.cnki.sjg.2026.08.028},
doi = {10.16791/j.cnki.sjg.2026.08.028},
abstract = {ObjectiveChemical engineering students often lack experience in complex engineering modeling, multi-objective decision-making, and integrated tool application. However, traditional experimental teaching is mainly limited to unit operations, resulting in limited exposure to full-process system cognition and insufficient capability to handle the coupling effects among multiple variables. To address these pedagogical gaps, this study develops an innovative virtual simulation teaching platform featuring an integrated modeling-optimization-evaluation framework. A 2-million-ton-per-year fluid catalytic cracking (FCC) fractionation and absorption-stabilization system is employed as the simulation case. The platform aims to systematically enhance students’ comprehensive engineering practice abilities, innovative thinking, and sustainability awareness.MethodsThe virtual simulation platform is constructed by integrating Aspen Plus for steady-state process simulation and MATLAB for advanced optimization algorithm implementation. It comprises three core models. The steady-state process simulation model, built on Aspen Plus, requires students to reconstruct the full FCC separation process based on actual refinery data. Based on the ASTM D86 data of the products, the pseudo-component method was applied to define the oil parameters. GRAYSON was selected as the property method for the fractionation column, while the RKS equation of state was applied to the absorption-stabilization system. The multi-objective optimization model employs the Nondominated Sorting Genetic Algorithm Ⅱ implemented in MATLAB. A bidirectional communication interface is established between MATLAB and Aspen Plus via COM technology, enabling automatic parameter transfer and iterative simulation calls. Through sensitivity analysis, nine key decision variables are identified from an initial set of twelve operational parameters based on their impact on product yield and system energy consumption. The Technique for Order Preference by Similarity to Ideal Solution is adopted to generate a Pareto optimal solution set that visualizes the inherent tradeoff between maximizing liquid petroleum gas (LPG) and stabilized gasoline yield and minimizing total system energy consumption. The comprehensive evaluation model systematically assesses optimization schemes from three critical dimensions: technical performance (product yield and quality specifications), economic feasibility (total annual cost, TAC), and environmental impact (CO2 emissions). This multi-dimensional assessment framework trains students to move beyond single-objective optimization thinking and develop balanced engineering judgment.ResultsSteady-state simulation results show that the yields of LPG, light diesel, stabilized gasoline, and dry gas agree well with industrial data, and all key product quality indicators meet the required specifications, validating the simulation accuracy. After multi-objective optimization, the combined yield of LPG and stabilized gasoline reaches 60.04% (by weight), representing a 2.44% increase over the pre-optimization value, while the energy consumption of the separation system is reduced to 42.3 MW, representing a decrease of 41.79%. These results demonstrate that multi-objective optimization effectively balances product yield and system energy consumption, achieving an overall optimal performance. Comprehensive evaluation further reveals that the TAC is reduced by 34.36%, with the total operating cost contributing the most significant decrease, while CO2 emissions are reduced by 41.85%, indicating a synergistic effect between energy conservation and carbon emission reduction.ConclusionsTeaching practice results indicate that the integrated platform can transform complex industrial engineering problems into systematic and operable experimental teaching projects. It enables students to develop skills in process simulation, parameter sensitivity analysis, multi-objective algorithm optimization, and multi-attribute decision-making, while fostering systematic engineering thinking and sustainable development awareness.}
}