In the ISA Mentor Program, I am providing guidance for extremely talented individuals from countries such as Argentina, Brazil, Malaysia, Mexico, Saudi Arabia, and the USA. This question comes from Bill Thomas.
Bill Thomas’ Question
What are your thoughts on distillation and reactor control modernization projects?
Greg McMillan’s Answer
Excellent temperature control is the key to both distillation and reactor performance.
I would make sure you have the best temperature measurement for both reactor and distillation control, which generally is a resistance temperature detector (RTD) in a tight fitting well with sufficient emersion length in the most representative location. Important general aspects of sensor location and selection are noted in the post How to Succeed – Part 4 from the Modeling and Control website. Additional Modeling and Control posts on “Reactor Cooling and Heating” and “Batch versus Continuous Control” and posts on “How to Succeed” discuss a wide variety of opportunities for reactor control and your career.
The tutorial from a life-long friend and associate Terry Tolliver, whom I consider to be one of the best column control experts, introduces the most important aspects of column control. Click this link to download a PDF of the slides from Terry’s tutorial: Fundamentals of Distillation Column Control.
A book given to ISA Mentor program participants, Advanced Temperature Measurement and Control, Second Edition, has chapters on reactor and distillation control.
Here is the key take-away from my perspective of Terry’s presentation.
The best tray (stage) for column temperature control is seen from changing the distillate to feed ratio (D/F) with the temperature controller in manual. This can be done rather readily in high-fidelity steady-state process simulations (e.g., HYSYS and AspenTech). Simulation results should be verified by field temperature measurements on several stages. The portability of wireless temperature measurements offers the opportunity to borrow plant spare transmitters and sensing elements for the plant tests if spare thermowell connections are provided near the optimum tray predicted by the simulation results.
Slides 26 through 28 of Terry Tolliver’s tutorial show how to select the best tray. On slide 26, the optimum location is tray 7 where the change in temperature is largest and most symmetrical. Since temperature is an inferential measurement of composition, the largest temperature change translates to the greatest sensitivity and the least composition error for a given temperature measurement error quantified on slide 27. By the temperature change being symmetrical, the process gain are about the same for an increase or decrease in the D/F. Even more significantly, using a significantly less than optimum tray you are left with a divergence of temperatures at other trays even if you do a great job of control at the chosen tray. Slide 28 shows that for tight control at tray 13 you have a large divergence at lower trays whereas tight control at tray 7 keeps the deviations at the other trays rather minimal. To me this is the most significant point because recent improvements in temperature measurement accuracy and adaptive tuning might be sufficient to deal with less sensitivity and linearity. Normally, we try to choose a measurement location that has the least deadtime. The best tray based on this analysis often has close to the smallest 63% response time but necessarily the least deadtime. The lesson here is that choosing the control location that provides best disturbance rejection throughout the unit operation (most representative location) is more important than minimizing deadtime.
I suggest considering feedforward and valve position control as ways of increasing reactor and column performance as summarized in the Control magazine article Don’t Over Look PID in APC. If you can add an analyzer of key components, you can do batch profile and continuous composition control. For columns there is a best tray for composition control. In a Control Talk interview with Jim Tatera titled Process Analyzers. Analyze This!, Jim says he would install sample connections on trays just above and below the optimum tray found by simulation. Borrowed analyzers were used to measure the tray compositions during the plant tests. Jim says simulations often cannot deal with non-ideal complex mixtures. Plant tests are needed to ensure you have the optimum location.
Both Terry Tolliver and Greg Shinskey state that column material balance control schemes are best. Terry’s tutorial showed types 1 and 4 direct material balance and types 2 and 3 indirect material balance control schemes.
ISA Mentor Program
The ISA Mentor Program enables young professionals to access the wisdom and expertise of seasoned ISA members, and offers veteran ISA professionals the chance to share their wisdom and make a difference in someone’s career. Click this link to learn more about the ISA Mentor Program.
Greg Shinskey discussed important cases when steam flow is manipulated for temperature control. My notes are:
- Low reflux ratio columns benefit from distillate receiver level manipulating both distillate (D) and reflux (L) in an L/D ratio (called reflux ratio control). The temperature loop manipulates boil-up (steam).
- Hi purity composition columns have a very low process gain at conservative setpoints. Increasing the impurity setpoint can dramatically decrease energy use if the PID is retuned for the higher process gain.
- For two point composition control, relative gain analysis can show you the best pairing
While material balance control is by far the most useful strategy (see the Hydrocarbon Processing article by Shinskey’s protégé titled Improve material balance in high-purity distillation control) it is worthwhile to know more about separation control. In both separation control (Terry’s slide 17) and indirect material balance type 3 (Terry’s slide 23), the temperature controller manipulates the reboiler steam flow (boilup). The difference is that in separation control the distillate flow is on local flow control and not on distillate receiver level cascade control.
When I asked Shinskey in a subsequent email for an example where separation control is used, he offered the following:
An example where separation control is used is a typical debutanizer: moderate purity specs, low reflux ratio. The D/F line is a diagonal, so relative gains (RGs) for material balance controls fall around 0.5—not good. The separation-boilup (L/D, V/F) RG is 2.5—that’s good; the (L/D, V/B) RG is even better at 1.5. (Some engineers have had excellent performance with that configuration. See page 360 of the fourth edition of Process Control Systems). RG numbers follow a geometric progression. I avoid anything >10 or <0.5.—they are about equally bad.
Additional Mentor Program Resources
See the ISA book 101 Tips for a Successful Automation Career that grew out of this Mentor Program to gain concise and practical advice. See the InTech magazine feature article Enabling new automation engineers for candid comments from some of the original program participants. See the Control Talk column How to effectively get engineering knowledge with the ISA Mentor Program protégée Keneisha Williams on the challenges faced by young engineers today, and the column How to succeed at career and project migration with protégé Bill Thomas on how to make the most out of yourself and your project. Providing discussion and answers besides Greg McMillan and co-founder of the program Hunter Vegas (project engineering manager at Wunderlich-Malec) are resources Mark Darby (principal consultant at CMiD Solutions), Brian Hrankowsky (consultant engineer at a major pharmaceutical company), Michel Ruel (executive director, engineering practice at BBA Inc.), Leah Ruder (director of global project engineering at the Midwest Engineering Center of Emerson Automation Solutions), Nick Sands (ISA Fellow and Manufacturing Technology Fellow at DuPont), Bart Propst (process control leader for the Ascend Performance Materials Chocolate Bayou plant), Angela Valdes (automation manager of the Toronto office for SNC-Lavalin), and Daniel Warren (senior instrumentation/electrical specialist at D.M.W. Instrumentation Consulting Services, Ltd.).