Open Access. Powered by Scholars. Published by Universities.®

Mechanical Engineering Commons

Open Access. Powered by Scholars. Published by Universities.®

Michigan Technological University

Series

Optimal control

Articles 1 - 2 of 2

Full-Text Articles in Mechanical Engineering

Optimal And Decentralized Control Strategies For Inverter-Based Ac Microgrids, Michael D. Cook, Eddy H. Trinklein, Gordon Parker, Rush D. Robinett Iii, Wayne Weaver Sep 2019

Optimal And Decentralized Control Strategies For Inverter-Based Ac Microgrids, Michael D. Cook, Eddy H. Trinklein, Gordon Parker, Rush D. Robinett Iii, Wayne Weaver

Michigan Tech Publications

This paper presents two control strategies: (i) An optimal exergy destruction (OXD) controller and (ii) a decentralized power apportionment (DPA) controller. The OXD controller is an analytical, closed-loop optimal feedforward controller developed utilizing exergy analysis to minimize exergy destruction in an AC inverter microgrid. The OXD controller requires a star or fully connected topology, whereas the DPA operates with no communication among the inverters. The DPA presents a viable alternative to conventional P−ω/Q−V droop control, and does not suffer from fluctuations in bus frequency or steady-state voltage while taking advantage of distributed storage assets necessary for the ...


Docking Control Of An Autonomous Underwater Vehicle Using Reinforcement Learning, Enrico Anderlini, Gordon Parker, Giles Thomas Aug 2019

Docking Control Of An Autonomous Underwater Vehicle Using Reinforcement Learning, Enrico Anderlini, Gordon Parker, Giles Thomas

Michigan Tech Publications

To achieve persistent systems in the future, autonomous underwater vehicles (AUVs) will need to autonomously dock onto a charging station. Here, reinforcement learning strategies were applied for the first time to control the docking of an AUV onto a fixed platform in a simulation environment. Two reinforcement learning schemes were investigated: one with continuous state and action spaces, deep deterministic policy gradient (DDPG), and one with continuous state but discrete action spaces, deep Q network (DQN). For DQN, the discrete actions were selected as step changes in the control input signals. The performance of the reinforcement learning strategies was compared ...