Sbir phase II: utilizing reinforcement learning to optimize ocean wave energy capture

This $999,977 Cooperative Agreement award from the National Science Foundation (CFDA 47.084 - NSF Technology, Innovation, and Partnerships program) will enable Ocean Motion Technologies Inc., a minority-owned small disadvantaged business, to develop an artificial intelligence system that optimizes energy capture from ocean wave energy devices. The key innovation is a system for training and fine-tuning AI models to the specific wave environments where the devices are deployed, which can increase energy output by 27% to 33%. Phase II of the project will extend this work by enabling remote fine-tuning of the AI system on local wave data and making the solution commercially accessible through a cloud platform. This will increase harvested power and provide better control and management of energy output for maritime data buoys, benefiting organizations like fisheries, research groups, government agencies, and coastal communities.

Abstract

The broader impact of this small business innovation research (sbir) phase II project will be to benefit companies and organizations ? Including fisheries, scientific research organizations, government agencies, and coastal security programs ? That collect data at sea by reducing the cost of powering data buoys. Power generation is typically one of the major costs for operating a data buoy. This project aims to develop a market-ready artificial intelligence system to increase the energy output of a wave energy device by dynamically changing the wave energy device?s settings based on incoming waves. This project enhances our understanding of the applications of artificial intelligence systems to increase energy output from renewable sources that have changing environments, specifically wave energy devices. As a result of lowering energy costs, this project will improve end-user capabilities to manage maritime assets and logistics, increase oceanographic and climatological data collection, and enable government agencies to collect more information. Additionally, this project enables local, state and federal agencies to deploy offshore water quality and environmental tracking buoys at lower cost, which is especially important for disadvantaged coastal communities. For example, an early intended beneficiary of this technology is an underserved coastal community that needs to track local wastewater discharge. This small business innovation research (sbir) phase II project will develop an artificial intelligence system that optimizes energy capture from incoming waves. The key innovation is a system for training and fine-tuning artificial intelligence models to ambient wave environments in which the wave energy device is deployed. This is important, since wave environments can vary dramatically. In phase i, the system was shown to increase energy output by 27% to 33%. Phase II will extend this work by enabling fine tuning of the system on local wave data, and by making the innovation commercially accessible on a cloud platform. By making the artificial intelligence system available for training and fine-tuning in a cloud environment, customers will be able to remotely update the artificial intelligence system for deployed wave energy devices on data buoys. This will increase harvested power and provide better control and management of energy output for data buoys. The research undertaken as part of this project will make improvements to power generation beyond those realized in phase i. The anticipated technical result of this research is a wave energy device with increased output power that uses artificial intelligence, as well as an innovative system for delivering this solution to market. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the foundation's intellectual merit and broader impacts review criteria.- subawards are not planned for this award.