Process parameter optimization of pumpkin combined drying based on genetic algorithm

Abstract

[Objective] To improve the drying quality of pumpkins. [Methods] This study conducted experiments on hot air drying, vacuum drying, and combined hot air-vacuum drying, these drying characteristics were evaluated and compared based on unit energy consumption, rehydration ratio, and color difference indicators. Combining BP neural network model with genetic algorithm, combined with entropy weight and weighted scoring method, a multi-objective comprehensive optimization was carried out for the combined hot air-vacuum drying of pumpkins. [Results] Under the same conditions, the highest drying efficiency was hot air drying; And the findings revealed that at drying temperatures was 55 ℃, with a moisture content transition point of 30%, the combined drying method reduced the drying time by 52.63%, compared to vacuum drying. The lowest unit energy consumption was vacuum drying; The worst rehydration performance was hot air drying. The best color was vacuum drying. The optimal drying parameters determined by the genetic algorithm combined with a BP neural network model were a hot air drying temperature of 65 ℃, conversion point moisture content of 50%, and vacuum drying temperature of 56.050 9 ℃. Verification experiments demonstrated that the average relative errors between the genetic algorithm optimized values and the experimental values for unit energy consumption, rehydration ratio, and color difference were 2.5%, 5.53%, and 4.84%, respectively, all lower than 6%. [Conclusion] The combined hot air-vacuum drying of pumpkin integrates the advantages of both hot air drying and vacuum drying, and combined with BP neural network genetic algorithm model can optimize the process parameters for pumpkin hot air vacuum drying.

Keywords

• pumpkin
• combined drying
• hot air drying
• vacuum drying
• genetic algorithm
• entropy weight method