Model Free Adaptive Control (MFAC) for Autonomous and Resilient Microgrids

Fran Li | The University of Tennessee

EW20-5331

Objective

A microgrid is an integrated system consisting of interconnected loads and distributed energy sources with a clear entity boundary, operated in either the islanded mode or the grid-connected mode. Among the many challenges in microgrid operation, the frequency and voltage (f-V) control (or, P-Q control) is one of the most challenging tasks. Many previous works in microgrid control use the trial-and-error approach to develop the Proportional-Integral-Derivative (PID) or Proportional-Integral (PI) control gains, which are critical to microgrid stability and control. However, this trial-and-error approach is very time consuming, and the control parameters for the optimal performance at a given operating point may not be effective at other operating points, such as those seen with a change in microgrid operation mode, a new solar Photovoltaic (PV) installation, or a reconfiguration of the network. With all these challenges, the principal investigator plans to investigate a model-free adaptive control (MFAC) for microgrid V-f regulation in islanded mode or P-Q regulation in utility-connected mode, in which the maximum power point and state of charging of the battery will be considered. The features and advantages of the proposed model-free adaptive control include:

  1. It is based on measurement, rather than the microgrid system model. Thus, it is model-free and suitable for real-time applications, scalable to different sizes of microgrid with different number of devices, and portable to other microgrids.
  2. The control gains are self-adjustable to track a pre-defined, desired trajectory such that a fast and smooth response can always be achieved in order to match the desired trajectory, regardless of external changes (load levels, distribution network, or solar PVs), designer experience for initial control gains, or any offline training studies. Therefore, it achieves the plug-and-play capability for microgrid V-f or P-Q regulation.

Back to Top

Technology Description

The project technology has four distinct features:

  1. A MFAC method is planned with autonomous and dynamic adjustment of its control gains to track a pre-defined desired trajectory, in which the trajectory tracking process is like adjusting the accelerator or decelerator of a moving object to follow the desired trajectory;
  2. The planned control method is independent of external conditions and can achieve a plug-and-play feature, which is highly suitable in microgrid operation that may include dynamic operation conditions with high uncertainty;
  3. The planned control is portable and scalable such that new installation of a distributed energy source (DER) in the same microgrid or transferring existing DERs to a different microgrid requires no re-tuning of the controls; and
  4. The control method is combined with solar PV model, battery model, maximum power point tracking, state of charging constraints, and so on, to achieve a fast, efficient, and scalable V-f regulation under islanded model and P-Q regulation under utility-connected mode, and the seamless transition between these two modes.

Back to Top

Benefits

The project MFAC approach for microgrids will apply to most, if not all, military microgrids with heterogeneous assets and mission-dependent loads for both islanded mode and grid-connected model. It will be an enabling methodology to smoothly accommodate new load additions and changing load configuration. The plug-and-play characteristic of proposed control will make the integration of the DERs and battery storage and coordinating different type of generators much easier than before. With the planned control approach, the U.S. military could save expenses by switching its bases from diesel backup generators to more efficient microgrids with distributed renewable energy generation and battery. Moreover, the military could protect its critical load, benefiting from the added resilience from the proposed method at a much-reduced cost.

Back to Top

Points of Contact

Principal Investigator

Dr. Fran Li

The University of Tennessee

Phone: 865-974-8401

Program Manager

Energy and Water

SERDP and ESTCP

Share