Each of these packages has its advantages and disadvantages in the process modelling and simulation. The primary objective of this project is to combine the special features of each software package, for example, co-simulating a distillation column using Aspen Plus, Aspen Dynamics, together with MATLAB Simulink toolbox.
To achieve this target, the following work has been covered throughout the project and is presented in the thesis. The comparison of different software packages used in modelling and simulation are presented. The scope and aim of the project are defined. The system will be tested in open loop and closed loop. The purity of distillate is controlled to achieve the quality of the product as required. The process is tested by the same method as in Section 4.
This section contains all necessary background and related information of the simulation; it covers an overview of the project, software, distillation column and Solvent entractant used in these simulations. Aspen Plus is a steady state simulation to obtain modelling of the extractive column, as the aim is to express dynamic processes according to the laws of conservation of mass and energy.
Therefore, it is very helpful especially in understanding the process behaviour and control design. Dynamic simulation is performed by Aspen Dynamics to understand system dynamics. Lastly, co-simulate the process of the distillation column is conducted using Aspen Dynamics and Simulink. Usually in the industry, extractive distillation process is a combination of some other processes, in which it can be divided into two parts refer to Figure 1 ; part one is the extractive column and part two is solvent recovery column.
For the extractive column part one , it is used to separate substance from composition mixture of the solvent that is hard to be separated by conventional distillation, producing a distillate product comprising of the substances with desired purity. This is done by using a third component entractant to give effect to the separation of the substances. The component use must non-volatile, higher boiling point, and miscible with mixtures, but it does not form an azeotrope in the mixture [1].
The difference in the interaction of the third component and the mixture causes a change in relative volatility. This allows the new mixture of components and solvent to be separated. The component with the highest volatility will separate as the main product as the top stream [2]. For the solvent recovery column part two , it is used to separate the low volatility solvent from the extractive column, with this the entrainer is circulating round and round.
Phenol is extracting Toluene from the mixture, and these two components go down to the bottom of the distillation column because they are heavy elements. Meanwhile, MCH is going to the top of distillation column as it is a light component [1]. In the solvent recovery column, pure Toluene is obtained at the top and Phenol entractant at the bottom stream, which is recycled to be fed to the extraction column as the entrainer.
It essentially charges the system with the amount of entrainer and that entrainer is recirculated in the system. This means under ideal conditions, none of the entrainer is lost. Figure 1 shows the system of the extractive column and solvent recovery column. Figure 1: Extractive Column and Solvent Recovery Column [3] In this case study only the extractive column is modelled, simulated and controlled.
While, the solvent recovery is left for future students. The extractive column will be tested starting with steady state process until dynamic process. Aspen Plus is used to perform the steady state process while Aspen Dynamics and Simulink are used to run the dynamic process. Results of both software were compared to determine which software is better for controlling the process.
Distillation is used to separate a mixture into one or more individual materials by using a heating medium. Therefore, the producing product will contain the desired purity by controlling the condenser and reboiler [5]. There are various types of advanced distillation techniques in the industry, such as Vacuum, Cryogenic, Reactive, Extractive, and Pressure Swing.
The extractive distillation technique has been used in this thesis to separate mixtures of the solvent that are hard to separate by conventional distillation. The third component solvent or entrainer used to give effect to the separation of the mixtures. This section will discuss the software that has been used to perform simulations for this project.
Aspen Plus is used for steady state simulation, in which it is used to identify the initial conditions for the dynamics simulation, determine material and energy balances, and conceptual design. This should be done before running in dynamics process. The steady state model is imported to Aspen Dynamics from Aspen Plus to allow the process runs in the dynamic process.
This is important to ensure that the Aspen Plus gets the correct steady state value before proceeding to the following steps. This means that if the value is false, the result of the process will be incorrect and will affect the whole process.
Unfortunately, Aspen Plus can only perform steady state simulation and requires other software to perform the Dynamic simulation [6]. The way DSTWU operates is by estimating the minimum number stages of the distillation column and the minimum value of reflux ratio. After that, it will calculate the required reflux ratio based on user input. At the same time, Aspen plus will estimate the best feed stages location, the condenser, and the reboiler.
All the results can be seen when Aspen plus is done with computation [7]. Distl Distillation column type Distl is designed for single feed process. Distl operates by using Edmister approach; it will calculate the product composition of the process.
