Autonomous Microgrids: Theory, Control, Flexibility and Scalability

U.S. Dept of Defense Office of Naval Research

Project Description and Research Objectives:
From large scale electric power grids and microgrids down to small scale electronics, power networks are typically deployed using a fixed infrastructure architecture that cannot expand or contract without significant human intervention. Mobile, monolithic power systems exist but are also not readily scalable to exploit surrounding power sources and storage devices. However, if a power network is constructed from physically independent and autonomous building blocks, then it would be infinitely reconfigurable and adaptable to changing needs and environments. The aim of this project is to integrate vehicle robotics with intelligent power electronics to create self-organizing, ad-hoc, hybrid AC/DC microgrids. The main benefits of this system would be the establishment and operation of an electrical power networks independent of human interaction and can adapt to changing environments, resource and mission. In the context of U.S. Naval platforms, this autonomous electrical network could be used in land, air or sea systems.

The focus of this work will be on land based autonomous microgrid systems, but the fundamental theory developed may be applicable to air and sea based systems as well. Investigators at Michigan Technological University have developed initial hardware and testbeds to study this problem. However, a more detailed theoretical foundation is needed to be developed to apply autonomous microgrids to a wide variety of operational scenarios with various resources. It is also hypothesized that given the flexibility of this approach that it could be equally applied over a vast scale of energy assets. A microgrid that grows in situ from 10 s to 100 s to 1000 s of energy assets can be equally managed, controlled and optimized through the highly scalable approach proposed in this project.

These applications are examples of the critical need for autonomous mobile microgrid capable of operating in highly dynamic and potentially hazardous environments. Our overall goal is to create a scalable architecture to develop a system that accounts for uncertainty in predictions and disturbances, is redundant, requires minimal communication between agents, provides real-time guarantees on the performance of path planning, and reaches the targets while making electrical connections. Such architecture provide a coherent layout for the interconnection between different disciplines on this topic and minimizes the integration concerns for future developments.

Description of the Proposed Work:
• Microgrid Planning and Control
• Microgrid Topology and Optimization
• Electrical Components and Power Flow
• Game-Theoretic Control
• Physical Autonomous Positioning and Connections

Investigator: Wayne Weaver, Rush Robinett and Nina Mahmoudian

Madhi Shahbakhti

MahdiShahbakhtiDr. Shahbakhti joined MTU in August of 2012. Prior to this appointment, he was a post-doctoral scholar for two years in the Mechanical Engineering Department at the University of California, Berkeley. He worked in the automotive industry for 3.5 years on R&D of powertrain management systems for gasoline and natural gas vehicles. Some of his past academic and industrial research experience includes system identification, physical modeling and control of dynamic systems, analysis of combustion engines, utilization of alternative/renewable fuels, vehicular emissions, and hybrid electric vehicles. Shahbakhti is an active member of ASME Dynamic Systems & Control Division (DSCD), serving as the trust area leader and executive member of the Energy Systems (ES) committee and as a member of the Automotive Transportation Systems (ATS) technical committee, chairing and co-organizing sessions in the areas of modeling, fault diagnosis, and control of advanced fuel and combustion systems.

His research focuses on increasing efficiency of energy systems through utilization of advanced control techniques. His current research involves the transportation and building sectors which account for 68% of total consumed energy in the United States. Dr. Shahbakhti’s research to optimize efficiency of energy systems centers on developing and incorporating the following research areas: thermo-kinetic physical modeling, model order reduction, grey-box modeling, adaptive parameter estimation, model-based and nonlinear control.

Areas of Expertise

  • Dynamic Systems Modeling and Control
  • Powertrain/Vehicle Control
  • Internal Combustion Engines
  • Alternative/Renewable Fuels
  • Vehicular Emissions and Aftertreatment Systems

Research Interests

  • Modeling and Control of Energy Systems
  • Hybrid Electric Vehicles
  • Fuel Flex Powertrains
  • Energy Control of Buildings in a Smart Grid

Energy Storage Design Research


From a controls point of view, energy storage systems are the “actuators” in the electrical power grid that enable the mitigation of the transient inputs of power supplies as well as uncontrolled loads. A goal is to optimize the location and amount of energy storage capacity needed to meet microgrid performance and stability constraints. This energy storage capacity can take on many forms from batteries to fly wheels to pumped hydro. Research is focused on integrated energy storage systems that utilize unconventional resources as much as possible. For example, buildings and parking lots full of PHEV’s and EV’s are good targets of opportunity when combined with PV on covered parking structures or distribution-scale PV systems.

Active Research Projects

Energy Storage Design

Energy Storage Design