Comparison of Two Types of Control Structures for Benzene Chlorine Reactive Distillation System s☆

2014-07-17 09:10CuimeiBoRidongZhangChenghaoZhangJihaiTangXuQiaoFurongGao

CuimeiBo,Ridong Zhang,Chenghao Zhang,Jihai Tang*,Xu QiaoFurong Gao*

1College of Automation and Electrical Engineering of Nanjing University of Technology,Nanjing 210009,China

2College of Automation,Hangzhou Dianzi University,Hangzhou,China,310028

3College of Chemistry and Chemical Engineering,Nanjing University of Technology,Nanjing 210009,China

4Department of Chemical and Bimolecular Engineering,The University of Science and Technology,Kowloon,Hong Kong,China

Comparison of Two Types of Control Structures for Benzene Chlorine Reactive Distillation System s☆

CuimeiBo1,Ridong Zhang2,Chenghao Zhang1,Jihai Tang3,*,Xu Qiao3,Furong Gao4,*

1College of Automation and Electrical Engineering of Nanjing University of Technology,Nanjing 210009,China

2College of Automation,Hangzhou Dianzi University,Hangzhou,China,310028

3College of Chemistry and Chemical Engineering,Nanjing University of Technology,Nanjing 210009,China

4Department of Chemical and Bimolecular Engineering,The University of Science and Technology,Kowloon,Hong Kong,China

A R T I C L E I N F O

Article history:

Received 27 December 2013

Accepted in revised form 19 January 2014 Accepted 19March 2014

Available on line 18 June 2014

Reactive distillation

Decentralized control structures

Optimal set point

Benzene chloride production

The“neat”operation of the two-reactant reactive distillation column has better steady-state economics,while it presents a challenge for design,optimization,and control of the process.Based on the optimal economic design, the dual-com position control structure and dual-temperature control structure are designed respectively for the benzene chlorine consecutive reactive distillation process.The effectiveness and robustness are analyzed comparably for the disturbance resistance in term s of changes of production rate and feed com position.Results show that dual-temperature control with propose selection of tray temperatures and the optimal profile of the set point can provide better transient process performance than the composition control structure.

©2014 Chemical Industry and Engineering Society of China,and Chemical Industry Press.All rights reserved.

1.Introduction

Reactive distillation(RD)combining reaction and multi-component separation in a single unit offers advantages over the catalyst usage,reactive heat utilization and cost saving[1,2].As reaction and separation take place simultaneously in one column,the strong interaction exists between the reaction and separation operations.The smooth operation and dynamic control of reactive distillation process are more challenge than conventional multi-unit processes[3].Kay mak and Luyben[4] compared the dynamic controllability quantitatively of a reactive distillation column against a conventional multiunit processes,and showed that its control was more difficult and its operability region was smaller than a conventional multiunit processas the reactive distillation column had less degrees of control freedom.Especially for a chemical reaction with two reactants,it is always expected to operate the reactive distillation column in“neat”case.In this case,a ratio of the two reactants needs to be maintained for the column to satisfy the stoichiometry.The“neat” flow sheet has better steady-state economics but presents a challenge in dynamic control and smooth operation because of the need to precisely balance the stoichiometry of the reactive section[5].Al-A rfaj and Luyben [6]proposed six types of alternative control structures for the quaternary or ternary reversible reaction process,in which a com position controller in the reactive section was used to satisfy the stoichiometric between the reactants.Some relevant process control structure shave been investigated with respect to the effectiveness of feed tray locations[7,8],relative volatility ranking[9],and plant-wide control structure design[10].

The above works are all confined to dynamic control for a quaternary or ternary reversible reaction process,while few relevant researches for consecutive reaction system have been reported.In this work,the dynamic design and simulation of two type decentralized control structures are studied for benzene chlorine consecutive reaction in a reactive distillation column based on the optimal economic design.The effectiveness of the proposed control structures are evaluated for the disturbances rejection of changes in the feed rate and com position.

