芳烃抽提过程多目标优化

芳烃抽提是芳烃生产过程的重要环节,其生产调优对提高整个芳烃联合装置的效益具有重要意义。基于流程模拟及响应面分析方法,得到了芳烃抽提过程的产品纯度模型及能耗模型。建立了以产品纯度最大化及过程能耗最小化的多目标优化模型。提出了一种改进的自适应加权求和算法,并用于多目标优化模型的求解。求解结果表明新算法在Pareto最优解分布的均匀性上与原算法相当,但求解效率要高于原算法。给出了不同产品等级下的最佳操作参数,采用优化后的操作参数可有效地提高产品纯度并降低过程能耗。提出的多目标优化模型及求解算法用于芳烃抽提过程的操作调优,可有效地提高决策的准确性。     

基于能量目标的芳烃萃取精馏溶剂评价模型

以NRTL活度系数模型为基础，利用Aspen Plus对不同单组分萃取剂回收芳烃的萃取精馏装置进行了全流程模拟和工艺操作参数优化。综合考虑各操作变量及其关联，提出了基于局部耦合参数迭代优化的整体协同优化策略，在保证分离要求的条件下，以能耗为目标，对萃取精馏塔（EDC）、溶剂回收塔（ERC）的进料位置、ERC回流比等关键操作参数进行优化，建立过程能耗的物性关联模型。通过分析不同溶剂对芳烃萃取精馏过程能耗和分离效果的影响，提出了基于能耗目标的芳烃萃取精馏溶剂评价模型，结果表明影响芳烃萃取精馏过程能耗的关键物性为溶剂的分子量及常压沸点，所建立的过程能耗关联模型具有较高的关联性，其R ²均大于0.9，可有效指导萃取精馏溶剂选择，为计算机辅助分子设计提供简化的目标函数。     

The dynamics of extraction processes: Part II. Comparison of dynamic testing methods and their application to an extraction process

A comparison of the theoretical and experimental difficulties and advantages of sinusoidal, pulse, and step testing procedures, as applied to a continuous stagewise extraction process, is made. An experimental extraction apparatus suitable for a variety of dynamic studies is described. For this apparatus step testing has been found to give results as reliable as those from pulse methods. The paper develops the theoretical background and describes experimental implementation of double pulse testing. The extraction column when used with the system water-acetic acid-methyl isobutyl ketone was found to behave linearly over the range of variables studied. The stages of the column were found to be “well-mixed”, and it was concluded that the column could be best represented by a lumped parameter model.     

Mass transfer studies on ternary systems in a bench-scale liquid-liquid extraction (LLX) column and a comparison with simulations

Mass transfer studies in a laboratory scale extraction column have been conducted for Toluene-Acetone-Water and MIBK-Acetic Acid-Water systems. From these experiments stage-wise solute (Acetone or Acetic-Acid) composition profiles have been obtained for both dispersed and continuous phase. These composition profiles have been compared with those obtained from ASPENPLUS, CHEMSEP and LLXSIM simulators. For liquid-liquid equilibrium calculations all these simulators use UNIFAC and UNIQUAC model. The binary interaction parameters for the UNIFAC are inbuilt in ASPENPLUS and CHEMSEP. UNIQUAC binary parameters were borrowed from DECHEMA. Error square analysis indicates that simulations based onnon-equilibrium option of LLXSIM matchclosely with experimental results. Temperature profiles and hydrodynamic features characterized by number of drops and static holdup on the stages have been compared between the LLXSIM simulated and the experimental results and these match well. However simulations on ASPENPLUS give sum of relative error-squares for all the experimental runs at leastten times higher, in spite of tuning the average stage efficiency.     

