Modeling and Control

Reflecting on why my discussions with engineers at Rosemount Analytical and Broadley-James Corporation (see Dec 22 entry) were so personally rewarding, I realized that I was getting access to knowledge of decades of experience from vintage specialists and that this knowledge might otherwise be lost.

Every year, hundreds of engineers with a wealth of experience are retiring. Most have never had the time to write. In 10 years, the knowledge gained will be so far gone the engineers left won’t even know what they missed. I am not talking about the specs, construction, and theory of measurements, control, and valves but what works and doesn’t work and why.

I have written a lot of articles (30), blogs (100), books (18), columns (68), and papers (20). What I got out of writing is a liberating feeling of having expressed and shared my understanding and experience. This is why I made several of my books available as E-books on the “Control Insights” website. Books also enabled me to organize my thoughts, bring closure, and give me the freedom to move on.

Ever since I got going on the “Modeling and Control” and “Control Insights” websites, my circle of readers has gotten increasingly larger. Books are nice and in fact people prefer to have a bound book for reading and reference but the web is the best way of sharing knowledge. A good strategy would be to put tidbits of new books out on the web to create interest and expose more people to what you offer.

So this is an invitation to avoid your knowledge be lost for eternity by creating your own website. If you are going to do a blog, don’t follow my example. The blog should be brief with pictures and be published the same day of the week every week.

What distinguish humans from other animals are the gifts through art and science to discover, create, and disseminate knowledge and beauty expanding our understanding and perception of the universe. Art and science can both get at the essence and create new entities that take on an essence of their own. Both improve the quality and level of life. For me, good technical writing is both art and science. Try doing a weekly blog on what you have learned. I bet if you stick with it you will find it rewarding and create something that takes on an existence of its own.

Technorati Tags: art science | education | technical writing |

My first experience with wet labs was about 35 years ago when I set up an acetic acid and water neutralizer and distillation column with conductivity, flow, level, pH, and temperature loops as part of a 6 week course to teach process control to new employees. I learned first hand how time consuming and expensive it is to set up and keep a lab running in top notch condition. Any compromises in hardware lead to headaches. I handed this off to someone willing to make the lab and the course a full time job.

About 10 years later, I managed to get a research technician to set up a lab to test pH electrodes from 9 different manufacturers. We got a lot of data on the effect of salt concentration and temperature on electrode performance but we had headaches with algae growth and dealing with undocumented features. The flat glass electrode developed large errors as the temperature and slat concentration was increase. Solid reference electrodes had acceptable performance but were not as accurate as some gel double junction electrodes. It would have been nice to study the effects of coating but the consequences in terms of cleaning out the equipment and lines were prohibitive. The lab was only used for one year but this still represented a cost of probably $50K in time and material. In retrospect, we should have studied the effects of velocity and mixing. I would love to have a lab today that is better than my 1980s lab to study the performance of new electrodes, diagnostics, and wireless reporting. It seems to me there is less information today than 20 years ago on the effect of process conditions on pH electrode performance. pHwetLab

While teaching a course on process dynamics and control at Washington University (WU) in Saint Louis, I was asked to teach a digital computer control lab for systems engineers. The lab was mostly a collage of instrumentation, valves, and controllers you would never see in industry. I arranged for the donation of the latest Fieldbus smart transmitters and smart control valves and a “state of the art” DCS system by Emerson. While teaching the lab would be neat, I knew from previous experiences it would an intensive effort that would take away from teaching, studying, and writing on modeling and control using the virtual plant I had set up at WU for my course for Chemical Engineers. Luckily I found the perfect choice in Bob Heider who enthusiastically has kept the lab in great shape and taught a course of great practical value for the last 5 years. The lab is unique in that engineers are taught how to calculate and evaluate process dynamics on a first principle basis.