Users need to enter the number of column specifications [7]. RadFrac Distillation column type RadFrac is designed for multiple feed process. This eliminated the need for the pumps, reflux tanks, heat streams, and heat exchangers [7]. Aspen Dynamics is firmly integrated with Aspen Plus, where the Aspen Plus is a simulator for steady state.
Then the model in the Aspen Plus is imported into Aspen Dynamics to run the dynamics simulation, and this allows the existing steady state in Aspen Plus simulation to create a dynamics simulation [1].
This software can identify the dynamic of the process, implement and control loop tuning controller. However, the control scheme in Aspen Dynamics is limited to conventional controllers [8]. Simulink can develop conventional and advanced control scheme, and it also can perform real-time simulation [9]. The real time simulation can be done when a physical device replaces the virtual device. Thus, costs are reduced when a replacement is carried out, and the quality of the physical system can be improved.
Besides, the simulation can be tested without having prototypes and tests can be conducted continuously. Realistic simulation means inputs and outputs in a virtual world simulation must be updated simultaneously with the real world. Therefore, real-time simulation is required to ensure that realistic simulation occurs. Simulink also has a special feature called control system toolbox and offers industry-standard algorithms in designing, tuning, and analysing the control systems.
Control System Toolbox provides facilities to examine the model [10]. Thus, the performance parameters can be checked, such as settling time and maximum overshoot. Aspen Dynamics and Simulink will be implemented from the same type of controller scheme and parameters values. Therefore, the results can be compared to verify the results of which are better for this process control strategy [11]. It is the process by which a third component or solvents are used to give effect to the separation of the chemical.
The third component will act to generate or increase volatility differences between components to be separated. The third component and the less volatile component will flow downstream of the distillation column so that the subsequent distillation process can recover the extracted components. On the other hand, the non-extracted component will be distilled at the top of the distillation column [2]. This case study is only involved with the extractive distillation column, the process modelling, and the steady state simulation in Aspen Plus, importing the steady state model from Aspen Plus to Aspen Dynamics, and the designing and testing the controllers in Aspen Dynamics and MATLAB Simulink toolbox.
The aim is to revise an example that has been done in with the new software versions and new operating system. Sensitivity analysis is conducted in Aspen Plus to find the mass flowrate of the phenol entering the distillation column so that the purity of Methyl Cyclo Hexane MCH leaving the distillation column can achieve at least 0.
Existing controllers are tuned in Aspen Dynamics to get the appropriate parameters. Then set point changes are made in each controller while disturbance changes are introduced in certain variables to examine the impact and response onto the purity of the product. Finally, a controller is setup to control the flowrate of phenol. The same method is used in co-simulation to test the purity of the product.
However, working with another different column is not in the scope of this project. This is because there was an error to link Aspen Dynamics version 8. PI controllers is used in Aspen Dynamics and Simulink to control the feed tank level, reboiler level, reflux drum level, and top stream pressure for an extractive distillation and a new controller in the phenol flowrate entractant. The development of a project plan is furthered with efforts focusing on some areas to attain the set goals.
Following to the very sophisticated software involved, it is important to understand every software used. The crucial purpose of choosing to explore in depth is due to the limitation of understanding in using the software. This is because of its ability to solve problems involving many calculations, where most of the equations used are very complicated. It is difficult and almost impossible to solve them by hand due to human errors and time constraints.
This software is often used in industrial oil production, refining, and environmental studies. With practical operation and reliable model, it allows process and control engineers to simulate process like an actual plant. Each process has its process model, and thus Aspen Plus is used for the process model. There are three steps to follow to obtain the process model; flowsheet, specifying the chemical components and operating conditions. Aspen Plus acts in regard to all specifications and simulations involved in different processes.
Also, it predicts the behaviour and calculates the results of the system. Aspen Plus will list the results for each of the streams and the unit when the calculation is complete [13].
Flowsheet The process flowsheet model will reflect the entire system. The flowsheet shows the inlet streams entering into the unit operation distillation column, reboiler, and heat exchanger and outlet streams from the unit operation, thus all inlet streams and outlet streams can be identified [13].
Chemical Components The components used in the Aspen should be defined before moving to the next step. Each of the components used must be explained in detail to facilitate simulation [13]. Operating Conditions Normally, all the operating units have specific operating conditions such as temperature, and pressure. It is determined according to the operating conditions of the process [13].
