Method for learning of power transaction model based of deep reinforcement learning and method for transacting power using the power transaction model

• Main Authors: SEON JOON HO, LEE SEONG WOO, KIM SOO HYUN, KIM JIN YOUNG
• Format: Patent
• Language: English; Korean
• Published: 31.03.2023
• Subjects: CALCULATING; COMPUTING; COUNTING; DATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FORADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORYOR FORECASTING PURPOSES; ELECTRIC DIGITAL DATA PROCESSING; MEASURING; METEOROLOGY; PHYSICS; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE,COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTINGPURPOSES, NOT OTHERWISE PROVIDED FOR; TESTING

A learning method of a deep reinforcement learning-based power trading model and a power trading method using a deep reinforcement learning-based power trading model according to a preferred embodiment of the present invention may learn a deep reinforcement learning-based power trading model for distributed power supply systems, and control the power trading of the distributed power supply system using the learned deep reinforcement learning-based power trading model, so as to improve the forecasting accuracy of power demand, power supply, and power trading revenue. 본 발명의 바람직한 실시예에 따른 심층 강화 학습 기반 전력 거래 모델의 학습 방법 및 심층 강화 학습 기반 전력 거래 모델을 이용한 전력 거래 방법은, 분산형 전력 공급 시스템을 위한 심층 강화 학습 기반 전력 거래 모델을 학습하고, 학습한 심층 강화 학습 기반 전력 거래 모델을 이용하여 분산형 전력 공급 시스템의 전력 거래를 제어함으로써, 전력 수요, 전력 공급 및 전력 거래 수익의 예측 정확도를 개선할 수 있다.