10/20/2026

Oscilla Power (OPI) is developing the Triton wave energy converter architecture. This system is a two-body, multi-mode point absorber consisting of a surface float connected to a submerged ring via three tendons. As the surface float is articulated by waves, the two bodies react against each other in all modes of motion to capture mechanical power. The resulting tension variation in each tendon is transmitted to power take-offs (PTO's) housed inside the surface float. The system has the ability to submerge in extreme waves to mitigate storm loadings, a strategy that has been validated both numerally and in physical experiments. OPI is developing different WEC variants based on the Triton architecture that will share these core operating principles. At the time of the TEAMER testing, there were two variants, the Triton (MW-scale) and the Triton-C (100kw), which are differentiated by their size and drivetrain design. The Triton (MW) has a linear input driving a hydraulic power conversion system, while the Triton-C has a rotary input driving an electrical conversion system.
Our work preceding the test plan on the Triton Utility system had been focused on numerical geometry optimization. This involved a collaboration with the University of Edinburgh. We had been exploring the use of a genetic algorithm to optimize the hull geometry to maximize power performance. Similar work has been done previously in the literature in the frequency domain (using WAMIT) with good success. We had taken this work further by developing a novel fast time-domain evaluation algorithm and incorporating it into the optimization routine. By using a time-domain approach, our expectation is that we can increase confidence in the results through a more accurate representation of the WEC.
Our initial work had generated a significantly different hull geometry that (numerically) provides a substantial AEP improvement over our existing hull shape. Further optimization with different evaluation conditions was planned to identify whether a global maximum had been found. The code used, WAMIT, is extremely well-validated for ?conventional? shapes typically found in the marine environment. However, the unconventional geometries produced through this GA approach are possibly less well explored. It is thus extremely important that a reference physical model validation of the generated shape is produced to confirm the results.
The approved test plan specified a minimum 1:50 scale physical model of the Triton with a representative PTO and an accurately modeled hull geometry in the multidirectional basin at the Hinsdale Laboratory at Oregon State University. The September 2021 campaign provided a side-by-side comparison of the existing baseline hull and the new candidate hull. Both models were 3D printed at OPI and fitted with representative power takeoffs and sensors for measuring power capture. The model-development programme selected 1:50 scale as an existing baseline model was available at this scale for direct comparison.