DESTINATION SETPOINT FROM THE EDITORS DESK CONTENTS OSU AUTOMATION SOCIETY NEWSLETTER SPRING Dear Reader,

Transcription

1 DESTINATION SETPOINT OSU AUTOMATION SOCIETY NEWSLETTER SPRING 2015 CONTENTS FROM THE EDITORS DESK PG. 1 FROM THE EDITORS DESK VIDHYA VENUGOPAL & AMEY THORAT ADVISOR S MESSAGE PG. 2 PRESIDENTIAL NOTE PG. 2 WHEN THEORY ISN T ENOUGH PG. 3 YOU DON T KNOW WHAT YOU DON T KNOW PG. 3 EXTRAMURAL LEARNING AND AUTOMATION PG. 4 SOFT SKILLS - AD ADDITIONAL POWER TO APC ENGINEERS PG. 5 EVALUATION OF AUTOMATED STEADY STATE & TRANSIENT STATE DETECTION ALGORITHMS PG. 6 SMART X = AUTOMATION + X PG. 10 FOUR OSU STUDENTS RECEIVE ISA SCHOLARSHIPS PG. 11 MEMBER S EXPERIENCE PG. 12 OSUAS TEAM PG. 12 LOOKING BACK PG. 13 HOW CAN YOU HELP WITH OUR CONTINUED SUCCESS? PG. 15 ACKNOWLEDGEMENTS PG. 15 AUTOMATION DAY SPONSORS PG. 15 Dear Reader, We are proud and delighted to present the Oklahoma State University Automation Society s (OSUAS) 4th annual newsletter, Destination Setpoint. OSUAS has been striving to provide insight into the automation and control industry, especially for OSU students, guiding them with information that would be useful for future careers in this field. We intend to use this newsletter to spread awareness of OSUAS s activities over the past year including, visits from professionals of reputed companies, several industrial visits, social activities, and a glimpse into what Automation Day 2015 will entail. Automation is the symbiosis between humans and machines. Human creativity, intelligence, and wisdom combined with the precision, accuracy, and robustness of machines can achieve wonderful results that are unattainable by either of them individually. Largely limited to the manufacturing industry in its earlier days, automation has now become an integral part of our everyday life. Intelligent systems are being developed for medical care, security, transport and communication, finance, education, and entertainment. With the advancements in computational resources, devices and control systems are becoming faster, versatile, and robust. This has given rise to a new generation of devices like smart phones, smart vehicles, wearable health monitoring systems, virtual reality devices, robots, etc. which have increased the human reach. Through this newsletter, OSUAS would like to provide an insight into the importance of automation and control in some of these areas and how OSUAS can serve as a launch pad for students aiming to explore the wonderful world of automation and control. We would like to thank all those who sponsored and contributed articles to Destination Setpoint. Our special thanks to Dr. Russell R. Rhinehart, OSUAS s faculty advisor, whose support and guidance has been key to the success of the group. Vidhya Venugopal (PhD, ChE) is the current Vice-President of OSUAS. She received her undergraduate degrees in Chemical Engineering and Chemistry from the University of Minnesota in Amey Thorat (MS, ChE) is the current General Secretary of OSUAS. He received his undergraduate degree in Chemical Engineering from the University of Pune in DESTINATION SETPOINT 1

2 ADVISOR S MESSAGE DR. R. RUSSELL RHINEHART About four-and-a-half years past, The OSU Automation Society received its charter from ISA. Now we have over 100 student members, monthly professional development events, Automation Day to celebrate the profession, a newsletter to support career understanding, a web site partly dedicated to matching job openings with applicants and partly to promoting automation, and income to support the initiatives. We received the 2013 ISA Section Excellence Award, and officers frequently receive ISA Scholarships. The activities and successes did not just happen. They are the result of the dedication and creativity of a sequence of officer sets and members that support the activities. You and your predecessors have created something good, strong, real, and impactful. This is an impressive organization. It is good for industry because we recruit, attract, and develop people for automation career needs. It is good for OSU, because the Automation Society activities and outcomes support the OSU mission, and provide evidence that OSU is playing its role in society. And, of course, it is good for the students that use the professional development insight to better prepare for careers, and who have employment opportunities open to them. But mostly, it is beneficial to the student officers. In the charter year, we were fortunate to have the ISA President, Mr. Leo Staples, working in Oklahoma City, and he came to OSU and personally presented the documents to the OSU AS officers. He told us what he had discovered by volunteering the volunteer gets much more than they give. The volunteer makes things work for other people. The volunteer gives. But, this is what an employer wants. Engineers don t just sit at a desk and do calculations. They are expected to make things better for the enterprise. As a volunteer in an organization, the volunteer develops skills that will enhance their future career success. More immediately, the volunteer develops a reputation for adding value that enhances their attractiveness to getting a job offer from an employer. And, it leads to opportunities for scholarships. And what the volunteer creates to help others, also serves self. But, most immediately, the volunteer sleeps tired and happy. Make volunteerism a lifelong activity. Dr. R. Russell Rhinehart (BP Professor and ISA Fellow) has been the Faculty Advisor of OSU Automation Society since its inception in October He was awarded the OSU Student Organization Advisor award for the year in by the President of OSU and Office of Campus Life. He is also the Past President of the AACC. PRESIDENTIAL NOTE SAMYUKTA KOTEESWARAN OSU Automation Society (OSUAS) is a student organization that unites students, alumni, and industries with common interest, Control and Automation. I joined OSUAS as a student member in 2012 after attending the international students orientation. From that day to now, being a part of OSUAS has been one of the most invigorating experiences. During my first year in OSU, I spent most of my time attending classes, doing homework, and research projects. I used to listen to my friends (then OSUAS officers) discuss and plan OSUAS events. Eventually I started being a part of all the activities such as field trips, social events, networking dinners and so on. I found myself an extracurricular activity that interested me, kept me engaged, and gave me an opportunity to interact with experienced professionals directly. The following year, I became the General Secretary of OSUAS in May When I was the General Secretary, I had the opportunity to talk with new graduate and undergraduate students about our activities, interact with them and enjoyed socializing with them. When we are looking for jobs our contacts and the way we network with helps us greatly in the long run. OSUAS helps students from various engineering departments by keeping them updated about any job and internship opportunities and also bring professionals from Automation industry to OSU. In the past many officers and members have attributed their success in professional life to OSUAS. We, here in the automation society, maintain a great relationship with our alumni. We reach out to them for our activities here and, in turn, they have been supportive and encouraging by contributing to OSUAS through sponsors, helping us set up networking events with their companies and also by writing articles in Destination Setpoint. OSUAS unites students interested in automation and instills responsible leadership. From the day of its inception till now the society has grown leaps and bounds in terms of industry participation, members and hosting the biggest event of the year The Automation Day. There is lot of planning and efforts that go into hosting the Automation Day every year and at the end of the day we get immense satisfaction when we see the students benefitting from our events. With over 100 student members and 50 active alumni members, with an average of 5 events every semester OSUAS is one the most active student organization in OSU with students and officers from diverse educational and cultural backgrounds. Our advisor, Dr. Rhinehart, has been the backbone for OSUAS ever since one of our alumni came up with the idea of starting the Automation Society. His encouragement and enthusiasm to do new events, to make the events better, to get professionals from industry participate is a big motivation for me and all the other officers. We look up to him and try to emulate him in our professional lives. We have been very fortunate to have a great working team always on their toes. With that note I would like to thank all the OSUAS members, alumni, past teams, industry professionals and our advisor Dr. Rhinehart without whom OSU Automation Society would still have been a dream. We hope to grow stronger, impact and mold many aspiring engineers in the years to come. Samyukta Koteeeswaran is a Ph.D student in Chemical Engineering. Her research is in the field of petroleum and has been a part of OSUAS for 3 years now. She aspires to work in the R&D of a refinery after graduating from OSU. 2 DESTINATION SETPOINT