These elements must be defined in the Aspen Plus to allow the components used. Aspen Plus has an extensive database of components used, including their physical characteristics. Therefore, it is capable of detecting the materials used and filling the required space automatically. The detailed information about feed stream and product stream can be seen in Table Feed stream and Table product stream respectively, in Appendix A — Results.
Details of information were collected as shown in Table 2 below. The total number of theory trays of the distillation column T-1 is 22, and reflux ratio is 8 with 16 psia and Nevertheless, the selection was studied to understand the reason behind of the selection and for the purpose of learning.
Each of them has different uses and capabilities. Moreover it is more rigorous than DSTWU columns and suitable for extractive distillation for allowing multiple product and feed streams. In the distillation column, the liquid and vapor compositions will therefore remain same. This means no matter how tall the tower is, or no matter how much reflux is used even if the column is operating at the total reflux, the best it can do is getting a distillate that is close to the azeotropic composition.
However, it simply cannot be produced using standard distillation because of this azeotrope, so no matter how much it is boiled; the vapor is not any richer in the light component. Therefore, no further separation is possible, no matter how tall the column is. There are various processing techniques to alter the vapor-liquid equilibrium and the system such that pure Methyl Cyclo Hexane MCH and pure Toluene can be obtained. There are two common ways; one is referred to as homogeneous extractive distillation, and the other is heterogeneous azeotropic distillation.
The former is a type of stage separation in which the mixture is not separated into different stages. However, the latter splits the mixture into aqueous and organic stages.
The mixture in the aqueous stage is richer in water, and the mixture is richer in organic for the organic stage. Hence, in heterogeneous azeotropic distillation - by adding an entrainer or a solvent - it will mostly cause a phase split, whereby the liquid separates into two stages which are water rich and organic rich, and because of this stage separation it crosses the azeotropic composition.
The idea behind homogeneous extractive distillation is to add an entrainer, which is heavy. This means Phenol is an attractor or also known as the stable node, in which it attracts all residue curves towards itself. The azeotropic composition at temperature These are called saddles depending on what the initial condition is. For a better understanding, Table 3 below shows the classification and temperature of each component [17].
Toluene Phenol MCH Therefore, to sufficiently break the azeotrope, it is necessary to have the heavy entrainer on almost all trays which require the heavy solvent Phenol fed some place near the top of the column above the feed. However, if the entrainer is mixed with the feed, the separation becomes infeasible. The full results can be seen in Table 12 in Appendix A — Results section. In this case study, the sensitivity analysis was conducted to determine the flow rate of phenol entering distillation column so that the purity of MCH can reach the best purity.
Two important steps need to be considered to perform sensitivity analysis: 1. State manipulated variable In this case study, Phenol flowrate is the manipulated variable. This step is performed so that Phenol flowrate can be varied. This is done so that the phenol flow rate can be specified either as equidistant points within an interval.
After the results were satisfied, then they will be exported to Aspen Dynamics to be tested in dynamic simulation. Aspen Dynamics is firmly integrated with Aspen Plus, whereby it is a simulator for steady state. This allows the existing steady state from Aspen Plus simulation to create a dynamic simulation [8]. Process dynamics means the situation is changing, in other words, the process changes over time.
Specifically, what it does mean is when the input of the process is changing, how the output variable would respond over the time. Mostly process dynamic deals with the systematic characterization of the time response of the affected variable to a change in the causal variable, the affected variable is also sometimes referred to as the output variable, and the causal variable is also usually referred to as the input variable [8].
Aspen Dynamics allows users to obtain a comprehensive understanding of the dynamics of processes. This knowledge could be exploited by users to design and operate with optimum safety achieving consistent product quality and operability of the process. A linear state space model can be extracted from Aspen dynamics using the control design interface in Aspen [18].
Before starting to simulate the dynamic process in Aspen Dynamics software, a steady state simulation must be done in advance in the Aspen Plus software. When the steady-state simulation was completed in the Aspen Plus, all the necessary results can be obtained, and tabulation graphs can be carried out where it will show behaviour corresponding to particular inputs.
This information is then exported into Aspen Dynamics. Sizing of the equipment such as column diameter, size of vessels, tray spacing, trays active area, weir length and height, reflux drum length and height, and reboiler length and height is the information needed for Aspen Dynamics. A tool called tray sizing provided by the Aspen Dynamics can be used to calculate the tray sizes based on the flow conditions in the column, but the sizing can also be done in Aspen Plus [19].
