AI Intelligence for the Grid 16 Years Later: Progress, Challenges and Lessons for Other Sectors
Abstract
How could the “new AI” based on neural networks and deep learning be applied to the electric power grid, so as to get maximum benefit from the new technology, and serve as a model for how to organize the new Internet of Things (IOT) in general? The first of these questions was already assessed in great detail in workshops organized jointly by NSF and the Electric Power Research Institute (EPRI) in 2002 [1], drawing on new technologies which included today's deep learning but also more advanced technologies in the same family [2]. The NSTC (White House) Smart Grid policy of June 2011 cited [1] in stating: “NSF is currently supporting research to develop a '4th generation intelligent grid' that would use intelligent system-wide optimization to allow up to 80% of electricity to come from renewable sources and 80% of cars to be pluggable electric vehicles (PEV) without compromising reliability, and at minimum cost to the Nation.” This paper gives some highlights of the progress made, the open challenges, and important connections to the larger needs of humanity, in that order. The synergy between new intelligence, new technology for cybersecurity [3] and new physical hardware [4] is essential to maximum success, and even to the very survival of our endangered species. Lessons from the power grid are essential to better understanding of urgent challenges central to the IOT in general [5].
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The data quality challenge is acute in legacy systems and developing nations.
“The challenge is acute in legacy systems and developing nations [35].”
From Addressing Challenges for the Effective Adoption of Artificial Intelligence in the Energy Sector (2025) · cites this paper as [35]
Foundational challenges such as data quality, cybersecurity, and high costs have persisted as core concerns throughout the 2018-2025 period.
“Foundational challenges such as data quality, cybersecurity, and high costs have persisted as core concerns throughout the 2018-2025 period (e.g., [11,24,35]).”
From Addressing Challenges for the Effective Adoption of Artificial Intelligence in the Energy Sector (2025) · cites this paper as [35]
Legacy infrastructure often lacks necessary digital capabilities, requiring costly retrofits.
“Legacy infrastructure often lacks necessary digital capabilities [35,36], requiring costly retrofits”
From Addressing Challenges for the Effective Adoption of Artificial Intelligence in the Energy Sector (2025) · cites this paper as [35]
Costs of integrating with legacy systems compound the financial barrier.
“which is compounded by costs of integrating with legacy systems (Section 3.3.1;[35])”
From Addressing Challenges for the Effective Adoption of Artificial Intelligence in the Energy Sector (2025) · cites this paper as [35]
Multi-energy coordination involves high modeling complexity due to diverse constraints and dynamics, overcoming data fragmentation from historically siloed operations, addressing interoperability deficits, and navigating regulatory/market designs often ill-suited for cross-sector optimization.
“However, this involves high modeling complexity due to diverse constraints and dynamics [35], overcoming data fragmentation from historically siloed operations ( [23]; linking to policy/governance, Section 3.2.3),addressing interoperability deficits [33], and navigating regulatory/market designs often ill-suited for cross-sector optimization [38].”
From Addressing Challenges for the Effective Adoption of Artificial Intelligence in the Energy Sector (2025) · cites this paper as [35]
Energy systems are safety-critical, and AI failures can have catastrophic consequences.
“Energy systems are safety-critical.AI failures can have catastrophic consequences [28,35].”
From Addressing Challenges for the Effective Adoption of Artificial Intelligence in the Energy Sector (2025) · cites this paper as [35]
Cites
Disruptive technologies: Advances that will transform life, business, and the global economy
Manyika, J., Chui, M., Bughin, J., Dobbs, R., Bisson, P., & Marrs, A. (2013) · Cited by 1,237 · see citing works
The Nerves of Government: Models of Political Communication and Control
McLuhan, H. M., & Deutsch, K. W. (1965) · University of Toronto Law Journal · doi:10.2307/825125 · Cited by 788 · see citing works
Cyber‐physical attacks and defences in the smart grid: a survey
He, H., & Yan, J. (2016) · IET Cyber-Physical Systems Theory & Applications · doi:10.1049/iet-cps.2016.0019 · Cited by 457 · see citing works
10-Qubit Entanglement and Parallel Logic Operations with a Superconducting Circuit
Song, C., Xu, K., Liu, W., Yang, C., Zheng, S., Deng, H., Xie, Q., Huang, K., Guo, Q., Zhang, L., et al. (2017) · Physical Review Letters · doi:10.1103/physrevlett.119.180511 · Cited by 449 · see citing works
Intelligence in the brain: A theory of how it works and how to build it
Werbos, P. J. (2009) · Neural Networks · doi:10.1016/j.neunet.2009.03.012 · Cited by 186 · see citing works
Self-citationLi, S., Fairbank, M., Johnson, C., et al. (2013) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2013.2280906 · Cited by 156 · see citing works
Computational Intelligence for the Smart Grid-History, Challenges, and Opportunities
Werbos, P. J. (2011) · IEEE Computational Intelligence Magazine · doi:10.1109/mci.2011.941587 · Cited by 118 · see citing works
Self-citationWhat Happened to Formal Methods for Security?