My most recent experience is initiating and guiding a wet lab for bench top and single use bioreactors (SUB). Fortunately, Broadley-James Corporation (BJC) has committed the resources to make this happen. As with all process research labs, especially bio-reaction labs using new analyzers, cell lines, and media, strict schedules are an impossibility particularly considering this is an extracurricular effort for BJC. We have suffered from prototype sample system and analyzer failures, an NIR probe company going bankrupt, media problems, cell line discontinuation, infections, and delivery delays. Despite the slow progression, we have adjusted the virtual plant’s model growth, death, and substrate kinetics to better match the lab runs and have tested closed loop glucose control, which is essential for an effective design of experiments (DOE) for parameter identification. We are also now testing wireless pH and temperature control on the SUB. PATLab

A successful wet lab requires patience, enthusiasm, and a focused "care taker." The practical experience and data gained for the development of sensors, control strategies, and dynamic models is well worth the effort. Experimentation in a wet lab can be more effective and is certainly much less expensive than experimentation in an actual plant even if permitted.

Technorati Tags: bioreactor modeling | bioreactor control | pH measurement | process control education | process control labs |

This is 100th anniversary of the glass pH electrode yet we still do not know much about what affects their life. Not much has been published about application problems and practices. There have been a few academic studies on the effect of process conditions on the glass electrode but these are over 20 years old. The best book on the theory of pH measurement The Determination of pH is over 35 years old. The glass electrode is more important than ever because of its extraordinary rangeability and sensitivity to hydrogen ion concentration. What other measurement can cover 14 orders of magnitude of concentration and detect changes as small as 0.00000000000001 (0 to 14 pH scale).

While the fundamentals of the glass pH electrode have not changed in a hundred years, there have been significant improvements in the glass formulation and construction so that it can handle high pH and high temperature fluids and repeated sterilizations. For example a new high temperature electrode increases the life expectance by 100% as seen in the attached test results. HighTemperatureGlassElectrodeLife The same company has developed a new electrode for the biopharmaceutical industry that can withstand 50 sterilizations. These new glass electrodes maintain a response time of seconds whereas other electrodes develop response times of several hours due to the premature aging of the outer glass surface.

As the number of pH applications increased dramatically due to the Clean Water Act and the growth of biopharmaceuticals and specialty chemicals, maintenance costs became a bigger issue. Filling and pressurizing reference electrodes and troubleshooting separate measurement electrodes became points of pain. In response, electrode suppliers developed the throwaway combination electrode that built a sealed reference, glass measurement, and temperature sensor into a single rugged probe body. Maintenance simplified to calibration and replacement. Some users unfortunately skipped the calibration part.

Reference electrodes maintain a low resistance path to the process for electrical continuity of the pH measurement by an internal electrolyte that is contact with the process fluid through a porous reference junction. The elimination of the flowing reference junction can lead to coating and contamination because there is no flushing action to prevent back flow (migration) of the process through the reference junction into the interior. To deal with these potential problems there have been major improvements in the construction of the reference electrode to reduce contamination (poisoning) by the use of multiple internal junctions and by the use of a porous solid with tortuous paths instead of a gel to slow down the migration of process ions and prolong the time it takes for the process to reach the internal silver-silver chloride element.

There have been attempts to develop alternatives to the glass electrode, such as the Iridium oxide, ISFET, and optical sensors. Yet we have not seen significant use.

The secret life of pH electrodes will be exposed in a series of Control Talk columns starting in the February issue of Control magazine. The series will include very candid views about the history and future of pH measurement from the most experienced people at major electrode manufacturers and the results from an online survey to find out what is really going on with pH applications. You can participate in the survey via the following link: http://www.zoomerang.com/Survey/?p=WEB228KQLJS4KT

If you want to learn more about pH measurement and control consider attending a short course. If you have a pH control problem, you can get it solved in the course and learn how to create a virtual plant of the pH system to demonstrate and prototype process control improvements. You can get info on my course via the following link:
http://www.emersonprocess.com/education/catalogrev/automationsystems/9060.asp

For more info on pH measurement check out previous entries in the pH and plant design categories on this web site. You can get more information on pH modeling and control from recent articles “Virtual Control of Real pH,” Control, p. 47 (Nov. 2007) http://www.ControlGlobal.com/articles/2007/385.html
and “Virtual Plant Provides Real Insights,“ Chemical Processing, (Jan 2009).

