基于gPROMS的间歇式过程的动态优化

分析了gPROMS(general Process Modelling System)软件的数值计算性能,并且以一类间歇式化学反应过程为例,基于作者提出的该反应过程的动态数学模型利用gPROMS软件对该反应过程进行了动态优化研究,检验了gPROMS的动态优化能力。     

Modelling and simulation of a hybrid process (pervaporation–distillation) for the separation of azeotropic mixtures of alcohol–ether

This work reports the modelling and simulation of a hybrid process, based on the combination of distillation and pervaporation, for the separation of azeotropic mixtures of alcohol–ether. After having selected the separation of methanol‐2‐metoxi‐2,2‐dimethyl ethane (ETHER) as a motivating example the mathematical modelling of the distillation column was achieved and used together with a mass transfer model previously reported for the pervaporation operation in order to simulate the behaviour of the hybrid process for different compositions of the feed stream (case 1: 3.2 wt% methanol, 55.4 wt% C₄, 41.4 wt% ETHER, and case 2: 5.2 wt% methanol, 42 wt% C₄, 52.8 wt% ETHER). Simulation tasks were carried out with the process modelling system gPROMS and the results of alternative process configurations that result from the relative location of the separation technologies have been compared on the basis of the required membrane area. Finally, the design of the pervaporation unit including the overall processing costs is reported. Copyright © 2001 Society of Chemical Industry     

A three-phase nonequilibrium dynamic model for catalytic distillation

A detailed three-phase nonequilibrium (NEQ) dynamic model for simulating batch and continuous catalytic distillation (CD) processes has been developed. In this model, both molar and energy holdups in liquid and vapour phases are taken into account. Multicomponent mass transfer and heat transfer between vapour and liquid phases as well as between liquid and solid (catalyst) phases are described by the Maxwell-Stefan equations. The resulting differential and algebra equations in this model are implemented in gPROMS and C++. The simulation results are in good agreement with the experimental data obtained from the batch and continuous CD processes for the production of diacetone alcohol (DAA) using Amberlite IRA-900 as a catalyst. Sensitivity analysis on the mass transfer and kinetics using the three-phase NEQ dynamic model indicates that the formation of DAA is controlled by solid-liquid mass transfer, whereas the formation of mesityl oxide is kinetically controlled under the simulation conditions.     

Entwicklungen bei der Lösungskristallisation in den letzten Jahren

Recent Developments in Crystallisation from Solution. Although there have been no great leaps forward in the design of plant for crystallisation from solution in recent years, our understanding of growth rate dispersion, the extension of the two-step model of crystal growth to a three-step model when required, and also convenient modelling approaches for crystal morphologies and the development of tailor-made additives have nevertheless greatly advanced our knowledge of crystallisation processes. Substantial divergencies in the measured values of crystal growth rates in the past decade can often be explained by paying due attention to the measuring methods and the purity of the substances as well as the nature of the solvent. Comparison of results obtained by modelling should always prove more satisfactory when heat effects are considered whenever this is necessary. The possibilities of crystallisation plant design are greatly enhanced by the availability of reliable data as a basis for calculation and when influencing crystal shape can provide downstream processes with a better product.     

Dynamic modelling of CO2 absorption for post combustion capture in coal-fired power plants

Power generation from fossil fuel-fired power plants is the largest single source of CO₂ emissions. Post combustion capture via chemical absorption is viewed as the most mature CO₂ capture technique. This paper presents a study of the post combustion CO₂ capture with monoethanolamine (MEA) based on dynamic modelling of the process. The aims of the project were to compare two different approaches (the equilibrium-based approach versus the rate-based approach) in modelling the absorber dynamically and to understand the dynamic behaviour of the absorber during part load operation and with disturbances from the stripper. A powerful modelling and simulation tool gPROMS was chosen to implement the proposed work. The study indicates that the rate-based model gives a better prediction of the chemical absorption process than the equilibrium-based model. The dynamic simulation of the absorber indicates normal absorber column operation could be maintained during part load operation by maintaining the ratio of the flow rates of the lean solvent and flue gas to the absorber. Disturbances in the CO₂ loading of the lean solvent to the absorber significantly affect absorber performance. Further work will extend the dynamic modelling to the stripper for whole plant analysis.     

