Research Of The Reactive Power Optimization On Distribution Network Considering PHEVs

In order to meet the needs of energy saving in the context of global warming,many countries have begun to develop electric vehicles. Parallel hybrid electricvehicle (PHEV) has a good prospect of application, but large scale of PHEVsconnecting to the distribution network will bring a huge security and stabilityrisk. The development of PHEV control programs and reactive poweroptimization can effectively solve the problem that the nodal voltage indistribution network is too low which is caused by large scale of PHEVs’connecting, it can also reduce the pressure of the operation and improve thebenefits of the distribution network.Firstly, the load of PHEVs’ connecting is studied with Monte Carlosimulation and the discharge of PHEVs to the grid is not considered. The peakload is coincident with the distribution system peak load and the differencebetween the peak load and neap load is too big, the delay charging based on thestate of battery capacity is proposed, there are different charging policies fordifferent states of the PHEVs, the PHEVs with poor state can charge rapidlywhile the others charge after a period of time. during the evening peak, allPHEVs take delay charging. Finally, the peak load of the distribution network isreduced. The simulation results show that the program can effectively reduce thepeak load, and improve the voltage level of the distribution network.Secondly, the influence of the PHEVs’ connecting for the power flowcalculation is researched, including distributed load and unbalanced powersupply in the distribution network, way of equivalent is proposed to deal with thedistributed load support on the branches, and the allocation factor of thedistributed load is introduced to pro rata the distributed load to both ends of thebranch. The virtual branch is proposed to deal with the unbalanced power supplyproblem which caused by large-scale PHEVs connecting to the distributionnetwork, let the single-phase (two-phase) power supply transform intothree-phase power supply, that the imbalance of the three-phase branchestransform into the imbalance of three phase load, effectively solve the problemthat the node admittance matrix is singular which caused by the imbalance of thebranches, examples show that the proposed scheme is valid.Finally, to solve the voltage limit problem caused by PHEVs’ connecting,the reactive power optimization of the distribution network is considered. Toexpand the search range of the population space and avoid the precocity of basicgenetic algorithm, transpose matrix and improved cross and mutation operator based on the research of the basic genetic algorithm are introduced. thetransposed matrix can improve its global search capability while the improvedcross and mutation operator can improve its local search capability. The fitnessanalysis of standard test function shows that the improved genetic algorithmbased on the transpose matrix operator has a stronger convergence and searchcapability. The distribution network capacitors’ switching optimization with theintroduction of PHEVs’ load is researched, the PHEVs’ load is proportionallyconfigured on the nodes, and take the penalty function method to deal with thevoltage constraints. Results of the IEEE69nodes system capacitor switchingoptimization show that the delay charging program combined with thedistribution capacitor switching optimization can solve the problem that the nodevoltage is too low when large scale of PHEVs connecting to the distributionnetwork, it can also reduce the system loss and improve the economic efficiencyof the distribution network.