【作者】 孙漾；

【导师】 顾幸生；

【作者基本信息】 华东理工大学 ， 控制科学与工程， 2012， 博士

【摘要】 煤炭是我国的基础能源和主要工业原料之一。化学工业每年的煤炭消费量在四亿吨以上。煤炭气化过程是对煤炭进行化学加工的一个重要方法,是实现煤炭洁净利用的关键技术。水煤浆加压气化技术是当今具有代表性的主流煤炭气化技术之一。作为煤化工的源头、全厂能量转换的核心,气化装置的运行状况十分重要。目前,采用先进的控制方法优化装置的运行状况的研究尚处于起步阶段。加快进行煤炭气化过程建模、流程模拟、控制优化等自动化关键技术的研究对提高装置的运转效率、提高煤炭资源利用率、实现节能降耗具有重要意义。本课题以水煤浆气化装置为研究对象,探讨复杂化工过程的建模和优化方法,提出了几种改进的智能优化算法应用于水煤浆气化装置过程建模与操作优化问题中。同时,设计并开发了水煤浆气化操作优化系统应用软件。本文的主要研究成果如下：(1)针对水煤浆气化装置优化配煤问题,建立了混煤质量指标预测模型。其中,采用最小二乘支持向量机(LS-SVM)建立混煤灰熔点预测模型,描述混煤灰熔点与灰成分之间复杂的非线性关系。同时,提出了一种多种群竞争型协同文化差分进化算法(MCCDE),算法中建立了基于差分进化算法的竞争型协同策略及竞争适应度评判方法,并引入了文化算法的部分思想。将MCCDE算法与其它五种变异策略的DE算法通过八种典型测试函数优化问题进行比较,验证了该算法的有效性。最后,将MCCDE算法用于优化LS-SVM模型的超参数。仿真结果表明,该方法建立的模型比其它3种比较方法建立的模型具有更好的泛化能力。(2)结合(1)中建立的混煤质量指标预测模型,建立了一个管理级视角下的配煤优化模型。模型综合考虑了原煤的采购成本、库存成本、实施配煤的操作成本等,在满足装置对煤质要求的前提下,实现用煤总成本最低控制。根据配煤优化模型的特点,将粒子群优化算法、文化算法和协同进化算法进行改进和结合,取长补短,提出了协同文化框架及其进化机制,并建立了基于粒子群算法的协同文化算法(CECBPSO)。采用正交试验的方法分析讨论了该算法的参数对其性能的影响。同时,将CECBPSO算法与其它四种算法优化八个典型测试函数的结果进行比较。最后,对某煤化工厂水煤浆配煤优化问题进行仿真分析,采用CECBPSO算法求解配煤优化模型,计算结果验证了模型和算法的可行性。(3)根据群搜索优化算法适于解决高维多峰函数优化问题的特点,在基本群搜索算法中引入了差分进化算法和混沌局部搜索的思想,提出了一种改进的群搜索差分进化算法(DEGSO),与其它四种算法通过标准测试函数的仿真比较验证了算法的性能。将DEGSO算法用于优化神经网络的权值和阈值,建立了“德士古合成气”CO、H2和CO2气体含量软测量模型(DEGSO-NN),实现以低成本及时、在线获得“德士古合成气”组分含量。采用某德士古气化装置现场实际数据进行仿真研究,结果表明,同另外两种方法相比,基于DEGSO的神经网络软测量模型具有最高的训练效率和泛化能力。(4)根据膜系统计算模型的分布式、极大并行性和不确定性特征,将粒子群算法引入到膜计算模型的框架中,提出了膜计算粒子群优化算法(MCBPSO)。该算法中,多个群体在膜系统的不同膜结构中并行搜索,同时,建立了协同和变异机制以提高其性能。采用正交试验的方法对MCBPSO的参数选择进行了研究,并与其它四种算法采用五种测试函数进行测试比较。针对德士古气化炉操作优化问题的特点,引入了区域优化的概念,建立了工况评判标准和操作优化模型,将MCBPSO用于优化模型的求解。最后,采用某化工厂德士古气化炉实际运行数据进行仿真,经过操作优化计算,能够获得优化的控制参数,并提高气化炉有效气产率。(5)针对某甲醇合成企业德士古水煤浆气化系统,开发了水煤浆气化操作优化系统应用软件。软件具有混煤指标预测、配煤操作优化计算、气化炉炉膛温度软测量、德士古合成气组分软测量、气化炉工况操作优化等功能,同时能够进行模型的自动更新以及记录报表的自动生成与存储。水煤浆气化操作优化系统应用软件能够实现将建模、控制、优化技术应用于实际生产中,以提高装置的经济效益。更多还原