In the future we can look forward to new electrode diagnostics, interrogation of the history of calibration adjustments for the electrode, and wireless transmission. CD Feng, receiver of the 2007 ISA Arnold O. Beckman Founder Award for his technical contribution and innovation has agreed to provide key chapters on pH sensors, diagnostics, and new technologies for the 4th edition of my book Advanced pH Measurement and Control. Maybe the life of the pH electrode won’t be so secret.

Technorati Tags: pH control | glass electrode | pH measurement | pH modeling |

Line “D” of a pet food plant never operates as well as the other lines. Line “A” has the best performance. The operators for line “D” say that line “D” is different and it can’t do better. When a line “D” operator gets sick, a line “A” operator fills in on Line “D”. Line “D” begins to do as well as line “A”.

A builder and operator of ethanol plants puts process metrics on the operator screens for each plant that are viewable by operations at all of the plants. The competitive nature of people kicks in and all of the plants start to do better.

The energy cost for a lime kiln is displayed online. Model predictive control (MPC) is installed and the energy costs drop by 10%. Projects are started to install MPC on all of the lime kilns.

Online process metrics can blow away war stories, motivate operators, increase the on-stream time of advanced controls, justify process control improvements, and develop correlations between key performance indicators and operating conditions. For example, processes may show daily and seasonal performance variations because of the change in feed and cooling water temperatures. Also, process may run better or worse at night and or weekends and holidays depending upon whether automation, maintenance, and process engineers are supporting or distracting and interfering with operations.

However, the implementation is not necessarily straightforward. Process metrics can show us something essential but we may not always like what we see. The president of an MPC company years ago was unequivocally against online process metrics because they may initially take a dive when the MPC is turned on.

I installed online metrics of base reagent cost to show the advantage of adaptive pH control for neutralization of an acidic waste stream. The tighter control increased the reagent costs for disturbances that drove the pH below set point or for increases in the pH set point because the addition of caustic was larger and sooner driving the pH to the set point faster. For disturbances and set point changes in the opposite direction the tighter control decreased the reagent costs. So is tight control right or wrong and are process metrics in this case not useful? If there is a penalty for being below set point, it should be added to the online cost metrics. If not, the controller should be tuned with a lower gain when the pH is below the set point.

Consider a batch operation where the process must be heated up before a reaction occurs. A control system that gets temperature to set point fast will increase the steam use per batch by overdriving the control valve past its resting position. The question is whether the reduction in batch cycle time is worth more than the increase in steam per batch.

You cannot control what you can’t measure. To control plant profitability we need to have the automation system and computations to put process metrics online. Undoubtedly, improvements will be needed to the metrics and to the automation systems that affect them. Filtering and averaging will be needed to screen out noise and delays added to make process inputs coincide with process outputs. New measurements and valves will be needed. Throttling valves with better deadband and resolution can reduce limit cycles. Coriolis meters can provide accurate flow measurements and inferential measurements of stream compositions important for yield, quality, and production rate calculations. Ambient, piping and equipment wall, feed, and coolant temperatures can help provide indications of previously unknown adverse effects.

I see a future where the cost and revenue per production rate, batch, shift, day, night, week, month, and season besides yield and on-stream time are displayed for each production line. Data analytics will be used to develop correlations for projections to latent structures or partial least squares (PLS) to provide predictions of process metrics online and to provide a drill down to contributions most affecting the metrics for better process understanding. The trends and future predictions of these metrics immediately translate to improvements and eventually "closing the loop" for plant profitability. I expect an MPC will be developed to use process metrics as controlled variables and the principal components as manipulated variables.

