CCB 3013 CHEMICAL PROCESS DYNAMICS, INSTRUMENTATION AND CONTROL MAY 2013

PROPOSAL REPORT GAS ADSORPTION COLUMN

GROUP

8 DANESKUMAR A/L MANOGARAN MOHD HUSAINNI BIN SAAID STEPHEN LIEW CHEE SENG TEE WEI HENG TUAN MUHAMMAD BIN TUAN YAHYA 15049 14745 14921 16063 14791

GROUP MEMBERS :

LECTURER

DR NOORYUSMIZA YUSOFF

0.0 ABTRACT This work presented the tray gas adsorption column as a selected unit process operation. This research work is to develop a steady state and dynamic models for the control process dynamic problem and this control process is to maintain a process at the desired operating conditions, safely and efficiently while satisfying environmental and product quality environment. The role of the control strategy will be presented by relating to variables and the assumptions of the adsorption column such as feedback and feed-forward control strategy. The developed models of the control strategy is programmed and simulated by using MATLAB and SIMULINK flow sheeting software using standards strategy by using the standards of S-Functions and a few assumptions have been made to simplify the derivation of the gas tray derivation of the tray gas adsorption dynamic model. The utilization of MATHLAB and SIMULINK successfully developed and derived the steady state and dynamic models for the control process dynamic problem. The implementation of a control strategy is important to achieve optimal operation although obvious uncertainties and disturbances of the plant behaviors are present. 1.0 INTRODUCTION Engineering information on the operation specifically on the gas absorption equipment was at present quite adequate to permit proper estimates and new design operation of the process itself. Existing data and analysis clearly indicate that in almost all the cases showed the rate of absorption of a solute was determined by the diffusion itself. Any other relative reaction usually makes a less effect to the absorption rate because of the equilibrium factor take into account (Lewk & Whitman , 1924 ).Our main interest was to learnt and study the physical and chemical part of this unit operation . So, the detail part of this introduction will outline the study of control and dynamic process in gas absorption column. Basically, the absorption column used in petrochemical industries processes like fertilizers, gas purifications and manufacture of nitric acid where produced in a large scale (Alguasane & Micheal, 2006).All this process can be visualized for controlling purposes and stimulation as mentioned by Edward & Wink (1998) in their studies Simulation of an absorption column performance using feed-forward neural networks in nitric acid production. The team managed to work on towards the modelling of gas absorption thus formulating mathematical model and stimulation by using computer software to obtain the rate of absorption .

Besides that, the idea of using computer was explained in detail by Menwer, Mazen & Hans in their research Dynamic and Control of sieve tray gas absorption . The team also argued powerful combination of using SIMULINK/MATLAB software were able to stimulate the steady state, open and closed loop, dynamic behaviour and stability of the unit operation itself. In an engineering calculation, a dynamic model can be used to characterize

the behaviour of a process for a wide variety of conditions whether steady state or unsteady state. For instance, steady state design of chemical equipment is confronted by dynamic and controllability issues in control and dynamics. In this regard, it is often easy to design a chemical process based on steady state conditions, which is practically uncontrollable and unrealistic. In order to avoid any wrong assumption during process synthesis and design, and to ensure safe start-up, shutdown and stable absorption column, the dynamic behaviour of the relevant units should be known. It is important to improve the overall design and optimize the operation of chemical processing units and is modelled through a system of mathematical equations to enable prediction of the process behaviour as well as to cope with the disturbances in a system (Menwer, Mazen & Hans, 2013). The next section would focus on stating clear and brief objective of this study and the method used as well. 1.1 Problem Statement The main objective of this research work is to develop a steady state and dynamic models for the control process dynamic problem, where an existing process is simulated as a real-life example. The developed models are programmed in MATLAB and SIMULINK flow sheeting software using the standards of S-Functions. The combination of SIMULINK and MATLAB is utilized to develop an industrial feedback control system for a general tray gas absorption column with only one solute transfer. This environment allows for a practical way to develop a control block diagram including measuring device, feedback controllers and final control element. In this section we present a couple of dynamic steady state model and mass transfer model for a general binary gas absorption column. The dynamic model is derived first and then followed by the steady state model derivation. To simplify the derivation of the tray gas absorption dynamic model, the following set of assumptions is used : 1. Liquid on the tray is perfectly mixed and incompressible. 2. Tray vapor holdup is negligible.