3 WHEN THEORY ISN T ENOUGH ANAND VENNAVELLI Have you ever wondered what makes a good automation engineer? Good theoretical understanding of automation principles is, in fact, a good answer. Without a good theoretical background or understanding, an automation engineer s career cannot go far. After all, many brilliant minds went into developing the theoretical methods. Standing on the shoulders of giants is a great strategy. However, more often than not, there is a gap between the theoretical methods and practical application. Then, a sound theory alone is not enough. A good theoretical methodology may not always be practically possible. For example, the Ziegler Nichols closed loop PID controller tuning method is widely taught in control courses. The method requires bringing the process to the brink of instability. Can you imagine using the Zeigler Nichols method to tune a critical refinery loop such as reactor temperature? Don t try it if you want to keep your job! Many other theoretical methods are quite well suited to practical application. Even when a theoretical methodology is practically well suited, challenges unique to the application always lurk around. What do you do when you have a beneficial but a relatively new solution? If the solution is practically viable, would you dare to try it? Very recently at FRI, we implemented Prof. Rhinehart s R-statistic method to detect steady state automatically from process variables. The method works on process variable (PV) time series data, and determines if a certain PV is at steady state or not. The R-statistic method had a strong statistical basis, was well tested, and was well suited for industrial implementation the perfect method, in theory. However, when the method moved from lab to process, there were new problems. Certain PVs showed significant autocorrelation an assumption that was quite valid, but had significant implications when violated. Certain other PVs had a discretized signal a feature unique to a particular FRI process. The method was ill-equipped to deal with those PVs. After a couple of iterations, a modified method was tried successfully. The modified method was theoretically less rigorous than the original R-statistic method, but worked more robustly with discretized PVs. As is the case with all methods, the new method also has its share of limitations. The new method requires constant maintenance in terms of adjusting tuning parameters frequently. Another modification is underway to improve both the methodology and implementation of the new method. Over the last two years, we have tried a sound theoretical method and failed several times. But, we have also learned from our failures and improved the method. Soon, we are confident that the new R-statistic method will provide FRI with automatic steady state identification a first in the automation world. Now, let s come back to our question about what defines an automation engineer. The dynamics of real world processes are quite complex that most methods in the automation engineer s toolkit are almost always imperfect solutions. As an automation engineer, you will be faced with difficult judgments several times in your career. Often times, there are two paths. A conservative path that will most likely succeed, but leaves opportunities and greater success behind, or a slightly aggressive path that could potentially lead to either failure or greater success. It is easy to be conservative and to not try something new. You might fear the risk of upsetting process or operators or management. That s what automation engineers do decide when to take that risk. It is tempting to avoid making such judgment calls but in the end, that s what defines an automation engineer. Dr. Anand Vennavelli is a Research Engineer at Fractionation Research Inc. in Stillwater, OK. He hold a BS in Chemical Engineering from Osmania University, Indian and MS and PhD degrees in chemical engineering from Oklahoma State University. YOU DON T KNOW WHAT YOU DON T KNOW DR. R. RUSSELL RHINEHART Apparently, I was not an easy kid to raise. Although Dad never said that, and Mom always said we were all wonderful children, I recall frequently being told, If you don t know, ask. One time, I was helping put groceries away, and put a cold thing in the refrigerator. I was proud to be able to help. But, that particular cold thing should have been put in the freezer, not the refrigerator. Days later they discovered that the thing had thawed and ruined. I didn t know that it had to be frozen. Dad said, If you don t know, ask. Another time, I had worn the rubber-soled boots in the house. I had been told not to wear them inside because they left black scuff-marks on the floor. When I saw the scuff-marks, and realized that I d be reprimanded, I decided to wipe them up. Paint thinner, I had learned was a great solvent; and expectedly, it made the wipe-up easy, and I could quickly get back outside, proud of my cleverness and solution. But, the paint thinner dissolved the floor finish. Dad said, If you don t know, ask. I did not know that I did not know. If I didn t know that I didn t know, what reason would I have to ask? I could not answer this question, and went to Dad for help. I remember it stumping him also. As a parallel message, with a very different story, here is a tale on one of my sons: I was sweeping the kitchen floor with a whisk broom, and my 4-year-old son wanted to help. Not even half as tall as the broom handle, Me do it, he said. I gave him the broom, so that he could OSU AUTOMATION SOCIETY NEWSLETTER 3