The process will then be tested in open loop and closed loop system. This is done to study the differences and implications between these two types of system in the process. Set point change will be made in each controller, and disturbance change will be introduced to individual variables to examine the impact of changes on the purity of the product.
Then, a new composition controller will be developed to control the flowrate of phenol as it affects the purity. This means the process information was visually monitored, and valve positions and pumps speeds were manually adjusted accordingly. Controller faceplate is a special tool provided by Aspen Dynamics, it is used to examine and monitored all the features of the controller.
Manual mode must be changed at controller faceplates as shown in Figure 4 to control in manual mode and Figure 5 below shows the Aspen Dynamics simulation was based on the open loop flowsheet. The results and impact of the open loop system against product purity was discussed in Section 6. The controller will react when the system senses the change in value of process variable, then loop back to the controller and compare with the reference value of the system.
If there is a difference, it will adjust the system to its reference value. For this case study, Aspen has developed four types of controllers to control the feed tank level, reboiler level, reflux drum level, and top stream pressure for an extractive distillation.
However, from the results obtained in Aspen Plus shows that phenol flow rate must be controlled to achieve a better purity. Therefore, a new controller was built to control the phenol entractant flowrate. Methyl Cyclo Hexane composition controller will manipulate phenol flowrate, and the purity product MCH will be the process variable. Table 5 shows the control strategies used in this project.
Therefore, all PI controllers were tuned individually using the 'Tuning' tool [14]. The 'Tuning' is a special tool available in Aspen Dynamics to perform auto tuning. After 10 seconds, the 'Calculate' should be pressed to the 'Tuning' start calculating. The 'Tuning' will provide the optimal parameter values to be able to respond the best control.
After that, the button 'Update Controller' must be pressed as shown in Figure 9 so the parameters can be updated into the controller. From Figure 9, the 'Tuning Rule' is an option so that users can choose which tuning method to use. Ziegler Nichols and Tyreus Luyben tunings are of popular methods. Ziegler Nichols tuning is quite aggressive compared to Tyreus Luyben tuning that is quite loose. Tyreus Luyben is typically used for distillation column to avoid the colossal, sudden increments or aggressive changes in the process.
For example, the reboiler duty could lead to hydraulic problems if the combative changes occur on the reboiler. So, Tyreus Luyben is a more conservative tuning method compared with Ziegler Nichols in this kind of process [20]. All controllers have been tuned using the same steps. Table 6 shows the parameter results obtained from the tuning. Through set point tracking, the performance of the controller being used can be tested, whether it is too aggressive or lagging when a set point change is made.
Besides, set point change is also used to study its impact of set point change towards the product purity. The testing of the controllers was done by performing steps on set points individually. The results of set point changes are discussed in Section 6. However, not all results have been presented, but only the results with visible change had discussed. However, the disturbance will only be introduced to individual variables, where it will have a significant impact on other variables.
Therefore, the phenol flowrate and coolant flowrate to condenser have been selected as the disturbance. Thus, the feed level, pressure, reflux level and reboiler level will be controlled by the controller when the disturbance is introduced in phenol flowrate, while feed level, reflux level, and reboiler level will be monitored by the controller when the disturbance change is made in the coolant flowrate. This allows control engineers to design, develop and to test the controllers for complex chemical processes.
The controller is developed in Simulink to control the flowsheet in Aspen Dynamics. The system in co-simulation is tested by the open loop and closed loop system, as was done in Aspen Dynamics. However, the controller in co-simulation is not tuned, but it uses the same parameters as in Aspen Dynamics.
This means that the reaction in Aspen Dynamics and co-simulation must have the same response and if the response is different further analysis will be done. The steps of implementing the test are exactly the same as in Aspen Dynamics. After that, step on set points and disturbances are introduced to the controllers and variables respectively to identify the impact on product purity, the same method as in Aspen Dynamics has been used. As the co-simulation can implement advanced controller, the Model predictive controller MPC was constructed to replace the conventional controller.
However, it has been studied to understand and learn how to implement it. Input port represents information sent from Simulink to the AMS, where only variables with fixed specification in the AMS can be used for inputs. These variables would represent those typically manipulated by a controller. While, output ports represents information received from the AMS by Simulink, where only variables with free, initial and rate Initial specifications in the Aspen Modeler simulation can be used for outputs.
These variables would represent those typically measured by a controller [21]. Figure 10 shows the AMS block which has been configured in this case study.
