專利授權區國立清華大學國際產學營運總中心 Operations Center for Industry Collaboration

搜尋專利授權區

關鍵字

» 新增關鍵字

選單

專利授權區

專利授權區
專利名稱(中)	基於強化學習的再生能源配置方法及裝置
專利名稱(英)	METHOD AND APPARATUS FOR RENEWABLE ENERGY ALLOCATION BASED ON REINFORCEMENT LEARNING
專利家族	中華民國：I767525 美國：11,861,702
專利權人	國立清華大學 100%
發明人	邱偉育,莊喻捷
技術領域	光電光學

專利摘要(英)
A method and an apparatus for renewable energy allocation based on reinforcement learning, adapted for an energy aggregator having an energy storage system (ESS) to determine renewable energy allocation between multiple energy suppliers and multiple energy demanders, are provided. In the method, historical power generation data of each energy supplier is collected and used to generate renewable energy indexes representing uncertainty of the renewable energy. Multiple market indexes associated with a state of a renewable energy market are collected and integrated with the renewable energy indexes and electricity information of the ESS to generate multiple states of Markov’s decision process. The states are inputted to a reinforcement learning model to determine prices bid for the energy suppliers and the energy demanders. According to supplies and demands submitted by the energy suppliers and the energy demanders in response to the prices, the ESS is adjusted to coordinate supply and demand of the renewable energy, and the reinforcement learning model is updated.

專利摘要(英)

A method and an apparatus for renewable energy allocation based on reinforcement learning, adapted for an energy aggregator having an energy storage system (ESS) to determine renewable energy allocation between multiple energy suppliers and multiple energy demanders, are provided. In the method, historical power generation data of each energy supplier is collected and used to generate renewable energy indexes representing uncertainty of the renewable energy. Multiple market indexes associated with a state of a renewable energy market are collected and integrated with the renewable energy indexes and electricity information of the ESS to generate multiple states of Markov’s decision process. The states are inputted to a reinforcement learning model to determine prices bid for the energy suppliers and the energy demanders. According to supplies and demands submitted by the energy suppliers and the energy demanders in response to the prices, the ESS is adjusted to coordinate supply and demand of the renewable energy, and the reinforcement learning model is updated.

聯絡資訊
承辦人姓名	李曉琪
承辦人電話	03-5715131 #31061
承辦人Email	hsiaochi@mx.nthu.edu.tw