In this paper, a Deep Neural Network (DNN)-based efficient interconnected energy management system and Reinforcement Learning (RL) technique are proposed. In this paper being investigated, interconnected M Gs are linked to a main grid, and to maintain local usage of power, they share power from the main grid. The objective from the viewpoint of the distribution system operator (DSO) is to reduce the demand-side peak-to-average ratio (PAR) but also to increase the profit from selling energy. DSO uses a DNN to train the Multi-MG response without having actual access to customer data in order to protect customer privacy. In addition, the DSO optimizes the prediction- based choice by applying the Monte Carlo method of reinforcement learning to determine its selling price strategy. Simulation findings from proposed computation deep learning method and comparison with conventional model-based techniques provide indication for the utility of the proposed method for resolving power system challenges with incomplete or ambiguous information.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Deep Neural Network Based Intelligent Multi-Microgrid Energy Management


    Contributors:


    Publication date :

    2023-08-09


    Size :

    609020 byte




    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Microgrid connection management based on an intelligent connection agent

    Rocabert Delgado, Joan / Azevedo, Gustavo / Candela García, José Ignacio et al. | BASE | 2010

    Free access

    Traffic management system applying deep neural network to intelligent traffic

    LIN XINGYE | European Patent Office | 2021

    Free access

    Intelligent microgrid ecological system based on new energy electric vehicle

    WANG YANSHEN | European Patent Office | 2020

    Free access


    Energy Management and Control for Islanded Microgrid Using Multi-Agents

    Hernandez, Frank Ibarra / Canesin, Carlos Alberto / Zamora, Ramon et al. | BASE | 2013

    Free access