Industrial Intelligence A structured map of industrial AIA structured map of industrial AI
LinkedIn
Technology
Status
Fit
Effect
Hover any cell to preview
Engineering & Simulation · Production
ML for Engineering / Physics-informed & Hybrid ML
Physics-informed process models for production engineering
Research Support low effect
Core capability
Because the model is constrained by physics as well as data, it can produce results that are more trustworthy in engineering contexts than purely data-driven models alone.
How it works
The model is trained not only to match historical data but also to respect physical laws, which helps reduce unrealistic outputs and makes the results more trustworthy in engineering applications.
Application here
Models embed process physics so predictions remain more robust when production conditions change or data is limited.
Business impact
This can improve prediction reliability where purely data-driven approaches become fragile under drift.
Limitations
It requires strong process-physics expertise, and wrong assumptions can make predictions worse rather than better.
In production
This is already valuable when pure machine learning is too unreliable but full first-principles simulation is too slow or difficult to operationalize.
Research
The frontier is toward systems that can recover or refine the hidden rules of a process from measurement data, making advanced modeling possible even when the full physics is not already written down.
Examples
Physics-Informed Machine Learning for Smart Additive Manufacturing
Description: PIML model integrates neural networks with heat transfer physics for laser metal deposition in additive manufacturing. Predicts temperature fields using limited data while satisfying governing physical equations, demonstrated through case studies showing improved accuracy over data-driven ML alone. Applied in production to enable real-time process monitoring.


Building hybrid virtual flow meters: ML + physics
Description: US oil & gas operator deployed hybrid VFM combining ML predictions of flow rates from sensor data (pressure, temperature) with physics-based fluid dynamics constraints. Achieves 60-80% cost reduction vs physical meters, handles multiphase flow across well lifecycles in production environments.


Physics-Informed and Physics-Guided Machine Learning in Industrial Rotating Equipment
Description: Refinery centrifugal compressor uses PIML embedding two-phase flow physics and thermodynamic constraints into ML models for early liquid carryover detection from sensor data. Prevents blade erosion and surge by producing physically consistent predictions even with unlabeled rare events.
https://arxiv.org/abs/2407.10761https://xenoss.io/blog/hybrid-virtual-flow-meters-ml-physics-modelinghttps://www.linkedin.com/pulse/physics-informed-physics-guided-machine-learning-industrial-bkw3c
Sources
NVIDIA PhysicsNeMo — ; Ansys SimAI — ; Raissi et al., PINNs —
https://developer.nvidia.com/physicsnemohttps://www.ansys.com/products/ai/ansys-simaihttps://doi.org/10.1016/j.jcp.2018.10.045