Research On Economic Optimal Operation Strategy And Method Of Microgrid Based On Multi-Agent

In recent years,with the social awareness of environmental protection and to reduce the limitation of fossil energy shortage,China has further put forward the goal of "Peak carbon dioxide emissions and carbon neutrality".The development of low-carbon economy and compliance with the green low-carbon trend are driving the power grid to develop in an efficient,flexible,intelligent and sustainable way.As the basic feature of the future power grid,sustainability is expressed as renewable power generation will be more and more widely connected to the grid.As a flexible and diverse organization form,the microgrid provides an effective way for the control of distributed generation.However,the characteristics of flexible topology,complex operation features and rich control objects of microgrid make the centralized control mode difficult to apply.In the actual operation of the microgrid,the communication delay and failure are inevitable.When solving the economic objective,it is necessary to receive/send information and calculate the optimal output value in a short time.In view of the important role of microgrid in smart grid,this paper carries out in-depth research on the economic optimal operation of microgrid and interconnected multi-microgrid.The main contents of this dissertation are shown below.Firstly,the optimal operation model of microgrid based on multi-agent hierarchical architecture is proposed.To fully reflect the specific operation mode of multi microgrid and distributed generations,a multi-agent based hierarchical control architecture is established.The architecture includes two layers as "multi microgrid layer-microgrid equipment layer".The different time scales,control objects,control range and implementation mode of the two-layer optimal operation strategy are analyzed.At the same time,the function,type and cost function of each agent are designed and described.On this basis,the optimal operation model of microgrid is established.The consensus algorithm is used to verify it.Secondly,aiming at the real-time economic optimal operation of microgrid,a distributed improved diffusion algorithm is proposed,which can achieve fast convergence.On the basis of the diffusion algorithm,a modification step is added and the gradient term is replaced,so that the active power balance in the microgrid can be ensured in the calculation.And the fully distributed operation can be supported without any "leader".At the same time,the convergence ability of the algorithm is analyzed theoretically,and the convergence rates of the consensus algorithm and the improved diffusion algorithm are compared.The results show that the algorithm has fast convergence speed and can stably converge to the optimal value even when the communication topology changes.Meanwhile,it can meet the requirements of "plug and play" capability for devices in microgrid.Thirdly,considering the influence of communication delay and failure on microgrid in the non-ideal communication state,the economic optimal strategy of microgrid is proposed.And the convergence and stability of the algorithm under delay are analyzed.The optimal operation model of microgrid in the presence of delay is discussed,and the proposed algorithm is modified to make it more suitable.The theoretical upper bound of the algorithm parameter is given,which can ensure that the algorithm converges to the theoretical optimal value.The results show that the communication delay will slow down the process of obtaining global information,which is also reflected in the algorithm as the number of iterations will increase.However,even if the communication delay exists,when the step size ? is less than the theoretical upper bound,the algorithm can still achieve stable convergence.Fourthly,aiming at the coordinated optimal problem of interconnected multi microgrid system considering the uncertainty of renewable energy output and load,a two-level coordinated optimal operation strategy is proposed.The proposed stratregy can take into account both the economic objectives of microgrids and the local consumption objectives of renewable energy.The day ahead collaborative optimization scheme of interconnected multi microgrid system is designed in detail,and a two-layer coordination optimization model is established.The upper layer of multi microgrid calculates the internal incentive price,and the lower layer takes the uncertainty into account.With the economic optimization objective,each microgrid calculates the output plan according to the incentive price.There are two-way and multi round interactions between the two layers.In order to solve the model,a hybrid interactive iterative algorithm based on bat algorithm is proposed.The results show the effectiveness of the proposed two-layer coordination optimization model.At the same time,it is proved that the coordination optimization among multiple microgrids can not only effectively improve the system operation characteristics such as peak valley difference and volatility,but also promote the local consumption of renewable energy and improve the economy of microgrid.