SBIR Phase I: Utilizing Reinforcement Learning to Optimize Ocean Wave Energy Capture

这项小企业创新研究（SBIR）第一阶段项目的广泛影响旨在促进蓝色经济向大数据范式的持续过渡。目前，对于海上离网、小规模的能源捕获应用，还没有具有成本效益的电力解决方案。该项目的可交付成果可能使商业海洋部门以及联邦政府和地方市政当局受益，使海上电力更便宜、更可靠。这项技术可以帮助规划人员和决策者预测和适应不断变化的海洋环境，最终降低成本，提高纳税人的可靠性。此外，为了实现其商业目标，参与的小企业致力于可持续发展计划，并致力于通过与当地供应商和当地采购的可回收材料合作来减少碳排放。这家小企业还将继续与当地技术培训/贸易学校和劳动力发展项目建立现有的合作伙伴关系，以指导服务不足的学生并创造就业机会。这项小企业创新研究（SBIR）第一阶段项目旨在利用先进的人工智能来优化电力输出。该项目旨在展示先进的机器学习技术的应用，以提高可再生海洋发电的效率和能源捕获，并减少间歇性。该项目通过使用先进的控制模型方法来实现适应性，该方法可以根据环境条件调整设备硬件以优化性能。由于部署环境的原因，本项目将在实验室环境下捕获训练数据，离线训练控制模型，并利用边缘计算将其应用于现场。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

The broader impact of this Small Business Innovation Research (SBIR) Phase I project seeks to facilitate the blue economy’s continued transition to a big-data paradigm. Currently, there is no cost-effective power solution for off-grid, small-scale, energy capture applications at sea. The project deliverables may benefit the commercial ocean sector as well as the Federal government and local municipalities by enabling cheaper and more reliable power at sea. This enabling technology may contribute to the ability for planners and decision-makers to anticipate and adapt to changing marine conditions, which will ultimately reduce costs and increase reliability for taxpayers. Additionally, to achieve its commercial objectives, the participating small business is committed to sustainability in its growth plan and aims to reduce carbon emission by working with local vendors and locally-sourced, recyclable materials. The small business will also continue its existing partnerships with local technical training/trade schools and workforce development programs to mentor underserved students and create jobs.This Small Business Innovation Research (SBIR) Phase I project seeks to leverage advanced artificial intelligence for optimizing power output. The project seeks to demonstrate the application of advanced machine learning techniques to improve the efficiency and energy capture, and reduce the intermittency, of renewable ocean-based power generation. The project enables adaptability by using an advanced control model methodology which adjusts the device hardware based on ambient environmental conditions for optimized performance. Due to the deployment environment, this project will capture training data under a laboratory setting, train the control model offline, and apply it in the field by leveraging edge computing.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.