Research On Influence Of Wind Power Access On Power System Reliability

As the main green energy in the 21 st century,wind energy plays a significant part in non-renewable energy.With the continuous renewal of wind power generation equipments and the continuous improvement of power generation technology,wind power generation has become the most mature,the most enormous scale development potential,the most commercial potential of new energy generation technology.With the sustained increasement of the proportion of wind power connected to the system,the adverse impact of wind power grid-connected operation on the power system has become increasingly prominent.Therefore,it is particularly important to evaluate the power system reliability of wind power access in order to realize the full use of wind energy and the unity of wind energy and power system reliability.Aiming at the problem of inaccurate prediction of wind power,this thesis proposes an improved fireworks algorithm(IFWA)to optimize the wind power prediction model of back propagation(BP)neural network.Firstly,the influence factors of wind power are analyzed and processed,and the prediction model of single BP neural network is established and expounded in detail.Then,the explosion radius,mapping rules and selection strategy of the fireworks algorithm(FWA)are improved and the single point crossover operation is added,the IFWA is used to optimize the BP neural network and establish the model.Finally,the output results of BP prediction model,FWA-BP prediction model and IFWA-BP prediction model are compared by simulation through the historical data of a wind farm.The results show that the IFWA-BP prediction model has faster convergence speed,higher prediction accuracy and better fitting effect,which provides an effective basis for quickly and accurately assessing system reliability.Due to the low efficiency of monte carlo state sampling,this thesis uses improved latin hypercube sampling(LHS)to change the probability distribution of sample space,optimize the sampling process,divide intervals and stratified sampling.The IEEE-RTS79 system simulation shows that the improved LHS optimization monte carlo method has higher accuracy,faster calculation speed and smaller sample space variance,which can meet the requirements of power system reliability assessment.Then,the reliability evaluation process of wind power access to power system is clarified,and the wind farm capacity,single node and multiple nodes of wind farm access system and wind power hybrid energy storage are analyzed in turn by using the improved LHS optimization monte carlo method.The four methods can improve the reliability of power system.Finally,aiming at the practical application of reliability evaluation of wind power connected to power system,this thesis studies a local power grid in Southwest China,and the relevant lines are improved according to the method of identifying the weak links of power system to further improve the system reliability.