生物柴油酯交换反应过程的优化与控制策略研究
文献类型：学位论文
| 作者 | 史洪岩
|
| 学位类别 | 博士 |
| 答辩日期 | 2014-11-26 |
| 授予单位 | 中国科学院沈阳自动化研究所 |
| 授予地点 | 中国科学院沈阳自动化研究所 |
| 导师 | 王天然 |
| 关键词 | 生物柴油 酯交换反应 动态优化 双层结构预测控制 响应面优化 |
| 其他题名 | Research on Optimization and Control Strategies of Biodiesel Transesterification Process |
| 学位专业 | 机械电子工程 |
| 中文摘要 | 近年来，生物柴油的开发和利用成为世界各国研究的热点。生物柴油是以动植物油以及废弃油脂等为原料制成的、可以直接替代石化柴油使用的最具发展潜力的生物质液体燃料。目前，世界各国广泛采用化学酯交换法生产生物柴油，酯交换反应过程是生物柴油生产过程的核心单元，过程本身具有非线性、时变性和不确定性等特点，一般很难对其进行优化和控制。因此，建立和发展与过程特点相适应的优化技术和控制策略成为必然，有助于提高生物柴油过程的整体生产水平和效率。本文在跟踪最新国内外研究成果的基础上，对典型生产工艺下酯交换反应过程的优化与控制等关键问题进行了深入研究，为实现复杂化工流程的最优操作和最优控制提供了新的技术和解决方案。主要内容和创新性成果如下：(1) 酯交换反应过程机理建模及动态特性分析分析了酯交换反应过程机理和动力学，联合酯交换反应过程中的物料平衡、能量平衡和动力学方程，建立了酯交换反应过程的动态机理数学模型。参考实际过程操作数据和工艺参数，对模型进行了动态仿真和特性分析，研究了每个状态变量在过渡过程的动态响应，同时，对主要操作参数进行了灵敏度分析，验证了所建模型的准确性，为后续不同生产工艺下生物柴油酯交换反应过程优化与控制策略的研究提供可靠的仿真模型。(2) 面向间歇工艺的生物柴油酯交换反应过程的动态优化为满足生物柴油小批量、多品种、多规格的市场需求，生物柴油生产通常采用间歇工艺。针对间歇工艺生物柴油生产过程具有的时变、强非线性和无稳态工作点等问题，提出基于最大值原理的动态优化策略。建立了以固定反应时间内生物柴油产品浓度最大化为目标、以非线性微分方程组和控制变量有界为约束条件的最优控制模型，得到了最优控制问题的伴随方程和目标函数的梯度，详细推导了状态变量和伴随变量对控制变量的全微分解析形式，并结合问题特点设计了基于有约束控制的梯度算法进行求解，给出了具体的实施步骤。仿真结果表明，所提出的优化方法提高了产品浓度和生产效率，验证了方法的有效性。(3) 连续生物柴油酯交换反应过程的双层结构预测控制为提高能源综合利用和降低生产成本，现代工业规模化生产常常采用连续工艺。针对连续工艺生物柴油生产过程具有的非线性、多变量、强干扰和存在约束条件等问题，提出了双层结构预测控制策略。上层的稳态优化主要解决局部过程的线性经济优化以及对上层实时优化结果的目标跟踪问题。下层的控制器采用动态矩阵控制算法，对上层的稳态目标计算所得的最优设定值进行快速跟踪及克服过程扰动，实现设定点偏差的调节。两层在同一采样频率内执行，同步实现酯交换过程的稳态优化与动态控制。在双层结构预测控制方案实施前，基于相对增益理论，对控制系统进行了稳态模型增益的定量计算，对控制结构进行了合理设计，用来指导多回路PID控制。仿真分析表明，所提出的双层结构预测控制比多回路PID控制有更优良的性能，体现在控制输出有更少的超调，更小的振荡和更快的响应时间，同时能较好地抑制干扰的影响。(4) 基于实验装置的超临界生物柴油酯交换过程的响应面优化为实现连续化生产和改善操作条件，在超临界工艺连续制备生物柴油酯交换反应的实验装置基础上，提出了以金属氧化物TiO2为催化剂强化文冠果油制备生物柴油的响应面优化方法。通过单因素实验考察了反应温度、反应压力、醇油体积比对生物柴油收率的影响，得到了可进一步进行优化操作条件的范围。在此基础上，通过响应面优化法中的CCRD实验设计和Design Expert 7.0软件分析，建立了各因素与目标值的二次多项式回归模型，通过方差及响应曲面分析了各因素之间的交互影响。最后，根据模型以及验证实验，确定了该工艺下生物柴油收率的最佳工艺条件。(5) 生物柴油酯交换反应过程优化与控制软件设计设计并开发了生物柴油酯交换反应过程的优化与控制软件。软件具有过程建模、系统响应分析、基础控制、多变量预测控制、双层结构优化控制等功能，可以实现过程模型分析、优化与控制算法等的数字仿真，为过程的工艺改造、先进控制方案的设计及应用、系统的实时优化提供技术支持和理论指导。 |
| 索取号 | TP273/S57/2014 |
| 英文摘要 | In recent years, the development and utilization of biodiesel energy has become a hot research around the world. Biodiesel which is made from animal fats and plants oil and waste cooking oil as raw material, can directly replace the petroleum diesel to use and is the most development potential of biomass liquid fuel. Currently, transesterification method is widely used to produce biodiesel all over the world, transesterification process of biodiesel production is the core unit, the process itself is nonlinear, time-varying and uncertainty, in general it is difficult to optimize and control. Therefore, the establishment and development of optimization technology and control strategies adapted to the characteristics of the transesterification process become the inevitable, which will help to improve the yield of biodiesel and conversion rate of raw material, thus improving the overall production level and efficiency of the biodiesel process. Aiming at the optimization methods and control strategies for transesterification process, this paper researches deeply through tracking the latest research results and analyzing the typical production process. The research provides new technologies and solutions for the optimum operation and optimal control for complex chemical processes. The main contents and innovative results are as follows: (1) Mechanism modeling and analysis of dynamic characteristics for transesterification process Dynamic mechanism mathematical model of the biodiesel transesterification reaction process is constructed by jointing the material balance, energy balance and kinetics equations bases on process mechanism analysis. Refer to the operation data of transesterification process, the dynamic and characteristics of the model are analyzed through simulation experiments to study the dynamic response of each state variable during the transition. As well as sensitivity analysis of the main manipulated paramters under different process conditions are discussed in order to verify the correctness of the model, which provide the reliable model for process optimization and control strategy research. (2) Dynamic optimization of batch biodiesel transesterification process To meet biodiesel in small batch, multi-species, multi-standard market demand, production of biodiesel is often used for batch process. Batch process