It seems to me online process metrics are the key for a manufacturer, process control group, and automation company to thrive in a competitive worldwide economy. Loop tuning and performance is just the beginning. As automation engineers we tend to think of the loop as the “end all.” We need to get outside of the box that is the loop to prevent islands of automation. We need to think in terms of unit operation control and how these units interact to affect the process as a whole. We need “oneness” guided by process metrics as introduced in my control talk column. This is the moment.

http://www.controlglobal.com/articles/2008/287.html

For a list of some items we need, see slide 57 in my presentation.

http://www.emersonprocessxperts.com/archives/2008/11/assessing_oppor.html

Technorati Tags: advanced process control | automation careers | process benchmarking | operator interface | opportunity assessment | process metrics | process control improvement |

One of my favorite cartoons is Einstein at a blackboard with his equations for the “General Theory of Relativity” all leading up to one concluding equation “Time = Money.” In this same vein, the great movie line comes to mind: “show me the money”. More than ever, if you want to do neat things in process control, you need monetarily decisive benefits.

What if a group of 5 modeling and control specialists working with the best automation engineers in the plants were given the freedom to find and improve the process control in 100 control rooms with Distributed Control Systems in a major chemical company without having to justify each venture for 2 years? Once upon a time this actually happened. The result was ongoing benefits of 75 million dollars a year. How did this start and finish and what does it say about the future?

The initiative began with a benchmark study of companies with the best track records in process control. The top three companies achieved a 8% reduction in the cost of good by a balanced application of 9 technologies in basic and advanced control and the use of data. The largest benefits came from better PID control (e.g. better tuning and control strategies), unit operation control (e.g. automated sequences), and advanced regulatory control (e.g. feedforward and override control). The next biggest source of benefits came from putting process metrics online via real time optimization and data analytics. The benchmarking study and other stuff supporting this discussion can be seen in a presentation I recently made to big chemical company viewable at the site:

http://www.emersonprocessxperts.com/archives/2008/11/assessing_oppor.html

We found in the benchmarking that each layer of technology was built on a solid foundation of supporting technologies to form a pyramid. The performance of regulatory control depended heavily upon the scope and sensitivity of the measurements and control valves and the tuning of the loops. In turn the success of model predictive control relied upon the performance of the regulatory loops. It is noteworthy that model predictive projects often report bigger benefits than those shown in the report because the improvements made in the regulatory control are attributed to the MPC project, which in a way is justifiable because the improvements to the loops probably would never have been made without the systematic approach used in well managed MPC projects. Finally it was obvious in the study that real time optimization worked best if was integrated into the predictive handling of constraints and interactions provided by an MPC.

The benchmarking study lead to 2 year process control improvement program developed by Vernon Trevathan, a recent inductee into the Control Magazine Automation Hall of Fame. You can read about the highlights of his career and his role in the program in my interview of him in my “Best of the Best – Part 3” series in Control Talk.

http://www.controlglobal.com/articles/2007/061.html

In the beginning of this initiative I conceived, trialed at an agricultural chemical plant, and documented in a internal report an opportunity sizing and opportunity assessment process, as noted in my Control Talk Column “Up for the Ashes.”

http://www.controlglobal.com/articles/2008/092.html

While I give myself credit for the original concept, the real success of this methodology and consequently the whole initiative was the result of Glenn Mertz who had a unique combination of process control and accounting skills and extensive plant experience applying the latest technologies. For each control room studied, Glenn went through the cost sheets of the production unit, found the best periods of operation, and quantified the demonstrable gaps between actual and practical process metrics such as production rate, raw material and energy cost, and rework. Glenn also dug out research reports and process simulation results to provide define theoretical goals and gaps. All of this could have been just an exercise if it wasn’t for the fact Glenn was able to get agreement from key process engineers in the plant to the goals and gaps to form an opportunity sizing that was then the basis of an opportunity assessment conducted for 3 days at the plant. Ultimately Glenn reported the benefits by again scouring the cost sheets and working extremely well with plant engineers in operations and process technology.

