All posts by Panagiotis Moutis

IET GTD paper on AC OPF machine learning solver, News

IET GTD Paper on Solving the AC Optimal Power Flow with Machine Learning

October 2022

I am elated to report the acceptance of my recent work at the IET Generation, Transmission & Distribution (GTD) open access (OA) journal of the Wiley publications! Before unpacking my paper with the (kind of long) title “Stochasticity Agnostic Solution to the AC Optimal Power Flow by Recursive Bound Tightening with Top-Down Heuristically Inducted Binary Decision Trees” (link here), let me rejoice in the fact that the IET GTD is, according to Scholar Google, the highest ranking Q1 (in Control & Systems Eng., Electrical & Electronic Eng., Energy Eng. & Power Technology) OA journal on Power Engineering. I have been an avid proponent of OA and Associate and Senior Editor in 2 such publications. Going through the process as an Author myself has reassured me that my heart is in the right place with OA! Read more about my position on OA here.

The AC Optimal Power Flow (OPF) is a non-convex optimization problem for resources or performance metrics within an electrical grid (see formulation above). The non-convexity of the AC OPF is due to the grid physics of power flows that introduce a non-convex equality constraint to the  optimization formulation. Since the AC OPF per se is – almost – never the end-problem we wish to solve, the sizes of problems encapsulating AC OPF are typically orders of magnitude larger than that and may include additional non-convexities, e.g. integer variables. The most typical approach used by several researchers in the recent years is the convex relaxation of the AC OPF to a larger dimension and – provided some conditions are met – projection of the relaxed solution back to the feasible space of the original  AC OPF problem. Much fewer approaches have considered machine learning for solving the AC OPF and even fewer of those come with guarantees of global optimality. The overall framework here described clearly means that the AC OPF cannot be solved in times that make sense for system operators and other electricity stakeholders when making decisions about how to operate and plan the electrical grid functions.

My paper introduces 2 perspectives to solving the AC OPF per se and another 1 in accounting for volatile resources, such as renewables like wind generators and photovoltaics.  But first of all, let me describe in a few words how the solver works (assuming we wish to minimize an objective function of energy cost). It starts by sampling the whole feasible space of the AC OPF instance. The samples are, essentially, random feasible dispatches for said problem. Then the method labels half of the samples as False, if the objective function is greater than the median cost of the samples, and True otherwise. Then a heuristically inducted binary decision tree (BDT) is trained with these samples and tightens/shrinks the feasible space (clearly towards the region of the feasible space with costs below the noted median). The method is recursively employed until the feasible space has tightened/shrunk around the global optimum (see “tightening” steps (a)-(e) in the the graph above).

Solving the AC OPF – Point 1
I prove that the method converges to the global optimum via Bayesian inference. Most typically, AC OPF solvers (and relaxations building upon them) pursue to solve the dual problem, relying on strong duality conditions that indicate that the solution of the primal and the dual problems are the same. I take a different path… I prove that a Bayes classifier for the global minimum in a small vicinity around it exists. This is true, due to the fact that the feasible space has non-random characteristics and I also explicitly label the global optimum (and some epsilon around it) as such, while the rest of the said vicinity as non-globally-optimum. From that point, iteratively, I can expand to the whole feasible space by properly labeling/splitting the vicinities of the feasible space as optimal/sub-optimal.

The existence of these Bayes classifiers means also the existence of the heuristically inducted top-down binary decision trees (BDTs) for the same classifications of optima/non-optima. The second part of the proof here could not have been possible if it had not been for Guy Blanc‘s, Jane Lange‘s & Li-Yang Tan‘s recent work on heuristically inducted BDTs. In 2020, Guy et al proved that if there exists a minimum (in the sense BDT-size; nothing to do with the underlying optimization problem here) classifier for a given function, then there exists a non-optimally-sized BDT for that same function. I am pointing this out, because the first time I have used a draft/reduced version of this method (back in 2010), I could not explain why/how the BDT would consistently converge to an optimum… Well, now I know – thanks Guy, Jane and Li-Yang!

