Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{202652,
        author = {Asst. Prof. Mohit Manmath Mitkari and Asst. Prof. Ankita Shankar Bawage},
        title = {Physics-Informed Deep Learning for Renewable-Dominated Power Systems: A Review of Forecasting and Grid Stability Applications},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {12},
        pages = {9004-9009},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=202652},
        abstract = {The increasing penetration of solar and wind energy has introduced significant uncertainty and nonlinearity into modern power systems, posing challenges to accurate forecasting and grid stability assessment. Conventional data-driven machine learning models, while effective in pattern recognition, often suffer from limited generalization and lack of physical consistency when applied to dynamic power system environments. To address these limitations, physics-informed learning approaches have recently gained attention by embedding physical laws, system constraints, and domain knowledge directly into neural network models. This paper presents a comprehensive review of physics-informed neural networks and related physics-aware learning techniques applied to solar and wind power forecasting, renewable energy integration, and power grid stability analysis. The review systematically examines model formulations, learning strategies, and constraint enforcement methods used to improve prediction accuracy, robustness, and interpretability. Key applications, including short-term and long-term renewable generation forecasting, transient and small-signal stability analysis, and grid-support functions under high renewable penetration, are discussed in detail. Current challenges such as computational complexity, scalability, data availability, and model validation are critically analyzed. Finally, emerging research directions and open issues are identified to guide future developments in physics-informed learning for renewable-dominated power systems. This review aims to provide researchers and practitioners with a structured understanding of recent advances and practical insights for deploying physics-informed models in real-world power system applications.},
        keywords = {Physics-informed neural networks, renewable energy forecasting, power system stability, solar and wind integration, physics-aware learning},
        month = {May},
        }