In this interview, we talk to Gregory L. Plett and M. Scott Trimboli, authors of the book, Battery Management Systems, Volume III: Physics-Based Methods. We discuss the motivation behind writing the book, the target audience, the most useful aspects of the book, the challenges of writing the book, and advice for other engineers who are considering writing a book.
Gregory L. Plett received his B.Eng. degree in Computer Systems Engineering from Carleton University and his M.S. and Ph.D. degrees in Electrical Engineering from Stanford University. He is currently a professor for the Department of Electrical and Computer Engineering at the University of Colorado, Colorado Springs and a senior member of the IEEE and life member of the Electrochemical Society.
M. Scott Trimboli received his B.S in Engineering Science from the United States Air Force Academy in 1980, his M.S in Engineering Mechanics from Columbia University in 1981 and his Ph.D. in Control Engineering from the University of Oxford. He previously served as an exchange scientist with the German Aerospace Research Establishment (DLR) in Göttingen, Germany. He is an Associate Professor of Electrical and Computer Engineering at the University of Colorado Colorado Springs.
1. What was your main motivation behind writing your book?
Battery management systems (BMS) must compute estimates of state of charge, state of health, state of energy, and state of power. In order to do so, they depend on sets of equations that describe the battery cells in the battery pack. These equations (or models) can be empirical, such as equivalent-circuit models, or they can be based on first-principles physics.
State-of-the-art BMS presently use equivalent-circuit models in their estimation algorithms. These models can predict cell voltage very well and so BMS methods based on these models can estimate state of charge and state of energy very well. However, empirical models do not predict the internal electrochemical variables in a cell—such as internal electrode potentials and lithium concentrations—and therefore they cannot anticipate the onset of conditions that will lead to premature degradation or rapid failure of the battery.
It is our strong belief that future BMS will depend on physics-based models since these models enable optimizing a proper trade-off between performance and degradation that will enable applications such as fast-charge based on physical limits instead of empirical observations and the calculation of power limits that correctly compute the maximum power available from the battery that will not prematurely age the battery.
The success of the first two volumes in this series on Battery Management Systems series pointed to the need for this third book focusing on physics-based lithium-ion battery modelling and controls methods. The two of us have collaborated closely for many years – the first author on modelling and the second author on controls. It seemed a natural fit to combine our efforts on a comprehensive volume addressing both physics-based battery models and their application in BMS control strategies.
Specifically, we wanted to address apparent significant roadblocks to using physics-based models: How to find the model parameter values? How to make reduced-complexity models that can be used on inexpensive microcontrollers? How to use these models for diagnosis and prognosis, or fast charge, or computing state and power limits? We wrote this book to present some approaches to overcoming these roadblocks. Our hope is that the book will help its readers take BMS to the next level of using physics-based models in their estimation algorithms.
2. Who is the main target audience for your book and what will they appreciate the most about the book?
The primary target audience comprises researchers and algorithm designers and implementers who have read the first two volumes on battery models and equivalent-circuit-based methods for implementing BMS estimation algorithms. The first volume in the series developed the foundation of physics-based models, upon which this third volume builds by showing how to estimate the parameter values of those models for a specific cell from laboratory tests and presenting some improved methods for converting those models into reduced-complexity versions useful in BMS algorithms. The second volume in the series set the foundation of BMS algorithms, upon which this third volume builds by showing how to estimate the state of an electrochemical model, how to use model variables for diagnosis purposes, and how to use these models in fast-charge and power-estimate applications. In summary, we believe that this book will be an important reference for researchers and industry designers interested in the latest advanced BMS techniques aimed at realizing high performance, extended life, and safety of operation. In addition, the volume serves as an excellent component in a graduate academic program on battery modelling and control.
We believe that the audience will appreciate the consistency in presentation among the volumes and how we share approaches that are computationally simple (even though they are conceptually challenging). We also believe that the reader will appreciate our careful step-by-step development of the background to the methods, teaching the concepts in an approachable manner.
