New-ACE

Andrew Ogden-Swift is currently Director of Technology Strategy for Honeywell Process Solutions Research and Development Department. Honeywell Process Solutions is a $2.5billion/year supplier of process instrumentation, control systems and advanced software applications and is part of Honeywell International Inc. Andrew graduated in 1980 is Chemical Engineering from the University of Leeds and has worked in process control and related areas for various companies including Esso and ABB.

Abstract: Process Control – Current Status and Industry Needs

Over recent decades huge strides have been made in process control. Very many sites have digital control systems with many different types of control algorithms such as PID. In some industries widespread use is made of more advanced methods such as model-predictive control and soft sensing. There is some limited application of non-linear control where the benefits warrant the expense. However, there is a continued loss of expertise in developed economies and lack of expertise in emerging economies where much of the new process plant construction is taking place. Furthermore competitive pressures are driving operating companies to want to achieve more performance at lower cost. This is then leading to new needs to drive expertise and cost out of deploying the systems while still focusing on generating ever better safety, reliability and efficiency.

Professor Barry Lennox School of Electrical & Electronic Engineering - Uni of Manchester

Barry Lennox is a Professor of Applied Control in the School of Electrical and Electronic Engineering at the University of Manchester. His research interests are in data analysis, condition monitoring and process control. Professor Lennox has been the PI on several industrial and research council funded projects with a value exceeding £1.5M. He has acted as a research consultant for many companies in this field of data analysis and process monitoring, including BP, Invensys, HBOS and BNFL and was the co-founder of Perceptive Engineering Ltd (www.perceptive-engineering.co.uk), an industrial process control and data analysis solutions provider. He is a Chartered Engineer, a Fellow of the IET, Senior Member of the IEEE and Member of the InstMC. He has published more than 120 journal and conference papers

Modelling and Control of Industrial Batch Processes There is an increasing emphasis in the process industries on producing small quantities of high value products. This is particularly the case in the pharmaceutical industry, where more materials are being produced in processes that operate in a batch manner. The task of ensuring optimal performance of industrial batch processes represents a very difficult challenge to systems and control engineers. Highly non-linear dynamics, large batch to batch variations, in addition to the difficulties in measuring certain quality variables means that maintaining consistent operation of such systems can be problematic. This can be a concern, where it is often imperative that operation be maintained within strict limits. Pharmaceutical systems can be highly sensitive to abnormal changes in operating conditions and to ensure consistent production it is necessary to detect any indications of abnormal operation as soon as possible so that corrective action can be taken before the conditions have a significant impact upon the batch. Furthermore, as part of the procedures to guarantee product chemical consistency, regulatory authorities (such as the FDA in the USA) demand proof, for many compounds, that consistent operation is adhered to. Without this the product cannot be sold. This presentation will describe the various methods that are available for modelling batch processes and will demonstrate how these techniques can be used to provide long range forecasting of batch properties and enable advanced control to be applied. The techniques will be demonstrated using applications from the steel, pharmaceutical and fine chemical industries.

Nina Thornhill, ABB/RAEng Professor of Process Automation
Centre for Process Systems Engineering
Department of Chemical Engineering
Imperial College London
South Kensington Campus, London SW7 2AZ
Tel: +44 (0)20 7594 6622, Fax: +44 (0)20 7594 6606
Web: http://www3.imperial.ac.uk/people/n.thornhill

Nina Thornhill holds the ABB/RAEng Research Chair in Process Automation in the Department of Chemical Engineering at Imperial College London. She read physics at Oxford and joined UCL as a lecturer after six years with ICI. She moved to Imperial in 2007 having been involved with the Imperial/UCL Centre for Process Systems Engineering for many years. Her research addresses industrial data analysis using time series analysis and signal processing with applications in oil and gas, chemicals and electricity supply, and has included placements with BP and ABB.