Solving the AC OPF – Point 2
Even though it kind of stems from Point 1, it is a remark that I wish to separately mention here. The steps of tightening the AC OPF feasible space towards the global optimum seem unaffected by the size of the grid, but follow the number of the decision variables. In the way I have structured the solver, the decision variables are the active and reactive power set-points of the generating resources. This can be particularly valuable when larger grids with fewer resources must be optimized. Given how typical solvers descend along gradients of a feasible space, the size of the latter will affect that descent. Hence, larger grids, meaning greater numbers of voltage angles, magnitudes and flows, will be slowing down the solving descent. In the AC OPF solution I propose, the size of the grid does not matter, provided that the BDT training samples are feasible and adequate. This is particularly interesting, because we can consider how to accelerate this solution via the control intervals of generators, which are typically not continuous and cannot take any value between minimum and maximum (looking for the exact NERC rule and I will cite it here).

Effect of Volatile Resources in Solving the AC OPF
Typically, renewable energy has negligible, if not zero, operating costs, hence, gets dispatched by priority in any electricity market. However, the exact amount of available renewable energy is usually impossible to determine ahead of time and only offered within some forecasted confidence intervals (see forecasting figure below). This means that as the proposed AC OPF solver converges to the global optimum by the successive median cost of the sampled feasible space, the commitment of renewable energy within the AC OPF increases towards the global optimum at a decreasing confidence. In other words, renewable energy will be committed within the AC OPF at lower set-points corresponding to higher forecasting confidence in the first few iterations of the method (higher objective function costs) and at higher set-points of limited confidence when the process converges to global minimum.

As in my last paper, I had the joy to work with another CMU MSc student, who put in some coding work for me; Parth joined me along for the ride on the paper authorship, too! It was fun and very rewarding to substantially advance my past work on machine-learning-driven optimization of electrical grids with volatile renewables, gain better understanding of the methodology and its characteristics, test it rigorously for performance and get it published OA. Next steps are determining appropriate BDT training sizes and focusing on the stochasticity aspect of this AC OPF solver. Feel free to reach out for any ideas for extensions or collaborations; I will be glad to see more applications of this tool!

Blog, Teaching Power Engineering

How I teach Power Engineering

September 2022

Past student (Nick Alexander) presenting his project in one of my courses (2019).

My first contribution to teaching was back in 2008 as an assistant to Prof. Korres and on the graduate-level subject on machine learning (ML) applications within power system control centers. The subject was more of a review on research ideas and publications exploring if and how ML could be useful in real-time operation of electrical grids, as also the planning of their infrastructure. As a junior PhD student under Prof. Hatziargyriou at that time I was just then dipping my toes in the vast sea of research on artificial intelligence (AI) in power systems. Helping teach that course offered some valuable insights.

Graph from this Verge article: https://www.theverge.com/2018/12/12/18136929/artificial-intelligence-ai-index-report-2018-machine-learning-global-progress-research

In those years, ML and AI were very appealing in the mainstream engineering education and research (buildings, components, power systems, circuits, etc), but were themselves going through some introspection within the computer science community. There are several articles pointing out to the relative plateauing of R&D on AI & ML in the later 2005-10 period, followed by exponential growth afterwards. Nevertheless, more traditional engineering fields were warming up to their “digitalization” and reinventing themselves as smart: smart grids, smart buildings, smart materials and so on. Even though this latter push was strong within research circles and several professional initiatives started popping up, in practice, AI & ML applications were taking baby-steps. The tide changed after bold start-ups adopted their value en masse in the mid 2010s.

Looking back to the period of 2005-10 through the lens of  the technology shifts of the later years, made me realize something ‘big’ about higher education teaching. We do not teach to only educate or provide necessary skill-sets; we teach to seed a vision. And I am very cautious with my words here. I do not proclaim that university teaching must be ‘persistently’ forward-thinking. I do not imply that the fundamentals should be in any way glossed over.  I do not believe that there is such a thing as either redundant or rudimentary knowledge. What I am saying though is that higher-ed teachers have a unique challenge: we must read the tea leaves or crystal balls of our “craft” and prepare students of what is to come, even though no one else might be seeing this yet.

