Are you driven by rigorous control design for large-scale dynamical systems? Do you want to develop scalable Model Predictive Control (MPC) architectures for complex multi-commodity energy networks operating under constraints and uncertainty? This PhD position focuses developing scalable Model Predictive Control architectures for complex multi-commodity energy networks operating under constraints and uncertainty within the BACH project, a real -life demonstration project on TU/e campus.
InformationFuture energy infrastructures integrate electricity, heat, gas, storage, and electric mobility into tightly coupled multi-commodity networks. These systems exhibit high-dimensional state spaces, strong interconnections, multi-timescale dynamics, mixed-integer operational decisions, and market-driven economic objectives. Traditional rule-based strategies are structurally inadequate, while centralized MPC approaches face scalability limitations.
In this PhD project, you will develop theoretically grounded and computationally scalable hierarchical and distributed MPC architectures for large-scale multi-commodity energy systems.
Research Objectives - Design multi-layer control hierarchies separating regulatory control, supervisory MPC, and real-time economic optimization.
- Analyze stability, recursive feasibility, and closed-loop performance under hierarchical coordination.
- Develop distributed and decentralized MPC formulations using primal and dual decomposition methods.
- Establish convergence guarantees and communication-efficient coordination mechanisms.
- Apply graph-theoretic and network-based decomposition methods for control-relevant system partitioning.
- Perform computational complexity and scalability analysis for real-time feasibility.
The project combines constrained optimal control, nonlinear and hybrid systems, large-scale optimization, network science, and energy system modeling. Results are expected to lead to high-quality publications in leading control and optimization journals.