COMPARISON OF KINETIC MODELS FOR THE AQUEOUS SOLID-LIQUID EXTRACTION OF TILIA SAPWOOD IN A CONTINUOUS STIRRED TANK REACTOR

The kinetic study of the aqueous extraction of Tilia sapwood was based on temperature, while all the other factors were kept constant. Several models were used and compared to explain the kinetics of the aqueous solid-liquid extraction of Tilia sapwood. For each model, the rate constant, the equilibrium extraction capacity, and the initial extraction rate were evaluated to analyze the suitability of these kinetic models to describe the leaching process for temperatures ranging from 313 to 363 K. The pseudo first-order model could not be applied during the whole of the process. The Elovich model provided a good degree of correlation (from 0.929 to 0.988) and could be applied all along the extraction process. The pseudo second-order model was satisfactorily applied, with coefficients of correlation over 0.998, showing that it perfectly described the process. A physical explanation of these models was finally proposed.     

换热网络运行模拟优化

换热网络系统大多数是按给定工况,以投资费用和运行费用最优为目标设计的。但在实际运行过程中,确定和不确定性的影响因素往往导致换热网络的运行工况偏离设计值。偏离设计工况运行的换热网络性能变差,导致运行费用增加,甚至不能满足工艺物流换热要求。在换热网络结构给定条件下,将满足物流目标温度和运行费用最优作为目标函数,以单体模型和Yee et al.（1990）提出的多级超结构为基础,建立换热网络运行模拟优化模型,并进一步去除恒定膜传热系数假设使模型贴近实际问题。针对提出的非线性数学模型（NLP）问题,以标准粒子群算法为基础建立求解策略。在论文的最后,4个来源于已发表论文的实例研究证明了该优化方法的有效性。     

Lithium Extracting from Zinnwaldite: Economical Comparison of an Adapted Spodumene and a Direct‐Carbonation Process

Lithium is used in a variety of everyday objects. The most important field of application is as components of battery electrolytes and electrodes. Next to salar brines, commercial lithium production relies on mineral ore sources such as spodumene, petalite, and lepidolite. This study provides an economic comparison of two processes for the extraction of lithium from zinnwaldite concentrate, a lithium mineral that can be found along the Czech/German border at Cínovec/Zinnwald. In contrast to the state‐of‐the‐art lithium extraction from siliceous ores, which is a hydrometallurgical extraction process with sulfuric acid, the novel direct‐carbonation process is based on digestion using supercritical carbon dioxide. An assessment of the capital and operating expenditures was also carried out.     

注塑机工艺参数的智能设置与优化

针对工艺人员的试模思路,混合使用实例推理、代理模型和模糊推理技术,建立一种描述注塑机工艺参数设置与优化全过程的混合智能模型。首先采用实例推理技术模拟工艺人员设置初始工艺参数时的“借鉴”思维,在实例推理失败的情况下,采用代理模型模拟工艺人员的“直觉”思维设置初始工艺参数,然后将初始参数用于试模,最后利用模糊推理技术实现工艺人员不断修正缺陷、优化工艺参数的思维过程。基于上述智能模型开发出了相应的软件系统,并通过与控制器的通讯实现与注塑机的集成,实际案例验证表明该系统正确有效,可应用于实际生产。     

常压反应-减压精馏耦合生产氯化苄的工艺优化设计

采用常压反应-减压精馏耦合装置应用于甲苯氯化体系生产氯化苄。为了找到集成装置的最佳结构参数,建立了以独立反应量为基础的模型方程,并开发了基于Powell法的多层优化设计法。结果表明,在塔釜上升蒸汽量一定的情况下,增加精馏塔分离段塔板数、反应器台数、相邻反应器间间隔塔板数,反应能力有所增加,然而当其超过一定值后,反应能力增加的趋势不明显;在优化得到的结构操作参数下,反应能力与分离能力达到最佳匹配。     

基于变长度移动窗口和HMM的工业过程故障在线诊断

An integrated framework is presented to represent and classify process data for on-line identifying abnormal operating conditions. It is based on pattern recognition principles and consists of a feature extraction step, by which wavelet transform and principal component analysis are used to capture the inherent characteristics from process measurements, followed by a similarity assessment step using hidden Markov model (HMM) for pattern comparison. In most previous cases, a fixed-length moving window was employed to track dynamic data, and often failed to capture enough information for each fault and sometimes even deteriorated the diagnostic performance. A variable moving window, the length of which is modified with time, is introduced in this paper and case studies on the Tennessee Eastman process illustrate the potential of the proposed method.     

