Research On Stochastic Power Flow Calculation And Optimization Strategy Of Microgrid Based On Multi-factor Feature Fusion

The microgrid contains many kinds of distributed generations.Its output power is volatile and intermittent.The power flow distribution and voltage quality of the distribution network are greatly affected.Therefore,it is necessary to deeply study the stochastic power flow calculation and optimal control strategy of microgrid.And the method of determining the grid-connected nodes of microgrid is discussed.At present,the traditional microgrid stochastic power flow calculation methods need to be further investigated in terms of characterizing the stochasticity and correlation of microgrid power flow,selecting the sensitivity function and considering the power flow influence of gridconnected nodes.Specifically,as follows: first,the traditional microgrid stochastic power flow method has the problem which is insufficient to characterize of the overall stochasticity of the system and the correlation of each stochastic factor;second,the position of the partial derivative needs to be further optimized when constructing the sensitivity function of the degree of influence between each stochastic factor and the distribution network power flow in the microgrid;third,based on the stochastic power flow algorithm,how to evaluate the grid-connected nodes of the microgrid needs to be further studied.To solve the above problems,the following studies are progressively carried out in this paper:Firstly,the problem of insufficient stochasticity and correlation portrayal in microgrid power flow is addressed.A novel microgrid power flow calculation and optimal control method based on MSFF-sensitivity is proposed in this paper.In this method,MSFF function is constructed using multivariate normal distribution.the stochastic factor power data are normalized transformed and standardized normalized.And it is effectively unified.As a result,the correlation of multiple stochastic factors in the microgrid and the randomness of power flow are extracted.The problem of insufficient integration of the distribution characteristics of multiple stochastic factors in the traditional stochastic power flow method is solved.Further,the F-sensitivity of the stochastic power flow in the microgrid merged into the distribution network is created.The deficiency of differential sensitivity is compensated.And based on this,the power flow of the microgrid merged into the distribution network is optimally controlled.The algorithm validation shows that the overall stochasticity of the microgrid and the correlation between each stochastic factor can be effectively characterized by the proposed method in this paper.And the power flow of the microgrid connected to the grid are effectively controlled.Then,for the MSFF-sensitivity algorithm,the position of partial derivatives needs to be further optimized.A microgrid power flow calculation and optimal control method based on RSIMSFF-sensitivity is proposed in this paper.The method is based on the MSFF-sensitivity.The method of determining the weights of sensitivity is improved by using rough sets.The problem of optimal selection of the position of partial derivatives is solved.Further,an improved stochastic factor influence weight is used for optimal control of the power flow.The accuracy of the implementation is improved.The algorithm validation shows that the position of the partial derivatives is effectively chosen in the proposed method in this paper.The degree of influence of each stochastic factor in the microgrid on the power flow of the distribution network is more accurately characterized.Further,the power flow of the microgrid connected to the grid is optimally controlled.Finally,the problem of how to evaluate the grid-connected nodes of microgrids for power flow optimization is addressed in the stochastic power flow algorithm.An evaluation strategy for grid-connected nodes of microgrids based on RSIMSFF-sensitivity is proposed.The feeder voltage theory is utilized by this strategy to determine the set of selected node locations for the microgrid grid connection.And the node locations that meet the specification are selected by steady-state analysis.Further,the system voltage deviation obtained from the method proposed in Chapter 4 is used for distribution network operation evaluation.A model for selecting the optimal node location for microgrid integration into the distribution network is established.The algorithm validation shows that the optimal evaluation strategy of grid-connected nodes can be used to maximize the voltage stability of the distribution network.The above algorithm verification shows that the correlation between various stochastic factors in the microgrid can be fully considered in the proposed method in this paper.And the appropriate position of the bias derivatives is selected.Optimization and evaluation of the node locations of the microgrid incorporated into the distribution network.The maximum voltage deviation of the distribution network nodes is 0.0266 pu.The total voltage deviation of the distribution network nodes is 0.208 pu.The voltage quality of the distribution network is improved.