Chemical Product and Process Modeling

A Hybrid Model of Solvent Hehydratic Distillation Column Based on Mechanism and LS-SVM

Qingtao Sun et al. — Wed, 21 Dec 2011 22:55:29 PST

Considering that there are many factors having highly nonlinear effect on the Murphree efficiency of column in petrochemical industry and that the obtained accurate value of the Murphree efficiency plays an important role in the performance of the column model, a novel hybrid model method which combines mechanism with least square support vector machine (LS-SVM) is proposed to model solvent dehydratic distillation column (SDDC). In the hybrid model, the mechanism method is used to develop the whole model of SDDC, and LS-SVM is employed to describe the relationship between the factors and the Murphree efficiency due to the nonlinear representing ability of LS-SVM. Further, the correlation between the factors and the Murphree efficiency is analyzed to select the input variables of LS-SVM, and leave-one-out cross-validation method is employed to demonstrate the performance of LS-SVM. The simulation result shows that the hybrid model has robust and good predicting performance and can simulate different conditions of SDDC.

Selection of Suitable Turbulence Models for Numerical Modelling of Hydrocyclones

Mohsen Karimi et al. — Wed, 07 Dec 2011 13:10:40 PST

The capability of Computational Fluid Dynamics (CFD) alternates the interest of researcher from the empirical models into the numerical approaches for studying hydrocyclones. This paper presents a comprehensive survey on the influences of turbulence model options in the 3D simulation of the hydrocyclone flow pattern. The required grid resolution was selected through a grid independency study. Four categories of turbulence models involving models based on the Boussinesq hypothesis, the Reynolds Stress Model (RSM), the Large Eddy Simulation (LES) model, and the Detached Eddy Simulation (DES) model were investigated for prediction of velocity components within the hydrocyclone. The methodology was validated by experimental data. The results confirm that both RSM and LES models are efficient turbulent model choices for the simulation of swirling flow of hydrocyclones.

Nonlinear Modeling for the Degradation of Aqueous Azo Dyes by Combined Advanced Oxidation Processes Using Artificial Neural Networks

Masroor Mohajerani et al. — Mon, 07 Nov 2011 19:09:27 PST

One-hidden-layer artificial neural networks (ANNs) using a back-propagation structure have been trained on different sets of experimental data to identify and evaluate the degradation of different azo dyes (Reactive Yellow 84, Reactive Blue 19, Direct Red 23, Direct Red 28, and Acid Blue 193) by photo-Fenton process and combined ozonation and ultrasonolysis processes. Different input variables such as pH, initial concentrations of dyes and ozone, reaction time, ultrasonic power density, and initial concentrations of hydrogen peroxide and ferrous in aqueous solution were employed to model the degradation rates of azo dyes based on the decolorization efficiency and the removal rate using chemical oxygen demand (COD) and total organic carbon (TOC). A new model expression is developed to find the effect of individual parameters and their interactions on the efficiency of organic degradation by advanced oxidation processes.

Modeling of a Fixed–Bed Reactor for the Production of Phthalic Anhydride

Amir Rahimi et al. — Sun, 06 Nov 2011 08:11:45 PST

In this study, the performance of a fixed–bed tubular reactor for the production of phthalic anhydride is mathematically analyzed. The conversion degree and reactor temperature values are compared with the measured one in a tubular reactor applied in Farabi petrochemical unit in Iran as well as reported data in the literature for a pilot plate. The comparisons are satisfactory. The effects of some operating parameters including reactor length, feed temperature, reactor pressure, and existence of an inert in the catalytic bed are investigated. The optimum value of each parameter is determined on the basis of the corresponding operating conditions.