2.Process Descriptions

2.1.Benzene chloride consecutive reactive distillation system

The benzene chloride production is atypical example of the exothermic consecutive reactions with two irreversible reactions,where chlorobenzene can be produced through the two reactions of toluene and chlorine,while chlorobenzene will react further with chlorine toproduce the by-product dichlorobenzene.The chemical reaction is as follows:

Table 1Kinetic and vapor-liquid phase equilibrium parameters

Fig.1.RD configuration for the benzene chloride production.

The benzene chlorination kinetics with ferric trichloride as catalyst was discussed in the Ref.[11].The reactions are highly exothermic and occur at a suitable temperature range of350 to 365 K.Some important kinetic and vapor-liquid phase equilibrium parameters for the system are tabulated in Table 1.

All the reaction takes p lace in the reactive section of the RD.The reaction rates are expressed by power function kinetic models on boiling states.The rate equation can be written as:

where r1and r2are the reaction rates of benzene and chlorobenzene,respectively,R=8.314 k J·km o l−1·K−1.The parameters in power function macro-kinetic model are shown in Table 2[12].A relatively higher selectivity to chlorobenzene can be obtained by maintaining a high concentration of reactant benzene and a low concentration of the product chlorobenzene in the reactive section.

2.2.RD configuration design

The optimal steady state configuration can be obtained using the economic optimum design of minimizing the annual cost,as reported in our previous research works[13,14].The optimal configuration of the benzene chlorine reactive distillation column is given in Fig.1.In the configuration,there are two sections:stripping section Ns=10 and reaction zone NRS=6.The feed stream s of chlorine and benzene (FC6H6=10.2 km o l·h−1and FCl2=10 km ol·h−1)are fed on the 6th tray and the second tray,respectively.In the simulated results,the purity of chloride benzene is 0.964 km ol·km ol−1and the selectivity is 0.964/(1−0.034)=99.8%.

3.Design of Bo th Control Structures

In this section,a systematic approach is proposed to design the composition control structures(CS1)and temperature control structures (CS2),respectively,based on the optimum steady-state design operated in“neat”m ode.The control objective is to maintain the product chlorobenzene purity at0.964 km ol·km ol−1and the selectivity of the product is more than 99%.

3.1.Design of the control structures

Firstly,some same based control loops are designed in both structures which are shown in Fig.2,such as the base level is controlled by manipulating the bottom flow rate(LB/FB),the reflux drum level is controlled by the re flux flow rate(LR/FR)and the top pressure is controlled by manipulating the hydrogen chloride gas rate(Ps/FD).

In the CS1 structure,the stoichiometry in reactive section is measured on line and controlled by manipulating the feed rate ratio(FCl2/ FC6H6)to keep the stoichiometric balance in the reactive section.And the bottom product com position(χB,C6H5Cl)is controlled by manipulating the vapor boil up VS,which are shown in Fig.2(a).

In the CS2 structure,the key issue is the location identification of the sensitive tray temperature and their pairing with the measurable handles.The relative gain array(RGA)method[14]is used to determine the control variable pairings for the controlled and manipulated variables.According to the RGA results,the temperatures of the 11th tray and the 13 th tray are the most sensitive to the changes in the feed rate FCl2and vapor boil up Vs,respectively,which is shown in Table 3. Since the temperature relevant in formation may be rich enough to in fer com position,the product composition and the stoichiometry in reactive section are controlled indirectly by controlling the tray temperatures T13and T11,as shown in Fig.2(b).

The sequential Tyreus-Luyben tuning method[15]is used to tune the controller parameters in both structures.The tuning parameters of controllers are shown in Table 3.For improving the robustness of the dual-temperature control structure,a new retuning based on stochastic dynamic optimization against uncertainty disturbances is applied, which was reported in our previous research works[16].