Open challenges in control loop performance assessment and diagnosis Tools for performance assessment of control loops are well established, however it is correct diagnosis that reduces maintenance costs. The talk will review the academic and industrial state of the art in control loop performance assessment, with a focus on methods for reaching a definitive diagnosis of control loop problems. It will show that diagnosis of control loop nonlinearity such as dead band or stiction in control valves is now a solved problem, however definitive signatures remain elusive for control loop tuning faults and for disturbances caused by process problems. The talk will present some case studies and ideas towards diagnosis of these remaining problems.

I joined Emerson in October 2008 where I work as Director of Global Chemical Industry Solutions. My remit is to bring a global, cross-divisional perspective to sales and marketing activities that relate to the chemical industries (which includes everything from petrochemicals to personal care products).

Prior to joining Emerson I spent nearly 20 years in various roles in the petrochemicals industry, mainly in the UK. I am a Chemical Engineer by training, and initially started work as a site control engineer for ICI Petrochemicals. I spent a couple of years seconded to the UK Government in London, where I was responsible for the policy and funding of engineering and science research in academia. Following that I returned to ICI as a business planner, which developed into a role in business strategy, before returning to automation as the manager of process control and manufacturing systems engineering for the UK. During this time I had the disturbing experience of having the business that I worked for sold twice, first to Huntsman and then to SABIC.

While my first degree is in chemical engineering from Cambridge, I have also got an MBA from Durham Business School in the UK where I continue to develop my interest in the application of complexity theory to innovation and the behaviour of economic systems as a Visiting Fellow.

Together with my wife Penny, young son Seth and clumber spaniel Custard, I live in the north of England near Hexham.

Process Control and Business Optimisation: A Heretical Approach An established convention has emerged in the process industries about how process control, process optimisation and business optimisation interact with each other. Typically a hierarchical structure is used in which ever more simplistic models are used to drive decisions of ever increasing value and impact as you move from the operation of the production facility to the management of the enterprise. This is a problematic formulation, albeit a internally consistent and self-fulfilling one. In my presentation I will examine the inherent problems and contradictions with the conventional approach, postulate a set of alternative constructions and propose a variety of potential research questions that remain unanswered.

Dr Stephen Kemp
Control, Systems Engineering and Robotics Portfolio Manager Materials, Mechanical and Medical Engineering Programme
Engineering and Physical sciences Research Council
Polaris House
Swindon SN2 1ET
tel: 01793444040
www.epsrc.ac.uk

Stephen's background is in organic chemistry, specifically natural product synthesis. Having carried out research to postdoctoral level in academic and industrial settings he decided that being at the coal face was a bit too close and realised that he wanted a broader perspective on research. This led to a move in 2009 to his current position at EPSRC as Portfolio Manager for Control, Systems Engineering and Robotics.

An introduction to EPSRC with a look at the support available, how to apply and the process used to choose which proposals are funded. This will also feature a discussion of EPSRC's perceptions of the UK control research base and opportunities that come from this.

Professor Sarah Spurgeon FREng
Professor of Control Engineering
Head of the School of Engineering and Digital Arts
University of Kent

Sarah K Spurgeon received the B.Sc. and D.Phil. degrees from the University of York, York, U.K., in 1985 and 1988, respectively.  She has held academic positions at the University of Loughborough and the University of Leicester in the UK and is currently Professor of Control Engineering and Head of the School of Engineering and Digital Arts at the University of Kent. She is a member of the Editorial Board of the International Journal of Systems Science, a member of the Editorial Board of the IET Proceedings D, a Subject Editor for the International Journal of Robust and Nonlinear Control and an Editor of the IMA Journal of Mathematical Control and Information. Her research interests are in the area of robust nonlinear control and estimation, particularly via sliding mode techniques in which area she has published in excess of 250 refereed papers.  Professor  Spurgeon received the IEEE Millennium Medal in 2000. She is a Fellow of the IET, a Fellow of the IMA, a Fellow of the InstMC, and was elected a Fellow of the Royal Academy of Engineering in 2008.

