Research On Control Strategy Of Intelligent Micro Grid Based On Cloud Service Platform

At present,the demand for energy in human society has reached an alarming level,and the contradiction between environmental protection and energy utilization has become increasingly acute.The clean energy utilization mode led by new renewable energy has become a global consensus.As a kind of distributed generation,wind and solar power has distinct advantages and disadvantages.In order to integrate the advantages of distributed generation and weaken its adverse impact on distribution network,microgrid technology emerges as the times require.With the emergence of a large number of data mining projects,micro grid can not be used independently.Cloud computing can realize massive data storage and real-time parallel computing of microgrid.Microgrid based on cloud service platform has broad application prospects in monitoring management,data analysis,forecasting and scheduling.Firstly,this paper introduces the research status of cloud computing and the application status of cloud computing services in power system and Microgrid.Through the research on the key technologies of cloud computing and the analysis of the classic control strategy of microgrid,the intelligent local controller suitable for the control requirements of microgrid on cloud platform is designed,namely the central controller of microgrid,micro power controller and load controller.Through the demand analysis of cloud platform,the micro grid control architecture and cloud platform system architecture are designed,and the generation prediction function of cloud platform is described.Secondly,this paper takes the No.2 optical storage microgrid which is universal in the microgrid control architecture as the research object,and proposes a microgrid cloud platform on-off and off-grid switching control strategy.The active off-grid control strategy,the forced off-grid control strategy and the off-grid to grid-connected control strategy are introduced.The application of Back Propagation(BP)neural network and Radial Basis Function(RBF)neural network in photovoltaic power prediction is studied and verified,and a multi-time scale energy management based on power prediction is proposed.Strategies,including real-time control strategy of tie-line power,real-time dispatch strategy of microgrid source storage under grid-connected state and off-grid state.Finally,this paper introduces the implementation scheme and composition of the optical storage cloud platform,designs and displays the system interface and power prediction function,and verifies the parallel off grid switching control strategy,photovoltaic prediction function and energy management strategy of the microgrid cloud platform.The results show that the above strategies can meet the stable operation requirements of the microgrid of the optical storage cloud platform.