The development of MPC techniques is established by utilizing the discrete-time convolution model. The most comprehensive techniques for model predictive control are those based on the objective function optimization which encompasses the error between the set point and predicted outputs. This is because there was an error to link Aspen Dynamics version 8. PI controllers is used in Aspen Dynamics and Simulink to control the feed tank level, reboiler level, reflux drum level, and top stream pressure for an extractive distillation and a new controller in the phenol flowrate entractant.
The development of a project plan is furthered with efforts focusing on some areas to attain the set goals. Following to the very sophisticated software involved, it is important to understand every software used. The crucial purpose of choosing to explore in depth is due to the limitation of understanding in using the software. This is because of its ability to solve problems involving many calculations, where most of the equations used are very complicated.
It is difficult and almost impossible to solve them by hand due to human errors and time constraints. This software is often used in industrial oil production, refining, and environmental studies. With practical operation and reliable model, it allows process and control engineers to simulate process like an actual plant. Each process has its process model, and thus Aspen Plus is used for the process model. There are three steps to follow to obtain the process model; flowsheet, specifying the chemical components and operating conditions.
Aspen Plus acts in regard to all specifications and simulations involved in different processes. Also, it predicts the behaviour and calculates the results of the system. Aspen Plus will list the results for each of the streams and the unit when the calculation is complete [13]. Flowsheet The process flowsheet model will reflect the entire system. The flowsheet shows the inlet streams entering into the unit operation distillation column, reboiler, and heat exchanger and outlet streams from the unit operation, thus all inlet streams and outlet streams can be identified [13].
Chemical Components The components used in the Aspen should be defined before moving to the next step. Each of the components used must be explained in detail to facilitate simulation [13].
Operating Conditions Normally, all the operating units have specific operating conditions such as temperature, and pressure. It is determined according to the operating conditions of the process [13]. These elements must be defined in the Aspen Plus to allow the components used. Aspen Plus has an extensive database of components used, including their physical characteristics. Therefore, it is capable of detecting the materials used and filling the required space automatically.
The detailed information about feed stream and product stream can be seen in Table Feed stream and Table product stream respectively, in Appendix A — Results.
Details of information were collected as shown in Table 2 below. The total number of theory trays of the distillation column T-1 is 22, and reflux ratio is 8 with 16 psia and Nevertheless, the selection was studied to understand the reason behind of the selection and for the purpose of learning. Each of them has different uses and capabilities. Moreover it is more rigorous than DSTWU columns and suitable for extractive distillation for allowing multiple product and feed streams.
In the distillation column, the liquid and vapor compositions will therefore remain same. This means no matter how tall the tower is, or no matter how much reflux is used even if the column is operating at the total reflux, the best it can do is getting a distillate that is close to the azeotropic composition. However, it simply cannot be produced using standard distillation because of this azeotrope, so no matter how much it is boiled; the vapor is not any richer in the light component.
Therefore, no further separation is possible, no matter how tall the column is. There are various processing techniques to alter the vapor-liquid equilibrium and the system such that pure Methyl Cyclo Hexane MCH and pure Toluene can be obtained. There are two common ways; one is referred to as homogeneous extractive distillation, and the other is heterogeneous azeotropic distillation. The former is a type of stage separation in which the mixture is not separated into different stages.
However, the latter splits the mixture into aqueous and organic stages. The mixture in the aqueous stage is richer in water, and the mixture is richer in organic for the organic stage.
Hence, in heterogeneous azeotropic distillation - by adding an entrainer or a solvent - it will mostly cause a phase split, whereby the liquid separates into two stages which are water rich and organic rich, and because of this stage separation it crosses the azeotropic composition. The idea behind homogeneous extractive distillation is to add an entrainer, which is heavy.
This means Phenol is an attractor or also known as the stable node, in which it attracts all residue curves towards itself. The azeotropic composition at temperature These are called saddles depending on what the initial condition is. For a better understanding, Table 3 below shows the classification and temperature of each component [17].
Toluene Phenol MCH Therefore, to sufficiently break the azeotrope, it is necessary to have the heavy entrainer on almost all trays which require the heavy solvent Phenol fed some place near the top of the column above the feed. However, if the entrainer is mixed with the feed, the separation becomes infeasible. The full results can be seen in Table 12 in Appendix A — Results section. In this case study, the sensitivity analysis was conducted to determine the flow rate of phenol entering distillation column so that the purity of MCH can reach the best purity.
