| Microgrids are designed to regulate energy generation and consumption within communities spreading over a limited geographical space.Typically,microgrids provide energy for isolated communities that are not connected to the main power grid.However,the presence of various stochastic energy sources such as PV arrays and wind turbines,and the varying structure of the control network present uncertainties and challenges.Potential communication link interruptions and delays can lead to various problems with the previously used distributed cooperative control scheme,therefore,a more mature control scheme needs to be found to mitigate the impact of communication faults and other problems.This paper proposes a hierarchical control method for AC microgrid under communication latencies and failures,which solves many existing problems of the original control method.In this paper,a detailed mathematical model is obtained through derivation and modeling,and a small signal model is established for each control method proposed.Using these mathematical models,the results of stability analyses are obtained,and the effectiveness of the proposed scheme is analyzed in detail.In addition,the effectiveness of the proposed control method is further validated through specific case studies.Finally,the proposed control scheme is tested and verified on the experimental MG platform developed using power inverters.The main contributions and innovations of this article are as follows:1)The hierarchical control structure of the microgrid under the operation of the island is introduced in detail(chapter 2).The control principle of microgrid,and the control method of parallel connected converters is discussed.Dynamic differential equations are derived,the mathematical model is introduced,and the basic experimental setup are described.In addition,the collaborative control under the networked control system and multi-agent system operation of microgrid is described using graph theory.2)Multiple link failures in the communication network of microgrids can form communication islands which may lead to unstable operation of AC microgrids.In order to resolve the problem,a multi-agent consensus-based method control scheme(Chapter 3)is proposed.Firstly,an effective method for fast detection of communication segmentation of the microgrid network is devised.This method provides detection of islanding by maintaining a connectivity graph at each node and dynamically updating it with the varying status of communication links.Using this method,the formation of communication island is analyzed and the information about the number of members of each sub-island is generated in few steps by detecting the cut sets of the graph.Secondly,these communication islands are treated as virtual microislands on electrical side for control implementation.The control gains in each virtual micro-island are updated through gain scheduling,which can effectively achieve the control of power sharing,voltage regulation and frequency regulation.Lastly,through the derivation of detailed mathematical models and stability analysis,the sensitivity of the control parameters and stability of the microgrid are analyzed,and the effectiveness of the proposed control method is verified by simulation studies on different communication link failure scenarios.3)The conventional multi-agent communication-based control systems mostly utilize consensus-based control strategies which perform poorly if large intermittent communication delays are encountered,due to their integrator-based structure.To solve this problem,a novel multi-agent quasi averaging observer is proposed without utilizing integrators or large gains,which makes it more resilient to communication delays(Chapter 4).The proposed observer is incorporated into a multi-agent based hierarchical control structure for AC-microgrids that collectively achieves sharing of active power,regulation of voltages and restoration of frequency.Through detailed mathematical analysis and specific simulation case studies,it is shown that in comparison with the existing consensus-based control methods,the method proposed in this thesis is more robust to communication delay and has better dynamic characteristics.4)In Multi-agent based control methods,the response speed of the parameter average estimator is the key factor in enhancing the control performance of the microgrid.In order to further improve the performance of the method mentioned above,a new multi-agent moving average estimation algorithm is proposed(Chapter5),which can estimate the average of the required parameters within a limited number of steps,which is used as a reference input for the control system,so as to minimize the deviation of the charged parameters(such as power and voltage).Based on the derivation of the whole system discrete time small signal model,the performance of the method is analyzed,and the effectiveness of the method is verified by case study simulations.Compared with the traditional consensus-based control scheme and the control method proposed in Chapter 4,this method can have better robustness and dynamic characteristics in the case of large delay in communication links.5)A distributed collaborative control scheme for reactive power sharing and voltage recovery(Chapter 6)is proposed to solve the problem of reactive power balance in the islanded power electronic converter based microgrid.In addition,the control scheme is adaptable to the time delay of the communication link and can solve the network problem of communication link delay in the "community grids",such as microgrids and microgrid clusters where the DGUs are located close to each other.The effectiveness of the control scheme was verified through case studies in different scenarios. |
PDF Full Text Request