A HYBRID CFD COMPARTMENTALIZATION MODELING FRAMEWORK FOR THE SCALEUP OF BATCH COOLING CRYSTALLIZATION PROCESSES

Crystallizer design is often hindered by the lack of scaleup rules, hydrodynamic information, and predictive crystallization modeling tools. A hybrid CFD compartmentalization batch cooling crystallizer model is proposed to take into account localized mixing, heat transfer, and fluid hydrodynamics, combined with key process engineering information obtained on a laboratory scale. The compartments were identified using CFD simulations based on the crystallizer geometry and operating conditions. The population, mass, concentration, and energy balance of each compartment is modeled separately as a well-mixed MSMPR unit with input and output streams. The software gPROMS (Process Systems Enterprise Limited) is a process-modeling tool that can facilitate compartmental modeling and will be used for the prediction of the crystallization behavior upon scaleup.     

Optimal seed recipe design for crystal size distribution control for batch cooling crystallisation processes

The paper presents a novel quality-by-design framework for the design of optimal seed recipes for batch cooling crystallisation systems with the aim to produce a desired target crystal size distribution (CSD) of the product. The approach is based on a population balance model-based optimal control framework, which optimises the compositions of seed blends, based on seed fractions that result from standard sieve analysis. The population balance model is solved using a combined quadrature method of moments and method of characteristics (QMOM-MOCH) approach for the generic case of apparent size-dependent growth. Seed mixtures are represented as a sum of Gaussian distributions, where each Gaussian corresponds to the seed distribution in a particular sieve size range. The proposed methods are exemplified for the model system of potassium dichromate in water, for which the apparent size-dependent growth kinetic parameters have been identified from laboratory experiments. The paper also illustrates the simultaneous application of in situ process analytical tools, such as focused beam reflectance measurement (FBRM) for nucleation detection, attenuated total reflection (ATR) UV/Vis spectroscopy for concentration monitoring, as well as the in-line use of laser diffraction particle sizing for real-time CSD measurement.     

Neurocontrol of a multi-effect batch distillation pilot plant based on evolutionary reinforcement learning

The time cost of first-principles dynamic modelling and the complexity of nonlinear control strategies may limit successful implementation of advanced process control. The maximum return on fixed capital within the processing industries is thus compromised. This study introduces a neurocontrol methodology that uses partial system identification and symbiotic memetic neuro-evolution (SMNE) for the development of neurocontrollers. Partial system identification is achieved using singular spectrum analysis (SSA) to extract state variables from time series data. The SMNE algorithm uses a symbiotic evolutionary algorithm and particle swarm optimisation to learn optimal neurocontroller weights from the partially identified system within a reinforcement learning framework. A multi-effect batch distillation (MEBAD) pilot plant was constructed to demonstrate the real world application of the neurocontrol methodology, motivated by the nonsteady state operation and nonlinear process interaction between multiple distillation columns. Multi-loop proportional integral (PI) control was implemented as a reduced model, reflecting an approach involving no modelling or significant controller tuning. Rapid multiple input multiple out nonlinear controller development was achieved using SSA and the SMNE algorithm, demonstrating comparable time and cost to implementation in relation to the reduced model. The optimal neurocontroller reduced the batch time and therefore the energy consumption by 45% compared to conventional multi-loop SISO PI control.     

