Power System Reliability Assessment Considering Random Fluctuations Of Source And Load

In recent years,with the increase of new power generation permeability,the reform of power market operation mechanism and the promotion of electric vehicles,the uncertainty of power system operation is further increased.And it poses a hidden danger to power system safe and stable operation.Power system unreliability may leads to voltage instability,and may causes large area blackout under severe conditions.This phenomenon not only produces a great threaten to personal safety,but also causes huge economic losses.In this paper,weak bus identification method under considering load random fluctuation is analyzed,and power system reliability is evaluated considering the relevance between bus load and photovoltaic power station output.This paper is supported by State Grid Shanxi Electric Power Company Science and Technology Project(SGTYHT/14-JS-188)“Development of large scale intermittent new energy grid connected technology”.In this paper,the existing probabilistic power flow considering load fluctuation limit is improved.The Latin hypercube sampling is used to sample the power system bus load in order to obtain the initial random load coefficients.The BP neural network and the Newton-Raphson power flow are combined,and is used to the tested the convergence of initial random load coefficients.Then the convergent random load coefficients are obtained.On this basis,the improved bisection searching technique is used to increase load in random,and the bus voltages are obtained.The simulation results show that the calculation accuracy is guaranteed,at the same time,the calculation speed of probabilistic load flow considering load fluctuation limit is improved.In this paper,the weak bus identification method is proposed based on bus voltage distribution characteristics.The normal distribution characteristics of bus collapse voltage are analyzed by normal distribution probability paper test method in this paper.And on this basis,two voltage stability assessment indicators are proposed to identify the weak bus in power system.In the process of weak bus analysis,the Bootstrap method is applied to the indicator calculation.The simulation results show that the analysis and identification method of weak bus based on bus voltage distribution characteristic has very important meaning on fast identification of weak bus in the case of load random fluctuation.And the value of weak bus indicator is calculated more accurately than the parameter point estimation method under the same sampling size.The photovoltaic power station output and bus load model considering the relevance between them is proposed in this paper.The probability distribution of light intensity and bus load are analyzed by nonparametric kernel density estimation method,and the bandwidths are obtained.The Kendall rank correlation coefficient is calculated between light intensity and bus load.The load buses are divided into strongly correlated buses and weakly correlated buses.For strongly correlated buses,the Copula function is used to establish photovoltaic power station and load correlation model.And for weakly correlated buses,the Latin hypercube sampling is used to establish load random fluctuation model.The proposed model is validated by simulation,and the result show that this model can be used to simulate the distribution characteristics of the bus load and output of photovoltaic power station,as well as the correlation between them.The reliability of power system with photovoltaic power station is assessment in this paper.On the base of photovoltaic power station and load model,the loss of load expectation(LOLE)and the expected energy not supplied(EENS)are used to evaluate the power system reliability under different permeability of photovoltaic power station.The effective load carry capability(ELCC)and equivalent conventional capacity(ECC)are calculated under different permeability levels,and the capacity credit of photovoltaic power station is estimated.The capacity credit of power system are estimated when the relevance between bus load and light intensity is not considered,and the results are compared with the relevance is considered.