Research On Hierarchical Intelligent Control Strategy For Active Distribution Network With Multi-microgrids Based On Sparse Communication Network Optimization

In recent years,with the rapid development of smart grid and energy internet,the number of micro-grids in distribution network is increasing.The generation and consumption of micro-grids only consider their own interests maximization,thus ignoring the economy as a whole.However,the traditional single microgrid and distribution network only transmit electricity through medium voltage lines.Because of their far geographical location,they often lead to large transmission losses.To this end,the geographically adjacent microgrids are interconnected to form a micro-grid cluster system,i.e.multi-microgrid clusters.As a relatively independent autonomous system,multi-microgrid clusters and active distribution network are managed by micro-grid operator and distribution network operator respectively.The operation status of multi-microgrid clusters will affect the scheduling strategy of distribution network operator,and the decision-making of distribution network will also affect the scheduling results of multi-microgrid clusters.In order to coordinate and optimize the operation of active distribution network and multi-microgrid clusters,this paper studies the network structure,optimal dispatching strategy,operation control strategy and multi-energy network modeling of active distribution network with multi-microgrid clusters.The main work and achievements are as follows:(1)The operation control Strategy of AC active distribution network based on radiation architecture is proposed.Considering the objectives of economic benefit,environmental protection and electric-thermal comprehensive benefits,the constraints of integrated energy system operation are analyzed and the energy management system on energy router is constructed.Secondly,multi-agent platform is used to classify the agents in AC active distribution network,and the functions of corresponding modules are defined.The optimization of communication network topology is used.To improve the droop control strategy of distributed equipment converter,the precise power control of AC active distribution network in different scenarios such as grid-connected,island-connected,grid-connected to island-connected,island-connected to grid-connected is realized.(2)The operation control strategy of DC active distribution network based on multi-level ring structure is proposed.Firstly,the typical structure of DC active distribution network is discussed,such as radial,hand-in-hand and ring.Then,an ring-type DC active distribution network is connected with natural gas network and thermal network to form a multi-energy network.Considering the characteristics of multi-time scale control,the control architecture of DC distribution network is divided into three layers: energy management system layer,bus voltage control layer and converter control layer,and the operation of DC active distribution network under three-terminal power supply,single-terminal power supply,microgrid splitting and so on is simulated.(3)The operation control strategy of AC/DC hybrid active distribution network based on central load structure is proposed.In the four-terminal AC/DC hybrid active distribution network,the control functions of each converter station are defined,including constant power control,constant DC voltage control and maximum power tracking control.At the same time,according to the power regulation capability,the converter station agents are divided into three categories,and the combined voltage control strategy between the converter station agents is discussed.The influence of communication delay on the distributed cooperative control process of the converter station agents is analyzed.(4)A control strategy for future active distribution network operation based on energy cell-tissue structure is proposed.Aiming at the physical relationship of regional energy interconnection system in the future,an active distribution network power allocation framework based on energy cell-tissue structure is constructed by using the basic theory of bionics and the control idea of artificial intelligence.At the same time,according to the different geographical location,control range and power generation type,the cells in energy tissue are divided into leaders and followers.The cost function of energy cells is analyzed,and the leader-follower control mode is discussed.Considering the factors of communication loss,communication noise and energy cell failure,the virtual consistency variable and variable convergence coefficient are introduced to solve the disadvantageous influence of the frequent change of communication topology due to the random input and exit of energy cells on the convergence of distributed cooperative economic algorithm.Compared with the centralized optimization algorithm,the advantages of the proposed strategy are further verified.(5)The interactive mechanism and hierarchical control strategy of active distribution network and multi-microgrid clusters are proposed.In the generalized active distribution network,the multi-energy network is modeled,including the thermal network,natural gas network and power network,as well as the coupling equipment of electrical and thermal network.At the same time,the interaction mechanism between active distribution network and microgrid is discussed,including the interaction between distributed generation and microgrid,the interaction between microgrid clusters and active distribution network.Finally,the edge network of communication system is designed.Aiming at the research hotspot in this field,according to the multi-agent(distribution network,micro-grid,components)in active distribution network with multi-microgrid clusters and the popular mobile multi-agent technology in the field of artificial intelligence,this paper puts forward the interaction mechanism and operation controller of different distribution network architectures,such as AC active distribution network,DC active distribution network,AC/DC hybrid active distribution network,etc.The validity of the proposed control strategy is verified by a large number of simulation results and practical operation results of some projects.The research results of this paper can effectively improve the reliability of active distribution network,improve power quality and economic benefits,and provide theoretical reference for the future distribution network and even the distributed and intelligent operation of energy internet.