【作者】 孔祥东；

【导师】 钱锋；

【作者基本信息】 华东理工大学 ， 过程系统工程， 2014， 博士

【摘要】 本文以GE气流床水煤浆气化炉为研究对象,深入了解气化反应过程机理,围绕工业气化炉的建模、控制与优化等方面的问题与技术进行了研究。分别建立了工业气化炉的热力学模型和动力学模型,考察了煤质组成和操作条件等工艺参数的变化对气化性能的影响,并以此为基础建立气化炉的动态模型,研究了过程的动态特性和系统控制结构的控制性能,采用智能优化算法进行了气化炉的操作优化,为实现工业气流床气化过程的优化与控制提供了新的技术指导。主要内容总结如下：1.建立了可预测碳转化率的三级气化炉平衡模型,采用热力学平衡假设,通过引入参与水煤气变换反应中H20的平衡分率来表征参与非均相气化反应的碳的数量,该方法克服了以往热力学模型需事先指定碳转化率的缺点,通过与实验数据和文献中的模型结果对比发现,该模型对气化过程关键指标具有较高的预测精度。研究了不同煤质组成对气化性能的影响,以煤中H/C、O/C摩尔比和灰分含量作为表征煤质的成分指标,通过模型计算发现,有效合成气产率随着氧元素相对摩尔量的增多而降低,随着氢元素相对摩尔量的增多而升高。随着灰分质量分数的增加,有效气产率降低,而比氧耗升高。作为气流床煤气化炉的原料,煤中的氧元素相对摩尔量越低,则气化炉气化性能越好,当原料煤中灰分质量分数发生变化时,应当调整气化操作参数使得气化过程能够高效稳定运行。2.通过对气化炉内流体流动的简化处理,综合考虑煤焦的气固非均相反应动力学,提出了一种简化的气化炉一维综合模型,沿气化炉内物流流动方向将气化炉分为裂解燃烧反应区和气化反应区,采用智能优化算法对焦炭气化反应、水煤气变换反应和重整反应的动力学参数进行了优化校正,获得了符合工业装置和特定煤种特性的气化反应过程模型。针对上述模型不能表征气化炉内物料回流的影响,通过流程模拟软件,在上述一维模型的基础上,结合气化炉内的流场分布,开发了考虑炉内回流的基于反应器网络方法的分区模型,与一维模型相比,该模型可更加准确的描述炉内的温度场和组分分布。根据模型研究了包含水煤浆浓度和氧煤比在内的操作参数对气化温度和出口合成气组成、碳转化率、有效合成气收率和比氧耗的影响,结果表明对于给定的煤浆浓度,存在不同的氧煤比使得有效合成气收率最大或比氧耗最小,应根据实际需求和装置操作特性对氧煤比进行优化和调整,该模型为开发过程的动态模型和操作优化提供了良好的基础。3.工业气流床气化炉操作变量存在着各种各样的和不同程度的波动、干扰以及操作条件的变化,同时气化过程具有停留时间短、反应速度快、各影响因素相互耦合、强非线性等特点,实际装置的运行是一个动态变化的过程。在稳态模型的基础上,建立了气流床气化过程的动态模型,并根据工业实际情况建立气化炉控制系统,研究了系统操作参数发生阶跃扰动时对过程动态特性的影响。根据工业气化炉现有的检测装置和控制现状,提出了一种新的控制系统,该控制系统以氧气流量作为负荷调节手段,气化温度由煤浆流量调节。通过阶跃扰动测试,结果表明该控制系统与原控制系统相比,在负荷出现变化时,气化温度和合成气收率的波动小,稳定时间短,提高了过程变负荷操作的安全性和经济性,为工业气化过程的稳定高效运行提供了理论基础。4.提出了一种改进的自适应多目标差分进化算法(CSADE),将混沌操作算子引入到差分进化算法的局部搜索功能中,以增加该算法的局部搜索功能。同时,算法控制参数的自适应调整功能加速了算法的收敛速度。标准测试函数的仿真结果表明,与文献中的优化算法相比,新算法对最优可行边界解的搜索能力强,在提高算法收敛速度的同时保持了解的分布性。然后,从现场气化炉操作优化的需求出发,结合气化过程的工艺操作约束,将本文提出的优化算法应用于水煤浆气化炉的操作优化中。结果表明,通过操作优化,工业气化炉可实现在提高有效合成气收率的同时降低比氧耗,对装置优化运行有较好的指导意义。更多还原

