Work Package 4

Compare advanced control algorithms and implement predictive and latching controls

 

Objective: Increase OWC turbine annual mean efficiency 50% and improve reliability of PTO Compare advanced control algorithms and implement predictive and latching controls.

Extensive theoretical and laboratory work indicates that, at virtually no additional cost, advanced control algorithms could greatly increase power production and device reliability, thereby significantly improving wave energy prospects. But these are yet to be tested in the open sea.

OPERA will conduct the first implementation at-sea of promising control algorithms that act throughout the power conversion chain: from the hydrodynamics of wave absorption, to turbine aerodynamic and electrical equipment efficiency. Algorithms will be assessed, compared and improved in terms of risk and failure, power production, reduction of peak loads, other drivers of component fatigue, and overall cost-reduction potential for wave energy.

After tuning in a dry power take-off laboratory, algorithms will be tested at the Mutriku shoreline wave power plant, and those which comply with performance and low-risk criteria will be tested on the floating WEC as well.

This Work Package will perform the first at-sea implementation of latching control and predictive control at the Mutriku wave power plant which will be fitted with a two-position fast acting valve between the air chamber and the turbine. This way, latching will be tested without involving unknown risk for the turbine, and allow to be removed/replaced without costly maritime operation. These control strategies are expected to increase energy production of 30% or more once technical challenges of open-sea implementation are addressed.

The Work Package will provide over 1.5 year of open-sea operating data of advanced control algorithms.

Algorithms used to devise the best in class adaptive control
  • Rotational speed control. Generator torque function of instantaneous rotational speed to maximize time-averaged turbine aerodynamic efficiency and hydrodynamic efficiency of wave absorption.
  • Mean pressure referenced adaptive control. Generator torque reference is a function of mean chamber pressure that can be determined off-line. Can include instantaneous chamber pressure to reduce turbine speed excursions.
  • Hourly sea-state adaptive phase control. PTO damping and latching duration function of hourly sea-state. Results indicate that performance can approach that of sub-optimal control while circumventing the need for future wave information.
  • Adaptive latching phase control. Instantaneous (non-predictive) values of turbine rotational speed (already adequately controlled) and chamber air pressure used to control valve open-close timings. Test at shoreline plant only.
Algorithms used to devise the best in class predictive control:
  • Simplified model predictive control. Adapts well developed model predictive control algorithm to wave energy applications, i.e. with limited wave observation, prediction and computation capabilities. Will determine the extent to which simplified next wave information can be used to enhance performance and reduce peak loads.
  • Near-optimal control. Using wave observation with 1-10 second lead time, turbine rpm is controlled to maximise efficiency across the power conversion chain.
  • Predictive latching phase control. Uses wave observation with 1-10 second lead time to operate the latching valve. Test only at shoreline plant, decisive for future floating application.

Task to be performed

Customisation of algorithms and instrumentation
Existing TRL3-4 versions of the control algorithms will be customised for Mutriku and the OCEANTEC WEC, including initial control parameters for turbine rpm and closure of its shut-off valve for protection. Data needs will be assessed in detail and provided to WP1, with particular attention to the placement of wave measuring instruments upwave of the device trading off observation lead-time and accuracy.
Dry power take-off lab tests to prepare open-sea operation
A physical model of the OCEANTEC WEC power conversion chain (PTO lab) will be available free of charge for the project. Control algorithms will be implemented in the logic controller and the test bench experimental data will be used to tune control parameters. In turn, information from open-sea testing will be used to enhance the realism of the PTO lab for applications in this and future projects.
Open-sea testing: Mutriku shoreline wave power plant
Algorithms will be compared regarding risk and failure data, desired trade-off between power-production and component lifetime, with the primary indicator being levelised cost of energy. A “best in class” adaptive control and a “best in class” predictive control will be devised and proposed. Recommendations will be proposed regarding selection of control algorithm for different applications, design requirement for control valves, wave measuring instruments and other control components, promising avenues for cost-reducing technological developments and innovations.
Open-sea testing: floating OWC at bimep
Methodology largely identical to the previous task, albeit with expected differences such as focus on safety and failure modes specific to floating applications, less accurate wave prediction, and more stringent operating range for power components as maintenance requires far costlier maritime operation.

Involved Partners

This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 654.444