Study Of Multi-scene And Comprehensive Index Reactive Power Optimization For Distribution System With Wind Turbines

With the shortage of global fossil energy and the enhancement of people's environmental awareness,wind power has received widespread attention and vigorously development because of its huge benefits showed in energy saving and emission reduction.As more and more wind turbines connected to the power grid,the structure of the distribution system changes from a single-supply radial structure to a complex network of multiple power supplies.And at the same time,the power flow of distribution network will change greatly.In addition,the power influx of wind turbines also has an important effect on the node voltage,power loss of branches and operational reliability of the distribution system.Wind turbines,as a source of reactive power,together with the traditional reactive power compensation equipment,participated in reactive power optimization of the distribution network,can greatly improve the operation stability of power grid,improve the voltage quality of power grid,improve the static voltage stability margin,and can reduce the network loss effectively,improving the operation economy of the power grid.Therefore,it is significant to study the reactive power optimization of distribution network with wind turbines.In this paper,about the reactive power optimization problem of the distribution after the wind power generating units connected,research work about the following aspects mainly has been done:Firstly,for the randomness problem of the wind turbine output power,the scene analysis method is introduced to solve the uncertainty problem.Based on the traditional reactive power optimization model,a reactive power optimization model of distribution network based on scene analysis is established.The model takes network active power loss,voltage deviation,static voltage stability margin,and the investment of the reactive power compensation equipment as optimization targets,and a multi-objective optimization model of comprehensive index is established.In the optimization process,the wind turbine adopts the PQ simplified model,as the PQ node participate in the power flow calculation.If the power flow does not converge or the variables exceed limited ranges,the penalty function is added to the objective function to reduce the chance that the bad solution will be left.Secondly,a hybrid algorithm——Adaptive harmony search particle swarm optimization(AHSPSO)algorithm is proposed,which combines the harmony search algorithm and the particle swarm algorithm,making full use of the global optimization ability of the harmony search algorithm and the fast convergence ability of the particle swarm optimization algorithm.Then the hybrid algorithm and several other algorithms are applied to the IEEE 33 node system to solve the reactive power optimization problem.The results show that the AHSPSO algorithm has the advantages of few parameters,simple structure,high convergence precision,fast calculation speed and easy programming,and can solve the problem of reactive power optimization effectively with certain practicability.Thirdly,under different scenarios with the same target,reactive power optimization of the distribution network including wind turbines is carried out,the results show that increasing the reactive power compensation capacity can effectively reduce the network active power loss,increase the node voltage level,and increase the static voltage stability margin in a certain range,when the reactive power output of wind turbines is involved in the reactive power optimization of the distribution.For power systems,wind turbines are better to operate at under-rated power mode.Besides,under the same scenario,the optimized objectives are different,the reactive power compensation schemes are different.Taking comprehensive index as optimization objective could obtain compromised scheme to coordinate every optimized target,which is more suitable for the reactive power optimization of the actual power system.What's more,from the study of influences of load variation on the optimization results,it is found that with the decrease of load rate,the net loss and voltage deviation index decrease gradually and the voltage stability margin gradually increases.With the gradual increase of the capacitive load in the system,the network loss and voltage deviation are gradually decreasing,but the voltage stability margin is gradually reduced.When all the loads are capacitive loads,the system is likely to lose stability.