Probabilistic Assessment Method Of Power Quality Of Rural Distribution Network Containing Photovoltaic Power Generation

With the continuous upgrading of the load side of the power industry and the innovation of the new energy power generation industry,the power system has become complex and random,and the security and quality of power supply to the grid have been challenged like never before.With the promotion o f photovoltaic poverty alleviation projects,more and more photovoltaic power plants scattered throughout the country are integrated into the power grid terminals by power sources,and there is a reverse flow in the power grid,which will lead to reduced power system reliability and difficult safety assessment.Series impact.Aiming at the random and variable influence of power system influencing factors,Probabilistic Load Flow(PLF)can comprehensively consider the uncertainty of multiple input variables,which can help to find the weak areas and potential faults of the power supply system,and more fully reflect the power grid.Operating status.Therefore it is widely used in wind power,photovoltaic and other new energy power generation.In this paper,the rural power distribution network is used as the object,and the power quality probability assessment method for photovoltaic power generation in the rural power distribution network is researched and practically applied.Due to the uncertainty and randomness of photovoltaic power generation,it directly affects the flow of power systems.In order to be able to accurately analyze the probabilistic power flow of power systems,accurate modeling of photovoltaic output is essential.In the photovoltaic power generation system,in order to solve the problem of low accuracy when fitting the photovoltaic output with Beta distribution,this paper uses a cubic spline function to fit the photovoltaic output,and then uses the polynomial normal transform technology to deal with the correlation of the photovoltaic output.The sampling algorithm generates initial samples and applies extended Latin hypercube sampling to the slice sampling technique.The algorithm in this paper not only considers the correlation of photovoltaic output,but also avoids the shortcomings of Gibbs sampling algorithm that requires a lot of complex iterative operations to obtain more accurate calculation results.Compared with MCSM,it greatly shortens the operation.time.Simulation results verify the accuracy and effectiveness of the proposed method.In order to analyze the impact of load more accurately,this paper proposes to establish a Gaussian mixture model and use the semi-invariant method to calculate the probability power flow.Firstly,in the aspect of photovoltaic output modeling,the photovoltaic output is discretely segmented,and the correlation coefficient of photovoltaic output is constructed by using the Nataf transform.Then,a load Gaussian mixture model is constructed according to the characteristics of the load.Second,the semi-invariant method is used for power flow calculation.Finally,for each type of system injected power,the semi-invariant method is used to calculate the expectation and variance of the corresponding state variables.Compared with Monte Carlo,this paper uses a method to simplify the power flow calculation process and improve the calculation efficiency.