Planning,scheduling and budgeting value-added chains

This paper addresses the implementation of financial cross functional links with the enterprise value-added chain including retrofitting activities at plant level when scheduling, planning and budgeting in short term planning in batch process industries. The main idea is achieving financial-supply chain integration into advanced planning and schedule (APS) enterprise systems (Badell & Puigjaner (2001a). Discover a powerful tool for scheduling in ERM systems. Hydrocarbon Processing, 80(3), 160; Badell & Puigjaner (2001b). Advanced enterprise resource management systems. Computers & Chemical Engineering, 25, 517). The platform built combines a deterministic cash flow management model with an advanced schedule algorithm using a MILP formulation. The modelling framework created can support the budgeting activity of the entire enterprise functionality, being now the budget the core document for enterprise management. The benefits of this work are shown through a case study that illustrates the modelling framework, the information flows and procedures necessary to implement a financial/supply chain scheduling methodology for the use of financial managers during planning and budgeting activities in process industries.     

A HYBRID CFD COMPARTMENTALIZATION MODELING FRAMEWORK FOR THE SCALEUP OF BATCH COOLING CRYSTALLIZATION PROCESSES

Crystallizer design is often hindered by the lack of scaleup rules, hydrodynamic information, and predictive crystallization modeling tools. A hybrid CFD compartmentalization batch cooling crystallizer model is proposed to take into account localized mixing, heat transfer, and fluid hydrodynamics, combined with key process engineering information obtained on a laboratory scale. The compartments were identified using CFD simulations based on the crystallizer geometry and operating conditions. The population, mass, concentration, and energy balance of each compartment is modeled separately as a well-mixed MSMPR unit with input and output streams. The software gPROMS (Process Systems Enterprise Limited) is a process-modeling tool that can facilitate compartmental modeling and will be used for the prediction of the crystallization behavior upon scaleup.     

Process optimisation in sunflower oil extraction by supercritical CO2

A continuous process for the extraction of sunflower oil using supercritical CO₂, featuring multiple extractors, one oil separator and three cascaded CO₂ recovery vessels operating at different pressures, was devised and studied. For every single equipment of the plant making up the process a mathematical model was built. Experimental tests—consisting in measurements of oil solubility in supercritical CO₂—were carried out in a laboratory-scale apparatus to characterise the behaviour of sunflower oil in the separation from the supercritical fluid. The mathematical model of the whole process was coded in the commercial gPROMS process modelling environment where both its simulation and optimisation—this latter assuming the overall oil production cost as the objective function—were carried out. The process- and economics-related results are discussed and compared with those obtained with traditional and cold-pressing extraction.     

Population balance and computational fluid dynamics modelling of ice crystallisation in a scraped surface freezer

Ice crystallisation in a scraped surface freezer is exceedingly complex and there is very limited fundamental understanding of the influences of fluid flow and heat transfer on ice crystal size, number and shape. This paper presents a computer modelling study that combines population balance method and computational fluid dynamics method. Ice crystal nucleation and growth kinetics has been described by discrete population balance equation. Algorithms for solving the coupled equations of fluid flow, heat transfer and discrete population balance of ice crystallisation have been developed and implemented into a commercial code. Demonstrated is a 2-D computer simulation of ice crystallisation in a scraped surface freezer. Although the simulation makes several assumptions including simplified fluid flow conditions, the predicted ice crystal size distribution is comparable to the experimental data. It is shown that with combined computational fluid dynamics and population balance modelling much insight into the interaction of fluid flow, heat transfer and ice crystallisation in the scraped surface freezer can be obtained.     

Influence of microstructure evolution on mechanical strength of blown poly(ethylene terephthalate)

Abstract

Evolution of the microstructure in poly(ethylene terephthalate) during the blowing process has an important influence on final mechanical properties. In order to simulate the blow moulding process it is necessary to take into account this evolution (molecular orientation, crystallisation, etc.). In the present paper, experimental results on free blown bottles are presented. These results are related to geometrical characteristics, mechanical characteristics, and physical properties. Global shape and thickness were measured to estimate strain distribution during the process. Axial and circumferential specimens were cut from several points on the bottle. Tensile tests were carried out to determine the distribution of mechanical properties. The crystallisation rate was measured on a small specimen cut from the bottle in order to correlate mechanical properties with thickness. This work is the first step in a more global study of the blow moulding process of injected poly(ethylene terephthalate) preforms.¹ A numerical simulation of free blowing is presented, using classical viscous behaviour law. This simulation clearly shows that the evolution of the microstructure is important in modelling the behaviour of poly(ethylene terephthalate). An initial model is proposed that takes into account the evolution of crystallisation rate.     