Wet Gas Measurement with Slotted Orifice Meter—Effect of Geometry of Slots and Pressure

Perumal Kumar et al. — Fri, 28 Oct 2011 11:51:13 PDT

Wet gas metering by differential pressure flow meters is gaining prominence in the oil and gas industry, owing to their simple construction. Slotted orifice, a modified version of the standard orifice meter has been found promising by many researchers. This novel flow meter is shown to be insensitive to the upstream flow profile with lower head loss and faster pressure recovery compared to the standard orifice. In the present work, the effect of geometry of slots and pressure on the performance of the slotted orifice has been studied by CFD modeling of the wet gas flow. The performance of slotted orifice with rectangular perforations (1.5 ≤ l/w ≤ 3.0) and circular perforations has been compared with that of the standard orifice having same β-ratio of 0.40. Simulation results reveal that the shape of perforation of the slotted orifice has no effect on the differential pressure. However, the pressure recovery with the rectangular slots is found to increase with increasing aspect ratio. Moreover, at low pressure, slotted orifice is found to be more sensitive to liquid presence than the standard orifice. The relatively higher over reading values obtained in this work is consistent with the results of Geng et al (2006) that for slotted orifice, a low beta ratio is more sensitive to the liquid presence in the stream and hence is preferable for wet gas metering. Homogeneous model, Steven’s and De Leeuw’s correlations, are found to be better than orifice correlations for wet gas mass flow prediction.

Simulation of an Industrial Linear Low Density Polyethylene Plant

Ali Farhangiyan Kashani et al. — Tue, 18 Oct 2011 12:40:43 PDT

In this paper, an industrial linear low density polyethylene (LLDPE) production process including two serried fluidized bed reactors (FBR) and other process equipment was completely simulated in steady state mode. Both of FBRs were considered like two serried continuous stirred tank reactors (CSTR). In this simulation, a kinetic model that is based on a multiple active site heterogeneous Ziegler-Natta catalyst was used for simulation of reactions in two FBRs. Simulator by using this model is able to predict the important attributes of LLDPE like melt flow index (MFI), density (ρ), polydispersity (PDI), numerical and weight average molecular weight (M_n, M_w) and co-polymer molar fraction (S_FRAC). On the other hand, this simulator can be applied in wide range of changing in inlet operating conditions. The results of the simulation are compared with industrial data of LLDPE plant. A good agreement is observed between the simulator predictions and actual plant data. Finally, by using of the simulator, the steady state operating conditions for producing different grades of polyethylene are obtained.

CFD Simulation of Water Gas Shift Membrane Reactor—Pressure Effects on the Performance of the Reactor

Byron Smith R.J. et al. — Wed, 12 Oct 2011 09:01:26 PDT

Membrane reactor is a process intensified equipment that carries out both the reaction and separation in a single vessel. The equilibrium limited water gas shift reaction is an ideal reaction to be carried out in a membrane reactor as it improves the conversion of the reaction and reduces the space requirement for the reactor. Computational fluid dynamics offers a virtual prototyping of the reactor and helps in design, optimization, and scale-up of the reactor. To obtain pure hydrogen from the membrane reactor, the pressure of the reactor needs to be optimized. Hence the water gas shift membrane reactor is subjected to computational fluid dynamic analysis to understand the role played by pressure on the performance of the reactor using three different gas mixtures. The CO conversion and H₂ recovery for the different operating pressures are simulated and the effects of pressure are discussed in this paper.

Microfluidics of Nanodrug Delivery: Effect of Reynolds Number Ratio and Particle Size

A. Venu Vinod et al. — Fri, 07 Oct 2011 08:49:27 PDT

In the present study dealing with nanodrug delivery, nanofluid flow in a microchannel is considered. The nanofluid considered is a solid (aluminum oxide) - liquid (water) mixture. A transient 3-D problem of controlled nanodrug delivery has been numerically solved. The study has been carried out using ANSYS CFX Version 11.0. The velocity profile and dimensionless nanodrug concentration profile for different Reynolds number ratios were obtained at different axial stations along the length of the microchannel (z = 0.5, 5, and 10mm). Effect of particle size on drug uniformity has been investigated.

