Optimization of vacuum belt drying process conditions for Citri Reticulatae Pericarpium using genetic algorithm and particle swarm optimization

Vacuum Belt Drying (VBD) is commonly applied in the pharmaceutical industry and its optimization is often required for different product processing. This study focuses on the optimization of process parameters for vacuum belt drying of Citri Reticulatae Pericarpium using Genetic Algorithm (GA) and Particle Swarm Optimization (PSO). VBD equipment control parameters are numerous, and lack of mature commissioning methods. Optimizing VBD process parameters can improve production efficiency and reduce production costs. The objective function was based on a numerical model of VBD. This model accurately predicts the moisture content variation during the drying process of Citri Reticulatae Pericarpium with a mean squared error (MSE) value of 0.0353. After that, the model introduces the formula for the income-expenditure ratio as the industrial optimization objective. The multi-objective problem of high yield and low energy consumption was transformed into a single objective by minimizing the income-expenditure ratio. After process optimization, the income-expenditure ratio was reduced by 3.80 and 4.02% compared to the original process conditions. A comparison between GA and PSO optimization results revealed that PSO performed better in optimizing the VBD process conditions, demonstrating superior convergence speed and optimization outcomes. In conclusion, this research emphasizes the effectiveness of utilizing GA and PSO algorithms to optimize process parameters for VBD. The algorithms provide optimized process conditions for VBD, replacing traditional empirical adjustment methods. This study contributes to the advancement of VBD technology by providing a systematic approach for optimizing process parameters.