等效滤饼过滤模型

Cake filtration has been widely used in many chemical processes with more non-Newtonian, highly viscous and compressible materials involved. Neither traditional nor modem filtration theory can be applied in practice ＂Equivalent cake filtration model＂ is a recently developed mathematical model to describe cake filtration for both Newtonian and non-Newtonian fluids, in either steady or unsteady filtration stages. This model has two strengths： （1） It can be used to determine equivalent capillary radii and predict filtration quality based on the properties of solid/liquid system and operation parameters; and （2） to calculate cake specific resistance and its variations with time at various cake thickness locations.     

Flexible and economical operation of chlor-alkali process with subsequent polyvinyl chloride production

Demand response (DR) can compensate for imbalances in variable renewable energy supplies. This possibility is particularly interesting for electrochemical processes, due to their high energy intensity. To determine the technical feasibility and economic viability of DR, we chose the chlor-alkali process with subsequent polyvinyl chloride production, including intermediate storage for ethylene dichloride. We estimate the maximum possible cost savings of implementing load flexibility measures. A process model is set up to determine the system characteristic. Subsequent optimizations result in the facility's best possible dispatch depending on additional and minimum power load, storage volume, and cost of a load change. Real plant data are used to specify model parameters and validate the system characteristic and the plant dispatch. An economic evaluation reveals the economic advantages of efficiency and flexibility. The approach can be used to analyze the DR potential of other chlorine value chains or facilities with high electricity demand in general.     

Optimal design of large‐scale chemical processes under uncertainty: A ranking‐based approach

An approach for the optimal design of chemical processes in the presence of uncertainty was presented. The key idea in this work is to approximate the process constraint functions and model outputs using Power Series Expansions (PSE)‐based functions. The PSE functions are used to efficiently identify the variability in the process constraint functions and model outputs due to multiple realizations in the uncertain parameters using Monte Carlo (MC) sampling methods. A ranking‐based approach is adopted here where the user can assign priorities or probabilities of satisfaction for the different process constraints and model outputs considered in the analysis. The methodology was tested on a reactor–heat exchanger system and the Tennessee Eastman process. The results show that the present method is computationally attractive since the optimal process design is accomplished in shorter computational times when compared to the use of the MC method applied to the full plant model. © 2014 American Institute of Chemical Engineers AIChE J, 60: 3243–3257, 2014     

MULTILEVEL PCA AND INDUCTIVE LEARNING FOR KNOWLEDGE EXTRACTION FROM OPERATIONAL DATA OF BATCH PROCESSES

A new methodology for monitoring batch processes is presented which is based on analysis of historical operational data using both principal component analysis (PCA) and inductive learning. Historical data of batch operations are analysed according to stages. For each stage, PCA is employed to analyse the trajectories of each variable over all batch runs and groups the trajectories into clusters. The first one or two PCs for all variables at a stage are then used in further PCA analysis to project the operation of the stage onto operational spaces. Production rules are generated to summarise the operational routes to produce product recipes, and to describe variables' contributions to stage-wise state spaces. A method for automatic identification of stages using wavelet multi-scale analysis is also described. The methodology is illustrated by reference to a case study of a semi-batch polymerisation reactor.     