4 explore the magic of organizing dust, and become proud of partnering with his parents in contributing to the home. He took the broom where he could hold it, just above the bristles, and pulled it across the dust pile, spreading dirt, not combining it. Then he figured, pushing the dust might be the solution, and rearranged the broom to push. While enjoying his exploration, what I mostly saw was the loose end of the broom handle flying around, banging things in the kitchen, trying its best to accidently knock things off of the counter and shelves. So, I corralled the loose end in a big circle of my arms to contain it. I did not hold it in one spot, just kept it out of trouble. This, of course, added difficulty to what my son was finding to be not an easy task. He was realizing that he was not yet the Big Boy! he aspired to be. And he got mad about it. There are two ends to every stick. When a novice is focused on one end of the stick, the consequences at the other end are seeking to make an accident happen. The two stories are the same, and the context of this third experience may help combine them. In my 13 years in industry, I worked on 8 wholly different projects. Not only different technical topics, but with the diverse contexts of process development, manufacturing, product development, and human supervision. I don t think this less-than-twoyears-per-assignment situation is unusual. Consider that it takes a year to begin to understand the technology and its context. What this means is that you spend a large portion of your career as a novice, not as an expert. As a novice you don t know what you don t know. As a novice you are not aware of the other end of the stick. As a novice, you always need to be cautious, observant, self-critical, and seeking awareness of the implications. Yet, you need to act, and act urgently with incomplete information. It is never as simple as it seems. You will be a novice throughout your career. More importantly, you will be a novice in personal financial decisions, in raising children (who change yearly with each of their development stages), in choices of where to live and what organizations to join, or in planning for your long term future. As you make decisions, be continually seeking for the other end of the stick, and engage others to discover what you don t know. Unfortunately, education misdirects your perspectives, and thereby undermines your success. Students are trained (taught and tested) on trivial problems. Teachers need to isolate concepts so that the topics can be learned, and they test the student s skill on easy-to-create and easy-to-grade problems. Accordingly, students tend to think the one end of the stick is the complete story. Further, students are selected and promoted on their individual, independent performance. Accordingly, once hired, they tend to want to demonstrate personal individual performance, and tend to not ask for other s input and advice. School dupes you into thinking that there is one end to a stick; and that once you know something, you know it all. Don t be snookered by your education. Be continually seeking for the other end of the stick, and engage others to discover what you don t know. EXTRAMURAL LEARNING AND AUTOMATION ANAND GOVINDARAJAN Recently, I heard a friend of mine, Ani, say that she learnt more engineering in the first few months of her first job as a process control engineer in a process industry, than she had ever learnt all her life sitting inside a classroom. Soon, she began quoting anecdotal evidences about how what is being taught at the university is all theoretical and the industry is all about being real. What Ani went through was nothing but a bout of extramural learning and, learning without being taught. I can imagine the readers, scratching their heads, already. Do you remember, when your friend taught you an important science or engineering concept, and all of a sudden you seemed to grasp it? Or the time when your friend taught you how to create that subtle embellishment while playing a piece on your piano? This, despite your professor/instructor explaining the concept several times in class. Yeah! Those childhood instances were a form of extramural learning. Now, how is extramural learning and automation connected? Consider this - automation is a ubiquitous field, and there are no college programs in the US that lead to a degree in automation. Therefore, all the young engineers, like Ani, involved in automation have received most of their training outside of a university environment, and much of it informally. In other words most of the automation training is extramural! While it is true that there is significant extramural learning involved in all disciplines, the absence of institutional training for automation, amplifies the importance of extramural learning in automation. At the entry level, there are fewer trained engineers in the market for automation than the number of jobs available. Although, the International Society of Automation and the Automation Federation are making strong efforts to pitch automation as a professional discipline, and institutionalize the profession, they cannot completely address (today s) the problem of lack of trained, entry level, automation engineers. Now, you may ask - are most of the automation jobs lying vacant for want of trained personnel? Not exactly, every year hundreds of young engineers, like Ani, are taking up automation as a career choice or research focus, by choice and chance. Most of these engineers, are electrical, chemical or mechanical engineers by training. Ani, herself is a chemical engineer by training, and holds a Bachelor of Science degree. Ani s company put her through a few weeks of classroom training and then on an informal probationary role, or on-the-job training. During this on-the-job training period, on some days, Ani would be interacting with a plant operator, Linear, who had worked in the plant for a few decades. On other days, her senior colleagues Quadratic and Optima, at the plant helped her program on the DCS, and monitor their APC software. When Ani was not interacting with Linear or doing DCS work with the help of Quadratic and Optima, she was trying to learn by doing. She had a simulation copy of their DCS and APC program on the engineer s station, and used to refer to the P&IDs and trace all the loops to understand their functioning, and try to think of opportunities to improve them. After all, she firmly believed in what her university professor, Dr. Leapfrogging, said Do, not study. Almost a year into her job, Ani asks Optima if it is a good idea to cascade loop PIC-1 with FIC-2. Optima asks Ani, why she thinks cascading the loop PIC-1 would be a good idea? Drawing from her conversations with Linear and her own understanding by not studying, Ani went on to reason out and explain to Optima. Soon, Quadratic joins the conversation and along with Optima is now thoroughly convinced that cascading PIC- 1 is a great idea and will lead to lesser upsets in the process. What 4 DESTINATION SETPOINT