Two important steps need to be considered to perform sensitivity analysis: 1. State manipulated variable In this case study, Phenol flowrate is the manipulated variable. This step is performed so that Phenol flowrate can be varied. This is done so that the phenol flow rate can be specified either as equidistant points within an interval.
After the results were satisfied, then they will be exported to Aspen Dynamics to be tested in dynamic simulation. Aspen Dynamics is firmly integrated with Aspen Plus, whereby it is a simulator for steady state.
This allows the existing steady state from Aspen Plus simulation to create a dynamic simulation [8]. Process dynamics means the situation is changing, in other words, the process changes over time. Specifically, what it does mean is when the input of the process is changing, how the output variable would respond over the time. Mostly process dynamic deals with the systematic characterization of the time response of the affected variable to a change in the causal variable, the affected variable is also sometimes referred to as the output variable, and the causal variable is also usually referred to as the input variable [8].
Aspen Dynamics allows users to obtain a comprehensive understanding of the dynamics of processes. This knowledge could be exploited by users to design and operate with optimum safety achieving consistent product quality and operability of the process. A linear state space model can be extracted from Aspen dynamics using the control design interface in Aspen [18].
Before starting to simulate the dynamic process in Aspen Dynamics software, a steady state simulation must be done in advance in the Aspen Plus software. When the steady-state simulation was completed in the Aspen Plus, all the necessary results can be obtained, and tabulation graphs can be carried out where it will show behaviour corresponding to particular inputs.
This information is then exported into Aspen Dynamics. Sizing of the equipment such as column diameter, size of vessels, tray spacing, trays active area, weir length and height, reflux drum length and height, and reboiler length and height is the information needed for Aspen Dynamics. A tool called tray sizing provided by the Aspen Dynamics can be used to calculate the tray sizes based on the flow conditions in the column, but the sizing can also be done in Aspen Plus [19].
The process will then be tested in open loop and closed loop system. This is done to study the differences and implications between these two types of system in the process.
Set point change will be made in each controller, and disturbance change will be introduced to individual variables to examine the impact of changes on the purity of the product. Then, a new composition controller will be developed to control the flowrate of phenol as it affects the purity.
This means the process information was visually monitored, and valve positions and pumps speeds were manually adjusted accordingly. Controller faceplate is a special tool provided by Aspen Dynamics, it is used to examine and monitored all the features of the controller. Manual mode must be changed at controller faceplates as shown in Figure 4 to control in manual mode and Figure 5 below shows the Aspen Dynamics simulation was based on the open loop flowsheet. The results and impact of the open loop system against product purity was discussed in Section 6.
The controller will react when the system senses the change in value of process variable, then loop back to the controller and compare with the reference value of the system. If there is a difference, it will adjust the system to its reference value. For this case study, Aspen has developed four types of controllers to control the feed tank level, reboiler level, reflux drum level, and top stream pressure for an extractive distillation. However, from the results obtained in Aspen Plus shows that phenol flow rate must be controlled to achieve a better purity.
Therefore, a new controller was built to control the phenol entractant flowrate. Methyl Cyclo Hexane composition controller will manipulate phenol flowrate, and the purity product MCH will be the process variable. Table 5 shows the control strategies used in this project.
Therefore, all PI controllers were tuned individually using the 'Tuning' tool [14]. The 'Tuning' is a special tool available in Aspen Dynamics to perform auto tuning. After 10 seconds, the 'Calculate' should be pressed to the 'Tuning' start calculating. The 'Tuning' will provide the optimal parameter values to be able to respond the best control. After that, the button 'Update Controller' must be pressed as shown in Figure 9 so the parameters can be updated into the controller.
From Figure 9, the 'Tuning Rule' is an option so that users can choose which tuning method to use. Ziegler Nichols and Tyreus Luyben tunings are of popular methods. Ziegler Nichols tuning is quite aggressive compared to Tyreus Luyben tuning that is quite loose. Tyreus Luyben is typically used for distillation column to avoid the colossal, sudden increments or aggressive changes in the process.
For example, the reboiler duty could lead to hydraulic problems if the combative changes occur on the reboiler. So, Tyreus Luyben is a more conservative tuning method compared with Ziegler Nichols in this kind of process [20]. All controllers have been tuned using the same steps. Table 6 shows the parameter results obtained from the tuning. Through set point tracking, the performance of the controller being used can be tested, whether it is too aggressive or lagging when a set point change is made.