3. Vapor and liquid are in thermal equilibrium (same temperature) but not in phase equilibrium. A Murphree tray efficiency is used to describe the departure from equilibrium. 4. Total gas flow rate (G) is constant. 5. The equilibrium relationship is linear. Using the above assumptions, the dynamic mathematical model for the tray gas absorption column is derived using unsteady state total material and component balances on the jth tray inside the column. This figure shows the schematic diagram of the tray gas absorption column under consideration with a focus on single cross-flow tray. An automatic control is design using an algorithm to keep the condition in the optimum range. This is to prevent any human error from occurring during the control process.

2.0 LITERATURE RIVIEW Separation processes play an important role in most chemical manufacturing industries. Streams from chemical reactors often contain a number of components (Mahmoud, 1992 ); some of these components must be separated from the other components for sale as a final product, or for use in another manufacturing process. A common example of a separation process is gas absorption (also called gas scrubbing, or gas washing) in which a gas mixture is contacted with a liquid (the absorbent or solvent) to selectively dissolve one or more components by mass transfer from the gas to the liquid. Absorption is used to separate gas mixtures; remove impurities, contaminants, pollutants, or catalyst poisons from a gas; or recover valuable chemicals. In general, the species of interest in the gas mixture may be all components, only the component(s) not transferred, or only the component(s) transferred (Lee, 1990) .

Absorption is frequently conducted in trayed towers (plate columns), packed columns, spray towers, bubble columns, and centrifugal contactors. A trayed tower is a vertical, cylindrical pressure vessel in which vapor and liquid, which flow counter-currently, are contacted on a series of metal trays or plates; see Fig. 1. Components that enter the bottom of the tower is the gas feed stream are absorbed by the liquid stream, that flows across each tray, over an outlet weir and into a down-comer, so that the gas product stream (leaving the top of the tower) is more pure.

B. Assumptions and definitions The basic assumptions used are:

A1) The major component of the liquid stream is inert and does not absorb into the gas stream.

A2) The major component of the gas stream is inert and does not absorb into the liquid stream.

A3) Each stage of the process is an equilibrium stage, that is, the vapour leaving a stage is in thermodynamic equilibrium with the liquid on that stage.

A4) The liquid molar holdup is constant.

We now introduce the following variable definitions: L = moles inert liquid per time: = liquid molar flow rate. V= moles inert vapor per time: = vapor molar flow rate M= moles liquid per stage: = liquid molar holdup per stage

W= moles vapor per stage: = vapor molar holdup per stage xj= moles solute (stage j) per mole inert liquid (stage j) yj= moles solute (stage j) per mole inert vapor (stage j) 3.0 METHODOLOGY 3.1 Introduction Generally in this specific chapter, we will be focusing on the study methodology in which covers several important aspects in conducting this study. It basically covers points such as the study location, study design, the instrument used, sample population,, the critical analysis and sources of data. 3.2 Location of Study This research is conducted throughout the area in Universiti Teknologi Petronas (UTP). This institution was a wholly owned subsidiary of PETRONAS, the national oil company of Malaysia. The campus is built on a 4 square kilometres (990 acres) site strategically located at Bandar Seri Iskandar, Tronoh,Perak, Malaysia. When UTP firstly moved to Tronoh, it took over the campus of Universiti Sains Malaysia (USM)'s engineering faculty, which has since moved to their new campus in Penang. In its initial years, lectures, lab sessions and tutorials were held in the old USM facilities, and UTP gradually shifted all operations to newer facilities once construction was completed. UTP was established in 10 January 1997 when PETRONAS was invited by the Malaysia Ministry of Education to set up the university. It was launched by honourable Tun Abdullah Ahmad Badawi, the fifth prime minister of Malaysia.