基于多数据空间全潜结构映射的化工过程性能评估方法

针对化工过程运行状态在线评估的问题,提出多数据空间全潜结构映射（multi-space total projection to latent structures,MsT-PLS）性能评估方法。该方法采用“离线建模,在线评估”的评估策略。首先对历史多数据输入空间进行全面分解,结合多数据空间基向量提取方法,剔除多数据输入空间中与质量变量无关信息的干扰。在与质量变量相关的多数据输入空间上,建立不同运行性能等级的离线数据网络分类模型,实现“离线建模”。“在线评估”阶段,以数据滑动时间窗为评估单元,将过程性能分为稳定和过渡性能等级,把在线数据与历史性能等级进行相似度匹配。利用过程变量相对贡献度,对性能变化起决定性影响的过程变量进行识别和贡献度分析,为系统性能劣化原因的识别提供了参考。最后,应用到乙烯裂解过程在线性能评估中,说明了本评估方法可以对系统进行准确的在线性能评估。     

A GENETIC NEURAL FUZZY SYSTEM AND ITS APPLICATION IN QUALITY PREDICTION IN THE INJECTION PROCESS

A genetic neural fuzzy system (GNFS) is presented and introduced to quality prediction in the injection process. A hybrid-learning algorithm is proposed, which is divided into two stages to train GNFS. During the first learning stage, the genetic algorithm is used to optimize the structure of GNFS and the membership function of each fuzzy term because of its capability of parallel and global search. On the basis of the first optimized training stages, the back-propagation algorithm (BP algorithm) is adopted to update the parameters of the GNFS to improve its predicting precision and reduce the computation time. The process of constructing a quality prediction model for an injection process based on GNFS is described in detail. The predicted weight of the molded part from the model based on GNFS demonstrates that the proposed GNFS has superior performance and good generalization capability in quality prediction in the injection process.     

A GENETIC NEURAL FUZZY SYSTEM AND ITS APPLICATION IN QUALITY PREDICTION IN THE INJECTION PROCESS

A genetic neural fuzzy system (GNFS) is presented and introduced to quality prediction in the injection process. A hybrid-learning algorithm is proposed, which is divided into two stages to train GNFS. During the first learning stage, the genetic algorithm is used to optimize the structure of GNFS and the membership function of each fuzzy term because of its capability of parallel and global search. On the basis of the first optimized training stages, the back-propagation algorithm (BP algorithm) is adopted to update the parameters of the GNFS to improve its predicting precision and reduce the computation time. The process of constructing a quality prediction model for an injection process based on GNFS is described in detail. The predicted weight of the molded part from the model based on GNFS demonstrates that the proposed GNFS has superior performance and good generalization capability in quality prediction in the injection process.     

Component content distribution profile control in rare earth countercurrent extraction process

Considering that the on-line measurement and automatic control of element component content (ECC) are dif-ficult to perform in rare earth cascade extraction process, the ECC distribution profile is dyn...     

稀土萃取过程组分含量分布控制方法(英文)

Considering that the on-line measurement and automatic control of element component content (ECC) are dif-ficult to perform in rare earth cascade extraction process, the ECC distribution profile is dynamical y regulated at al stages to assess the effect of product purity control. Focusing on the theory of countercurrent extraction, the technology parameters and pre-setting flow-rates during the extract process are designed. Under varying process parameters, a novel step by step model is also proposed for each stage to analyze the impact on the distribution profile change. Combining the mass balance model and ECC changing trend at the monitoring stage, the ECC dis-tribution profile can be automatical y regulated by dynamical y compensating the related extract or scrubbing liquid flow-rate. To this end, the required product purity at the two outlets is achieved. Based on Wincc and Matlab dynamic simulators, a specific Pr/Nd cascade extraction process is used to illustrate and demonstrate the application of the present approach.     

兼顾求解速度和精度的换热网络动态模拟方法

以换热器集总参数动态模型为基础,首次建立了基于Grossmann分级超结构的换热网络动态模型和求解方法,可自动识别换热网络结构并建立相应的求解策略,实现了换热网络的动态模拟。模型中考虑了流体温度变化对物性和换热性能的影响,有效地提高了换热网络动态解的求解精度;采用内外双层解法,提高了换热网络动态解的求解速度。实例应用表明,基于Grossmann分级超结构的换热网络动态模型及求解策略可靠、准确,能够获得扰动后换热网络中流体参数的真实变化过程,为换热网络的动态优化和控制奠定了基础。