Besides, set point change is also used to study its impact of set point change towards the product purity. The testing of the controllers was done by performing steps on set points individually.
The results of set point changes are discussed in Section 6. However, not all results have been presented, but only the results with visible change had discussed. However, the disturbance will only be introduced to individual variables, where it will have a significant impact on other variables. Therefore, the phenol flowrate and coolant flowrate to condenser have been selected as the disturbance. Thus, the feed level, pressure, reflux level and reboiler level will be controlled by the controller when the disturbance is introduced in phenol flowrate, while feed level, reflux level, and reboiler level will be monitored by the controller when the disturbance change is made in the coolant flowrate.
This allows control engineers to design, develop and to test the controllers for complex chemical processes. The controller is developed in Simulink to control the flowsheet in Aspen Dynamics. The system in co-simulation is tested by the open loop and closed loop system, as was done in Aspen Dynamics.
However, the controller in co-simulation is not tuned, but it uses the same parameters as in Aspen Dynamics. This means that the reaction in Aspen Dynamics and co-simulation must have the same response and if the response is different further analysis will be done.
The steps of implementing the test are exactly the same as in Aspen Dynamics. After that, step on set points and disturbances are introduced to the controllers and variables respectively to identify the impact on product purity, the same method as in Aspen Dynamics has been used. As the co-simulation can implement advanced controller, the Model predictive controller MPC was constructed to replace the conventional controller.
However, it has been studied to understand and learn how to implement it. Input port represents information sent from Simulink to the AMS, where only variables with fixed specification in the AMS can be used for inputs. These variables would represent those typically manipulated by a controller. While, output ports represents information received from the AMS by Simulink, where only variables with free, initial and rate Initial specifications in the Aspen Modeler simulation can be used for outputs.
These variables would represent those typically measured by a controller [21]. Figure 10 shows the AMS block which has been configured in this case study.
The development of MPC techniques is established by utilizing the discrete-time convolution model. The most comprehensive techniques for model predictive control are those based on the objective function optimization which encompasses the error between the set point and predicted outputs. The DMC has a number of design parameters that can be manipulated to achieve the desired response: the model horizon, sampling period, control horizon, prediction horizon, and two weighting matrices for predicted errors and control moves respectively.
Model horizon along with sampling period are both defined first, as they are needed to obtain step response data. A general rule is that the model horizon multiplied by the sampling period should exceed the time taken for the process to be ninety nine percent complete. The model horizon needs to be large enough so that it captures enough data on the dynamics of the system [24]. Increasing the phenol entractant flowrate will increase MCH concentration in the distillate as shown in Table 7.
However, after the sensitivity analysis it was possible to increase the MCH purity in the distillate to 0. It can be clearly seen from the figure that increasing phenol flowrate will increase the distillate purity.
This means that if it is necessary to develop another control loop to control the distillate purity by manipulating the phenol flowrate can be selected as MV, which confirms the control system proposed in Table 5 on page Table 8 shows the results of the purity product was obtained when the step was introduced in manipulated variable.
The purity of the product was at the lowest level of 0. Most of the product purity above than the minimum purity 0. Therefore, the closed loop system is developed to maintain product quality. The testing of the controllers was done by performing step on set points individually. All tests were run with the same parameters as shown in Table 6 in Section 4.
The period of the simulation was kept constant for all controller performance tests; units. In the other hand, the purposes of set point changes were implemented are to investigate the effect of set point changes in each variable towards distillate and purity of the product. For the full results of the set point changes, please refer to Appendix C. The minimum requirements to achieve excellent purity must be at least 0. From the tests that have been carried out, the results obtained in Aspen Dynamics stand-alone and co-simulation is the same.
The results have been illustrated in Figure 12 a b , and it is clearly shows the purity of distillate is at the minimum requirement with 0. Besides, the change in feed level has affected the pressure, reflux drum and reboiler level but the controller managed to control them back to the desired set point. This shows that the link is successfully made to connect the co-simulation. Furthermore, the performances of the controllers in both strategies are at satisfactory levels.
The method for performing disturbance changes was slightly different than the set point tracking, where Phenol flowrate and coolant flowrate to the condenser were set as a disturbance in order to investigate the effect of disturbances toward purity of the product. The feed level, pressure, reflux level and reboiler level have been set to automatic mode when the disturbance in phenol flow rate was introduced.
The feed level, reflux level, and reboiler level have been set to automatic mode when the disturbance in coolant flowrate to the condenser was introduced.