5 was a rather uneventful afternoon, turned out to be Animated. At the end of the conversation, Ani realized, she had been learning through all her conversations and discussions with colleagues around her. She had been learning without being taught. An exhilaration of learning happened. Thanks to extramural learning. Anand Govindarajan is a Development Engineer with Optimized Gas Treating, Inc. at their Technical Development Center in Buda, TX. Anand develops new capabilities for ProTreat simulation software. He received his B. Tech. (2008) from Anna University (India), and M.S.(2011) and Ph.D.(2014) degrees from Oklahoma State University, all in chemical engineering. He completed internships with Sun to Market Solutions, Sunnyvale, CA and Fractionation Research, Stillwater, OK. Anand was the founding member and President of the ISA-OSU Student Section, and President of the ChE Graduate Student Association. He was recognized as Young Innovator on the ISA InTech magazine, and received the ISA Foundation Scholarship, OSU Distinguished Graduate Fellowship. SOFT SKILLS AN ADDITIONAL POWER TO APC ENGINEERS UPASANA MANIMEGALAI SRIDHAR Soft skills, as the name does not suggest, is a powerful tool that supplements an Engineer s skill set. While almost universally true, for an Advanced Process Control Engineer (APC) this is even more a valued skill set to possess. Let me dig a little deeper into the experiences of three of my friends A, P, and C who were aspiring APC Engineers. A, a Chemical Engineering graduate student, wanted to secure an internship position during a summer as an APC engineer in an industry, to understand how it is to be one. It is already February, time is ticking and nothing is working out for A. One fine day, A receives a call from his senior at grad school, now working for an industry. As it happens, A misses the call and gets a voice message where he hears Hi A, I am calling to check if you are interested in working as an APC Intern for XYZ Inc. A was happy but restrained and calls his senior and says, Yes! I would definitely love to work as an APC intern this summer. A understands from his senior that his presentation skills were noticed by his senior s bosses during various events at the grad school and they were thoroughly impressed. At a chance meeting with the leadership team of XYZ Inc, A was recommended by his senior s bosses. Things moved forward and A had a telephonic interview with XYZ Inc. The rest, as they say was A s story. A aced the interview and secured an internship position at XYZ Inc. A could not thank his senior and his bosses enough for this opportunity. A realized the importance and impact of having developed his presentation skills, that lead to an internship position and helped experience being an APC Engineer. P s professor Dr. R. Cubed decided to send P and her friend to the ISA Automation Week. At the conference hotel, while looking for the registration desk, P and her friend heard a voice from behind calling out their names. With a surprised look on their faces they turned back to see two gentlemen who shook hands and introduced themselves. Understanding the question mark on their faces, the gentlemen explained that they read a few articles written by P and her friend in a magazine of the ISA. Totally impressed with these articles, they were scouring around the conference trying to meet P and her friend to congratulate them and appreciate their work. P and her friend later on got to know that these gentlemen were renowned professionals from a leading industry. They realized that through good writing, people recognize you and your skills. You can become known, without having to be introduced. They thanked their professor for mentoring their writing skills and providing them with opportunities to write articles. C was close to graduating with his graduate degree in Chemical Engineering. Last year, C worked as an APC Co-Op in a company. Thinking that he will take more than a year to graduate, C missed the interview process to be hired as a full time employee after graduation. This company had a process of hiring summer interns and Co-Ops for full time positions through an interview process that was conducted during the last week of their internship program. C expressed interest to his boss for returning as a Co-Op again next year. C s boss was happily willing to hire him again. C left the industry on a positive note, and kept in touch with his boss, regularly updating him about his progress at school. Due to a certain course of events, C decided to graduate earlier than the one year he had estimated. Once C finalized his change in plan of graduation, he apprised his previous boss about the recent developments and his graduation status. C had no hope about full time opportunities at the company where he worked as a Co-Op as he did not attend the interview process. To his surprise, C s previous boss decided to set up an interview process to talk about the possibilities of hiring C as a full time employee. C impressed everyone at the interview and was hired by the company. All C did was to stay in touch with his boss and communicate his progress periodically, even though he had completed his Co-Op. Through this constant communication, C s previous boss continued to remember C and hired him as a full time employee. A, P and C were armed with the power of soft skills, early in their career. This is truly essential. As APC Engineers, it is important to convince the production units in the industry with our work, as they believe in If it isn t broken, don t fix it. We need to present and convince the production units with our work which is to fix (even if not broken) to improve production. Our soft skills will come in handy at this point. As APC Engineers we also need to interact constantly with operators who are the primary personnel in-charge of operating the production units. It is important to communicate to them the importance of APC work, and how it helps the entire plant be more safe, stable and economic in a way that they understand and appreciate. Our communication skills help them understand and writing skills guide them through operating procedures which are easier for them to understand and follow. OSU AUTOMATION SOCIETY NEWSLETTER 5

6 Soft skills will definitely supplement the technical skill set of an APC Engineer who needs this skill set on a day to day basis. This is a secret to the success of being an APC Engineer. The secret of success for an APC engineer in additional to the technical skill set is: A Articulating views clearly P Penning thoughts lucidly C Communicating regularly Upasana Manimegelai Sridhar is a Dynamic Specialist in the Process Dynamics and Optimization group at Bayer MaterialScience, Baytown, TX. In her current role she is monitoring the performance of PID control loops in a production area. In 2013, she worked as a summer intern for the same group. Upasana received her B.Tech. (2009) from Sri Sivasubramania Nadar College of Engineering, Anna University (India), MS (2010) and PhD (2014) from Oklahoma State University (OSU), all in Chemical Engineering. Upasana was one of the founding members of the student section of ISA at OSU and she led the team that received the Section Excellence Award of ISA in She was recognized as Young Innovator on the ISA InTech magazine, and received the ISA Foundation Scholarship, OSU Distinguished Graduate Fellowship. EVALUATION OF AUTOMATED STEADY STATE & TRANSIENT STATE DETECTION ALGORITHMS SARVESH GORE AND R. RUSSELL RHINEHART ABSTRACT A comprehensive comparison of two statistical methods for automated identification of probable steady state and probable transient state in a noisy process signal is performed. The effectiveness of both methods with respect to performance, computational burden, ease of understanding, storage, etc. is analyzed for both step and ramp changes in a noisy process signal. INTRODUCTION Chemical processes are inherently noisy and non-stationary, but identification of steady state (SS) is often very important for online control actions related to both process analysis and model development. SS triggers process condition monitoring, model validation, economic adjustment, or point of transition. TS triggers data gathering for transient models, recording for historians, and diagnosis. Since the processes produce noisy data, statistical based methods are required for detecting probable steady and transient state (TS) conditions. An automated approach for SS/TS ID is preferred to human interpretation. However, ease of understanding the method is also important for implementation acceptance. Two different identification approaches are explored in this paper: First, an approach that calculates variances using first-order filters [1,3]. Second, an approach that calculates variance from mean values and uses an array for storage [2]. Both the approaches are easily implemented statistical methods with well-defined critical values for determining points of change [4,5]. METHOD Both approaches used in this paper utilize the R-statistic method, a ratio of two types of variances measured on the same set of data [3]. In order to reduce computation burden, one of the approaches uses digital firstorder filters to provide unbiased estimates of the average and variance (exponentially weighted moving average and exponentially weighted moving variance) instead of conventional average and variance. The other approach incrementally estimates average and variance using conventional methods. The filter concept is illustrated in Fig 1, where the dots represent the noisy measured data. The trend remains at steady state until samplecounter (time) of 20 is reached. Then it ramps from 15 to 10 and holds steady state after a time of 70. The filter approach estimates the filtered values of the process measurements represented by the solid line in Fig 1. Two types of variances are estimated. First, deviation d2, indicated in the upper left part of the figure, is the difference between measured value and the filtered value; and second, deviation d1, indicated in lower right corner, is difference between sequential data. If the process is at steady state, the filtered value is almost the average of the measured value, as shown in the 0 to 20 and 70 to 100 time period. Then, the variance estimated by d2, σ 2d2, will be ideally equal to variance estimated by d1, σ 2d1, and the ratio of the two types of variances, will be approximately unity, Alternately, if the process is at transient state, as in the ramp part of Fig 1, the filtered value lags behind and the variance estimated by d2, σ 2d2, is much larger than the variance estimated by d1, σ 2d1. So, the ratio of the variances given by R will be much greater than unity, Figure 1: Filter Concept Illustration [1] Three first-order digital filters are used in the filter approach in order to reduce computational burden in determining the variances. The first filtered value provides an estimate of the mean value of the data: X f,i =λ 1 X i +(1-λ 1 )X f,i-1 (1) X = the process variable X f = Filtered value of X λ 1 = Filter factor i = Time sampling index The variance obtained using another filter utilizes exponentially weighted moving variance based on the square of difference between data and the average: v 2 =λ (X -X f,i 2 i f,i-1 )2 +(1-λ 2 )v 2 (2) f,i-1 6 DESTINATION SETPOINT

