Exothermic Reactor Unit (ERU) Operating Manual

Transcription

1 Exothermic Reactor Unit (ERU) Operating Manual 1/13 11/21/2017

2 1. Background Complex industrial processes must be controlled well to ensure an economic return on investment, the safety of operating personnel and the local community, and to preserve the environment. Though steady state operations may often be the ideal condition for such processes, the behavior of industrial processes is often not static but very dynamic, requiring control systems to ensure safe and profitable operations. Even what appear to be simple processes can present difficult control challenges. The exothermic reactor process is one such seemingly simple process. 2. Description of Equipment A schematic of the test facility is shown in Figure 1. Figure 1 Exothermic Reactor Apparatus (as shown with Honeywell Station software) In this simulated constant-volume reactor, both the reactor and the cooling jacket are perfectly mixed, the product and feed have the same constant density and heat capacity, the cooling water density and heat capacity are constant, and the following irreversible firstorder exothermic reaction occurs: A B 2/13 11/21/2017

3 Automatic control valves regulate the flow rates of the feed and cooling water streams. The feed valve is air-to-open; the cooling water valve, air-to-close. Flow rates are measured by differential pressure (orifice) flow rate transmitters, temperatures by thermocouples, and the concentration of component A in the product stream by an analyzer. The experiment is interfaced to Honeywell s Experion PKS (Process Knowledge System) distributed control system (see below). The system receives process measurements and makes adjustments to the control valve positions to control the feed and cooling water flow rates, which then control the reactor and cooling jacket temperatures and product concentration. The experiment utilizes cascade control structures commonly used in the process industries. Low level regulatory loops (secondary or slave controllers) are used to control the feed and cooling water flow rates. A higher level (primary or master) controller maintains the product composition by manipulating the setpoint of the feed flow rate controller. The other cascade structure has three levels: the highest (primary) level maintains reactor temperature by manipulating the setpoint of the cooling jacket temperature controller (the upper secondary), which maintains cooling jacket temperature by manipulating the setpoint of the cooling water flow rate controller (the lower secondary). What is not apparent upon a superficial view of this simulated reactor system is that the simulation effort is split between the Honeywell Experion PKS and compiled C++ code. The C++ program solves all the material and energy balance equations required for the modeling of the reactor system but the process and valve dynamics for the cooling water and feed flows are modeled in the Honeywell itself. The compiled code executes in nearreal-time (once per second) to retrieve flow information from the Honeywell side of the simulation and return state variable values to the Honeywell from the C++ side of the simulation. This split responsibility is not unusual in the industrial simulation world. 3. Using the Honeywell Experion PKS Control System Honeywell s Experion is a distributed control system (DCS) widely used in industry and on many lab experiments here at LSU. In Experion, each process variable is represented by an entity called a Control Module (CM). Each CM is a collection of Function Blocks (FB). And each FB consists of many values called parameters. Within a CM (and sometimes between CM s), the FB s are wired together in various ways to monitor and control the process. Desired values of many parameters may be entered via the computer keyboard. The purpose of the next few sections is to explain how to use Experion to run this equipment. 4. Logging in to Honeywell Access to the Honeywell Experion DCS application is through a virtual machine using VMware Horizon View software. After logging into one of the computers in the UO Control Room, among the icons on the desktop shown is the VMware Horizon View Client icon. Double-clicking this icon should bring up the following popup with the che-view.lsu.edu icon showing. If that icon does not appear, then click New Server; complete the requested information to add it. The following should show: 3/13 11/21/2017

4 Double-click the che-view.lsu.edu icon. If an additional login menu appears, login using your LSU ID and password. At this point the following view should appear: 4/13 11/21/2017