Polyhedral Particles Hopper Flowrate Predictions using Discrete Element Method

Feras Y. Fraige et al. — Tue, 04 Oct 2011 09:31:56 PDT

Discrete, or Distinct, element modelling, DEM, is used to investigate the effect of particle shape on hopper flowrate. Firstly, simulation results using a model based on spherical particles are validated against solid discharge rates predicted by the modified Beverloo equation. The validated model is then used to investigate the shape effects on discharge rate from a flat bottomed rectangular hopper. Two regular rounded polyhedral particles are studied. The discharge rate varies significantly with particle shape. A reduction on flowrate of up to 49% and 41% is observed for tetrahedral and octahedral particles, respectively. The effects of friction, cohesion, and damping on flowrate are also investigated with DEM.

Model Order Reduction and the Heat Transfer Modelling of a Heat Exchanger by Using Artificial Neural Approach

Yahya Chetouani — Mon, 19 Sep 2011 12:48:32 PDT

The main aim of this paper is to establish a reliable model of a process behavior under the normal operating conditions. The use of this model should reflect the true behavior of the process in the whole way and thus distinguish a normal mode from the abnormal modes. In order to obtain a reliable model for the process dynamics, the black-box identification by means of a NARMAX model has been chosen in this paper. It is based on the neural networks approach. The main advantage of the proposed approach consists in the natural ability of neural networks in modeling non-linear dynamics in a fast and simple way and in the possibility to address the process to be modeled as an input-output black-box, with little or no mathematical information on the system. This paper will show the choice and the performance of the neural network in the training and the test phases. A study is related to the number of inputs, and of hidden neurons used and their influence on the behavior of the neural predictor. Three statistical criterions, Aikeke’s information criterion (AIC), Rissanen’s Minimum Description Length (MDL), and Bayesian information criteria (BIC), are used for the validation of the experimental data. In order to illustrate the ideas proposed concerning the dynamics modelling, a heat exchanger is used. The outlet temperature is modeled according to the inlet temperature. The model is implemented by training a Multilayer Perceptron artificial neural network with input-output experimental data. Satisfactory agreement between identified and experimental data is found and results show that the model successfully predicts the evolution of the outlet temperature of the process.

Controllability of Distillation Sequences for the Separation of Ternary Mixture

Rajasekhar Nadendla et al. — Wed, 14 Sep 2011 10:22:28 PDT

The present work aims at finding the best distillation sequence in terms of control properties for the separation of a ternary mixture of n-pentane, n-hexane and n-heptane. The study has been carried out using ASPEN PLUS. The different sequences have been compared for regulatory problem (step change in feed concentration and feed flow rate) and servo problem (step change in product composition). PI controller has been used to control level and composition in the various configurations studied. Auto tuning of the controller has been carried out using relay feedback test. Ziegler–Nichols method was used to calculate controller gain (KI) and integral time (τI) of PI controller. Controller parameters have been fine tuned using these tuning maps.

A Note on a Double Temperature-Cascade Control Strategy for Distillation Columns

Galo R. Urrea-Garcìa et al. — Tue, 06 Sep 2011 22:59:50 PDT

A double temperature-cascade control strategy for controlling distillate composition is proposed. The control structure is based on two secondary loops that are designed for tracking the temperature of one tray in the rectifying and stripping section by manipulation of the reflux rate and reboiler heat input, respectively. Simulation results show that when saturations are imposed on controller outputs (reflux rate and heat input) both cascade controllers can act in a switching-like scheme without interaction problems. It is also shown that an adequate setting of a saturation parameter allows a low energy regulation of distillate composition under disturbances in feed composition.

Study of the Particle Separation Performance of a Dust-settling Hopper

Bofu Wu et al. — Mon, 29 Aug 2011 13:24:21 PDT

This paper presents a numerical study to predict the particle separation performance of a dust-settling hopper using computational fluid dynamics. The Euler-Lagrange approach was employed to analyze the particle separation efficiency and the outflow particle concentration of the dust-settling hopper under different inlet airflow velocities. The calculation results obtained reveal that the overall particle separation efficiency and the outflow particle concentration decrease with the increase of the inlet airflow velocity, and the particle grade efficiency increases with particle size. Since there is a paradox between the particle separation performance and the particle removal performance for a street vacuum sweeper, it is necessary to counter-balance the effects of the inlet airflow velocity on them. According to the simulation analysis, an appropriate inlet airflow velocity is provided for the design of the dust-settling hopper.