【Abstract】 Coal is China’s basic energy and one of the main industrial raw materials. Over four hundred million tons of coal is used by chemical industry each year. Gasification is one of the most important methods and key technology of coal for clean use. Currently, coal water slurry (CWS) gasification technology is a representative technology of coal gasification.As the headstream and energy conversion core, the running status of gasification unit is very important. Nowadays, researches on process modeling and operation optimization for CWS gasification unit using advanced control methods are still at the initial stage. It is of great significance for improving the unit operation efficiency and enhancing coal resource utilization ratio to speed up the researches on coal gasification automation technology, such as process modeling, process simulation, and optimization. In this dissertation, for CWS gasification process, complex chemical process intelligent modeling approaches and optimization algorithms are studied. Based on these methods, several improved intelligent optimization algorithms are proposed and used in coal gasification process modeling and operation optimization problem. Meanwhile, the CWS gasification operation optimization system software is designed and developed. The main results in this dissertation can be summarized as follows:(1) It appears that the coal-blending optimization is a vital part of the optimal operation of coal gasification process. For this problem, a series of mixed coal quality target prediction models are constructed. One of them is mixed coal ash fusion temperature prediction model, which is established based on least squares support vector machine (LS-SVM) to describe the complex nonlinear relationship between ash fusion temperature and ash components. Meanwhile, a multi-population competitive co-evolutionary cultural differential evolution (MCCDE) algorithm is proposed. In MCCDE, a competitive co-evolutionary strategy based on differential evolution and a fitness value evaluation method are designed. And some ideas from cultural algorithm are also introduced into MCCDE. Five differential evolution algorithms with different mutation strategy and eight typical benchmark functions are adopted to verify the performance of MCCDE algorithm. Finally, MCCDE algorithm is used to optimize the hyper-parameters of LS-SVM. The simulation results indicate that the model based on MCCDE-LS-SVM has stronger generalization.(2) Combined with the mixed coal quality target prediction models established in (1), a process model from a management and decision-making perspective is constructed to solve the problem of CWS gasification coal-blending optimization. The model takes into account mixed-coal indicators, inventory costs, market prices, operating costs and consumptions of stockpiling and transit. According to the process model, a co-evolutionary mechanism between two cultural algorithms is established, and a hybrid co-evolutionary cultural algorithm based on particle swarm optimization (CECBPSO) is proposed in order to fully use the advantages of co-evolutionary algorithm (CEA), cultural algorithm (CA) and particle swarm optimization (PSO). Factorial Design (FD) approach is used in this paper in order to get a guideline on how to tune the designed parameters in CECBPSO. Meanwhile, extensive computational studies are also carried out to evaluate the performance of CECBPSO on eight benchmark functions, compared with other four intelligent optimization algorithms. Finally, CECBPSO algorithm is employed to solve the problem of coal-blending optimization process model of a fertilizer. The calculation results validate the feasibility of the coal-blending optimization model and CECBPSO algorithm.(3) The Group Search Optimization (GSO) algorithm has good performance in high-dimension multi-modal optimization problems. In this paper, differential evolution and chaotic local optimizer are introduced into basic GSO algorithm, and an improved Differential Evolution Based Group Search Optimization (DEGSO) algorithm is proposed. Compared with other four evolutionary algorithms on function optimization problems, the performances of DEGSO are satisfactory. In addition, DEGSO algorithm is applied to optimize the weights and thresholds of neural network, and three DEGSO-NN based soft sensor models are established for estimating the percentage compositions of CO, H2 and CO2 of Texaco gasifier syngas. The simulation results indicate that soft sensor modeling method based on DEGSO-NN has higher training efficiency and stronger generalization than the other two compared methods.(4) Membrane computing is a new branch of natural computing with the features of distribution and great parallelism. Considering the features of membrane computing and PSO, a hybrid algorithm MCBPSO is proposed. In MCBPSO, PSO is introduced into the computing model of membrane system. Multi-populations iterate in different membrane structures in parallel. Meanwhile, cooperation and mutation strategy are also established in the hybrid algorithm to improve the performance. Influencing rules of parameter selection of MCBPSO are studied through orthogonal design experiments. Also, the performance of MCBPSO is evaluated by five test functions. For Texaco coal gasification operation optimization problem, the concept of regional optimization is introduced, running status evaluating standard is designed, and operation optimization model is constructed. MCBPSO algorithm is used to solve the problem of operation optimization model. Finally, simulations with a Texaco gasification unit for example testified that optimized operation variables can be found and the effective gas rate can be increased by the optimization model and algorithm.(5) For a real-world Texaco CWS gasification process of a methanol synthesis system, the CWS gasification operation optimization system software is designed and developed. The software provides functions of mixed-coal qualities prediction, coal-blending optimization, soft sensing of furnace box temperature and syngas percentage compositions, gasifier running status optimization. Process models can be updated automatically. Record reports also can be created and stored periodically. Modeling, control and optimization technologies can be used in practical production to gain more economic benefits through the use of the CWS gasification operation optimization system software.更多还原