Artificial Intelligence-Driven Computational Physics: Machine Learning Models for Predictive Analysis, Simulation, and Complex Physical Systems
Computer and Decision Making: An International Journal · 05.11.2026 · 3 (6S) · 2892-2904
DOI: 10.1922/comdem.v3i6s.19037 ↗Kurz gefasst
Fachartikel in der Zeitschrift Computer and Decision Making: An International Journal, online erschienen am 05.11.2026. Verfasst von Ashok Kanade und 4 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Themen: Künstliche Intelligenz in der Pflege, Simulation in der Pflegeausbildung
Abstract (Originalsprache)
Computational physics has seen a major boost from the use of artificial intelligence (AI) in predictive modelling, numerical simulation, and analysis of complex physical systems. This paper is a narrative review of the principles, machine learning techniques, applications, challenges, and progress made in the field of AI in computational physics. Machine learning approaches including supervised learning, deep learning, physics-informed neural networks, graph neural networks and hybrid models offer potential for the prediction of physical properties, understanding of nonlinearities and efficiency of computation. AI tools can be used with traditional numerical ones to build surrogate models, reduced order models, and approximate complex physical phenomena. A wide range of fluid dynamics, materials, thermodynamics, climate modelling and multi physics systems illustrate the ability of intelligent computational methods to aid science and engineering optimization. Furthermore, AI-assisted digital twin technologies can be applied for prediction monitoring, adaptive modelling and simulation-based decision making. Still, there are major issues concerning the quality of training data, computational demands, model interpretability, physical consistency, uncertainty quantification and generalization. To ensure scientific credibility, it is important to have reliable validation protocols and standardized benchmarking to guarantee that the AI's predictions are trustworthy. Additional enhancements in computational power may come by means of new generative AI, scientific foundation models, explainable AI and high-performance computing. Lastly, this review highlights the power of using machine learning together with physics-based models to develop more reliable, efficient, interpretable and accurate computational tools for scientific discovery and physical and engineering problem solving in several fields.
Zitieren
APA
Kanade, A., J, P. G., MV, P., Lanke, N. P., & Dr.AL.Nachammai (2026). Artificial Intelligence-Driven Computational Physics: Machine Learning Models for Predictive Analysis, Simulation, and Complex Physical Systems. Computer and Decision Making: An International Journal, 3(6S), 2892-2904. https://doi.org/10.1922/comdem.v3i6s.19037
Vancouver
Kanade A, J PG, MV P, Lanke NP, Dr.AL.Nachammai. Artificial Intelligence-Driven Computational Physics: Machine Learning Models for Predictive Analysis, Simulation, and Complex Physical Systems. Computer and Decision Making: An International Journal. 2026;3(6S):2892-2904. doi:10.1922/comdem.v3i6s.19037
BibTeX
@article{Kanade2026,
author = {Kanade, Ashok and J, Poornima G and MV, Pooja and Lanke, Neha P. and Dr.AL.Nachammai},
title = {Artificial Intelligence-Driven Computational Physics: Machine Learning Models for Predictive Analysis, Simulation, and Complex Physical Systems},
journal = {Computer and Decision Making: An International Journal},
year = {2026},
volume = {3},
number = {6S},
pages = {2892--2904},
doi = {10.1922/comdem.v3i6s.19037}
}
EUVIMED