Scholarlink

<p>The proposed work is the outcome of the application of, advanced optimization techniques like Response Surface Methodology (RSM) and Artificial Neural Networks (ANN) to predict and optimize the flaxseed oil yield and antioxidant activity through supercritical CO₂ (SC-CO₂) extraction. A Box-Behnken design (BBD) of RSM has been used to design and evaluated the impact of five most important key variables that are temperature, pressure, co-solvent concentration, CO₂ flow rate, and particle size. The proposed ANN model, comprising five input variables and one hidden layer, demonstrated superior predictive performance compared to RSM, with higher accuracy and lower mean square error. It predicted that the optimal conditions of 300&#xa0;bar pressure, 60&#xa0;°C temperature, 8% ethanol, 6&#xa0;g/min CO₂ flow, and 0.50&#xa0;mm particle size, yielding 0.33&#xa0;g/g oil and 36.73% antioxidant activity, both closely aligned with experimental results. Flaxseed oil is rich in bioactive compounds such as total phenolic content (67.13&#xa0;mg GAE/100&#xa0;g), total flavonoid content (24.00&#xa0;mg QE/100&#xa0;g), and total tocopherol content (29.16&#xa0;mg/100&#xa0;g), enhancing its nutraceutical value. It also comprises a huge quantity of polyunsaturated fatty acids (67.4%), oleic acid and predominantly linolenic acid (51.57%). The combined use of RSM and ANN not only enables efficient extraction but also establishes a scalable, eco-friendly model for future industrial applications. This study advocates the advances in green extraction by enabling accurate, scalable, and sustainable processes through predictive modeling. The approach is versatile and applicable to other plant-based bioactive. Future work may explore hybrid modeling, real-time control, and life-cycle assessment to enhance industrial relevance.</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

Modelling and optimization of flax seed oil extraction and antioxidant activity via supercritical CO₂ using RSM and ANN

  • 期刊论文

  • 发表日期：

    2025年9月24日

查看全文
  • Sunil Kumar,
  • Amit Rai,
  • Kamlesh Prasad

摘要

The proposed work is the outcome of the application of, advanced optimization techniques like Response Surface Methodology (RSM) and Artificial Neural Networks (ANN) to predict and optimize the flaxseed oil yield and antioxidant activity through supercritical CO₂ (SC-CO₂) extraction. A Box-Behnken design (BBD) of RSM has been used to design and evaluated the impact of five most important key variables that are temperature, pressure, co-solvent concentration, CO₂ flow rate, and particle size. The proposed ANN model, comprising five input variables and one hidden layer, demonstrated superior predictive performance compared to RSM, with higher accuracy and lower mean square error. It predicted that the optimal conditions of 300 bar pressure, 60 °C temperature, 8% ethanol, 6 g/min CO₂ flow, and 0.50 mm particle size, yielding 0.33 g/g oil and 36.73% antioxidant activity, both closely aligned with experimental results. Flaxseed oil is rich in bioactive compounds such as total phenolic content (67.13 mg GAE/100 g), total flavonoid content (24.00 mg QE/100 g), and total tocopherol content (29.16 mg/100 g), enhancing its nutraceutical value. It also comprises a huge quantity of polyunsaturated fatty acids (67.4%), oleic acid and predominantly linolenic acid (51.57%). The combined use of RSM and ANN not only enables efficient extraction but also establishes a scalable, eco-friendly model for future industrial applications. This study advocates the advances in green extraction by enabling accurate, scalable, and sustainable processes through predictive modeling. The approach is versatile and applicable to other plant-based bioactive. Future work may explore hybrid modeling, real-time control, and life-cycle assessment to enhance industrial relevance.