One-Day Faculty Development Program
Physics-Informed Machine Learning for Engineering Applications — 25 September 2026.
Program Overview
This one-day Faculty Development Program introduces the foundations, methods and engineering applications of PIML. Four conceptual sessions in the morning are followed by two practical laboratory sessions in the afternoon.
Morning Session 1 — Foundations of PIML
Limitations of purely data-driven models; meaning of physics-informed learning; data, equations and scientific constraints; conventional ANN, PINNs and hybrid models.
Morning Session 2 — Governing Equations
Ordinary and partial differential equations, initial and boundary conditions, conservation laws, residual formulation, physics-based loss functions and automatic differentiation.
Morning Session 3 — Physics-Informed Neural Networks
PINN architecture, collocation points, training, forward and inverse problems, parameter estimation, validation and common optimisation difficulties.
Morning Session 4 — Engineering Applications
Examples from mechanical, civil, chemical, electrical, electronics, biotechnology, computer science and allied engineering disciplines.
Laboratory 1 — Building a Basic PINN
Define a governing equation, create the neural-network approximation, impose initial and boundary conditions, construct the loss and train a basic model.
Laboratory 2 — Engineering Case Study
Compare physics-informed and data-only predictions, examine residuals, assess physical consistency and identify possible research extensions.
Participation & Enquiries
Venue, timings, registration and software requirements will be announced after confirmation.
Register InterestAdvance interdisciplinary research with the PIML community.
Contact us about research collaboration, programs, membership or institutional participation.