Schaffer, K., & Voas, J. M. (2016) · Computer · doi:10.1109/mc.2016.228 · Cited by 28 · see citing works
Putting more brain-like intelligence into the electric power grid: What we need and how to do it
Werbos, P. J. (2009) · doi:10.1109/ijcnn.2009.5179088 · Cited by 24 · see citing works
Self-citationFisch, E. A., & White, G. B. (1999) · CRC Press eBooks · doi:10.1201/9781420049091 · Cited by 23 · see citing works
Neural networks and the experience and cultivation of mind
Werbos, P. J. (2012) · Neural Networks · doi:10.1016/j.neunet.2012.02.026 · Cited by 18 · see citing works
Self-citationAnalog quantum computing (AQC) and the need for time-symmetric physics
Werbos, P. J., & Dolmatova, L. (2015) · Quantum Information Processing · doi:10.1007/s11128-015-1146-2 · Cited by 15 · see citing works
Self-citationProtecting the Internet from Dictators: Technical and Policy Solutions to Ensure Online FreedomsPreprint
Bowman, W. M., & Camp, L. J. (2012) · SSRN Electronic Journal · doi:10.2139/ssrn.2101677 · Cited by 7 · see citing works
New Technology Options and Threats to Detect and Combat Terrorism
Werbos Paul, J. (2017) · NATO science for peace and security series. Sub-series E, Human and societal dynamics · doi:10.3233/978-1-61499-748-1-34 · Cited by 6 · see citing works
Werbos, P. J. (2018) · Elsevier eBooks · doi:10.1016/b978-0-12-815480-9.00008-6 · Cited by 5 · see citing works
Self-citationExample of lumped parameter modeling of a quantum optics circuit
Werbos, P. J. (2014) · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · doi:10.1117/12.2048317 · Cited by 4 · see citing works
Self-citationThe new AI: Basic concepts, and urgent risks and opportunities in the internet of things
Werbos, P. J. (2023) · Elsevier eBooks · doi:10.1016/b978-0-323-96104-2.00006-3 · Cited by 4 · see citing works
Self-citation
Flagged first, then by citations
- Authors
- Werbos, P. J.
- Affiliations
- U.S. National Science Foundation (US)
- Corresponding author
- Werbos, Paul J.
- ORCID
- Werbos, Paul J.
- Venue
- 2018 International Joint Conference on Neural Networks (IJCNN)
- Publisher
- IEEE
- Published
- 2018-07
- Pages
- 1-8
- ISSN
- 2161-4407 (ISSN-L)
- OpenAlex
- W2897962492
- Cited by
- 6
- References
- 18
- Citations by year
- 2025: 1 · 2024: 1 · 2020: 1 · 2019: 2 · 2018: 1
- FWCI
- 1.60
- Type
- conference paper
- Language
- en
- Open access
- closed
- Locations
- International Joint Conference on Neural Networks (IJCNN) (publishedVersion)
- Topics
- Physical Sciences › Engineering › Electrical and Electronic Engineering › Smart Grid and Power Systems
- Physical Sciences › Engineering › Electrical and Electronic Engineering › Smart Grid Energy Management
- Physical Sciences › Engineering › Control and Systems Engineering › Smart Grid Security and Resilience
- Keywords
- power grid, deep learning, artificial neural networks, smart grid, Internet of Things, renewable energy integration, plug-in electric vehicles, grid cybersecurity
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
