Control of WECs based on dielectric elastomer generators

Overview

This project examined the control of Wave Energy Converters (WECs) based on Dielectric Elastomer Generators (DEGs). DEGs are a class of Power Take-Off (PTO) systems made of polymeric rubber-like materials, which exploit a variable-capacitance electrostatic principle to directly convert wave energy into electrical energy. Previously investigated in the framework of two WES-PTO projects (Stage 1 and 2), DEGs feature promising performance in terms of efficiency, adaptability to different WECs (e.g., oscillating water columns and pressure differential WECs), LCOE and architectural simplicity.

Compared to other PTO systems, DEGs show some peculiar features:

• DEGs contribute to the dynamics of coupled WECs not only through controllable (electrostatic) loads, but also through their rubber-like elastic response, which is non-linear.
• Their power electronics (namely, a bidirectional DC-DC converter) is different from the electrical machines usually involved in the operation of other PTO systems, and require dedicated drivers.

As a consequence, DEG PTOs require dedicated control logics and control systems, able to practically drive the electrical state of these devices in a way to guarantee robust, reliable and efficient performance of coupled DEG-WEC systems.

This project aims at investigating the feasibility of new control logics and control system architectures for DEG-based WECs. Two levels of control will be investigated:

• High-level control. Here the objective is to identify optimal control strategies that maximise the DEG performance while enforcing the respect of operating constraints based on DEG survivability and on the feed-back on the DEG’s state monitoring. In doing that, a multidisciplinary approach will be pursued. The identification of the most appropriate/performant control algorithms will be performed using model-based dynamical optimization or machine learning techniques.
• Low level control. Here the objective is to optimally drive the power electronics in order to successfully implement the control strategies investigated at high-level while maximising the efficiency of the electronics. This problem will be tackled using hybrid control techniques.

The most promising high-level and low-level control solutions will be used to define integrated control system architectures and identify key control variables, sensing equipment, and hardware requirements for the practical implementation of the control systems.

Cheros Srl also presented a poster on their Stage 1 Control Systems project at the 2017 WES Annual Conference.