复杂体系的过程模拟研究
文献类型：学位论文
| 作者 | 张香平 |
| 学位类别 | 博士 |
| 答辩日期 | 2004 |
| 授予单位 | 中国科学院过程工程研究所 |
| 授予地点 | 中国科学院过程工程研究所 |
| 导师 | 张锁江 |
| 关键词 | 过程模拟 复杂体系 碳酸二甲酯 重质油热裂化 蛋白质溶解度 相图 相平衡 |
| 其他题名 | Studies on the Simulation of Complex Systems |
| 学位专业 | 化学工艺 |
| 中文摘要 | 过程模拟技术的出现和发展总是与实际工业的发展紧密相关的。随着过程工业的发展重点逐渐从大宗产品向高附加值产品的转变，精细化学品、医药工程、新材料等的研究开发越来越受到重视，过程模拟技术也将重点转向精细化工、医药工程、环境工程以及一些特殊体系过程的模拟和分析。与传统的石油化工体系相比，这些体系常常表现为多相多组分及复杂热力学性质，如共沸物的反应精馏祸合过程，蛋白质或生物医药大分子的结晶过程，以及一些多组分体系的物理化学转化过程等。对这些体系进行系统的研究，将有助于充实和发展过程模拟技术的理论和方法。本研究以复杂体系的模拟为主线，主要进行了如下研究：（1）对绿色产品碳酸二甲醋的酷交换法工艺进行了过程模拟。根据该物系强的非理想性的特点，首先建立了基于UNIQuAC一RK方程的甲醇一碳酸二甲醋一碳酸丙烯醋一丙二醇四元体系的热力学汽液平衡模型。该模型不仅可用于常压下该共沸物体系汽液相平衡的预测，在较高压力（＜1.5MPa）下也得到了非常好的预测结果。其次采用平衡级模型计算了醋交换法合成碳酸二甲醋的反应精馏塔，对各关键参数进行了灵敏 度分析，计算结果与工业实际数据进行了对比。依次对该流程中的变压分离提纯精馏塔，间歇减压精馏丙二醇提纯塔进行了模拟计算与分析，最后采用序贯模块法对醋交换法进行了全流程模拟。（2）对重质油热裂化复杂过程进行了模拟计算。基于“虚拟馏程”方法和“集总模型” 的方法对重质油热裂化复杂体系进行了组分划分，建立了基于集总模型的动力学方程，根据实验数据回归了动力学参数。建立了基于平推流模型的焦炭塔的反应器模型，对重质油热裂化过程进行了模拟计算。建立了平衡级的馏分分离塔模型，对热裂化产品的分离过程进行了模拟，得到了裂化产品，即汽油、瓦斯油1、瓦斯油2 和焦炭。在此基础上，完成了重质油热裂化过程的全流程模拟。该方法也为重质油或原油的催化裂化过程的模拟计算提供了可行的方法。（3）分别采用机理热力学模型和智能方法挤卜经网络）对盐溶液体系中蛋白质的溶解度进行了模拟计算。首先改进了计算蛋白质一水体系相图的Lattice model模型，引入可调参数，使得该模型能较好地用于计算蛋白质一盐一水体系的相图，得到不同温度下 的溶解度值。由于基于了严格的热力学模型，该模型具有很好的预测性。其次采用神经网络（ANNs）对蛋白质盐水体系中蛋白质的溶解度进行了模拟计算，预测结果的误差远小于经验关联式、热力学模型的预测值，表明ANNs方法能有效地用于求解复杂体系的热力学性质，可以避免一般热力学方法繁琐且需要回归或确定许多参数的限制，为复杂体系热力学性质的预测研究提供了新方法。 |
| 英文摘要 | The appearance and development of process simulation technology always have the deep and strong background of industrial engineering. With the developing emphases of process engineering is transfer from the commodity chemistry to the special and high-value chemistry, such as special fine chemistry engineering, biochemical engineering, environmental chemical engineering, pharmacology engineering, and so on. Comparing to the traditional petrochemical systems, the characteristic of those complex systems mentioned above are multi-phase and multi-component, such as reactive-distillation with azeotropic system, crystallization process of protein or some other macro-molecular. In this report, to the simulation of complex systems, the following works have been researched. Modeling and simulation of synthesis process of green product- Dimethyl carbonate (DMC) with transesterification method. As it is a typical non-ideality system, we developed vapor-liquid equilibrium thermodynamic model of the methanol- Dimethyl carbonate- Propylene carbonate-Propanediol quaternary system based on the UNIQUAC-RK equation. The model could predict the middle-pressure system (<1.5MPa) accurately. Then, simulating the reactive distillation column with the equilibrium-stage model and analyzing the critical parameters. Finally, the flowsheet of synthesis DMC with transesterification method was simulated with the sequential-module approach. Simulating of heavy oil cracking process. Based on the "Pseudo-distillation region" and "lump model" approach, the heavy oil mixture and the cracking products are divided into nine lump and the reactive kinetics equations are established. An equilibrium-stage model was used to separating the cracking products, and the whole flowsheet of cracking process was simulated. The calculated yields of cracking products of gas, gasoline, gasoill, gasoi!2, coke were agreed well with the industrial design data. Simulation and prediction of solubility of protein in salt aqueous with lattice model and artificial neural network(ANN). After introducing the adjusting parameters, the phase diagram of protein-salt-H20 is calculated and the solubility curve in different temperature is obtained. This model has the good ability of prediction and extrapolation. A back propagation (BP) network was used for predicting the solubility of lysozyme in lysozyme-NaCl-water systems. It was found that a properly selected and trained neural network could fairly represent the dependence of protein solubility on the pH, salt concentration, and temperature. It demonstrated tliat the ANNs could hold promise as a strategy for solving the thermodynamic properties in complex systems, which can avoid the limitations of conventional thermodynamic methods that is tedious and requires determination of "parameters" which is arbitrary in many ways. |
| 语种 | 中文 |
| 公开日期 | 2013-09-16 |
| 页码 | 101 |
| 源URL | [http://ir.ipe.ac.cn/handle/122111/1390]  |
| 专题 | 过程工程研究所_研究所（批量导入） |
| 推荐引用方式 GB/T 7714 | 张香平. 复杂体系的过程模拟研究[D]. 中国科学院过程工程研究所. 中国科学院过程工程研究所. 2004. |
入库方式： OAI收割
来源：过程工程研究所
浏览0
下载0
收藏0
其他版本
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。