Massively Digitized Power Grid: Opportunities and Challenges of Use-Inspired AI

Xie, Le; Zheng, Xiangtian; Sun, Yannan; Huang, Tong; Bruton, Tony

doi:10.1109/jproc.2022.3175070

Cited by 12 publications

(4 citation statements)

References 189 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Because ( 2) is reduced to a tractable convex problem with fixedinteger variables. To achieve online regulation, we propose to predict the integer variables of s * dns based on the input parameters θ to achieve MIP-based DNS solution acceleration, as shown below in (3).…”

Section: P-dns Modelmentioning

confidence: 99%

“…To keep the system safe and economic, DNS is solved for different input scenarios to reach the proper system decisions repeatedly, where the upcoming scenario could be similar to the historical scenario. However, repeatedly solving MIP for DNS could be time-consuming and energy-wasting, because solving similar scenarios of DNS requires solving the corresponding MIP, which is intractable and time-consuming [2][3][4][5]. For MIP, due to the complex features of integer variables, current popular algorithms leverage the idea of branch and bound or benders cut, and the corresponding solution time is unstable.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Online Distribution Network Scheduling via Provably Robust Learning Approach

Wang,

Cai,

Sang

et al. 2024

Energies

View full text Add to dashboard Cite

Distribution network scheduling (DNS) is the basis for distribution network management, which is computed in a periodical way via solving the formulated mixed-integer programming (MIP). To achieve the online scheduling, a provably robust learn-to-optimize approach for online DNS is proposed in this paper, whose key lies in the transformation of the MIP-based DNS into the simple linear program problem with a much faster solving time. It formulates the parametric DNS model to construct the offline training dataset and then proposes the provably robust learning approach to learn the integer variables of MIP. The proposed learning approach is adversarial to minor perturbation of input scenario. After training, the learning model can predict the integer variables to achieve online scheduling. Case study verifies the acceleration effectiveness for online DNS.

show abstract

Section: P-dns Modelmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Online Distribution Network Scheduling via Provably Robust Learning Approach

Wang,

Cai,

Sang

et al. 2024

Energies

View full text Add to dashboard Cite

show abstract

“…Recent scholarly works offer insights into the complex nature of SG technologies. They illustrate the impact of digitalization and renewable energy integration on power systems and consumers [5,[7][8][9][10][11][12]. From decentralizing renewable energy generation to leveraging broadband over power line infrastructure for unified data exchange, these technologies balance supply and demand and enhance grid intelligence.…”

Section: Introductionmentioning

confidence: 99%

Digitalization Processes in Distribution Grids: A Comprehensive Review of Strategies and Challenges

Aghahadi,

Bosisio,

Merlo

et al. 2024

Applied Sciences

View full text Add to dashboard Cite

This systematic review meticulously explores the transformative impact of digital technologies on the grid planning, grid operations, and energy market dynamics of power distribution grids. Utilizing a robust methodological framework, over 54,000 scholarly articles were analyzed to investigate the integration and effects of artificial intelligence, machine learning, optimization, the Internet of Things, and advanced metering infrastructure within these key subsections. The literature was categorized to show how these technologies contribute specifically to grid planning, operation, and market mechanisms. It was found that digitalization significantly enhances grid planning through improved forecasting accuracy and robust infrastructure design. In operations, these technologies enable real-time management and advanced fault detection, thereby enhancing reliability and operational efficiency. Moreover, in the market domain, they support more efficient energy trading and help in achieving regulatory compliance, thus fostering transparent and competitive markets. However, challenges such as data complexity and system integration are identified as critical hurdles that must be overcome to fully harness the potential of smart grid technologies. This review not only highlights the comprehensive benefits but also maps out the interdependencies among the planning, operation, and market strategies, underlining the critical role of digital technologies in advancing sustainable and resilient energy systems.

show abstract

“…G RAPH neural network and its variant, graph attention network (GAT), have achieved state-of-the-art performance in multiple power system analysis tasks by effectively recognizing and processing power system topologies [1]- [3]. In the context of deepening interconnection of modern power grids, the abilities of AI models to scale for large-scale power systems are vital for their adoption in the power industry [4].…”

Section: Introductionmentioning

confidence: 99%

Emergent Abilities of Graph Neural Networks for Large-scale Power System Analysis

Zhu,

Zhou,

Yan

et al. 2024

Preprint

View full text Add to dashboard Cite

The scale-up of AI models for analyzing largescale power systems necessitates a thorough understanding of their scaling properties. Existing studies on these properties provide only partial insights, showing that loss function decreases predictably with increased model scales; yet no scaling law for power system AI models has been established, and model performance remains unpredictable due to “emergent abilities”. This study pioneers the discussion on the emergent abilities of graph neural network (GNN) for analyzing large-scale power systems, revealing that model performance improves dramatically once model scale exceeds a threshold. Furthermore, we introduce an empirical power-law formula to quantify the relationship between this threshold and the power system size. Our theory accurately predicts the threshold for the appearance of emergent ability in large-scale power systems, including a synthetic 10,000- bus and a real-world 19,402-bus systems.

show abstract

Massively Digitized Power Grid: Opportunities and Challenges of Use-Inspired AI

Cited by 12 publications

References 189 publications

Online Distribution Network Scheduling via Provably Robust Learning Approach

Online Distribution Network Scheduling via Provably Robust Learning Approach

Digitalization Processes in Distribution Grids: A Comprehensive Review of Strategies and Challenges

Emergent Abilities of Graph Neural Networks for Large-scale Power System Analysis

Contact Info

Product

Resources

About