Table 2Power function parameters of the macro kinetic model

Fig.2.Two control constructs for the reactive distillation column.

3.2.Optimization of the temperature controller

When the feed composition χFchanges,the set point of the separating degree is changed.To satisfy quality requirement but prevent over distillation,the optimization is used to change the set point of the tray temperature controller in response to the change of feed com position[17].The steady state optimization is applied to reach a more economic operation point based on the rigorous reactive distillation model.The optimization problem can be described as

whereη*,T*11and T*13are decision variables and J is the objective function of optimization,such as positive energy consumption;column MESH equation and reactive kinetics equation are reported in our previous research works[14].The least square optimization algorithm is used to solve the on-line optimization problem.The optimal temperature values T*11and T*13are the set points of the temperature controllers.

4.Simulations and Discussion

The dynamic systems for the benzene chlorine process with CS1 and CS2 structures are simulated using the Aspen Plus software and Aspen Dynamic modules.In the simulation of the CS1 structure,supposed that the com position analyzers have the 10 m in dead-time.The effectiveness and robustness of the two structures are analyzed comparably for the disturbance resistance in term s of changes of production rate and feed com position.

4.1.Dynamic performance of the production rate changes

After the process reaches steady-state with one hour,the step changes in the benzene feed stream with ranges of−5%,10%,15%and 20%are introduced to the dynamic simulation system,respectively. Their system response curves are given in Fig.3.It shows that the feed stream FCl2is increased with the increase of the production rate by the ratio control strategy.More amount of the vapor boil up VSis required to maintain vapor-liquid equilibrium in the column.The results show that both structures can handle well the product rate changes and recover to the ideal steady state of0.964 km o l·km ol−1,while the dynamic responses ofCS2 are greatly improved than those ofCS1,such as the faster response(the CS2 structure takes about 1 h to recover the ideal steadystate,while the CS2 takes 5 h for the bottom product com position),a smaller overshoot for tray temperatures T13and product com positions χB,C6H5Cl,and a higher accuracy,as shown in Fig.3.

Table 3Relative gain array and tuning parameters for the benzene chlorine RD

Fig.3.Dynamic response under step changes in the feed stream of benzene ΔFCl2.

4.2.Dynamic performance of the feed composition changes

The com position changes with ranges of 5%and 10%in both feed stream s are added in the simulating systems,respectively.Fig.4 gives the system dynamic responses against the feed com position disturbances. The results show that both control structures can settle dynamically to the ideal steady state(χB,C6H5Cl=0.964 km ol·km ol−1).The com position control structure has the larger transient time(about 8 h)and larger overshoot(1.45%).For the separating degree can be changed with the feed com position change.The set point of the temperature needs to bead justed slightly in order to maintain the product quality.The suitable set point of the 13th tray temperature controller is given according to the on-line economic optimization.The simulations show that the CS2 structure has better transient performance than the structure CS1,such as fast dynamic response,smaller overshoot and steady-state time,as shown in Fig.4.

The effectiveness and robustness of both structures are analyzed and compared for the disturbances resistance in term s of the production rate and the feed com position stream s.It has been shown that the dual-temperature control with propose selection of tray temperatures,and their pairing and optimization of the set point can provide better transient process performance than the com position control structure.

Fig.4.Dynamic response under step changes in feed composition ΔzCl2.

[1]D.B.Kaymak,W.L.Luyben,A quantitative comparison of reactive distillation with conventional multi-unit reactor/column/recycle systems for different chemical equilibrium constants,Ind.Eng.Chem.Res.43(2004)2493-2507.

[2]Yeong-Tarng Tang,Yi-WeiChen,Hsiao-Ping Huang,Cheng-Ching Yu,Design of reactive distillations for acetic acid esterification,AICHEJ.51(2005)683-1699.

[3]Shih-Bo Hung,Ming-Jer Lee,Control of different reactive distillation con figurations, AICHEJ.52(2006)1423-1441.

