首页> 外文会议>International Symposium on Supercritical Fluids Tome 1: Natural Products, Process amp; Equipment Development; 20030428-20030430; Versailles; FR >APPLICATIONS OF ARTIFICIAL NEURAL NETWORKS IN SUPERCRITICAL FLUID EXTRACTION MODELLING AND SIMULATION
【24h】

APPLICATIONS OF ARTIFICIAL NEURAL NETWORKS IN SUPERCRITICAL FLUID EXTRACTION MODELLING AND SIMULATION

机译:人工神经网络在超临界流体萃取建模与模拟中的应用

获取原文
获取原文并翻译 | 示例

摘要

In this work, the artificial neural networks (ANN) technology was applied to the simulation of supercritical fluid extraction (SCFE) process of vegetable oil. For this technology, a 3-layer BP network structure is applied, and the operation factors such as pressure, temperature and extraction time are used as input variables of the netwotk, whereas the oil yield of extraction is treated as output values of the network. Optimization of the topological structure of the net, 6 neurals of middle hidden layer had been proved to be the optimum value. With the normalization pretreatment of the initial input data, not only the convergent speed and accuracy has been improved greatly, but also the problem of derivative at zero point has been solved. Therefore, the method is an improved model based on Fullana. In addition, we investigated the impact of the output variables on the net properties, and found that it is very convenient for the data correction when the extraction rate-time curve is set as training sample. An ANN-SCFE simulation system has been programmed. For the first time, the simulation for the SCFE procees of Hippophae Rhamnoides L. seed oil has been made, and the results show that the average absolute relative deviations (AARD) are lower than 6%.
机译:在这项工作中,将人工神经网络(ANN)技术应用于植物油超临界流体萃取(SCFE)过程的模拟。对于该技术,应用了三层BP网络结构,并且将压力,温度和提取时间等操作因素用作netwotk的输入变量,而将提取的石油产量视为网络的输出值。优化网络拓扑结构,证明中间隐藏层的6个神经网络是最佳值。通过对初始输入数据进行归一化预处理,不仅大大提高了收敛速度和精度,而且还解决了零点导数的问题。因此,该方法是基于Fullana的改进模型。此外,我们研究了输出变量对网络属性的影响,发现将提取率-时间曲线设置为训练样本对于数据校正非常方便。已对ANN-SCFE模拟系统进行了编程。首次对沙棘种子油的SCFE程序进行了仿真,结果表明平均绝对相对偏差(AARD)低于6%。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号