Reaserch On Reconfiguration Of Distribution Network Considering Unbalanced Conditions

At present,most of the research on the reconfiguration of distribution network is carried out under the condition of three-phase balance of distribution network,and only the equivalent single-phase network is analyzed independently.However,with the development and progress of social level in our country,new loads such as electric vehicles and distributed generators have developed rapidly with the advantages of low pollution,high substitution rate and low noise.As a result,the actual operation of distribution network is usually accompanied by large single-phase and two-phase loads,which do not meet the balanced conditions.Therefore,a reasonable distribution network needs to expand the traditional single-phase network analysis—in the condition of three-phase unbalanced operation,reconstruct the network and eliminate the impact of unbalanced conditions.And this will have important practical significance.Based on the unbalanced distribution network,the work in this paper is as follows:The three-phase power flow calculation method for unbalanced distribution network is studied,according to the generation characteristics of distributed generation,it is divided into four main types of nodes.Then,based on a direct matrix method for three-phase power flow calculation,the calculation steps of three-phase power flow with DG are derived.Under this method,combined with the improved IEEE33 node three-phase example,the influence of DG access location,capacity and phase on the unbalance of distribution network is analyzed.Under the background of information big data era,a more practical segmentation method is introduced.First,the charging curve of electric vehicle is obtained through monte carlo simulation,which is superimposed with the daily load curve to form the equivalent daily load curve.On the basis of fisher-optimal segmentation,this paper proposes a new method of dividing time segments by information entropy theory to ensure the rationality and rigor of time segments.Secondly,choose the lowest degree of three-phase unbalance as the optimal target,establish the corresponding dynamic reconstruction model in the three-phase distribution network.In the reconfiguration of distribution system,an improved imperialist competition algorithm is used to solve the problem.In the link of colonial revolution,compound differential evolution is introduced.At the same time,an adaptive empire merging mechanism is proposed to improve the convergence speed and accuracy of the algorithm.Simulation resultsverify the rationality of the proposed dynamic reconstruction method.A comprehensive optimization strategy for distribution network reconstruction and node phase switching is proposed.Based on the day-ahead decision scheme,a three-phase current unbalance degree constraint is introduced,and the subjective unbalance degree of the whole day is defined and transformed into the corresponding constraint.According to the above single constraint and all-day unbalanced constraint as the basic limit,three different optimization methods are effectively developed.Choosing the minimization of the subjective three-phase unbalance in the whole day as a comprehensive optimization,a compound coding solution is proposed combined with the comprehensive strategy.Finally,the simulation of EV access scenarios with different degrees of equilibrium is taken into account in the IEEE33 three-phase node calculation example,which verifies that the method proposed in this paper can effectively reduce the degree of three-phase unbalance of the system and,in most cases,simultaneously reduce the network loss of the system.It has practical application value to ensure safe and stable operation of distribution network.