Advancements and Applications of Physics Informed Machine Learning (PIML) For Complex Systems Modelling and Prediction

Dr. Ananya Mehta, Prof. Raghav Singh

Abstract


Authors: Dr. Ananya Mehta, Prof. Raghav Singh

ABSTRACT: Physics Informed Machine Learning (PIML) represents an emerging interdisciplinary approach that integrates classical physical laws with modern data-driven machine learning techniques. By embedding prior knowledge of physics into machine learning models, PIML provides a framework that improves prediction accuracy, reduces computational costs, and enhances generalization in scenarios where conventional models struggle due to limited or noisy data. This paper reviews the principles of PIML, highlights key methodologies, discusses recent applications across engineering and scientific domains, and outlines current challenges and future scope. PIML has shown promise in fluid dynamics, climate modeling, structural health monitoring, and biomedical simulations, among other areas. This paper aims to provide a comprehensive perspective on the current landscape and potential of PIML as a transformative tool in computational modeling.

KEYWORDS: Physics Informed Machine Learning, PIML, Data-Driven Modeling, Partial Differential Equations, Computational Simulation, Hybrid Models, Scientific Machine Learning.


Full Text:

PDF 84-95

Refbacks

  • There are currently no refbacks.