Modeling and optimization of a high-pressure ethylene polymerization reactor using gPROMS

A gPROMS implementation of a comprehensive steady-state model of the high-pressure polymerization of ethylene in a tubular reactor is presented. Model outputs along the reactor length include the complete molecular weight distribution and branching indexes, as well as monomer conversion, average molecular weights, reactants’ compositions, and reactor temperature and pressure. A detailed calculation of physical and transport properties, such as the reaction mixture density, heat-transfer capacity, viscosity and global heat-transfer coefficient is also included. The reactor model is included in an optimization framework that is used to determine the best operating conditions for producing a polymer with tailor-made molecular structure in terms of the complete molecular weight distribution, branching and polydispersity.     

Integrated design and control of semicontinuous distillation systems utilizing mixed integer dynamic optimization

Semicontinuous distillation systems are notoriously difficult to design and optimize because the structural parameters, operational parameters, and control system must all be determined simultaneously. In the past 15 years of research into semicontinuous systems, studies of the optimal design of these systems have all been limited in scope to small subsets of the parameters, which yields suboptimal and often unsatisfactory results. In this work, for the first time, the problem of integrated design and control of semicontinuous distillation processes is studied by using a mixed integer dynamic optimization (MIDO) problem formulation to optimize both the structural and control tuning parameters of the system. The public model library (PML) of gPROMS is used to simulate the process and the built-in optimization package of gPROMS is used to solve the MIDO via the deterministic outer approximation method. The optimization results are then compared to the heuristic particle swarm optimization (PSO) method.     

High-order iterative learning model predictive control for batch chemical processes

In order to address two-dimensional (2D) control issue for a class of batch chemical processes, we propose a novel high-order iterative learning model predictive control (HILMPC) method in this paper. A set of local state-space models are first constructed to represent the batch chemical processes by adopting the just-in-time learning (JITL) technique. Meanwhile, a pre-clustered strategy is used to lessen the computational burden of the modelling process and improve the modelling efficiency. Then, a two-stage 2D controller is designed to achieve integrated control by combining high-order iterative learning control (HILC) on the batch domain with model predictive control (MPC) on the time domain. The resulting HILMPC controller can not only guarantee the convergence of the system on the batch domain, but also guarantee the closed-loop stability of the system on the time domain. The convergence of the HILMPC method is ensured by rigorous analysis. Two examples are presented in the end to demonstrate that the developed method provides better control performance than its previous counterpart.     

Soft sensor based on multi-phase stacking ensemble model with self-selected primary learner for batch processes

In batch processes, existing soft sensing methodologies encounter substantial challenges when confronted with nonlinearity and multi-phase issues. In response to these challenges, an innovative soft sensing framework known as the multi-phase stacking ensemble model with self-selected primary learner is proposed. The main innovation of this framework lies in the solution to the primary learner selection issue within the stacking model. To commence, the batch process is divided into multiple phases employing a Gaussian mixture model, thereby establishing local ensemble models for each phase. Subsequently, the iterative self-selection of primary learners strategy is proposed, which iteratively selects suitable primary learners for these models, optimizing their combination of primary learners for each local model. This primary learner selection strategy effectively enhances the accuracy of predictions in the stacking ensemble model. To further enhance the performance, Bayesian optimization is utilized to tune the hyperparameters of each local ensemble model. This step guarantees optimal performance of the model across diverse phases. Extensive simulation experiments are conducted on an industrial penicillin fermentation process to validate the effectiveness of the proposed framework. According to the findings, the model demonstrated superior performance compared to existing single-learner soft sensing methods and commonly utilized ensemble-based soft sensing methods in terms of both R² score (0.97584) and RMSE (0.0513). Overall, this framework offers a novel approach for selecting primary learners in stacking ensemble models and enhancing the predictive performance in batch processes for soft sensing.