5 Virtual machines are available for the microscope in the UO Analytical Lab (shown as Microscope) and for access to the Honeywell DCS (shown as Uolab). Double-click the Uolab icon. A splash screen for the virtual machine should appear with an OK button. Click OK. The VMware software should start up a virtual machine and show you the desktop of that machine. Open the Honeywell Station software either by double-clicking the Station icon if visible or by navigating using the following pathway: Start>All Programs>Honeywell Experion PKS>Client Software>Station If login credentials are proper, the Station program will appear and exercise patience here the following default splash screen within it: The Honeywell login process is now complete. 5. Overview of Operations The simulated process runs continuously and is controlled from a computer schematic. From the Unit item on the Station menu, select ERUn (where n is 1, 2 or 3 for control studies and 4 for kinetics studies). The ERUn P&ID schematic will appear. This schematic is much like a Process and Instrumentation Diagram (P&ID) with four controller faceplates. To enable the full simulation beyond that of the two flows emulated in the Honeywell, one must run the compiled C++ code from the desktop of the virtual machine where the Honeywell application is running. You will find the executable for this within the Simulators folder on the app / X drive at X:\CHE4162\Simulators\ERU1, for example. You must quit execution of this program before logging out, by bringing up its window and typing Q and enter to quit. 5/13 11/21/2017

6 6. Controlling From The P&ID Schematic Every controller is represented by a small colored circle containing the tagname of the CM, with the values of the setpoint (backlit in green) and the process value (backlit in cyan) near the circle. The first letter indicates the type of measurement - T for temperature, F for flow, or A for analyzer (concentration of component A in this case). To change any analog value from a schematic, you must click on it (if it can be changed, its background will change), and then enter the new value. The changeable objects in this display are the SP and tuning constant values near each controller, and there are additional objects on the faceplates, all of which are explained below. The main value associated with a controller is the measured input, or process value (PV). In addition to the PV, controllers have many more values, the most important of which is the setpoint, or SP. A common example of a SP is the speed setting on a cruise control - the controller will manipulate its output (the throttle position in this case) to move the PV to the SP and hold it there. The SP (in green) and the PV (in cyan) are shown immediately to the side of the circle representing the controller. Just below these values are the tuning constants (in tan): Kc (proportional gain), i (integral time constant, or reset), and d (derivative time constant, or rate). The four small windows to the left and right of the P&ID are the faceplates for four of the five controllers. Honeywell schematics can display only four faceplates at once. To see the fifth faceplate (either the reactor temperature or the cooling water flow rate), click on the Display Xnnn Faceplate button just above and to the left of the cooling water temperature controller. From a controller faceplate, you can see the tagname, description, engineering units, instrument range, and several of the most important parameters on the controller. You can also change many of these parameters. Near the bottom of the faceplate is a drop-down menu box labeled MD, which can be used to select the mode of the controller. Immediately above the mode are the OP (changeable), the PV (not changeable) and the SP (changeable). To change the OP or SP, single click the value, type in the new value and press ENTER. The SPs and tuning constants on the schematic can be changed the same way. 7. Controller Modes As mentioned above, a controller has an SP and an OP. The OP is always given in percent (0-100%) and the SP has the same engineering units as the PV (in this experiment, the flow rates are in m 3 /min, the temperatures are in degrees K, and the concentration is in kmol/m 3 ). When the controller mode is MANual, the OP is held until the operator changes it. When you want to change it, simply click the OP in the faceplate, type in the new value, and press ENTER. The new OP will be held until you change it again. Note that you may enter an OP only while a controller is in MAN. When the mode is not MAN, the controller uses the PV, SP and tuning constants to calculate the OP. When the mode is AUTOmatic, you may enter a new SP to be used for control. Note that changing an SP affects the OP only while a controller is in AUTO. 6/13 11/21/2017