7 v 2 = Filtered value of a measure of variance based on differences f, i between data and filtered values v 2 = Previous filtered value f, i-1 Then the ratio of the variances is estimated as: (8) Another approach is to obtain exponentially weighted moving variance (third filter) based on sequential data differences: δ 2 =λ (X -X f,i 3 i i-1 )2 +(1-λ 3 )δ 2 (3) f,i-1 δ 2 = Filtered value of a measure of variance f,i δ 2 = Previous filtered value f,i-1 Using Eqs (2) and (3), ratio of variances, R-statistic can be estimated as: (4) This method assumes that there is no auto-correlation in the process measured data. The filter tuning factors, lambda values can be related to number of data points as it influences the variance estimation. Lower values of lambda are equivalent to higher number of data points used for data analysis. The three sums are decremented and incremented at each sampling, and the data are stored in an array with a pointer that is incremented at each sampling. Regardless of the approach (filter or array) the value of the resulting ratio statistic needs to be interpreted. Fig 3 shows the distribution of R-statistic at steady state. The mean or average value is 1. However, the R-value will have variability to it due to random fluctuations in the measured data. If the R-statistic value is greater than R1-α, say1-α = 0.95, then it can be claimed with 95% confidence that the process is not at steady state condition. The filter approach reduces computational burden, but has more tuning factors. The storage required for the filter approach is also low, as only 9 variable values are to be stored. However, the filter structure is not very easy to understand for many who are familiar with conventional variance estimation. Also, a sampling window of width N is easier to understand compared to interpretation of three filter tuning factors, lambda values. The concept of the array approach is illustrated in Fig 2. The data used for the array approach is identical to the filter approach. The solid line represents the mean of the data. The variance estimated using d2, indicated in the upper left of Fig 2, is the difference between measured data and the mean of the data in the particular sampling interval. The variance estimated using d1, indicated in lower right of the figure, is the difference between sequential data. Figure 3: R-statistic distribution at steady state Fig 4 shows the R-statistic distribution at transient state. For a process which is not at steady state there is a high chance, almost 70% in this illustration, of transient state condition. So, for R> R1-α the odds of the R value coming from SS distribution to that of TS distribution are 1:15, in which case TS condition can be claimed. However, R< R1-α does not necessarily indicate SS condition. Here, the odds of R value coming from SS to that of TS distribution are 95% to 30%, i.e almost 3:1. This is not very definitive. Figure 2: The Array Concept Illustration First, the average is estimated for the array approach, (5) N = Sampling interval X = average The conventional variance is shown in an expanded form in Eq(6): (6) Figure 4: High probability of not being at steady state (TS condition) Fig 5 shows another critical value Rβ. If the process is at SS, then there is about 40% chance of the R value to be less than Rβ critical value. However, for TS condition, the chance of having R< Rβ is about 1%. So if R< Rβ the odds of the process being at SS are 40:1. If Rβ<R< R1-α, then one cannot confidently state that the process is either at SS or TS. But the chances of the process being at its previous condition are high in this case. Assuming no auto-correlation the variance estimated from the difference of the sequential data gives: (7) OSU AUTOMATION SOCIETY NEWSLETTER 7

8 array method was varied from 20 to 75 for each set of lambda for step test as well as ramp test. RESULTS AND DISCUSSION Figs 6 and 7 show for ARL and T-II vs. signal to noise ratio (s/n) for both methods. The diamond markers represent filter data and triangular markers represent the array method. The N and lambda values are selected so that the ARL is almost same for both the methods. Figure 5: Steady state identification METHODOLOGY Due to random variations and vagaries of the statistical methods, the automated SS/TS identification methods might show wrong outputs. Some of the major aspects that can be considered as performance metrics of the approaches mentioned above are: Type-I error: It is the trigger of not at steady state claim when the process is actually at steady state. This trigger might happen if due to noisy measured data, the R-statistic value crosses the upper critical value. This error is dependent on the choice of α value for determining the R1- α critical value. The obvious way to reduce the Type I error is to choose a smaller value of α. Type-II error: It is the trigger for at steady state claim when the process is actually at transient state. If a very small value of β is selected then, the Rβ value will shift to the left; and even if the process is at TS, the condition might not be triggered since R value did not cross the critical value. Average Run Length (ARL): This gives the number of data sample points the method takes to recognize the point of change. ARL depends on the length of the sample window. For the array method, average ARL is usually equal to twice the size of the sample window used. The approach for comparison of performance of the filter method and the array was to compare the negatives of the two methods. A noisy process signal was generated using VBA-Excel and the two methods were tested on that signal. The variables in the process were signal to noise ratio, critical values, N for the array method, λ 1, λ 2, λ 3 for the filter approach, T-I, T-II errors and ARL value. Following set of steps were used in order to carry out the simulations and generate data for comparing the two approaches: 1. Select signal to noise ratio. 2. Select N and λ 1, λ 2, λ 3 values. 3. Use a true Steady state data to estimate the R-critical values. 4. Perform step tests and ramp tests to replicate runs in order to reduce the effect of variability of the process on findings. 5. Obtain data for T-II and ARL for each method. 6. Change N(or λ 1, λ 2, λ 3 ) values and repeat steps 3,4 and Repeat the whole procedure for a different signal to noise ratio. 8. Choose lambda and N values, so that the ARL of both the methods is almost same and analyze the probability of T-II error vs s/n data. For the sake of this study, T-II and ARL were compared keeping the third negative, T-I almost negligible and uniform for all runs. The 1-α and β values selected for determining the critical values were 99.9% and 25%. A very high value of 1-α was selected for reducing the probability of T-I error to a negligible value. For ramp test the signal to noise ratio represents the ratio of slope of the ramp to the variance of the noise. For the step test, the signal to noise ratio represents the ratio of step change to the variance in the noise (which was kept constant at 0.5). N for the Figure 6: ARL vs signal to noise ratio for N=30 and λ1=0.2, λ2=0.1, λ3=0.1 for a step test Fig 7 illustrates that for almost the same ARL values, the array approach gives a better performance than the filter approach in terms of T-II error, meaning the array method has a lower probability of triggering the at steady state condition when the process is at transient state. So, for a step test, the array method is more sensitive to change in the process compared to the filter approach. For this data, the lambda values selected were λ1=0.2, λ2=0.1, λ3=0.1. The probability of T-II error depends on sample window used. A larger window length leads to a lower probability of T-II error. In order to verify the performance of the two methods over a range of N and lambda values, the same step test was performed for lower lambda values, λ1=0.1, λ2=0.05, λ3=0.05, thus increasing the number of data points considered for filter approach. Figs 8 and 9 show that for the larger sample window, the array method again shows better performance in determining TS compared to the filter method Figure 7: T-II error vs signal to noise ratio for N=30 and λ1=0.2, λ2=0.1, λ3=0.1 for a step test 8 DESTINATION SETPOINT