Sliding mode control and observation techniques with application to a fermentation process

Sliding mode control and estimation systems use unique properties of differential equations with discontinuous right-hand sides which provide total robustness to a substantial class of parameter changes and external disturbance signals. Essentially, the output response of the differential equation is invariant despite the presence of uncertain parameters and external disturbance signals. Dynamic performance requirements are met by prescribing an appropriate manifold and ensuring the trajectories of the system of interest are constrained to lie on the manifold by selection of an appropriate discontinuous injection signal. This presentation will first briefly describe the fundamentals of sliding mode theory. An observer based control using sliding mode techniques will then be developed to obviate the need to measure biomass concentration in a fermentation system.  A new variable, the substrate consumption rate, consisting of a combination of substrate concentration, biomass concentration, specific growth rate and yield production coefficient is introduced to simplify the nonlinear differential equations of the fermentation process. A sliding mode observer is then developed to estimate this substrate consumption using readily available measurements of substrate concentration. It is shown that the sliding mode exhibited by the corresponding observer error dynamics is exponentially stable. This parameterisation and the resulting estimate of biomass concentration is then utilised within a feedback control strategy. Nonlinear simulation results demonstrate the robustness of the observer based control in the presence of both parameter uncertainties and external disturbances.

Prof. Weiss George 
Department of Electrical Engineering - Systems School of Electrical Engineering, Faculty of Engineering
Tel Aviv University, Ramat Aviv, Israel 

George Weiss received the Control Engineer degree from the Polytechnic Institute of Bucharest, Romania, in 1981 and the Ph.D. degree in Applied Mathematics from the Weizmann Institute,Rehovot, Israel, in 1989. He has been working at Brown University (Providence), Virginia Tech (Blacksburg), the Weizmann Institute (Rehovot), Ben-Gurion University (Beer Sheva), the University of Exeter, Imperial College London, and currently he is with Tel Aviv University. His research interests are distributed parameter systems, operator semigroups, control applied in power electronics, repetitive control and periodic linear systems. He is also doing work as a consultant in power electronics. He is a co-author (with Marius Tucsnak) of the book "Observation and Control
for Operator Semigroups" (Birkhauser, 2009).

In many areas of science and engineering it is important to estimate the initial (or the final) state of a linear distributed parameter system (DPS) from its input and output functions measured over some finite time interval (this is needed, e.g., in oceanography, meteorology and medical imaging). An infinite-dimensional system is called exactly observable in time if the problem of estimating the initial state from input and output data measured over a time interval of length is well-posed. For linear systems, the problem can always be reduced to an equivalent problem where the input function is zero, and we want to estimate the initial state based on the output function.
Suppose now that we have an exactly observable linear DPS. The formula for expressing the initial state from the measured segment of the output function involves inverting the Gramian operator of the system (see, for instance, [7, Section 6.1]), and this may be numerically very challenging. Today we have advanced numerical solvers for PDEs and it would be good to be able to use these in order to estimate the initial state. If we find a stabilizing output injection operator for the system (its existence follows from exact observability), then we can use a numerical solver to implement an observer for the DPS in order to estimate its current state. The estimation error tends to zero exponentially as
time goes to infinity. From the final state we can, in principle, recover the initial state if the dynamics are time reversible. The purpose of this research is to describe a way in which we can estimate the initial state of a linear DPS by operating only on a finite segment of output data. In short, the idea is to scan the same segment of data back and forth several times, using two observers,
one working in forward time, and one in backward time. For a general analysis of this idea and for its history we refer to [3]. An abstract formulation of a related algorithm, suitable for skew-adjoint generators and bounded observation operators, has been given in [1]. Linear DPS often have unbounded control and/or observation operators. This is often the consequence of boundary control and/or boundary observation (see Tucsnak and Weiss [7] for an elementary introduction to this topic). To make our basic ideas more easily understandable, we give in this extended abstract a short presentation of the simple particular case when the observation operator is bounded, the output injection operators
are also bounded and the semigroup is invertible (i.e., the observed system is time reversible). However, we emphasize that our results do not require the observation and output injection operators to be bounded (not even admissible), and we also do not require the observed system to be time reversible.

Professor Murray Dalziel

Director of the University of Liverpool Management School

Murray Dalziel is Professor of Management and Director of The University of Liverpool Management School (ULMS).  He is leading ULMS in its next stage of expansion as a world-class center for developing leaders. 