【Abstract】 Gasification technology is being widely developed in the chemical and energy processes as a practical coal-utilizing technology using coal more efficiently and cleanly. Based on the mechanism models, research on gasification process modeling, control and optimization are conducted in this study. Thermal model and kinetic model are both established to give deep insiht of behaviors for the industrial entrained flow coal gasifier. Effects of feed coal composition and operating conditions on gasification performances have been investiged in light of the developed models. Dynamic models are set up to analyse the transient dynamic reponses and evaluate the effectivenesses of control structures. Furthermore, the operation optimization for the gasifier is carried out by using the proposed new intelligence alrorithm. All these provide a new technical guidance for optimization and control in the industrial gasification process. The main contents of this paper are summarized as follows:1. A novel three stage equilibrium model which can be used to predict carbon conversion, is developed for coal gasification on the basis of thermal equilibrium theory. The model is divided into three stages including pyrolysis and combustion stage, char gas reaction stage, and gas phase reaction stage. Steam participation ratio expressed as a function of temperature is introduced to estimate carbon conversion by assuming that only part of the water produced in the pyrolysis and combustion stage is involved to react with the unburned carbon in the second stage. The model overcomes the shortcome of the troditional themal model which need a specified carbon conversion in advance. Model results show a high prediction accuracy compared with published experimental data and models found in literatures. Effects of the amount of element C, H, O and ash in dry coal on the performance of gasifier are investigated by means of changing H/C and O/C molar ratios and ash content. The simulation results show that at the same operating temperature, the syngas productivity and oxygen consumption increase with the O/C molar ratio. However, with the increase of H/C molar ratio, the syngas productivity increases slightly and oxygen consumption remains unchanged. The relative amount of element O in coal has a more significant effect on the performance of gasifier. The syngas productivity reduces and the oxygen consumption raises with the increase of ash content. The above simulation results indicate the effects of major constituents of coal, i.e. C, H, O and ash on the performance of gasifier should be concerned and operating parameters need be adjusted in the industrial operating to optimize the production process and enhance the economic benefits.2. Cosidering the gas-char heterogeneous reactions kinetics, a one-dimensional partition gasifier model is proposed by simplying the flow behaviours in coal gasifier. The gasifier is divided into pyrolysis-combustion zone and gasification zone along the flow direction. The pyrolysis-combustion zone is modeled using the stoichiometry method. Detailed investigation was carried out on the gasification reaction rates in the reduction zone. The particle swarm optimization technique is introduced in this paper to address the lack of heterogeneous reaction kinetic parameters based on the random pore model for the specific feed Shenfu coal, and the huge deviation of the industrial product gas composition from the theoretical composition at equilibrium state. With the evaluated optimum kinetic parameters, robust agreement is achieved between the model outputs and the industrial data. Since the materials recirculation is not considered by one dimensional model, an equivalent compartment model (CM) is presented using the Aspen Plus process simulator. The CM blocking is established based on gasifier flow field analysis, using a number of compartments. A simple configuration of these compartments involving material recirculation should be able to simulate the main flow and provide the temperature and gas component distributions. The model predictions exhibit good agreement with industrial data in the model validation. The influences of the oxygen-to-carbon ratio (ROC) and the coal slurry concentration on the gasification performance are discussed. According to the intended final use, however, choosing a reasonable ROC to obtain a higher efficient syngas yield and lower oxygen consumption can be flexible.3. A dynamic model derived from the steady state model mentioned above is constructed for evaluating different control structures based on the disturbances rejection capabilities. From the sensitivity analysis, the optimal oxygen to coal ratio is obtained. The selection of an appropriate control structure is the most important decision when designing gasification control systems. In industrial practice, the gasifier is controlled by gasifier temperature, which manipulates the oxygen to coal ratio. The temperature is controlled at a suitable value slightly higher than the melting temperature of feed coal so that the operation can accomplish the slag discharge target. Two control structures are studied. The first control structure (CS1) uses the coal feed rate as the throughput manipulator (TPM). The other control structure (CS2) uses the oxygen feed rate as the TPM. The dynamic responses for feed flow rate and composition disturbances are evaluated in the two control structures. Although both control structures can handle the disturbances and hold the gasification temperature very close to the specified value, the results show that the the CS2control structure solve the disturbance issues effectively with smaller deviations.4. Since entrained flow gasification is such widely used, even a slight improvement in the operation of the gasifier can increase the economic benefits significantly. Different goals are to be reached depending on the downstream demands. Hence the objective of the operation optimization of the gasification process is to maximize the yield rates of efficient syngas and the H2product rate as well as minimizing the oxygen consumption. However, the three optimization objectives cannot be achieved simultaneously because of conflicts between them. An Chaos self-adaptive multi-objective differential evolution (CSaDE) algorithm is proposed to solve this multi-objective problem. In order to overcome the problems of premature convergence and falling into the local optimum, a chaotic migrate operator is introduced to the SaDE algorithm to strengthen the local search ability and optimization accuracy. The proposed algorithm is successfully tested on various benchmark test problems, and the performance measures such as convergence and divergence metrics are calculated. Multi-objective optimization problem of an entrained flow coal gasifier is then solved using the proposed CSaDE algorithm. Many sets of operating conditions that will yield such an end result are provided. Operating under the conditions predicted will enhance productivity and reduce the consumption and thereby increase profit. Since the CSADE method is a general algorithm, the described procedure is suitable for maximizing the benefits of any operating industrial gasification plant.更多还原