9 Figure 8: ARL vs signal to noise ratio for N=70 and λ1=0.1, λ2=0.05, λ3=0.05 for a step test Similar simulations were carried out for ramp tests. Figs 10 and 11 show the data for ramp simulations for lambda values, λ1=0.2, λ2=0.1, λ3=0.1 and N = 25. Fig 11 illustrates that for almost same ARL values, the performance based on probability of T-II error of both the methods for a ramp test is similar for lower signal to noise ratio. However, for higher s/n ratio, the array approach shows improved performance and more sensitivity to TS detection. Figs 12 and 13 show the data for a sample window for a ramp test. For same ARL, the two methods show almost similar performance. Figure 11: T-II error vs signal to noise ratio for N=25 and λ1=0.2, λ2=0.1, λ3=0.1 for a ramp test Figure 12: ARL vs signal to noise ratio for N=60 and λ1=0.05, λ2=0.05, λ3=0.05 for a ramp test Figure 9: T-II error vs signal to noise ratio for N=70 and λ1=0.1, λ2=0.05, λ3=0.05 for a step test Figure 13: T-II error vs signal to noise ratio for N=60 and λ1=0.05, λ2=0.05, λ3=0.05 for a ramp test Figure 10: ARL vs signal to noise ratio for N=25 and λ1=0.2, λ2=0.1, λ3=0.1 for a ramp test COMPARISON BASED ON FUNCTIONING Performance of the two approaches based on the ability to determine change of state is very important. However, there are also some in-use issues which must be considered for smooth execution: Ease of Understanding: The concept of conventional average used in array approach is comparatively easier to understand as compared to concept of three first order filters as used in filter approach Storage: The filter approach has less than 10 storage variables as OSU AUTOMATION SOCIETY NEWSLETTER 9

10 compared to the array approach which stores more variables than the length of window used. Computational burden: The filter approach takes computational time as it uses simple first order filters. The array approach computes square of number of terms, sum of squares as well as squared deviation. On an average, filter approach is twice as fast as the array approach Ease of tuning: The array approach can be tuned using one tuning factor, window size, N and is less complicated to tune. The filter approach has three tuning factors, λ1, λ2, λ3, which can be difficult if the impact of each factor on window size is unknown. CONCLUSION A comparison of two statistically based methods for automated identification of steady state and transient state was made based on the performance of the methods in successfully determining transient state in the process. The two techniques were investigated and analyzed based on the computational burden, storage, ease of understanding the techniques and flexibility of the techniques. The array method showed improved performance (lower T-II errors for equivalent T-I and ARL) for a step test. However, the performance difference of the two methods was inconsequential for the ramp tests. The filter method is computationally less expensive and requires less storage compared to the array method. Also, the filter method has flexibility in tuning the method due to three factors, λ 1, λ 2, λ 3. However, the conventional variance measure and the definite window length, N, of the array method makes it easier to understand, apply, and interpret. REFERENCES 1. Huang, Ting; Rhinehart, R.Russell, Steady state and Transient State Identification for flow rate on a pilot-scale absorption column, American Control Conference (ACC), 2013, vol., no., pp.4498,4503, June Rhinehart, R.Russell, Automated steady and transient state identification in noisy processes, American Control Conference (ACC), 2013, vol., no., pp.4477,4493, June Cao, S., and R. R. Rhinehart, An Efficient Method for On-Line Identification of Steady-State, Journal of Process Control, Vol. 5, No 6, 1995, pp Cao, S., and R. R. Rhinehart, Critical Values for a Steady-State Identifier, Journal of Process Control, Vol. 7, No. 2, 1997, pp Shrowti, N., K. Vilankar, and R. R. Rhinehart, Type-II Critical Values for a Steady-State Identifier, Journal of Process Control, Vol. 20, No. 7, pp , Sarvesh Gore is currently a Technical Consultant at Aspen Technology. He received his BS in Chemical Engineering from Visvesvaraya Technological University, India and MS in Chemical Engineering from Oklahoma State University. Sarvesh can be contacted at sarvesh.gore@ okstate.edu and Dr. Rhinehart can be contacted at rrr@okstate.edu. SMART X = AUTOMATION + X SOWMYA NELLUTLA The views expressed in the article belong to the author alone and don t necessarily represent any individual or organization s positions, strategies or opinions. Automation and Smart are the buzz words now. In the title of this article, x symbolizes a variable. Be it a conventional technology or a process addition of automation leads to a smart system. A smart system leads to a smart industry. A smart industry builds cities and further leads to smarter nations. Integrating automation into a process leads to a smarter technology discussions, articles, inventions - all are bent on proving this fact. Having followed the automation news in India very closely, it was clearly a year of investments from top-notch companies in home automation, power automation, healthcare, and media to name a few. In Indian power sector, Solar took the major chunk of headlines and investments. In the energy industry, efficiency and reliability play a major role. In a huge country such as India with an estimated 1.22 billion people, maintaining reliability in the power delivery process is a huge task. Several large companies are currently investing to improve efficiencies through automation. Smart investments. One of the impressive recent headlines from The Hindu, a leading newspaper of India read Automated, self-cleaning e-toilets for Tirupati (01/14/2015). Tirupati in south India is a pilgrimage temple city. An estimated million domestic tourists visited in 2013 (The Hindu, 23 July 2014). With a huge average influx of 60,000 visitors daily, the basic amenities have to be reviewed on a timely basis. The temple city caught up with times with use of biometrics for ticketing system, automated sweets making machines for the pilgrims. Hygiene here is of the utmost importance. The introduction of the automated e-toilets here is a huge step for a developing nation such as India, with a huge rural population. These installations will help maintain the cleanliness in the cities and give access to clean toilets in the rural areas. Smart installations. Automatophobia (not in dictionary, should be added) - Fear of automation and its threats to jobs. Not everything is written positively about automation in the media which is a little depressing though. Automation threatening employment and jobs also hit the headlines around the world. It needs to be understood that in order to automate automated installations are required installation parts are requiredinstallation parts manufacturing is required which further might be another automated process tree initial automated design is required education is required. Education will then become a basic amenity. According to me, Automation instead of slowing the progress, speeds it up. It is upon the bigwigs and enthusiasts of the Automation stream to explain the pros. Smart Thinking. Summation of smart investments, smart installations and smart thinking will lead to smart living. REFERENCES Sowmya Nellutla is an Associate Engineer at CG Power Solutions, Inc. She received her BS in Electrical Engineering from Jawaharlal Nehru Technological University, India and MS in Electrical Engineering from Oklahoma State University. 10 DESTINATION SETPOINT