It would be nice if we could have a Glenn on each our teams, but given this is improbable what can we do? Can we get Glenn to give up his days on the golf courses in Door County Wisconsin in the summer and Fort Meyers Florida the rest of the year and travel on crowded flights in coach to not so gorgeous places? Can we create a virtual Glenn? Not likely. What is possible? For my answer and the conclusion to this series, see my entry on Dec 5.

Technorati Tags: advanced process control | automation careers | process benchmarking | operator interface | opportunity assessment | process metrics | process control improvement |

I think the future is advanced process control (APC). My definition of APC is any technology that puts process knowledge on the line online. Feedforward control is APC when the feedforward gain and dynamic compensation are based on process knowledge. On-demand and adaptive auto tuners, such as DeltaV Insight, are APC tools because these tuners identify the process dynamics that are useful for process diagnostics and training besides model based tuning. For example, the process deadtime can be monitored as an indicator of heat transfer surface fouling in temperature loops and the dynamics can be inserted in simulations for operator training and scenario testing and prototyping of PID enhancements (e.g. set point filtering and structure) or Model Predictive Control (MPC). There are many higher level technologies. In a recent presentation I made to a major chemical company I showed these technologies, the results from a benchmarking study of the top ten companies in the use of process control, and practical tips on how to conduct an opportunity assessment. The presentation can be seen at:

http://www.emersonprocessxperts.com/archives/2008/11/assessing_oppor.html

Slide 8 shows the pyramid of technologies that includes process performance monitoring (data analytics and process metrics), abnormal situation prevention, property estimators (inferential composition or quality measurements), model predictive control (MPC), rampers and pushers to maximize or minimize a controlled variable (e.g. feed rate), linear programs (LP) for optimization given defined constraints and economics, and real time optimization (RTO) for variable constraints and economics. The importance of process knowledge in all of these technologies is obvious. Slide 9 gives a straightforward “easy to remember” relationship between controller tuning for loop performance. The equation indicates before, during, and after APC implementation, the controllers should be tuned.

The amount of effort and the performance of the upper level technologies rest upon the strength, breadth, and integrity of the foundation of basic control. As you improve the number, type, and sensitivity of the measurements and control valves, the performance of these systems improve by reducing the number of unidentifiable disturbances and enabling more first principle calculations and inferential measurements, such as frosting rate, fouling rate, crystallization rate, and reaction rate important for diagnostics and batch profile control as discussed in a recent article in Control magazine.

http://www.controlglobal.com/articles/2008/230.html

Decades ago, field pressure and temperature gages were installed. These were not very accurate. prone to be broken, and obviously were not visible in the control room or historized. With wireless, we can afford to get many more measurements into the control system. Wireless measurements offer the opportunity to provide many of these missing measurements at a reasonable cost. However, the choice of measurements for data analytics (principal component analysis and projection to latent structures) must be judicious. Randy Reiss, the developer of online data analytic algorithms for Emerson, says “more measurements for analytics means more correlations. However, it introduces the possibility of dominate correlations that do not relate to product quality. That would skew the model for the worse. So there is a double edge sword there.”

For portable bioreactors, laboratory analyzers, and sterilization systems, wireless adds flexibility and utility. Wireless access to process and loop performance monitoring systems in the field makes troubleshooting much smarter. Wireless access anywhere to virtual plants with process performance scores for university courses on process control makes learning almost like a video game. There are many more applications for wireless than the monitoring of remote tanks and pipelines. The following Control Talk column slated for the December issue of Control magazine discusses the role of wireless in APC.

WirelessControlTalkColumn

Randy Reiss’s list of the “Top Ten Reasons You Will Go Wireless Next Year” in the above column provides a reality check in case we are thinking of making everything wireless. This list has the insight, bite, and humor typical of the lists Randy has contributed to my column in recent months. Upon reading the draft of the column, Randy said “it’s the best argument I have heard for wireless.” Randy agreed to the post of this quote after checking with his PR agent.

