Research On Reactive Power Optimization Of Distribution Network With DG Based On Hybrid Intelligent Algorithm

With the rapid development of economy in Belarus,higher requirements have been proposed for smart grid construction.Owing to promotion and operation of power market,it has become impossible to meet practical needs by traditional modes of power supply.The introduction of distributed generation(DG)not only has handled many deficiencies of large power system such as poor economical properties,long construction cycle,inflexible operation and maintenance modes,but also has solved some problems like insufficient energy sources supply and environmental pollution faced on Belarus society.Based on this,a brief introduction is made about definition and classification of DG sources as well as characteristics of different ones in this thesis.Their impacts upon power distribution networks are analyzed as well.Meanwhile,four common energy source's structures and working principles are described respectively.Power flow calculation is an important part of steady state analysis for power system,and it is also a foundation to solve many unsteady state problems of power system.Through detailed description of power flow calculation method used in the power distribution network,the flow calculation for power distribution networks is validated by some calculation instances through Forward and Backward Sweep Method.The results show that it is scientific and effective to introduce Forward and Backward Sweep Method into the flow calculation for power distribution networks.At last,aiming at reactive power optimization of the power distribution networks with DG,a hybrid algorithm based on the Genetic Algorithm(GA)and Ant Colony Optimization(ACO)algorithm is presented and used in reactive power optimization of one calculation example.The optimization results are analyzed and compared with ones obtained by single GA or ACO.The results shows that the hybrid algorithm has features of quick convergence speed and good optimization performance.