Search

• A method for modeling and optimizing processes with limited data is proposed. • Three swarm intelligence algorithms are studied to develop the method. • SI-M achieves high accuracy and reasonable computational time for process modeling. • SI-M with SMA balances exploration ability and computational cost well. • SI-O maximizes productivity and ensures the continuous operation simultaneously. Processes with unknown reactions and limited data are commonplace in the chemical industry. However, modeling and optimizing these processes are challenging tasks. In this paper, we propose the Swarm Intelligence-based Modeling and Optimization (SI-M/O) algorithm to address these challenges. The SI-M/O algorithm integrates swarm intelligence with chemical process fundamentals. This fusion empowers SI-M/O to navigate complex chemical landscapes effectively. Swarm intelligence excels at exploring vast solution spaces and adapting dynamically. When combined with first-principle knowledge of chemical reactions and thermodynamics, SI-M/O not only finds optimal solutions but also considers chemical feasibility and physical constraints. To validate its effectiveness, we applied SI-M/O to optimize a production plant, achieving a substantial 5.3 % productivity increase during preliminary testing. We also designed a user-friendly graphical interface for SI-M/O, enhancing accessibility for researchers and practitioners. [ABSTRACT FROM AUTHOR]

• Hybrid modelling showed higher accuracy than the first-principle model. • Polynomial statistical functions perform better than MRF function. • Differential evolution search is more accurate than the Particle Swarm Optimisation. • The training strategy has a significant impact on the models accuracy. • The complexity of the cost function showed a crucial role in the model training. Applying machine learning (ML) techniques is a complex task when the data quality is poor. Integrating first-principle models and ML techniques, namely hybrid modelling significantly supports this task. This paper introduces a novel approach to developing a hybrid model for dynamic chemical systems. The case in analysis employs one first-principle structure and two ML-based predictors. Two training approaches (serial and parallel), two optimisers (particle swarm optimisation and differential evolution) and two ML functions (multivariate rational function and polynomial) are tested. The polynomial function trained with the differential evolution showed the most accurate and robust results. The training approach does not significantly affect the hybrid model accuracy. However, the main effect of the training approach is on the robustness of the parameter predictions. The coefficients of determination (R2) on the test batches are above 0.95. In addition, it showed satisfactory extrapolation capabilities on different production scales with R2>0.9. [ABSTRACT FROM AUTHOR]