Murray’s current research interests focus on innovation strategies.  He has also published in the areas of change management, leadership development and human resource management.

Before joining ULMS, Murray Dalziel was Group Managing Director of Hay Group and in charge of Global Practices. He was based in Hay Group's corporate headquarters in Philadelphia. From 1991 he was a member of the Hay Group Executive and served in a variety of senior leadership roles in Europe and North America.

 In his nearly 30 years in professional services, Murray worked with large global clients in every continent on their key leadership and organization development issues.   His most recent clients include Microsoft, Novartis, TD Bank Financial Group.  He has also worked with companies such as Unilever, PepsiCo, IBM, BHP Billiton, ICI and General Electric.  Murray is also actively involved with a number of North American and European venture capital funds on the development of emerging businesses.

Murray grew up in Glasgow.  He received his MA in Sociology from the University of Edinburgh, and has a PhD. in Sociology from Harvard University.  He is also a Fellow of Royal Society of Arts.  He is a member of Academy of Management and European the European Academy of Management.

Yes you can shepherd cats!  Lessons from flocking for leaders of professionals

Conventional wisdom is that it is especially difficult to manage highly trained professionals whose professional identity and personal direction may transcend the organisation in which they work.  The paper questions typical advice that these leaders receive such as centrality of providing vision or avoiding doing the professional work yourself.  Using concepts from “flocking” and “swarming” theory the paper analyses the dynamics of professional organisations and how leaders can learn to build change in a dynamic environment.  The paper addresses a number of key questions:  why you only need a few people to support change but where they are situated is critical (and it’s not at the top); why leading from the front is dangerous; why you need to do professional work yourself as a leader. 

Prof. Pedro Albertos
Universidad Politecnica de Valencia (UPV)
Department of Systems Eng. and Control (DISA)
C/ Vera s/n, 46022 Valencia, Spain

e-mail: pedro(at)aii.upv.es

Pedro Albertos, past president of IFAC (the International Federation of Automatic Control) in 1999-2002, and Senior Member of IEEE, is a world recognized expert in real-time control, leading several projects in the field.
Full Professor since 1975, he is currently at the Systems Engineering and Control Department of UPV. He is Doctor Honoris Causa from Oulu University (Finland) and Bucharest Polytechnic (Rumania). Invited Professor in more than 20 Universities, he delivered seminars in more than 30 universities and research centers. Authored over 300 papers, book chapters and congress communications, co-editor of 7 books and co-author of "Multivariable Control Systems" (Springer 2004) and "Feedback and Control for Everyone" (Springer 2010). He is also associated editor of Control Engineering Practice and Automatica and Editor in Chief of the Spanish journal RIAI. His research interests include multivariable control and non-conventional sampling control systems, with focus on time delays and multirate sampling patterns, being involved in the ARTIST2 Network of Excellence on Embedded Control Systems

Control Design for Industrial Processes with Transport Delays

In this talk different industrial processes with time delays are introduced as a motivation to the design of control systems coping with delays. Different process models are reported and classical approaches are reviewed. Some recently proposed design methodologies dealing with unstable and MIMO plants will be reviewed and some application will illustrate the proposed solutions. Still there are some open issues to research and, mainly, for applications.

Department of Aeronautical and Automotive Engineering
Loughborough University
Leicestershire LE11 3TU
UK

Richard Stobart is the Head of Department and Professor of Automotive Engineering at Loughborough University in the Department of Aeronautical and Automotive Engineering.  Richard’s professional interests are in the engine technology and the application of controls systems to engines and vehicles.  He is active in the Society of Automotive Engineers and the Institution of Mechanical Engineers.  Before joining Loughborough he was Professor of Automotive Engineering at the University of Sussex (Brighton, UK) from 2001 to 2007.
Richard is a graduate from the University of Cambridge with a first class honours degree in Mechanical Engineering.  He was elected a Fellow of the Institution of Mechanical Engineers in 2000.

In this talk, he will take a brief historical look at the reason for engine control and then uncover the trends and explain the implications for the control community.