3.3 Study Design

We will be using the quantitative method to obtain our data in which the data taken will be in numerical value rather than qualitative method. This method is much more accurate as it will show a specific group of answers that are easy to interpret. The main purpose for this survey is to obtain feedback from our topic based on the study case and discussion according to the course study .In short, our study or research will

reveal the performances of gas absorption column by displaying the graph pattern with variables, review of study case and data collection with appropriate tools. 3.4 Research Instrument Throughout this study, we choose the SIMULINK, Matlab for making graphs, stimulations and study on the effects of regulating and disturbance variable. Other than that, we are also interested to describe the looping system, material and energy balance involved by constructing the block diagram.

3.5 Sampling Our data and results are mainly being taken from our study case from industry that has been used gas absorption column. From there, we analyse this unit process in term of feedback from the industry and its stability, effectiveness and efficiency 3.6 Sources of Data In order to obtain such reliable and practical facts to complete this study, we were mainly utilized our primary sources which is the study case from past research paper and issues relating to gas absorption column. 3.8 Data Analysis Obtain the data was gathered and analysed thoroughly and well-portrayed in tables, charts and graphs mainly on the variables and process step as shown in the result section. 3.9 Conclusion For this part, after summarising the main points, we highlight important process dynamic and control in this gas absorption column. Therefore we were able to interpret, analyse and compile the data effectively and efficiently.

4.0 CASE STUDY The gas absorption column shown in the schematic diagram below is used to adjust the mass fraction of A in the outlet stream to 0.6 and deliver it at 150kg/min to the downstream process. The mass fraction and mass flow rate of stream2 at the design condition are 0.4 and 90kg/min, respectively. The volume of the gas at the steady state (design) condition is 4.2 m3.

Gas Absorption Column

Feedback controller

M.D. Gas Analyzer

Desired Set Point

Control Objective: To control the volume at and outlet concentration at .

Variables: Controlled Manipulated Disturbance = = = ,

Assumption: Perfect mixing

4.1 DEVELOPMENT OF NON-LINEAR EQUATION MODEL

Overall Mass Balance

( )

.. (1)

Component Balance ( )

For constant ,
( )

(2)

By using chain rule,

.. (3)

Substituting (3) into (2), .. (4)

Substituting (1) into (4),

After cancelling common terms and rearranging, ( ) ( )

4.2 DEVELOPMENT OF LINEAR EQUATION MODEL Overall Mass Balance .. (1)

Linearization:

).. (2)

Substitute (2) into (1)

).. (3)

Steady State: (4)

Deviation: (3) (4) ( ) ( ) ( )

Laplace: ( ) ( ) ( ( ) ( ) ) ( ) ( ) ( ) ( ) ( ( ) ( ) ) ( ( )

Substituting value for ( ) ( )

and

, ( ) ( )

( )

) (

) (

( )

( )

( )

Component Balance ( ) ( )

Linearization:

) )

( )

)
(

) (1)

) .. (2)

) )

( )

)
(

) (3)

( )

) . (4)
(

) (

)
(

( )

)
(

)
(

) (5)

Steady State equation:

) .. (6)

Deviation equation: (5) (6), ( )( ( ( ) ) )( ( ( ) ) ) ( ( ) ( ) ( )( ( )( ) ) ( ) )

Laplace Transform: ( ) ( ( ) ) ( ) ) ( ( ) ) ( ) ( ) ( ) ( ( ) ) ( ) ( ) ( ( ) ( ) ) (

( )

( )

( )

( )

( )

5.0 SIMULINK AND CONTROL

5.1 Simulink: Linear Model

Figure 5.1 Shows the working steps of gas absorption column Graphs of Simulink for Linear Model

Fig 5.2 : Output response with positive step input 5.2 Simulink: Non-Linear Model Subsystem

Graph for Non-Linear Model Subsystem

Figure 5.3 : Output response with negative step input Simulink: A Linear Subsystem A ( Fuzzy Controller)