has time-varying, strong nonlinear and no steady working point features. Dynamic optimization strategy based on maximum principle is proposed on the basis of batch transesterification process model, an optimal control model is constructed aiming at maximizing the biodiesel concentration within the fixed reaction time and subjecting to the nonlinear differential equations and boundary constraints of control variables. The adjoint equations of optimal control problem and the gradient of the objective function are obtained, and then the differential analytical forms of state variables and adjoint variables are deduced in detail. A gradient algorithm as well as its concrete implementation is given. Simulation results show that the proposed method can improve the biodiesel product concentration and production efficiency and it is an effective method. (3)Two-layer structure predictive control of continuous biodiesel transesterification process To improve energy utilization and reduce production costs, biodiesel production of modern industrial is often continuous process. Continuous biodiesel process is a typical complex industrial process with nonlinear, multivariable, strong interference and constraints etc. A novel two-layer structure predictive control scheme for a continuous biodiesel transesterification process is presented. Among them, steady-state optimization of the upper layer is mainly to solve the linear economic optimization of local process and realize target tracking of real-time optimization results. Dynamic controller of the under layer based on dynamic matrix control algorithm is mainly to track fast the optimal setting value of steady-state target calculation and overcome disturbance to realize deviation adjustment. Two layers executed in the same sampling frequency, steady-state optimization and dynamic control of transesterification process are realized synchronously. Before the implementation of the two-layer structure prediction control scheme, the steady-state model gain of control system is calculated quantitatively based on the relative gain theory, and control structure is designed to establish a reasonable pairing control loop for realization the multi-loop PID control. Simulation results show that the two-layer predictive control strategy is more efficient in term of minimal overshoots, lower oscillation, faster response times and disturbances rejection as compared to conventional multi-loop PID controller. (4) Response surface optimization of supercritical biodiesel transesterification process based on experimental device Response surface methodology (RSM) based on experimental device of supercritical continuous biodiesel transesterification production from xanthoceras oil using TiO2 as the catalyst is carried out. Through the single factor experiment, the effects of reaction temperature, pressure and ratio of methanol/oil to biodiesel yield is investigated, the further optimization range of operation conditions is obtained. On this basis, central composite rotatable design (CCRD) of RSM and Design Expert 7.0 software are used to design the experiment and analyze operating parameters. The quadratic polynomial regression model is established and the interaction effects between factors are analyzed by variance and response surface of the model. Finally, the optimum conditions of biodiesel yield based on the verification experiment are concluded. (5) Optimization and control software design of biodiesel transesterification process The optimization and control software of biodiesel transesterification processes is designed and developed. The software provides functions of process modeling, system response analysis, basic control, multi-variable predictive control and two-layer structure prediction control, which can realize dynamics model analysis of process, digital simulation of optimization and control algorithm. Using this software can provide technical support and theoretical guidance for process design and application of advanced control scheme and real-time optimization of system. |
| 语种 | 中文 |
| 产权排序 | 1 |
| 页码 | 123页 |
| 源URL | [http://ir.sia.ac.cn/handle/173321/16759]  |
| 专题 | 沈阳自动化研究所_信息服务与智能控制技术研究室 |
| 推荐引用方式 GB/T 7714 | 史洪岩. 生物柴油酯交换反应过程的优化与控制策略研究[D]. 中国科学院沈阳自动化研究所. 中国科学院沈阳自动化研究所. 2014. |
入库方式： OAI收割
来源：沈阳自动化研究所
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。