The Probabilistic Assessment Of Wind Power Accommodation Considering Uncertianties

With the rapid development of wind power in recent years,wind power generation has become one of the most mature power generation modes by new energy.However,with the increasing of wind power installed capacity,the phenomenon of wind power curtailment is more and more serious,which cause the wind power curtailment has become the major factor restricting the development of wind power at present.With the increasing complexity of the power system,the reasons of wind power curtailment become more and more.Therefore,considering many random factors to evaluate the capacity of wind power accommodation is of great significance for the study of wind power accommodation.In order to analyze the influence of uncertain factors on the wind power accommodation of electric power system comprehensively,this thesis focuses on the research of probability assessment for wind power accommodation of power system,the main research work is as follows:Because of the low accuracy of the traditional probability model of stochastic factors,a probability model based on the weighted Gaussian mixture distribution(WGMD)is proposed,and used to simulate the stochastic fluctuation of load and wind power.Moreover,as the commonly used parameter estimation algorithm of the weighted Gaussian mixture model,EM(Expectation Maximization)algorithm is easily converging to local optimization.According to this problem,the DAEM(Deterministic Annealing Expectation Maximization)algorithm improves the defects of EM algorithm by introducing the temperature coefficient.Then,the DAEM algorithm is applied to construct the probabilistic model of load and wind power.The accuracy and effectiveness of the proposed model and algorithm are verified by choosing different models and different parameter estimation algorithms to fit the probability density curves of the measured data of load and wind power.In the selection of analysis method for probabilistic assessment of wind power accommodation,because the traditional probability analysis method has the defect of high computational cost,this thesis proposes a Markov Chain Monte Carlo(MCMC)simulation method based on slice sampling algorithm.As a widely used sampling algorithm in the MCMC simulation method,the Gibbs sampling algorithm needs a lot of simulation calculation to get the more accurate calculation results,which limits the application.Compared with the Gibbs algorithm,the slice sampling algorithm is more flexible with less computation,and it is suitable for various kinds of distribution.Therefore,the MCMC method based on slice sampling algorithm is applied in probabilistic power flow calculation for evaluation of wind power accommodation.The simulation results show that the slice sampling method is more stable and more accurate than the Gibbs algorithm in the same sampling scale.According to the shortage of traditional evaluation model of wind power accommodation,a probabilistic assessment model of wind power accommodation based on conditional risk constraint is established.By considering the error of the traditional evaluation model,the wind power acceptance risk function is defined,and the conditional value at risk(CVaR)is introduced,as risk indicators,to quantify wind power accommodation risk.Then,the evaluation model of wind power accommodation based on condition risk constraints is constructed.By combining the evaluation model of wind power accommodation and the probability analysis method mentioned above,the probability distribution of wind power accommodation is obtained,at the same time,the risk value of each wind power consumptive level is calculated.In this thesis,the simulation results from IEEE 39 bus system show that the proposed methods can provide more accurate information for effective wind power accommodation under the premise of safe operation of the system.