Comparing Usage of Temperature Control and Composition Control for Reactive Distillation Column in Methyl Ester Production

Kulchanat Prasertsit et al. — Thu, 11 Aug 2011 11:54:01 PDT

The temperature control and the composition control for the production process was used to evaluated a method to control the quality of methyl ester production in a reactive distillation column. The optimum condition at steady state was selected as the base case condition for the control study by using Aspen Dynamic Simulation. In this work, two controller settings for PID controller of each control loop were based on the Zeigler-Nichol and the Tyreus-Luyben tuning formulae. The controllability and performance of both control structures were compared. The result shows that the control loop using the Ziegler-Nichols setting gave better control performance than the other using the Tyreus-Luyben setting. It also found that the control structure based on temperature control gives the same control performance as the other based on composition control, but less expensive and less complicated to implement.

CFD Modelling of Reverse Osmosis Channels with Potential Applications to the Desalination Industry

S. Wardeh et al. — Fri, 05 Aug 2011 14:16:36 PDT

Desalination by reverse osmosis is receiving increasing attention due to recent improvements in membrane technology in particular. This technique is now used for small to very large scale applications, and has been identified as a key contributor to water treatment and provision not only in Middle Eastern countries but also closer to home in the United Kingdom for example. Computational Fluid Dynamics (CFD) is increasingly used for complex modelling applications in the chemical process and water industries. Previous work by the writers and others for spacer-filled desalination channels have yielded very good results in terms of modelling the flow and salt characteristics and history in the channels, as well as the overall performance of the membrane system. However, transient features imply that such simulations can be very costly and, if they can approximate experimental layouts with horizontal membranes, the investigation of a full desalination performance and operation remains a very distant prospect. CFD would therefore appear to be unsuited to industrial design at the present time. It is necessary to reduce the cost of the numerical simulations whilst maintaining high fidelity in key design parameters such as pressure loss, shear stress and salt concentration on the membrane, and through flow across the membrane for the method to be useful in design. A momentum loss approach based on a porous medium model is presented to this effect and benchmarked against detailed CFD results with success. It is shown to be capable of meeting key design parameters at a fraction of the full CFD cost.

Novel Heuristic for Low-Batch Manufacturing Process Scheduling Optimisation with Reference to Process Engineering

S. Maqsood et al. — Fri, 05 Aug 2011 14:16:25 PDT

Scheduling is an important element that has a major impact on the efficiency of all manufacturing processes. It plays an important role in optimising the manufacturing times and costs resulting in energy efficient processes. It has been estimated that more than 75% of manufacturing processes occur in small batches. In such environments, processes must be able to perform a variety of operations on a mix of different batches. Batch-job scheduling optimisation is the response to such low batch manufacturing problems. The optimisation of batch-job process scheduling problem is still a challenge to researchers and is far from being completely solved due to its combinatorial nature. In this paper, a novel hybrid heuristic (HybH) solution approach for batch-job scheduling problem is presented with the objective of optimising the overall Makespan (C_max). The proposed HybH is the combination of Index Based Heuristic (IBH) and the Finished Batch-Job (FBJ) process schedule. The heuristic assigns the first operation to a batch-job using IBH and the remaining operations on the basis FBJ process schedule. The FBJ process schedule gives priority to the batch-job with early finished operations, without violating the constraints of process order. The proposed HybH is explained with the help of a detailed example. Several benchmark problems are solved from the literature to check the validity and effectiveness of the proposed heuristic. The presented HybH has achieved batch-job process schedules which have outperformed the traditional heuristics. The results are encouraging and show that the proposed heuristic is a valid methodology for batch process scheduling optimisation.

