Scholarlink

<p>Vacuum frying is a potential method for developing fried products with reduced fat content and enhanced quality. However, its application on restructured products, especially those based on millet, remains understudied. This study is the first to apply an artificial neural network coupled with a genetic algorithm (ANN-GA) optimization for vacuum frying of millet-based snacks. The aim was to examine the influence of vacuum frying on product quality and optimize the process parameters. A Box–Behnken design was employed with independent variables, frying temperature (120, 130, and 140 °C), time (4, 5, and 6 min), and pressure (35, 45, and 55 cm of Hg). Optimization and model comparison were performed using response surface methodology (RSM) and ANN–GA. Findings revealed that increasing temperature and time reduced moisture and increased fat absorption. Additionally, oil uptake may be influenced by the cooling phase suction effect, where pressure differences draw oil into the pores. Minimal color changes (1.01–7.09) were observed in vacuum-fried millet-based snacks. The snacks' hardness (3.28–30.99 N) increased with elevated temperature and pressure due to crust development. The overall acceptability of 7.7/9 was obtained at 130 °C, 45 cm of Hg, and 5 min. The ANN-GA showed better predictive performance than RSM, with higher average R<sup>2</sup> (0.96) and a reduced 12.6% prediction error. Vacuum frying effectively produces high-quality millet-based snacks with desirable texture and appearance, demonstrating favourable opportunities for the snack industry.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Modeling and optimization of vacuum frying parameters for gluten-free millet-based ready-to-eat snack using ANN-GA and RSM

  • 期刊论文

  • 发表日期：

    2026年6月20日

查看全文
  • Heena Sudhakar Kamdi,
  • Sivaranjani. S,
  • Jayasree Joshi T,
  • P. Srinivasa Rao

摘要

Vacuum frying is a potential method for developing fried products with reduced fat content and enhanced quality. However, its application on restructured products, especially those based on millet, remains understudied. This study is the first to apply an artificial neural network coupled with a genetic algorithm (ANN-GA) optimization for vacuum frying of millet-based snacks. The aim was to examine the influence of vacuum frying on product quality and optimize the process parameters. A Box–Behnken design was employed with independent variables, frying temperature (120, 130, and 140 °C), time (4, 5, and 6 min), and pressure (35, 45, and 55 cm of Hg). Optimization and model comparison were performed using response surface methodology (RSM) and ANN–GA. Findings revealed that increasing temperature and time reduced moisture and increased fat absorption. Additionally, oil uptake may be influenced by the cooling phase suction effect, where pressure differences draw oil into the pores. Minimal color changes (1.01–7.09) were observed in vacuum-fried millet-based snacks. The snacks' hardness (3.28–30.99 N) increased with elevated temperature and pressure due to crust development. The overall acceptability of 7.7/9 was obtained at 130 °C, 45 cm of Hg, and 5 min. The ANN-GA showed better predictive performance than RSM, with higher average R2 (0.96) and a reduced 12.6% prediction error. Vacuum frying effectively produces high-quality millet-based snacks with desirable texture and appearance, demonstrating favourable opportunities for the snack industry.