Research On Load Transfer Control Strategy Of Urban High Voltage Distribution Network

The rapid development of the national economy is accompanied by the rapid growth of urban load.Then,the load density of the city is greatly improved.However,the lag of the construction of high-voltage distribution network(HVDN)and the actual growth of the load has been out of touch.During the peak summer or winter load,load distribution of city H VDN is not balanced.It is easy to cause the 220kV transformers overload and power congestion.Thus,the safe and stable operation of HVDN is affected.In such a situation,regional dispatchers often carry out frequent or even large area load transfer operations.Due to the lack of adequate theoretical guidance,they are usually based on experience to transfer the load of some 220kV substations with high load rate to other 220kV transformers with low load ratio.But,when faced with the increasing complex structure,more and more heavy load in the urban HVDN,dispatchers seem inadequate.Therefore,an efficient load transfer strategy for HVDN is urgently needed to assist dispatchers in decision-making.In this paper,the load transfer problem of HVDN is studied from the model and algorithm.In model,load transfer is belong to reconfiguration optimization,the traditional reconfiguration problem is mainly based on the mid-low voltage distribution network.Due to structural differences,traditional reconfiguration model is not suitable for HVDN.In this paper,based on the structure of HVDN,the concept of functional unit is abstracted.The topology structure of HVDN based on functional unit is proposed.The functional units that are related to each other are divided into a unit group Changing the topology of the group indirectly changes the load rate of each power point.The functional unit group is the smallest operational object for load transfer.It is used as the object of the reconstruction modeling.In this way,the switching problem of reconstructing the whole network is transformed into a topology selection problem of a unit group,and the dimensionality reduction is realized.In order to select a suitable feasible topological state.Finally,a two layer optimization model of load transfer is put forward.In the upper layer optimization model,a load pre distribution scheme for a network is formed.In the underlying optimization model,the feasible topological states of the unit groups are enumerated,and finding the topology state closest to the upper layer.Finally,by using the characteristics of iterative updating of the optimization algorithm,actual topological state closest to load balancing scheme for main transformer is found.The load transfer model of HVDN is a complex optimization problem.It is an effective strategy to choose intelligent algorithm to solve this problem.This paper introduces a new intelligent optimization algorithm-teaching and learning optimization(TLBO)algorithm.This algorithm having less parameters,simple principle,easy programming and fast convergence,in solving the optimization problem,shows full advantage.In order to avoid falling into local optimal solution in solving complex high dimensional optimization problems,in this paper,the crossover operation of differential evolution algorithm(DE)is introduced into the algorithm to improve the search precision.The test results show the effectiveness of the improved algorithm.Then the algorithm is used to solve the optimal transfer model.Finally,the local distribution system of a high voltage power distribution network is tested and analyzed.Satisfactory results are obtained,and the effectiveness of the strategy is verified.