Knowledge Based System Implementation for Lean Process in Low Volume Automotive Manufacturing (LVAM) with Reference to Process Manufacturing

N.M.Z.N. Mohamed et al. — Thu, 04 Aug 2011 17:06:24 PDT

Global manufacturing industry mostly depends on new product development and processes to become competitive. The product development process for automotive industry is normally complicated, lengthy, expensive, and risky. Hence, a study of lean manufacturing processes for low volume manufacturing in automotive industry is proposed to overcome this issue by eliminating all wastes in the lengthy process. This paper presents a conceptual design approach to the development of a hybrid Knowledge Based (KB) system for lean process in Low Volume Automotive Manufacturing (LVAM). The research concentrates on the low volume processes by using a hybrid KB system, which is a blend of KB system and Gauging Absences of Pre-requisites (GAP). The hybrid KB/GAP system identifies all potential waste elements of low volume process manufacturing. The KB system analyses the difference between the existing and the benchmark standards for lean process for an effective implementation through the GAP analysis technique. The proposed model explores three major lean process components, namely Employee Involvement, Waste Elimination, and Kaizen (continuous improvement). These three components provide valuable information in order for decision makers to design and implement an optimised low volume manufacturing process, but which can be applied in all process manufacturing, including chemical processing.

Numerical Analysis and Optimization of Oxygen Separation from Air via Pressure Swing Adsorption

Seyyed M. Ghoreishi et al. — Thu, 04 Aug 2011 17:06:11 PDT

A pressure swing adsorption air separation process in a commercial aircraft using 13X zeolite with a more complex cycle than the classic Skarstrom was simulated via a predictive dynamic model to evaluate and optimize oxygen generation system. The coupled mass, energy, and momentum differential equations were discretized using the implicit central finite-difference technique and the obtained equations were solved by Newton-Raphson method. The validated model in conjunction with an optimization procedure (Successive Quadratic Programming) was utilized to investigate the oxygen separation efficiency as a function of β (ratio between the bed time constant and the particle diffusion time constant), C_fp (purge orifice coefficient), θ_cycle (cycle time), C_ff (feed valve), C_fe (exhaust valve) and p_H* (high pressure operation). A set of optimum values (β=150, C_fp=0.7, θ_cycle=1.5, C_ff=31, C_fe=52 and p_H*=3.8) was obtained and recommended to achieve maximum recovery (0.26) at 98% purity.

Numerical Simulation of Fluid Flow and Heat Transfer in a Counter-Current Reactor System for Nanomaterial Production

Cai Y. Ma et al. — Tue, 02 Aug 2011 17:06:25 PDT

Continuous hydrothermal flow synthesis (CHFS) systems can provide high quality fine nanoparticles. However, optimisation of the CHFS system including the reactor and heat exchanger design, and their scaling-up for commercial applications have not been studied and cannot be achieved only through laboratory and pilot plant experiments. CFD modelling techniques are being widely used to simulate fluid field, heat and mass transfer in a lot of industrial process equipment. However, the application of CFD to model CHFS systems is still rare. This paper employs CFD methodology to simulate fluid flow and heat transfer in a counter-current reactor and a tubular heat exchanger of a laboratory-scale CHFS system for the production of TiO₂ nanoparticles. The distributions of flow and heat transfer variables such as velocity and temperature in both units are obtained using ANSYS Fluent package. The tracer concentration profile is also simulated via solving the species equations to investigate the mixing behaviour in the counter-current reactor. Temperature distributions at different locations in a counter-current reactor and a tubular heat exchanger of a CHFS system were obtained experimentally. The simulated temperatures in both the reactor and the heat exchanger are compared with the available experimental data, which reveals that a good level of agreement is achieved.

Fast Pyrolysis of Biomass in Bubbling Fluidized Bed: A Model Study

Priyanka Kaushal et al. — Tue, 02 Aug 2011 17:06:10 PDT

This study proposes a model for the fast pyrolysis of biomass. A reaction scheme of a set of three parallel reactions followed by a set of two parallel reactions has been used to describe the primary and secondary reactions of biomass pyrolysis in a stationary bed reactor. A simple first-order kinetic approach has been applied to predict the product yields. The bed hydrodynamics, the mass transfer between phases and the reaction kinetics have been mathematically formulated. The effects of the operating parameters on the biomass pyrolysis product yield were simulated; the results show that the reaction temperature and nitrogen flow rate plays an important role in the yield of bio-oil. Good agreement between the predicted and measured results was obtained.