7 For practice, change the output of the feed flow rate controller to 50% (wait a few seconds for the PV to reach a steady flow rate), then to 100% (wait a few more seconds), and then back to 6.9% (-6.9% is known as tight shutoff in the Experion system). Notice the small bar under the control valve on the schematic its length is proportional to the output. Sometimes controllers are stand-alone (i.e. there are no connections to any other controller), and sometimes they are in a cascade structure (like the concentration-to-feed flow rate controllers and the reactor temperature-to-cooling jacket temperature-to-cooling water flow rate controllers in this equipment - cascade connections are indicated by solid lines with small circles on the schematic). The upper controller in a cascade (the one which can receive its SP only from a human) is called the primary and the lower controller (the one which can receive its SP from another controller) is called the secondary. Secondary controllers require an additional mode so the computer system will know when to close the cascade (i.e. put the cascade structure fully on control). In the Experion system, this additional mode is called CAScade. When the mode of the primary is AUTO and the mode of the secondary is CAS, the cascade is said to be closed and the primary is sending its OP to the SP of the secondary. When the mode of the secondary is AUTO, the operator is responsible for changing the SP of the secondary to control the process. The circle representing a controller is filled with a color which indicates the current mode of the controller. For all controllers, yellow means the mode is MAN. For primary controllers, cyan means the mode of the primary is AUTO, but the mode of the secondary is not CAS (i.e. the primary is ready to start sending new SPs as soon as the secondary goes to CAS). For secondary controllers, cyan means the mode is AUTO (i.e. any new SP must come from you, not from the primary). And finally, for all controllers, white means the mode is in its normal state - fully on control, which is AUTO for primary controllers and CAS for secondary controllers. For practice, set both the concentration and feed flow rate controllers to AUTO. Notice that the background of both controllers changed from yellow to cyan. Now set the feed flow rate controller to CAS (i.e. close the cascade). Notice the change in background colors of both the primary and secondary controllers. Now put both controllers back in MAN and set the flow rate OP back to -6.9% to stop the flow. 8. Sampled Data Control Unlike dedicated analog instrumentation for process control, the Honeywell Experion uses microprocessor-based digital computers to perform regulatory control tasks, and workstation and server PC s to configure the system, build control strategies and schematics, gather and display data, etc. The calculations in such systems are typically performed at regular intervals of time. The time between two consecutive data readings is referred to as the sampling time and such systems are referred to as sampled data systems. The sampling time used in this Experion system is one (1) second. All data are collected, control algorithms processed, and outputs sent each second. 7/13 11/21/2017

8 9. Display Navigation When you logged into Flex Station, you used an item from the menu bar to call up the main ERU schematic. There are several additional ways to go from one display to another. For example, you can enter the tagname of a controller in the Command field at the top of the screen and press F12 to call up the detail display. Try it with your feed flow rate controller (F700 for section 1, or F710 for section 2). For a controller, the detail display has 7 tabs. The one labeled LOOP TUNE will probably be useful for tuning your flow rate controllers. Most of the toolbar buttons are used for navigation some require a name or number to be entered, and some go directly to the display. Most of the same functions are on the function keys. For example, to return to the previous display, click, or press F8. To return to the display before that, do it again. From most displays (both system displays and custom schematics such as ERUn), double clicking any value associated with a CM will take you to its detail display. From a detail display, click or press F2 to return to the main ERU schematic. On most custom schematics there may also be buttons to quickly get you from one display to another. 10. Using Trends There are two buttons on the main ERU schematic to call up trends. The button labeled Temps & CW displays the PVs, SPs and final OP of the temperature and cooling water flow rate controllers. The button labeled Conc & Feed displays the PVs, SPs and final OP of both the concentration and feed flow rate controllers. Either trend can be used to show the interactions between the two primary controllers. With all controllers on computer control, a well-tuned system will bring the system back to steady state (i.e. both primary controllers within 0.5% of SP) within 15 minutes of any reasonable disturbance or setpoint change to the reactor temperature and/or the product concentration controller(s). Click the Conc & Feed button. At the bottom of the trend is the legend with all the tag.block.parameters associated with the traces. The checkboxes in the Pen column indicate which traces are currently on the trend. Click on the chart area of the trend and a white hairline cursor appears on the chart and the values at the hairline cursor appear in the Reference Value column of the legend. Along the bottom of the chart area is a horizontal scroll bar which allows you to scroll the chart area back and forth. Along the left axis you ll find the low and high range of the selected trace. These allow you to change the range of the trace for the selected parameter. Practice by changing the range of the product concentration controller to 1-5 kmol/m 3. Immediately above the left side of the chart area is a drop-down list which allows you to select one of the traces (you may also click anywhere on the line for this trace in the legend area). When you select an active trace, it is highlighted (i.e., becomes thicker) in the chart area. Above the right side of the chart area is the Period drop-down list. This drop-down list allows you to select the period of time for which you want data to be displayed in the 8/13 11/21/2017