Image taken from GSMArena article https://www.gsmarena.com/asymco_pricing_doesnt_affect_smartphone_adoption_in_the_us-news-8982.php

Think about the example of the course I mentioned earlier and the ‘smartening up’ of many economy sectors. The actual computer science field was kind of taking a breather on AI and ML at that time, yet all other engineering fields were bracing confidently about a tech boom relying exactly on AI and ML! If we seek to dissect this paradox a bit, we will see that there were – actually – no tea leaves or crystal balls necessary. Computers and mobile devices were becoming widely available and used a ton by consumers, practically giving educators the low-hanging fruit of inspiration about advanced computing tools.

Image taken from this Economist article https://www.economist.com/graphic-detail/2017/01/16/china-powers-ahead-with-a-new-direct-current-infrastructure

Fast forward to 2018, when I developed, proposed to CMU and started teaching a course, then titled, “Optimization Modeling in Power Systems”. I was reading my “crystal ball” and realizing that the ever present discussion about the failing and ageing grids (first pointed out in the 1990s) was actually spilling dangerously into reality. Add another 3-4 years to that and here we are in 2021 and 2022 witnessing the passing of multi-billion dollar bills that will expand the US electrical grid, several start-ups monetize efficiency for end-customers globally, while China has already taken strides ahead in its high-voltage transmission system backbone. Optimizing all electrical grid operations and planning, which used to be the expertise of engineers with PhDs, is now expected by the MSc graduates of energy engineering programs. My personal story here is that in 2018 and the next couple of years, the “Optimization Modeling in Power Systems” course was attended by barely 5 students – mostly in their PhDs; now I teach a roster of 16 students – more than half of whom are MSc students.

What I mean to point out with the above story is that power and energy engineering education is not straight-forward, since it links to multiple other subjects and fields. I am sad to attest to statements of scholars I admire saying that “power engineering is not science”.  And yet here we are, following a “hunker down pandemic” (which should have reduced energy demand) and a 6-month localized war (which should have not hurt international energy security), which are crippling multiple economies by threatening electricity markets everywhere. Even more ironically, most of these electricity markets could have already had energy independence, had they built out their grids and/or made them much more efficient. Put simply, energy and power educators have to look deep and wide in their research and expertise, and aspire to new subjects and angles (decentralized? autonomous? self-sufficient? resilient?). The energy and power work force is crumbling, several places around the world are still not electrified consistently (let alone have energy security), and energy dependencies are heavy and stretch the globe.

I conclude this blog with a final thought. The energy sector comprises entities and processes that are complex, large and slow to adapt, while the challenges they are faced with and the tools to address them are fast and with much impact. This means that the teaching aspirations of power and energy engineering faculty must be also realistic and delivered with confidence and persistence. As instructors in this field, we must develop the necessary intuition and retain adequate humility in keeping our ear to the ground for the right signs of change. And this last thing is not easy; personally, I have acquired these “skills” with years of practicing engineering and humbling disappointment across several projects I have worked on. Others before and around me are trying through broad involvement in committees, initiatives and working groups. Whatever the way, educating the next generations of power and energy engineers is an urgent duty hanging already over our heads, especially, if we wish to be honest to our vision for the clean energy transition.

Appointed Energy & Power Co-Lead CCAI, News

Power & Energy Community Co-Lead at the CCAI

Aug. 2022

It gives me immense joy to announce that I have joined the core team of the Climate Change AI (CCAI) organization. I will be serving the role of Power & Energy Community Co-Lead, alongside the tireless, devoted and unimaginably energetic CCAI’s co-founder, Dr. Priya Donti,

CCAI has identified climate change as the humankind’s most imminent existential threat, affecting primarily those who are underprivileged – communities of color, impoverished, without access to advanced technology and modern infrastructure. Within CCAI it is also understood that the climate change is a multi-faceted problem, at varying scales and with unique nuances across different sectors. In this sense, Artificial Intelligence (AI) and the computational tools, approaches and frameworks it brings to the table represent the means to implement the coordinated and demanding efforts required to address this threat. However, with AI’s meteoric rise, its adverse impacts come also into scope within a world that attempts strides towards a fairer, more inclusive and more equitable environment and society. CCAI seeks to be the organization that brings together researchers, engineers, entrepreneurs, policy and decision makers, and all stakeholders from the public and private sectors to promptly and consistently create the community, educate the society, inform the infrastructure planning and serve as the global forum that will put AI to the service of climate change mitigation for the sake of each and every life on the planet.