11 FOUR OSU STUDENTS RECEIVE ISA SCHOLARSHIPS The International Society of Automation (ISA) has announced over $3,400 in scholarship to four OSU students. Three of the students received the ISA Educational Foundation scholarship and are all seeking PhDs in their diverse research fields. Samyukta Koteeswaran and Suresh Jayaraman are from the School of Chemical Engineering, and Sunil Lakakulla is from the School of Electrical and Computer Engineering. Sandeep Vysyaraju, the fourth scholarship recipient was awarded the ISA Process Measurement and Control Division scholarship for He is pursuing an MS from the School of Electrical and Computer Engineering. Samyukta Koteeswaran sees this as a positive sign. She is a PhD student in Chemical Engineering and her research is on Interaction of Oil and Gas Shale with Drilling Mud in Fracturing Fluids. She has been closely associated with the OSU Automation Society (an ISA Student Section) since 2012, the year she joined OSU as a Master s student. She served as General Secretary from and currently is the student section President. Samyukta says, Working in research related to petroleum I was initially skeptical about being a part of OSUAS, but the then officers and Dr. Rhinehart explained to me about how each industry, even if it is petroleum, the backbone is Automation. Now I encourage hundreds of other students with the same skepticism to be a part of our society understand and learn the importance of Automation and Process Control in our everyday lives. ISA scholarship has boosted my confidence and reinstated the fact that you can contribute to the field of Automation in many other ways. She also states that she will continue being a part of the Automation Society even after she graduates; and that she considers herself as an integral part of the society. Samyukta is very enthusiastic about OSU Automation Society and its students being identified at the national level and sees this as a welcoming sign for the new student members. Suresh Jayaraman, who is receiving this scholarship for the second time, attributes it to the active participation in OSU Automation Society and his research. Suresh says, I have been continuously associated with OSU automation society, a student chapter of International Society of Automation (ISA) in various leadership roles. I was serving as Public Relations Officer (PRO) in , PRO and Alumni Networking officer during My association with the OSU automation society has given me an opportunity to co ordinate with fellow graduate students and people from automation industry. My involvement with the OSU automation society played a vital role in receiving Educational Foundation Scholarship from ISA for two consecutive years. I am currently working on Modeling and Optimization of Algal growth in order to make the production of bio fuel & other bio products from algae, economically viable. Suresh believes that teaching is his forte and he has a future in it. Suresh is also teaching Engineering Computer Programming course (ENGR 1412) for 2 sections of 20 students each. He likes interacting with students and also believe teaching is a great way to pass on knowledge. Suresh regards teaching as a two way processes where you just don t stop with teaching but also learn from students. Sunil Lakkakula is a PhD Student in Electrical Engineering. He is an active participant, and was Vice-President and Design Contest Chair for the OSU Automation Society for the year Sunil shares with us, Since I became a member of the International Society of Automation through the OSU Automation Society at Oklahoma State University, I have developed a special interest in the field of automation. I have served as the Vice-President and Design Contest Chair for the OSU Automation Society for the year and am a continuing member of the society. I thank OSU Automation Society for giving me a better exposure and knowledge towards the automation field through its activities and field trips to various automation industries. ISA recognizes its student members and student chapters by giving various scholarships and awards. Our student section at Oklahoma State has received the Section Excellence Award for the year I am very happy that I am a recipient of the ISA Educational Foundation Scholarship recipient for the year I encourage all the students interested in automation career to take advantage of the opportunities provided by the ISA and excel in your area of study. In my research I am working on CMOS circuit optimization for speed and power by determining the optimum transistor widths using a simplified RC delay model which will help the circuit designers to quickly and efficiently meet the design performance specifications. Sandeep Vysyaraju s interests in the field of automation began during his undergraduate days, and this motivated him to pursue his masters in the same field. His interests lie in process control and measurement. Sandeep is the current Public Relations Officer and Alumni Affairs Coordinator for OSUAS for the year Sandeep says, I am very much thankful for the experiences that I gained being the officer for Automation Society as well as the lessons and support I got from Dr. Russ Rhinehart. I received ISA PMCD Scholarship award for the year 2015 from ISA which indeed makes me very much honored and privileged. My core interest lies in automation and am ready to serve the field of automation. I encourage all the students interested in automation career to take advantage of the opportunities provided by the ISA and excel in your area of study. OSU AUTOMATION SOCIETY NEWSLETTER 11

12 MEMBER S EXPERIENCE SURESH JAYARAMAN As an active member of OSU Automation Society (OSUAS) for the past three years, it s my pleasure to share my experience with the society. My interest towards Automation and Process control made me to opt for the Advanced Process Control course in spring The course taught me various controller techniques, and Dr.Rhinehart s homework helped me big time to understand the application of controllers in a wide spectrum ranging from a car s cruise control to a pilot scale packed absorption tower in the Unit Operations Lab. And my interest to learn more about Automation out of my discipline motivated me to become a part of the Automation Society since my first semester in the OSU. I have held various positions in the OSUAS board for the past three years and every event has been nothing but a learning experience for me. OSUAS served as a platform for many of the members to interact with people from various automation industries like Yokogawa, GE, Exxon Mobil, and Georgia-Pacific. Every guest lecture has motivated and inspired in every possible way to achieve in the field of Automation. The industrial visits has truly helped to bridge the gap between academe and industry. I still remember our visit to Phillips 66 controller room last semester and it was really amazing to see how such a big refinery is being controlled and operated in a room by a bunch of operators. Apart from these technical events, OSUAS also organizes social events like potluck, picnic, and bowling. These social events are helpful to take a break from school work and make some good friends outside one s department. I was awarded ISA Educational Scholarship by International Society of Automation for two consecutive years, and my association with OSUAS has played a vital role in this. I am pretty sure that the experience and knowledge gained through the society will take me a long way to achieve my goals in the field of Automation and Controls. I would strongly encourage everyone to become a part of the Automation Society and to gain the most of it. Suresh Jayaraman is a PhD student at Oklahoma State University. He holds an MS degree in Environmental Engineering from University of Houston and a BTech in Chemical Engineering from Anna University. OSUAS TEAM OSUAS Board Members from L to R: Rakesh Ravishankar (CS), Webmaster; Ram Kumar Issaki (BAE), Social and Networking Chair; Vidhya Venugopal (ChE), Vice-President; Samyukta Koteeswaran (ChE), President; Dr. R. Russell Rhinhart, Faculty Advisor; Sandeep Vysyaraju (ECE), Public Relations Officer and Alumni Affairs Coordinator; Shreyas Sule (ECE), Program Chair; Amey Thorat (ChE), Secretary. Not in picture: Ryan Todd Hinchey (MechE), Treasurer; Savyasachi Gupta (ECE), Automation Contest Chair 12 DESTINATION SETPOINT

