Session: 10-02: Alternative Energy Conversion Technology (including Wind, Geothermal, Hydro, and Ocean)

Paper Number: 131343

131343 - A Reinforcement Learning-Based Hierarchical Speed Control of an Infinitely Variable Transmission for Tidal Current Energy Converters 

Abstract: 

A tidal current energy converter (TCEC) is specifically designed to harness the kinetic energy present in tidal currents and efficiently convert it into a continuous and stable mechanical rotational energy, thereby enabling power generation. Tidal flows always have very low speeds that rarely exceed 3 m/s. Lower tidal speeds result in lower rotational speed, which is in a range of 4 – 7 rpm. Therefore, if conventional generators are used to produce electricity, a drivetrain of the TCEC is necessary to achieve higher rotor speeds. An infinitely variable transmission (IVT) is a novel solution for the drivetrain design of the TCEC, which can continuously adjust speed ratios with respect to variable input speeds induced by varying tidal speeds to achieve stable speed output for a generator. To ensure the high-efficiency energy harvesting performance of a tidal turbine at any variable input speeds and high operation performance at a desired output speed of the IVT, an accurate and stable control design for continuously varying the IVT's speed ratio is required. This work presents a novel model-based closed-loop control strategy for output speed control of the IVT, aiming to track its speed ratio while reducing fluctuations in its output speed through data-based feedback compensation control. Firstly, based on a simplified dynamic model, a crank-length controller and a forward-speed controller are developed to adjust the IVT's speed ratio according to its varying input speeds, achieving the desired output speed of the IVT. Additionally, the robustness of the speed controller can be enhanced by introducing a reinforcement learning-oriented hierarchical speed control compensator based on monitoring data values of input and output speeds of the IVT. To achieve the best operation performance of the TCEC over most of the working range of the tidal turbine, it should operate at a particular value of the tip-speed-ratio. However, speed fluctuations in the output speed of the IVT from its operation occur due to various disturbances. This reinforcement learning-based hierarchical speed control model effectively reduces fluctuations in IVT output speed while ensuring accurate tracking of the desired output rotational speed under high-torque and low-speed conditions for tidal current energy harvesting. Furthermore, theoretical and experimental studies are conducted on power regulation of the entire IVT system under variable-speed-ratio conditions based on the previous speed control strategy. Experimental results of the variable tidal speed show that speed ratios of the IVT with the proposed control strategy can achieve excellent tracking performances for the desired constant output speed and reduce speed fluctuations of the output speeds of the IVT. These promising results can directly contribute to future research aimed at enhancing efficiency in tidal energy harvesting.

Presenting Author: Gang Li Mississippi State University

Presenting Author Biography: Dr. Gang Li is an Assistant professor in the Department of Mechanical Engineering at Mississippi State University.
From 2021 to 2023, he was an Assistant Research Professor in the Department of Mechanical Engineering at the University of Maryland, Baltimore County (UMBC). From 2017 to 2021, he was a Postdoctoral Research Associate in the Department of Mechanical Engineering at UMBC. From 2016 to 2017, he was a visiting Ph.D. student in the Department of Mechanical Engineering at UMBC. He received the B.S., M.S., and Ph.D. degrees from the University of Shanghai for Science and Technology, Shanghai, China, in 2010, 2013, and 2017, respectively, all in mechanical engineering.
His research interests include the fields of renewable energy technologies, dynamics and vibration, control theory, condition monitoring algorithms, structural health monitoring, and life cycle assessment, and involve analytical development, numerical simulation, experimental validation, and industrial application.

Authors:

Zhichang Qin Tianjin University of Technology
Gang Li Mississippi State University
Weidong Zhu University of Mayland, Baltimore County