PSO-Optimized Data-Driven and Mechanism Hybrid Model to Enhance Prediction of Industrial Hydrocracking Product Yields Under Data Constraints
Open Access
Abstract
The accurate prediction of hydrocracking product yields is crucial for optimizing resource allocation and improving production efficiency. However, the prediction of product flowrates in hydrocracking units often faces challenges due to insufficient data and weak correlations between input and output variables. This study proposes a hybrid framework combining a Convolutional Neural Network–Long Short-Term Memory (CNN-LSTM) model, mechanism modeling, and Particle Swarm Optimization (PSO) to address these issues. The CNN-LSTM captures spatiotemporal dependencies in operational data, while the mechanism model incorporates domain-specific physical constraints. The hybrid model is structured in both series and parallel configurations, with PSO optimizing key hyperparameters to enhance its predictive performance. The results demonstrate significant improvements in prediction accuracy, with determination coefficients (R2s) reaching 0.896 (kerosene), 0.879 (residue), 0.899 (heavy naphtha), and 0.78 (light naphtha). Shapley Additive Explanations (SHAP) and Mutual Information Coefficient (MIC) analyses highlight the mechanism model’s role in improving feature interpretability. This study underscores the efficacy of integrating kinetics modeling, deep learning, and metaheuristic optimization for complex industrial processes under data constraints, offering a robust approach to enhance hydrocracking yield prediction.Keywords
Funding Information
- Science and Technology Development projects of SINOPEC (323063)
This publication has 49 references indexed in Scilit:
- Prediction of railroad track geometry change using a hybrid CNN-LSTM spatial-temporal modelAdvanced Engineering Informatics, 2023
- Comparative study of single- and two-stage slurry-phase catalytic hydrocracking of vacuum residue for selective conversion of heavy oilCatalysis Today, 2023
- Predicting occurrence of liquefaction-induced lateral spreading using gradient boosting algorithms integrated with particle swarm optimization: PSO-XGBoost, PSO-LightGBM, and PSO-CatBoostActa Geotechnica, 2023
- A novel PSO algorithm for DG insertion problemEnergy Systems, 2022
- A comprehensive review on multiple hybrid deep learning approaches for stock predictionIntelligent Systems with Applications, 2022
- Machine-learning-guided reaction kinetics prediction towards solvent identification for chemical absorption of carbonyl sulfideChemical Engineering Journal, 2022
- Hydrocracking, hydrogenation and hydro-deoxygenation of fatty acids, esters and glycerides: Mechanisms, kinetics and transport phenomenaChemical Engineering Journal, 2022
- Kinetic mechanism research of an industrial hydrocracker based on strict calculation of stoichiometric coefficientsFuel, 2013
- Prediction of engine performance for an alternative fuel using artificial neural networkApplied Thermal Engineering, 2012
- Statistical estimation of the number of hidden units for feedforward neural networksNeural Networks, 1998