Load Frequency Control in Microgrids using target adjusted Model Predictive Control

Problem and Research Question

Microgrids must keep generation and consumption balanced despite changes in aggregate load. A persistent, unmeasured disturbance shifts the equilibrium state and the control effort needed to restore nominal frequency. The central question is: can an online disturbance estimate be used to update the steady-state targets followed by an MPC frequency controller?

Method

This paper developed a target-adjusted Model Predictive Control (MPC) formulation for load frequency control in islanded microgrids. The approach combines:

• Receding-horizon optimization — solving a constrained optimal control problem at each time step over a prediction window, so the controller re-plans as new measurements arrive.
• Disturbance estimation — estimating a lumped, approximately constant disturbance from the observed frequency response.
• Equilibrium target calculation — converting that estimate into steady-state state and input targets before solving the MPC problem.

The controller was evaluated in simulation on a linear three-area system containing a tie-line and two generating units. The reported simulations used load-disturbance time series; they did not use renewable forecasts, forecast-uncertainty ranges, or realistic wind and solar profiles.

Personal Contribution

Frederik Banis was the first author. He developed the target-adjusted MPC formulation, designed and ran the simulation experiments, and wrote the manuscript. Co-authors Daniela Guericke, Henrik Madsen, and Niels Kjølstad Poulsen contributed supervision, methodological guidance, and review.

Validation and Key Results

• Both the classical and target-adjusted MPC formulations regulated frequency after changes in the aggregate load disturbance.
• The target-adjusted formulation demonstrated how a disturbance estimate can be translated into updated equilibrium references.
• The study explicitly found no advantage for target-adjusted MPC over classical MPC in the reported simulations. It should not be read as evidence of reduced deviation or faster settling in general.

The method was published in IET Renewable Power Generation (2019), a peer-reviewed journal. As of March 2026, the paper has been cited 20 times (Google Scholar, retrieved 2026-03-24).

Related Outputs

• Conference paper: Utilizing flexibility in Microgrids using MPC, MEDPOWER 2018. DOI: 10.1049/cp.2018.1856
• PhD thesis: Efficient Operation of Energy Grids, DTU, 2020. DOI: 10.11581/DTU.00000334
• Related software: SINDYc and MCMC Framework (Zenodo) — used in subsequent prosumer modeling work

Collaborators and Institutions

• DTU Compute, Technical University of Denmark — Henrik Madsen, Niels Kjølstad Poulsen (supervisors)
• Daniela Guericke (co-author, DTU Compute)

Status and Next Steps

Status: Published (2019). The target-adjusted MPC framework was extended in subsequent work on hierarchical microgrid control (Beus et al., 2020) and SINDYc-based prosumer modeling (Banis et al., 2020).

Transfer to current research: The sequential-decision framework — repeated constrained optimization with online state and disturbance updates — is part of the methodological base that now drives adaptive waveform design for biomedical sensing.

See the related project page for the broader research context and the Methods page for the methodological framework.