13 LOOKING BACK Every year, OSUAS maintains a traditional potluck dinner at Dr. Rhinehart s house, marking the first event for a new OSUAS team. This event gives a great opportunity to the new and old officers to interact with each other and with Dr. and Mrs. Rhinehart in a social environment. The first social event was a picnic at Boomer lake. We were joined by Dr. and Mrs. Rhinhart and their grandson, Kain. Several new students also attended the event, where we played games and had an interactive session to get to know one another. The first site visit organized by the new OSUAS team was to the Braum s factory in Tuttle, Oklahoma. The students got a chance to see the applications of automation in the food industry. The facility houses the ice cream production and bakery. In September 2014, OSUAS and its members were fortunate to visit the Fractionation Research Inc. (FRI) in Stillwater, OK. They had a tour of the distillation facility and the control room. They were able to interact with the engineers at FRI and get an in-depth look into the applications of automation and control in the chemical industry. The OSU Automation Society (OSUAS) attended the International Students and Scholars (ISS) graduate orientation of Fall 2014 to publicize the events and activities of OSUAS. Around 80 students, both undergraduate and graduate from various disciplines, became new members of the OSUAS. OSU AUTOMATION SOCIETY NEWSLETTER 13

14 Dr. Rhinehart shared his experiences on working in the industry as well as in an academic setting. His presentation entitled, Bridging the gap gave insight into the difference in problem solving approaches required by academic and industrial environments. OSUAS team experienced a different dimension of automation through a visit to Dr. Joe Cecil s laboratory at OSU. His research work encompasses development of virtual reality applications for educational and industrial training purposes. OSUAS also visited the control room of Philips-66, which processes 1.8 million barrels per day of crude. The control engineers explained and presented on the applications of advanced process control in refining. OSUAS hosted a networking event with Dr. Maurice Wilkins, Vice- President, Global Marketing Center, Yokogawa Corporation of America. Dr. Wilkins gave a presentation titled, Are machines better than humans in a crisis? The networking was conducted at Hideway Pizza in Stillwater, OK. OSUAS was proud to host Automation block during Research Week 2015 at OSU. We had presentations on the application of modeling, control, and automation in areas like biofuels, wireless technology, embedded systems, renewable energy, and non-stationary dynamical systems. Dr. Prabhakar Pagilla (MAE, OSU), Dr. Fred Striz (AME, OU), and Mr. Marc Trammell (Senior plant engineer, OG&E) were the three judges.

15 HOW CAN YOU HELP WITH OUR CONTINUED SUCCESS? OSU Automation Society is open to conducting and hosting networking events with industry professionals, which can be used as a platform to recruit aspiring professionals interested in these fields. If you and/or your organization would like to schedule an event with OSUAS, please contact Dr. R. Russell Rhinehart at rrr@okstate.edu. Donations to the OSU Automation Society can be made through multiple channels: 1. By visiting the website and selecting a gift for the Account Process Control and Optimization Engineering Workforce Development Program. Recurring gifts starting and ending anytime can be made. Several employers make matching contributions to encourage their employees to donate to higher education. Please make sure to check on the site, if your employer is part of the matching contribution program. 2. By Check to the OSU Foundation 400 S. Monroe St., Stillwater, OK to be placed in the OSU Foundation Account Process Control and Optimization Engineering Workforce Development Program. 3. By Check to the OSU Automation Society Attn: Dr. R. Russell Rhinehart, 423 Engineering North, School of Chemical Engineering, Oklahoma State University, Stillwater, OK ACKNOWLEDGEMENTS The success of OSUAS events for the year would not have possible without the support, guidance, and assistance from several individuals and organizations alike. We take this opportunity to thank all of them. We would like to thank the following organizations for the field visits: Braum s Bakery and Processing Plant, Tuttle, OK; Fractionation Research Inc., Stillwater, OK; Phillips 66, Ponca City, OK. We would also like to thank: Dr. Maurice Wilkins, VP, Global Marketing Center, Yokogawa Corporation of America Dr. Harold Wade, Principal, Wade Associates, Inc. Mr. Michael Moan, CRAD Program Engineer, Raytheon Dr. Alfred Striz, Aerospace and Mechanical Engineering, OU Dr. Prabhakar Pagilla, Mechanical and Aerospace Engineering, OSU Mr. Marc Trammell, Senior Plant Engineer, OG&E Mr. Justin Dean, Managing Partner and President, Sahoma Controlware Dr. Joe Cecil, Industrial Engineering and Management, OSU Mr. Paul Galeski, CEO and Founder, Maverick Technology Dr. Craig Satterfield, Associate Director, Graduate College, OSU Ms. Andrea Payne, Chair, Research Symposium Committee, OSU Mr. Michael Louis Gumbs Betancourt, SGA AFAP Chairman, OSU Mr. Lance Millis, Director, Student Academic Services, OSU Mr. Cole Griffin, President of Student Council, CEAT - OSU Our supporters and well-wishers: OSUAS team Office of Campus Life, OSU Chemical Engineering Graduate Student Association (ChEGSA) Dr. Rob Whiteley, John Hendrix Chair of Engineering, Professor and Head, School of Chemical Engineering, OSU Dr. Clint Aichele, Assistant Professor and Harold Courson Chair of Petroleum Engineering, OSU Dr. Josh Ramsey, Assistant Professor, School of Chemical Engineering, OSU Dr. Sundar Madihally, Associate Professor, Edward Joullian Chair in Engineering, and Graduate Program Director, OSU Our logistics, marketing, and publicity support: Kassidy Conway, Student Union Marketing Communications Coordinator, OSU Wravenna Bloomberg, Marketing Manager, CEAT, OSU Lora Polson, Admin Support Specialist, CEAT, OSU Debbie Clemons, Stewardship and External Relations Coordinator, OSU Mr. Sunil Kumar Lakkakulla, Graduate Research Assistant, Electrical and Computer Engineering, OSU OSU-IT HireOSUgrads AUTOMATION DAY SPONSORS Dr. Harold Wade Dr. R. Russell Rhinehart Mr. Michael Moan Mr. Rohit Kadam Dr. Anand Govindarajan Dr. Upasana Manimegalai Sridhar Dr. Yash Tamhankar OSU AUTOMATION SOCIETY NEWSLETTER 15

16