9 chart area. To the right of that is the Interval drop-down list. This drop-down list allows you to select the interval between points in the chart area. Practice changing to a different period and interval. Leave the period set to 1 day and the interval at 1 minute for now. For practice, scroll back until some variation in some of the traces appears. Notice that the timestamps below the chart area change as you scroll. Find some local max or min in one of the traces and click on it. Now change the period back to one hour and notice that the cursor is centered on (or at least near) the local max or min. If necessary, move the hairline so it is exactly on the peak or valley and notice that the values, as well as the date and time, are shown in the Reference Value column in the legend. Now return the trend to the current time by clicking. All changes you make to the trend can be saved by clicking the familiar Windows Save icon just above the right end of the chart area next to the word (Modified). 11. Changing Design and Operating Specs for the ERU 1, 2, or 3 Some studies in ERU 1, 2, or 3 require changing design specifications other than feed flow and cooling water flow. Figure 2 identifies those that can be changed live and those which require a simulation engine.exe restart to force transfer of those values to the engine. Figure 2 Operating Screen for ERU 1 or 2 or 3, with indications of changeable design parameters. 9/13 11/21/2017

10 12. Saving Data into Excel Tuning the level and ph controllers will require analysis of a great deal of data in the time domain. To collect this data, an Excel workbook containing a Visual Basic Add-In is provided. On the Desktop, look for a folder named Excel SpreadSheets Within that folder, open the folder named snr and double click on ERURecordern.xls (where n is your section number). The workbook will open with a Start button, the experiment name, a collection frequency drop-down menu box, and a Stop button on the top line. Click Start, and the workbook will start collecting the relevant data at the specified collection frequency. These data will be extremely useful in analyzing your results. While the workbook is collecting data, it may be scrolled, but you should not attempt to do anything else in this instance of Excel until after you click Stop. If you do, the collector may stop and you may lose valuable data. When you finish a run, click Stop and cut or copy whatever data you need to your daily workbook in a separate instance of Excel. Let the workbook collect data while you complete the following step for practice. 13. Manual Control (Single Stable Steady State) If the study to be conducted WILL NOT include examination of multiple steady states then manual control should be achievable over a wide range of reactor temperatures. Practice controlling the experiment manually from the Experion Flex Station as follows: Start Excel data collection. Place all five (5) controllers in MAN if these are not already so. Adjust the OP of the feed flow controller to deliver 3.2 m 3 /min of feed, which is 50% of the full range of that flow controller (OP 75.79%). Adjust the OP of the cooling water flow controller to deliver 1.65 m 3 /min of cooling water, which is about 75% of the full range of that flow controller (OP 25%). The process should come to steady state at a particular set of values for the reactor temperature, cooling water temperature, and reactor concentration. While leaving the feed rate at its current value, adjust the OP of the cooling water flow controller to deliver 0.55 m 3 /min of cooling water, which is about 25% of the full range of that flow controller (OP 75%). The process should again come to steady state at a particular set of values for the reactor temperature, cooling water temperature, and reactor concentration. Click on the Stop button in the Excel workbook. Compare the values of both temperatures and concentration from these two trials. 10/13 11/21/2017

11 The relative values should make sense, given the nature of the reaction. Practice saving the data to another workbook in a separate instance of Excel (do not copy or cut the buttons or the drop-down list box), then save that workbook to the C:\Temp directory on this workstation, and finally save the copy in C:\Temp to your group s directory using COPY/PASTE (not SAVE AS ). 14. Manual Control (Multiple Steady States) If the study to be conducted WILL include examination of multiple steady states then use the method described below to practice controlling the experiment manually from the Experion Flex Station. Using the design cooling water flow rate and design feed flow rate from supplied project specifications, the system will have three steady states, corresponding to three different reactor temperatures. We ll call these P1, P2 and P3, in increasing order of temperature. States P1 and P3 are open-loop stable, but to hold the system at P2 requires automatic control. We ll start at P1. Start Excel data collection. Leave all controllers in MAN, set the cooling water flow rate OP to 75%. Adjust the OP of the feed flow controller to deliver the project specification flow rate. When the reactor temperature rises above 350 K, adjust the cooling water flow rate OP so the PV is at the project specification rate. The process will settle at the P3 steady state in a few minutes. Set the cooling water flow rate OP to 25%. When the reactor temperature falls below 360 K, adjust the cooling water flow rate OP so the PV is at the project specification rate. The process will settle at the P1 steady state in a few minutes. Now see how closely you can hold the process to P2 by manipulating the OP of the CW flow rate controller. You may stop trying after 10 minutes. Click on the Stop button in the Excel workbook. Practice saving the data to another workbook in a separate instance of Excel (do not copy or cut the buttons or the drop-down list box), then save that workbook to the C:\Temp directory on this workstation, and finally save the copy in C:\Temp to your group s directory using COPY/PASTE (not SAVE AS ). 15. Shutdown Procedure Before leaving for the day: Put all controllers in MAN and set the PID parameters on all controllers to 1, 0, 0 (ERU 1, 2, and 3 only). Set the feed flow rate OP to 75.79% and the cooling water flow rate OP to 50% (ERU 1, 2, and 3 only). 11/13 11/21/2017