[4]D.B.Kaymak,W.L.Luyben,Quantitative comparison of dynamic controllability between a reactive distillation column and a conventional multi-unit process,Com put. Chem.Eng.32(2008)1456-1470.

[5]M.A.Al-Arfaj,W.L.Luyben,Comparison of alternative control structures for an ideal two-product reactive distillation column,Ind.Eng.Chem.Res.39(2000)3298-3307.

[6]M.A.Al-Arfaj,W.L.Luyben,Comparative control study of ideal and m ethyl acetate reactive distillation,Chem.Eng.Sci.57(2002)5039-5050.

[7]Yu-Cheng Cheng,Cheng-Ching Yu,Effects of feed tray locations to the design of reactive distillation and its implication to control,Chem.Eng.Sci.60(2005) 4661-4677.

[8]G.Sunar,D.B.Kaymak,Effect of Feed Tray Location on Temperature-based Inferential Control of Double Feed Reactive Distillation Columns,Ind.Eng.Chem.Res.48 (2009)11071-11080.

[9]Chin-Shih Chen,Cheng-Ching Yu,Effects of relative volatility ranking on design and control of reactive distillation systems with ternary decomposition reactions,Ind. Eng.Chem.Res.47(2008)4830-4844.

[10]I-Kuan Lai,Shih-Bo Hung,W an-Jen Hung,Cheng-Ching Yua,Ming-Jer Lee,Hsiao-Ping Huang,Design and control of reactive distillation for ethyl and isopropyl acetates production with azeotropic feeds,Chem.Eng.Sci.62(2007)878-898.

[11]O.Sokolov,M.D.Hu rley,T.J.Wallington,E.W.Kaiser,Kinetics and mechanism of the gas-phase reaction of Cl atoms with benzene,J.Phys.Chem.A 102(1998) 1671-1681.

[12]Zezhang Sheng,Shan Li,Mifen Cui,Study on macrokinetics of benzene chlorination boiling state,J.Nanjing Univ.Chem.Technol.19(9)(1997)8-13(in Chinese).

[13]LianghuiDing,JihaiTang,Mifen Cui,Cuim eiBo,Qiao Xu,Optimum design and analysis based on independent reaction amount for distillation column with side reactors:production of benzyl chloride,Ind.Eng.Chem.Res.50(2011)1143-1152.

[14]B.O.Cuimei,T.A.N.G.Jihai,B.A.I.Yangjin,Q.I.A.O.Xu,D.I.N.G.Lianghui,Z.H.A.N.G.Shi, The design and control of distillation column with side reactors for chlorobenzene production,Chin.J.Chem.Eng.20(6)(2012)1113-1120.

[15]Yu-De Lin,Hsiao-Ping Huang,Cheng-Ching Yu,Relay feedback tests for highly nonlinear processes:reactive distillation,Ind.Eng.Chem.Res.45(2006)4081-4092.

[16]Cuimei Bo,Jihai Tang,Mifen Cui,Kangkang Feng,Xu Qiao,Furong Gao,Control and profile setting of reactive distillation column for benzene chloride consecutive reaction system,Ind.Eng.Chem.Res.52(49)(2013)17465-17474.

[17]L.Ü.Wenxiang,Z.H.U.Ying,H.U.A.N.G.Dexian,J.I.A.N.G.Yongheng,J.I.N.Yihui,A new strategy of integrated control and on-line optimization on high-purity distillation process,Chin.J.Chem.Eng.18(1)(2010)66-79.

☆Supported by the National Natural Science Foundation of China(61203020, 21276126),Jiangsu Province Natural Science Foundation(BK2011795),National Key Technology Research and Developm ent Program of the Ministry of Science and Technology of China(2011BAE18B01).

*Corresponding author.

E-mailaddresses:jh tang@n jut.edu.cn(J.Tang),kefgao@ust.hk(F.Gao).