EUVIMEDEUVIMED entdecken
Crossref

Integrated Artificial Intelligence and Stochastic Optimization Framework for Resilient and Low Carbon Renewable Energy Manufacturing Systems

Energy, Environment, and Economy · 01.01.2023 · 1 (1) · 1-8

DOI: 10.25163/energy.1110684 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Energy, Environment, and Economy, online erschienen am 01.01.2023.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

As the power grid embraces renewable energy, systems for manufacturing these are being confronted with more and more uncertainty from both fluctuating market demand and sporadic supply of variable renewables to equipment breakdowns as well as strict low carbon emission requirements. Although AI and stochastic optimization appear promising in isolation, their combined application for improving the resilience and low-carbon level of production is still scarce.

Methods

An AI-based stochastic optimization framework that combines machine-learning predictions with two-stage stochastic programming was designed. Artificial neural networks anticipated demand, and Random Forest models predicted equipment failures. Probabilistic outputs were implemented in order to optimize production, energy usage, and emissions with operational (2018–2022) and survey data provided by 210 U.S. solar manufacturing professionals evaluated by means of reliability, factor, and scenario analyses.

Results

The resulting framework achieved substantial improvement over the results of deterministic planning models. Total operating costs were decreased by 16.2%, with energy expenditure reduced by around 20%. Carbon emissions were down by almost 20%, and renewable energy use went up from 41% to 56.8%. Operational resilience also increased as unmet demand decreased from 9.8 to 7.3% and recovery time following a severe disruption decreased from 7.2 to 4.9 days.

Conclusions

AI-integrated prediction together with stochastic optimization makes a self-sustainability system possible in terms of both economic efficiency and resilience along with low-carbon performance, providing such scaled solutions for the sustainable renewable energy manufacturing systems.

Zitieren

APA

(2023). Integrated Artificial Intelligence and Stochastic Optimization Framework for Resilient and Low Carbon Renewable Energy Manufacturing Systems. Energy, Environment, and Economy, 1(1), 1-8. https://doi.org/10.25163/energy.1110684

Vancouver

Integrated Artificial Intelligence and Stochastic Optimization Framework for Resilient and Low Carbon Renewable Energy Manufacturing Systems. Energy, Environment, and Economy. 2023;1(1):1-8. doi:10.25163/energy.1110684

BibTeX
@article{euvimed2023,
  title = {Integrated Artificial Intelligence and Stochastic Optimization Framework for Resilient and Low Carbon Renewable Energy Manufacturing Systems},
  journal = {Energy, Environment, and Economy},
  year = {2023},
  volume = {1},
  number = {1},
  pages = {1--8},
  doi = {10.25163/energy.1110684}
}
Kostenlos starten →