12 Exit Station application (all ERUs). Turn off the simulator software (all ERUs) by hitting Q and then Enter on that app. Log off the computers (all ERUs). 16. Tags, Descriptions, Ranges, and Units Tag Name Service Description Instrument Range Engineering Units A700 R1 Product Concentration 0-to-6 kmol/m 3 F700 R1 Feed Flow Rate 0-to-6.4 m 3 /min T700 R1 Reactor Temperature 250-to-450 K T701 R1 Cooling Water Temperature 250-to-450 K F701 R1 Cooling Water Flow Rate 0-to-2.2 m 3 /min A710 R2 Product Concentration 0-to-6 kmol/m 3 F710 R2 Feed Flow Rate 0-to-6.4 m 3 /min T710 R2 Reactor Temperature 250-to-450 K T711 R2 Cooling Water Temperature 250-to-450 K F711 R2 Cooling Water Flow Rate 0-to-2.2 m 3 /min A720 R3 Product Concentration 0-to-6 kmol/m 3 F720 R3 Feed Flow Rate 0-to-6.4 m 3 /min T720 R3 Reactor Temperature 250-to-450 K T721 R3 Cooling Water Temperature 250-to-450 K F721 R3 Cooling Water Flow Rate 0-to-2.2 m 3 /min A730 R4 Product Concentration 0-to-6 kmol/m 3 F730 R4 Feed Flow Rate 0-to-6.4 m 3 /min T730 R4 Reactor Temperature 250-to-450 K T731 R4 Cooling Water Temperature 250-to-450 K F731 R4 Cooling Water Flow Rate 0-to-2.2 m 3 /min For the ERU4 in kinetics mode, the following additional tag exists in the Excel data collector for the purposes of displaying elapsed time (Hours-On-Oil) since fresh catalyst was supplied to the unit: Tag Service Instrument Engineering Name Description Range Units ACT730(2) R4 Hours-On-Oil Timer 0 to hours 17. Safety Considerations (ERU 1, 2, and 3) Even though this is a simulated reactor, you must still fill out a JSA before beginning 12/13 11/21/2017

13 operations. Chemical Hazard Assume that the feed and product are 0.1 molar HCl, and that the product has approximately the same toxicity as formaldehyde. Slipping Hazard Any process equipment (particularly control valves and pumps) may leak from time to time, so be on the lookout for and mop up any liquid. Temperature Hazard Even though the jacket is insulated, some of the piping is not. Identify the piping which may be a burn hazard. 18. Safety Considerations (ERU 4) Even though this is a simulated reactor, you must still fill out a JSA before beginning operations. Chemical Hazard Assume that the feed and product are similar to linear alkylbenzene with di-olefin contaminants, with the hazards associated with that general chemical class (see the MSDS list on the uolab.com website). Also note that hydrogen gas is part of the feed as well. Slipping Hazard Any process equipment (particularly control valves and pumps) may leak from time to time, so be on the lookout for and mop up any liquid. Temperature Hazard Even though the jacket is insulated, some of the piping is not. Identify the piping which may be a burn hazard. 19. Relevant Literature References 1. Smith, C.A. and A.B. Corripio. Principles and Practice of Automatic Process Control, 2 nd Ed. New York: John Wiley and Sons, Seborg, D. E., T. E. Edgar and D. A. Mellichamp. Process Dynamics and Control, 2 nd Ed. New York: John Wiley and Sons, Romagnoli, J.A. and A. Palazoglu. Introduction to Process Control, Boca Raton: CRC Press - Taylor and Francis Group, /13 11/21/2017