My first couple of weeks within CCAI have been absolutely amazing. The organization has shown immense attention to detail, there are multiple outcomes of hard and coordinated work, the processes are defined clearly and the roles are crisp. Having being part of many volunteer and non-profit organizations for the past 15 years of my work on renewable energy research, CCAI is among the very few that have been so meticulously designed and operated. Combining this with the CCAI’s community’s passion for the cause, the members’ unwavering work ethic since the organization was first rolled out in 2019 and all the plans already in motion across multiple venues, I am certain that CCAI heads out to build and inspire many great accomplishments in this space! If you are reading these lines, you should definitely reach out and engage with the CCAI.

I want to thank from the bottom of my heart Dr. Priya Donti, Marcus Voss, Raphaela Kotsch, Dr. Kasia Tokarska, Dr. Evan Sherwin and the many other CCAI team members for the warm welcome. Let’s do this!

Inertia and Frequency Control by RES at UNIFI Sep. 19, News

Inertia Emulation & Frequency Control at the UNIFI Fall 2022 Seminar Series

August 2022

I am sincerely grateful to the many wonderful colleagues at the UNIFI consortium for having me present my older work on methods to emulate inertia and perform frequency control with wind and photovoltaic generators. My seminar is scheduled on Monday Sep. 19th at 4 pm ET, as part of the Fall 2022 Seminar Series (more information here).

The UNIFI project aims to conduct advanced research, design testing and develop standards on grid-forming inverters. Inverters are the power electronics devices that enable the efficient connection of many renewables to the electrical grids. With the gradual replacement of conventional units by renewables, the roles of the former pass on to the latter. “Grid-forming” is the functionality necessary to establish and maintain a standardized three-phase alternating current that serves some load demand. Inverters have been typically able to perform “grid-forming” at limited off-grid scales, but are now expected to expand it at the level of large interconnected grids.

Typical stages of Load-Frequency Control (LFC) in power systems

In my seminar I will focus mostly on the side of the sources, specifically wind and photovoltaic energy. Traditionally, these generators have been operated on strategies of maximum power absorption. Even though these strategies optimize the use of renewable sources, they are inflexible when load demand varies or the generation from other resources fluctuates. In fact, due to electricity physics any generation-demand imbalance, reflects to changes in the electrical frequency of the alternating currents; this change of frequency can control generators to respond to the load-generation imbalances. With wind and photovoltaic generators at maximum power absorption, responding to frequency signals that requires them to contribute additional power is impossible; hence, the requirement to procure reserves arises. I will review the methods, challenges, some results of real-world testing and expand on how grid-forming functionalities might be affected by inertia emulation and frequency control by wind and photovoltaic units.

JEPE Special Issue on protection in 0 carbon grids, News

Special Issue at the Intl. Journal of Electrical Power & Energy Systems

June 2022

I am kindly inviting you all to submit your works to the Special Issue on “Novel Protection and Control Methodologies towards Electrical Grids with Net-Zero Carbon Emissions” at the International Journal of Electrical Power and Energy Systems of the Elsevier publications. Here is the link to the call for papers. You may submit your novel contributions (full manuscripts) starting July 1st and by 30 Sep, 2022, on any of the following subjects:

  • Online/real-time monitoring and situational awareness solutions (detection and location of system oscillation, fault level monitoring and quantification, inertia measurement, etc.),
  • Analyzing and characterizing fault behavior and the novel protection strategies and solutions,
  • Converter control-based solutions to support protection operation,
  • Electrical grid design and assessment for robust point of common coupling impedance behavior,
  • New coordinating control solutions,
  • Methods for assessment of resilience,
  • New protection and control solutions during extreme weather/operating conditions and
  • New ICT technologies for protection.