Scott Broadley, the president of Broadley-James, is participating in a beta test with Emerson on the use of wireless transmitters on portable single use bioreactors (SUB) whose size is steadily growing from pilot plant (100 liter) to production (1000 liters) scale. Scott is also looking forward to the elimination of ground loops and noise by wireless pH transmitters particularly where the solution ground is not used or where AC noise gets through the power supply. Scott says, tongue-in-check, “we could hook the pH and DO transmitter up wirelessly to a Twitter account so your cell phone is getting constant text updates on how your bioreactor is “feeling”. Scott offers the following additions to the top ten list for going wireless.....(11) Each bioreactor can have its own Face book page where operators from different shifts can leave their comments......(12) Each transmitter can be on Twitter and send you instant text messages on your phone when it is moody...”

Technorati Tags: advanced process control | automation careers | opportunity assessment | process control improvement | wireless measurements |

In the early 1990s the chemical company I was at for decades decided the way to capture and put online process and control expertise was to get corporate licenses to the most powerful real time expert system available for real time process control applications. Some of the most experienced people at the plants developed applications for industrial and agricultural chemical production. We even had a well attended and enthusiastic user’s group meeting centered about the biosphere. In retrospec the biosphere story itself was prophetic about how unaddressed non-idealities and chaos can make a seemingly good idea impractical.

The expert system allowed the entry of rules without any sort of order or overall logic. What we ended up with was a hodgepodge of rules whose interrelationship and execution was difficult to decipher. It was hard to demonstrate that the advisory messages resulted in quantifiable benefits or an increase in knowledge. Even when the expert system did correctly diagnose fault, operations said they had also identified the problem. When the expert behind the expert system moved on, the expert systems fell into disuse.

There was an exception. An online real time expert system was developed using first principal calculations and rules to detect the onset of flooding in a distillation column that was being pushed beyond its design limit to maximize production. When flooding was eminent, the expert system initiated online automatic actions to override the control system. The expert system was a success and routinely prevented flooding problems and is in service to this day.

You might judge another expert system to detect measurement problems to be a success based on the fact it was widely employed at the plants and eventually commercialized by a third party application software company. However, the system did not endure the test of time in the control room. The utility of the system was severely diminished from the attempt to provide inferential flow measurements based on the position of control valves without positioners or with pneumatic positioners whose calibration was questionable. Even if the valve positions were accurately known, the lack of pressure measurements meant the installed valve characteristic was not accurately modeled. There were false alarms. I don’t think any of these systems are really being used.

There were also attempts at using neural networks and Fuzzy logic. Networks lead to some knowledge discovery in terms of previously unsuspected process relationships and was able to predict batch cycle time in some cases but these did not stay online and had no quantifiable benefits. A fuzzy logic system for optimization of reagent usage in a waste treatment system is still automatically trimming upstream pH loop set points and achieves a reduction in reagent cost. However, operations can not figure out the actions or diagnose any possible problems. In other words, the logic is fuzzy. Also it is a one of a kind. A model predictive controller has been demonstrated to be able to accomplish the same goal with much more definable and understandable functionality. http://www.controlglobal.com/articles/2007/385.html

In a separate case, the highly touted use of fuzzy logic controllers and its special tuning of scaling factors for temperature loops are now in question because if you increase the gain of a standard PID controller and add set point filtering, you can do about as well.

So what are the lessons? Most importantly the technology needs to take corrective automatic action without operator intervention and provide benefits that are definable and observable. Second, the system must be able to be maintained and be adjusted without the presence of the author. Third, operations needs to understand and appreciate the actions taken by the system. Training and simplicity are keys here. Fourth, the solution should use standard supported tools wherever possible. Fifth, the performance realized and effort required greatly depends upon the scope and technology of the measurements and valves and is only as good as the inputs and the feedback correction. Finally, the system must use documented algorithms, logic, and calculations based on solid process and control principles. If we can do this, we can make advanced control a bigger part of the future.

Technorati Tags: advanced control | automation careers | expert systems | fuzzy logic | process control improvement |