Also, the effects of disturbance toward the distillate and the purity product were considered and analyzed. The minimum level of purity is 0. This has affected the reboiler level because the flow has been affected by the feed stream which remains at the initial steady state and disturbance changes in phenol flowrate. Due to the change in the reboiler level, then it causes changes in pressure, reflux level, distillate and purity of product. Observe Figure 14 a b where the controllers managed to eliminate the disturbance to achieve the desired set point.
Besides, although the purity of phenol was affected by the changes in the flowrates at 0 s to 20 s , but the purity of product was below the minimum condition with result of 0.
Overall disturbance rejection is as expected; the purity and amount of distillate are at the effective level. The controllers on all variables managed to control and eliminate the disturbance when it was introduced in the phenol flowrates. As can be seen in Figure 15 a b , the purity of the product 0. The result was expected because as discussed in Section 6. Therefore, the purity of distillate is increase if the Phenol flow rate increased as shown in Figure 14 a b , and the purity will decrease if the phenol flow rate is decreased as illustrated in Figure 15 a b.
Therefore, the changes in phenol flowrate must be considered in order not to affect the purity. Lastly, the feed flowrate was not affected by the disturbance change in the phenol flowrate nor water input flowrate at the pressure.
This would inadvertently affect the reflux level and reboiler level. The response in reflux and reboiler did overshoot at the beginning then they started to reach the desired set point after 10 s. From the results obtained in Figure 16 a b , the purity of the product decreased to 0.
The reflux was changed to get the appropriate temperature at the top tray to get the absolute purity of product. Some of the condensate in reflux drum is recycled back into the column, while the remaining is discharged as top stream product. The purity of the product was reduced from 0. However, the purity is still considered good because it is above the minimum level of purity. From the results obtained in Figure 16 a b and Figure 17 a b , it can be seen that the purity of the product produced a better response when the step down of disturbance was introduced compared to the step up.
However, the flowrate of distillate product has decreased from 2. The result was as expected because as discussed earlier, the output stream of reflux drum is divided into two streams.
The first stream is withdrawn as a final product, and another stream is recycled back to the distillation column. Most of the condensate in the reflux drum was frequently recycled for multiples times to get a better purity before the condensate is withdrawn as a final product. To sum up, the results of disturbance rejection is observed the same as in Aspen Dynamics stand-stand alone and co-simulation. The advantages of co-simulation are able to develop advanced controller strategy such as Dynamic Model Control DMC , and perform real time simulation.
Besides, the results of the disturbance change of phenol flowrate in Section 6. Therefore, a lot of time was spent in constructing phenol flow rate controllers in Aspen Dynamics and co-simulation. The process of identifying problem and solution resulted in the redesign of the controller code and configuration.
After several attempts, the closed loop system has been successfully developed. From Figure 18 it can be seen the product purity achieved the desired set point and yielded good results. However, the results of phenol flowrates to achieve the desired purity in Table 7 in Section 6.
The comparisons between these results are as shown in Table 9. However, Aspen Dynamics and co-simulation only requires Out of nine status cases, only the first status cases have the same phenol flowrates to achieve 0. At first, it was expected to get the same flowrate for all product purity. As the results arisen with some problem, some assumption had been made. The first assumption was: probably a mistake occurred while performing some changes during the redesign and configuration of the controller code.
The second assumption: maybe the dead time need to be installed on composition controller, this is to avoid the limitation of the doable response in the system. However, a full analysis was unable to be performed due to time constraints. However, after repeatedly trying to develop DMC ended unsuccessfully because of the system was showing a "linearization" error as illustrated in Figure Further analysis was not done because of time constraints and will be recommended for future work.
The aims will be reviewed again, and achievements will be presented. Studying and understanding the Aspen Plus, Aspen Dynamics and Co-Simulation has been done since the beginning of this project, as they are sophisticated software. The results indicate the software used are very helpful especially in developing and testing process simulation by providing a comprehensive system. Aspen Plus was used to obtain the results of the process model as it can predict the behaviour and calculate the steady state value.
The first task is to conduct sensitivity analysis in Aspen Plus. Steady state simulation has given good results. The results show the flowrate of phenol was substantially affecting the composition of the product. Increasing the phenol flowrate will increase composition of the product. Therefore, the new controller has been developed to control the product composition. It is possible for a complex system such as extractive distillation to perform dynamics simulation in Aspen Dynamics.
0コメント