Master Thesis – Reinforcement Learning Control of a Liquid–Liq...
Hybrid
Core
Develop an offline reinforcement learning controller for a liquid-liquid gravity settler using physics-informed neural network models as a virtual training environment.
Role type
Master thesis researcher (Reinforcement Learning & Process Control)
Builds
RL control policies for industrial separation processes
Domain
Chemical Engineering / Energy Systems / Machine Learning
Deliverable
production ML models
Required skills
Python, Physics-Informed Neural Networks (PINN), Modeling, Simulation, Control Engineering
Preferred skills
PyTorch, Reinforcement Learning
Technologies
Python, PyTorch, PINN
Responsibilities
Formulate control problems and design reward functions, Handle process constraints during learning, Evaluate trained controllers on detailed process models
