超临界萃取工艺参数优化及其控制方法研究

发布时间：2018-11-16 09:11

【摘要】：超临界C02萃取(Supercritical CO2 Extraction)是一种新型的常温下进行的提纯技术,具有萃取效率高、产品质量好、能量损耗低、安全和可回收等优点；且萃取过程中超临界C02既有液体的特性同时也有气体的特性,因而有很好的溶解能力、传质性质,特别适用于热敏系数高、化学性质不稳定的天然产物的萃取方面,目前已经在天然香精香料和中草药有效成分的提取、食品工业、石油化工和轻工等各行各业中得到了广泛应用。本文采用五味子作为实验原料,在超临界C02萃取装置进行萃取分离实验,萃取实验结果对工艺参数与产品指标之间进行了讨论,建立基于支持向量机和遗传算法的萃取工艺条件与得率的数学模型,对萃取得率进行预测及超临界萃取条件进行优化,主要工作与研究结果概括如下：(1)研究超临界C02分离技术的原理、设备及应用,并结合影响萃取得率的主要指标,运用超临界萃取装置进行参数测试,根据实验数据,对超临界c02萃取技术工艺参数对萃取结果的影响进行了分析和讨论。(2)系统的阐明了支持向量机的基本原理和学习训练过程,针对超临界CO2萃取过程的复杂性,利用支持向量机全局最优、训练效率高特点与遗传算法全局搜索能力强的特点,建立了GA-SVM的预测模型,在此基础上,建立了基于该预测模型的多目标遗传算法的优化系统工艺参数模型,对得率进行了加权处理。运用该优化模型,可以获得以得率为优化目标的萃取压力、萃取温度等优化工艺参数,为企业以及后续试验提供合理的优化策略。(3)研究基于加权因子自调整的模糊控制方法,实现了该算法对超临界萃取温度的控制。以超临界萃取温度作为输入量,以热水箱为被控对象,设计了基于萃取温度的模糊控制规则自调整系统的仿真模型并对仿真结果进行了分析。
[Abstract]:Supercritical CO2 extraction (Supercritical CO2 Extraction) is a new purification technology at room temperature, which has the advantages of high extraction efficiency, good product quality, low energy loss, safety and recoverability. The supercritical CO2 has both liquid and gas properties during extraction, so it has good solubility and mass transfer properties, especially for the extraction of natural products with high thermal sensitivity and unstable chemical properties. At present, it has been widely used in the extraction of natural flavor, Chinese herbal medicine, food industry, petrochemical industry and light industry. In this paper, Schisandra chinensis was used as the experimental material, and the extraction separation experiment was carried out in the supercritical C02 extractor. The results of the extraction experiment discussed the relationship between the technological parameters and the product indexes. A mathematical model of extraction process and yield based on support vector machine (SVM) and genetic algorithm (GA) was established to predict the extraction yield and optimize the supercritical extraction conditions. The main work and research results are summarized as follows: (1) the principle, equipment and application of supercritical CO2 separation technology are studied. The influence of the technological parameters of supercritical CO2 extraction on the extraction results is analyzed and discussed. (2) the basic principle and learning and training process of support vector machine are systematically explained, and the complexity of supercritical CO2 extraction process is pointed out. Based on the characteristics of support vector machine (SVM) global optimization, high training efficiency and global search ability of genetic algorithm, the prediction model of GA-SVM is established. The optimization system process parameter model based on multi-objective genetic algorithm is established, and the yield is weighted. Using this optimization model, the optimum process parameters such as extraction pressure, extraction temperature and so on, with yield as the optimization target, can be obtained. It provides a reasonable optimization strategy for enterprises and subsequent experiments. (3) the fuzzy control method based on weighting factor self-adjustment is studied and the control of supercritical extraction temperature is realized by the algorithm. Taking supercritical extraction temperature as input and hot water tank as controlled object, the simulation model of fuzzy control rule self-adjusting system based on extraction temperature is designed and the simulation results are analyzed.
【学位授予单位】：长春工业大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TQ028.32;TP273.4

【参考文献】