3. What do you see your book being most useful for?
Those interested in pushing the state-of-the-art in advanced battery management will ultimately need to understand and apply computationally compact mathematical models of lithium-ion batteries that capture important electrochemical behaviours not available in present-day empirical models. This work, together with its predecessor volumes, provide both the background and the detail necessary for a battery engineer to apply these emergent techniques.
We foresee this volume in the series being especially useful to two categories of reader. The first comprises researchers in physics-based battery-management systems. Readers from this category will learn methods that we have found to be helpful and effective and will then develop improvements of their own. We look forward to learning what they are able to do! The second category comprises algorithm designers and implementers. We anticipate that readers from this category will learn what they need to know to understand the premise of physics-based BMS models and algorithms, and will be able to prototype their own versions to compare against the present state-of-the-art methods. They will advance the technology readiness level of the methods we present to develop robust implementations. Again, we look forward to learning what they are able to do!
4. How did you find the writing of the book? Do you have a specific process or are you quite methodical in your writing approach?
No-one should underestimate the enormity of a task like this. The sheer amount of background research, detailed mathematical development, and computational verification is intimidating indeed. Perhaps most challenging is exercising the discipline to keep the project on schedule amidst one’s day job and other lifetime responsibilities.
We started by making a very detailed point-form outline of the whole book. These points were extracted from articles that our research team has produced in the past or from other articles that we felt were seminal in the field. After editing this outline to arrange it in a format that had logical structure, we used it as the basis for teaching a trial offering of a graduate-level course to discover which aspects of our approach worked and which did not work from a learner perspective. We then spent considerable time revising the outline into its present form. Finally, we chose the examples we would use in the book, made figures, and converted the point-form outline into paragraph form. Overall, the creation process was methodical, but was followed by many rounds of editing and revising and refining that perhaps might have appeared to an onlooker as being less methodical.
5. What challenges did you face when writing the book and how did you overcome them?
We found the writing of this book to be very challenging. The first author promised in 2015 that this third volume in the series would be forthcoming in the near future but it has been delayed considerably more than expected. This was primarily due to that author being overly optimistic regarding the state of readiness of physics-based methods at that time. It has taken the intervening years for our research team and other key research teams across the world to develop practical physics-based models and methods that were ready to share.
Because much of the material contained in this volume is quite new and evolving, one recurring challenge was the fact that we were often chasing moving targets. It was important to set a “line in the sand” periodically to freeze development at a meaningful point to avoid persistent scope creep. Settling on consistent notation – or having to modify notation – also added to the fun.
We have been very fortunate to be able to work with some truly outstanding students and to have been able to focus their attention on solving problems that previously had unsatisfactory and deficient solutions. At last, we have been able to assemble a complete beginning-to-end approach that uses physics-based methods for battery management. There is a lot of work still to be done in this field, but we hope that what we are able to present in this book will form a helpful foundation for the next generation of researchers and partitioners in BMS algorithms.
6. What advice would you give to other engineers who are considering writing a book?
There are two primary lessons that we have learned. Our first recommendation to any prospective author is that s/he have the complete set of results to be presented in the book on hand before beginning to write the book. We were overly optimistic that we would quickly find solutions to some unsolved problems in our outline, which caused unforeseen delays in writing the book since it took longer than anticipated to find solutions to those problems.
Our second comment is that writing a book is not something to be undertaken lightly; it needs to be a labour of love. You must be in a position to commit to the task – heart and soul. Progress needs to be made every day (or at least every week); otherwise, it will never be completed. Sometimes it can be hard to find the proper motivation. But, in the end, we are both pleased that we went through the process and are delighted with the result.
7. What are you working on next?
The reader may be aware that the first author of this book has previously created a specialization on Coursera titled “Algorithms for Battery Management Systems,” which has been very well received. He is presently working on developing a new Coursera specialization on “Applied Kalman Filtering,” which is expected to launch in the fall of 2024.
The second author would like to write a very approachable and practical book on model predictive control – one of the key methods applied in the present BMS volume. Of course, he will consider this decision carefully.
*****
ARTECH HOUSE USA : Battery Management Systems, Volume III: Physics-Based Methods
ARTECH HOUSE U.K.: Battery Management Systems, Volume III: Physics-Based Methods
More Power Engineering content here Power Engineering – Artech House Insider