Fig 4 . Controller is tuned to interfere the natural decay of the system

Fig 5 : The controller is tuned to improve the response of the system

6.0 DISCUSSION Our project hereby is aimed to develop a steady state and dynamic models for the control process dynamic problem. This control process is to maintain a process at the desired operating conditions, safely and efficiently while satisfying environmental and product quality environment. The project presented the tray gas adsorption column as a selected unit process operation. The role of the control strategy will be presented by relating to variables and the assumptions of the adsorption column such as feedback and feed-forward control strategy. To study the dynamic behaviour of the process in the tank, dynamic models (mathematical model) of the process are developed from both the mass balance and component balance, to address how the output variable(s) changes with time in response to the changes in the input variable(s). The developed dynamic model which is non-linear is needed to be solved in order to quantify the changes in the output variable that occurs. This can be done in two ways in which the is solving by using computer software such as MATLAB and SIMULINK flow sheeting software using standards strategy by using the standards of S-Functions and a few assumptions have been made to simplify the derivation of the gas tray derivation of the tray gas adsorption dynamic model. Solving of the dynamic models using SIMULINK, differs from the method of using Laplace domain in term of linearity. Laplace domain model actually involves the solving of the linear approximation of the non-linear model but however SIMULINK models, simulates and analyses the non-linear dynamic systems without the need for linearization. The Laplace domain model leads to the development of transfer functions which relates the changes of a process output with respect to the changes in the process input. Therefore, it can be devised that the Laplace domain method of solving dynamic model of a process is the approximation method of the models solved using SIMULINK. Laplace domain models represent best the non-linear models, at a very narrow point of linearization. Large changes in the operating conditions for a dynamically non-linear. For control part, we use the method of Frequency analysis responses in order to determine the effect of using with or without controller. As we can see from the figure 4 and 5, it is clear that fizzy controller helps in improving the responses of the system. By fine-tuning of the controller parameters, it is shown that the gas absorber system has always a guaranteed stability.

Numerical simulation of a six-order gas absorber is carried out and the obtained results show clearly that the proposed estimator-fuzzy controller scheme yields excellent performance.

In conclusion, the main objective of this work that to develop the steady state and dynamic models for the control process dynamic problem has been successfully developed and derived by MATHLAB and SIMULINK flow sheeting software using the standards of SFunctions. The assumptions that have been made helped to simplify the derivation of the tray gas adsorption dynamic model. By using these assumptions, the dynamic mathematical model for the tray gas adsorption column is derived using unsteady state total material and component balances on the tray inside the column. 6.1 Recommendations Recommendations for further improvement of the operation of the absorption column include: Connect a de-ionized water feed to the still pot for easy charging. Currently, the water must be charged by hand Insulate top stage to improve the accuracy of the TD reading. Use sensitive measurement method to detect low concentrations in samples Upgrade column with real-time computer-monitoring and control systems Quantify the effect of heat loss on the internal reflux ratio stage by stage Develop a more accurate model to predict the performance of the separation.

7.0 CONCLUSION In conclusion, the main objective of this work that to develop the steady state and dynamic models for the control process dynamic problem has been successfully developed and derived by MATHLAB and SIMULINK flow sheeting software using the standards of S-Functions. The combination of MATHLAB and SIMULINK is a great combination to simulate steady state, opened, and closed loop dynamics of an adsorption column. This condition helped to

develop a control block diagram including measuring device, feedback controllers and final control element that led the achievement of the main objective of this work. Besides that, the assumptions that have been made also gave a big contribution for the success of the development of the model. The assumptions that have been made helped to simplify the derivation of the tray gas adsorption dynamic model. By using these assumptions, the dynamic mathematical model for the tray gas adsorption column is derived using unsteady state total material and component balances on the jth tray inside the column.

7.0 REFERENCES C. C. Lee, Fuzzy Logic in Control Systems: Fuzzy Logic Controller, Parts I& II", IEEE Systems, Man and Cybernetics, Vol. 20, No.2, March/April 1990, pp.404-435.

M. S. Mahmoud, S.Z. Eid and A .A. Abou-Elseoud, "A Real Time Expert System for Dynamical Processes", IEEE Transactions Systems, Man and Cybernetics, Vol. SMC-19, No. 5, September/October 1989, pp. 1101-1105.

M. S. Mahmoud, S. Kotob, and A. A. Abou-Elseoud, "A Learning Rule- Based Control System", Journal of Information and Decision Technologies, Vol. 18, No. 1, January 1992, pp. 55-66.