I am grateful to my friend Dr. Qiteng Hong (University of Strathclyde, Glasgow), as also, Dr. Botong Li (Tianjin University), who are the Guest EiCs of this special issue and kindly invited me to serve with them on the editorial board. You may contact me for any additional details for works you would like to submit.

News, Power & Energy at 2022 IEEE WF IoT

Power & Energy Vertical Track at the 2022 IEEE World Forum on Internet of Things

May 2022

I am sincerely excited to co-chair the Power & Energy Vertical Track at the 2022 IEEE World Forum on Internet-of-Things (WF IoT), in Yokohama, Japan, coming November. I have happily chaired the same track in the last installment of the WF and I look forward to putting together multiple sessions of researchers and experts on all things (“Internet of… things” – see what I did there?) energy and power systems.

My track co-chair Sérgio Ivan Lopes, Technology and Management School of the Polytechnic Institute of Viana do Castelo (ESTG-IPVC), and I will be reaching out to many of you who can contribute to the subjects of interest. The contributions may also be remote/online. A paper track is planned, too, and I will be updating this announcement with submission and deadline details soon.

If you want to nominate yourself or someone you know as a contributor to the Energy & Power Vertical Track of the 2022 IEEE WF on IoT, please reach out. I will be delighted to have you!

Appointed Co-Lead NASPI DisTT, News

Appointed co-lead of NASPI Distribution Task Team

May 2022

I am particularly happy and honored to join Daniel Dietmeyer from San Diego Gas & Electric in leading the Distribution Task Team (DisTT) at the North American Synchro Phasor Initiative (NASPI).

NASPI was founded in 2003 as the Eastern Interconnection Phasor Project, it is funded by the US Dept. of Energy, and is supported by the Pacific Northwest National Laboratory (PNNL) and the Electric Power Research Institute (EPRI). It is the largest collaboration of academics, industry practitioners and standardizing bodies for the development, use, understanding and promotion of methods and technologies based on synchronized measurements of voltage and current waveforms in power systems. These measurements with granularity of at least 30 per second and which are time synchronized via satellite across large grids, allow us to better analyze and control the stability and the security of the electrical grid.

Within the framework of DisTT, synchronized measurements enable the detection of faults, increase of hosting capacity of renewables, monitoring equipment health and others functions. At the current stage, DisTT focuses on the medium voltage beyond the substation.

This great opportunity and responsibility could not have been possible without the mentorship, support and inspiration that Sascha von Meier from UC Berkeley has gracefully offered me. I take over her role in leading DisTT in the hopes I can achieve half of what she did! Also, many thanks to Jeff Dagle (PNNL and chair of NASPI) for welcoming me on board.

 

Appointed IET RPG SS Editor, News, Uncategorized

Appointed Special Issues Editor at the IET RPG

March 2022

I am thrilled to announce that the Editors in Chief of the IET Renewable Power Generation (RPG) journal, Prof. Infield and Prof. Tricoli, have invited me to serve as the inaugural Special Issues Editor for this publication. I assume this role immediately and further to those of the Regional Editor for North America and Associate Editor on the subject of Hybrid Renewable Energy Systems for the same Open Access (OA) journal. It is particularly indicative of the Institute’s priorities that this is only the second IET publication in the field of energy and electrical power systems that is assigned with a Special Issues Editor.

Since the decarbonization of the energy sector is an aim long-overdue and  particularly complicated, it requires the mobilization of several stakeholders in the academia, the industry and among the decision and policy makers. The role and the positioning of the IET RPG in this discussion is central in bringing stakeholders together as authors, reviewers and adopters of publications disseminating how renewable energy at the microscopic and macroscopic levels can fulfill the clean energy transition.

In my role as Special Issues Editor, I will be soliciting, organizing, overseeing, editing and managing thematic calls for papers at the intersection as also the periphery of IET RPG topics, supporting their Guest Editors and attracting prospective Authors. For those of you who know me, understand that my involvement will be hands-on, the Special Issues will be appropriately curated, and that the Guest Editors will have the full support of the RPG journal staff and the IET organization. I urge you to contact me with ideas and proposals, even though I will also be reaching out to many of you.

As I have stated previously, OA to the concepts and results of academic and industrial R&D is the corner-stone of promoting and disseminating crucial ideas and important scientific results in the times of urgent calls to action. In my view, the OA publications by the IEEE and the IET have been serving  this mission with respect to the Authors and their work at the highest level of quality and with a dedicated pursuit for academic excellence. I am very proud to serve publications for both the IEEE and the IET!

IEEE TPWRS paper on fire-detecting digital twin of overhead line, News

IEEE TPWRS Paper on Digital Twin of Overhead Lines for Fire Detection

March 2022

Extending some of my previous work, I developed a digital twin for overhead conductors that detects an approaching forest fire and de-energizes the affected lines in a timely manner and not preemptively. The work has just been accepted in the IEEE Transactions on Power Systems (preprint here).

In California (CA) and elsewhere, the risk of overhead conductors igniting forest fires or adding seats to on-going ones is very real and extensive. In CA, PG&E’s overhead conductor equipment was determined to be the reason for the 2018 Camp fire, leading to law suits that caused the utility’s bankruptcy. After restructuring, the company updated its practices with preemptive disconnections of large parts of its grid during days of high risk of fire. The new practice disrupted service to thousands of customers, in most cases unnecessarily. Hundreds of new suits threatened PG&E with a second bankruptcy in 3 years.

Phasor Measurement Units (PMUs) have been widely adopted across grids. PMUs may be installed along a line in distances as close as a 1-2 miles in between. This gives rise and basis to the idea of real-time monitoring of line impedance for any reasons of variation. As resistance increases with ambient temperature (not proportionally), steep decreases in the inductance/resistance ratio (tangent of the impedance phasor – tanδ in the figure) of an overhead conductor may indicate that a forest fire burns near said conductor and it should, thus, be disconnected.

Behavior of moving average of impedance phasor as a forest fire approaches an overhead conductor and affects its resistance. Such a behavior should control the disconnection of this conductor.

The in silico testing under numerous worst case scenario conditions (no solar heating effect, broad measurement error intervals, synchronization errors, etc.), showed that the proposed method detects some cases of a forest fire approaching a conductor, in sub-second times and at extremely low false positive rates. In the next steps, I plan a collaboration with interested utilities and the US Forest Service for field testing.

I want to thank CMU ECE’s MSc student (at that time) and co-author Uday Sriram for his help in setting up the tests, Dan Dietmeyer from SDG&E for informing me about PMU deployments in CA, Farnoosh Rahmatian from NuGrid Power for lending his expertise on instrument transformers and Jeff Dagle from PNNL for his crucial comments in the earlier stages of this work.

News, PMUs and Inverters Panel IEEE SGSMA 2022

Panel on Synchrophasors in Zero Inertia Grids at the IEEE SGSMA 2022

February 2022

I am grateful to the Technical Program chairs of the 2022 IEEE International Conference on Smart Grid Synchronized Measurements and Analytics (SGSMA 2022) for accepting our panel proposal titled “Towards a Zero Inertia Grid thanks to Synchrophasor Measurements”. I have been delighted to have Prof. Yilu Liu (University of Tennessee at Knoxville), Dr. Evangelos Farantatos (EPRI),  Dr. Deepak Ramasubramanian (EPRI, on behalf of UNIFI), Dr. Qiteng Hong (University of Strathclyde, Glasgow) and Dr. Krish Narendra (Electric Power Group) accept my invitations to join this panel and contribute their expertise and experiences on the matter.

What we will be talking about revolves around how the electrical grid shifting to renewables and batteries, entails the shift to resources interfacing with the power system via power electronics – inverter, rectifiers and converters. As these devices and the sources they interface are characterized by fast dynamics, the traditional control paradigm followed to the day cannot suffice. The reason is that the phenomena that used to span seconds (thanks to large rotating inertias of conventional generators), will now be unfolding in milliseconds. Hence, the operators’ response times in the control rooms will be very limited. Thankfully, synchrophasors and the applications they enable can match these time-frames and allow for the transition to a new control paradigm.

I look forward to the conference and hope to be attending it in person in the beautiful town of Split in Croatia.