Probabilistic Load Flow Calculation For Power Grid Containing Photovoltaic Plants

In recent years,photovoltaic(PV)generation gets rapid development,and the penetration rate in the grid is gradually increased.To develop photovoltaic generation is of great significance to the country’s energy supply and energy security,and it is one of the development trend of the power grid in the future.But PV generation is affected by the weather conditions such as light intensity,which makes the power outputs of PV generation have strong fluctuation.It is difficult for traditional determinative load flow calculation to assess the fluctuation of the system,while probabilistic load flow(PLF)calculation can consider the fluctuation of the power outputs of PV generation,and probabilistic load flow calculation can also consider the random fluctuation of load as well as conventional generator fault outage etc.The probability distribution of node voltage and branch power flow can be obtained by probabilistic load flow calculation,and it can provide important reference information for the planning and operation of power system as well as relay protection setting.The traditional probabilistic load flow is calculated and analyzed by the cumulant method,and it requires that the inputs of the nodes are independent of each other,but this condition in the actual situation often can’t be met.The power outputs of PV generation system close in geographical location possess correlativity,which will affect the accuracy of probabilistic load flow calculation.The modified Nataf transformation is adopted to handle the correlativity of PV generation,combining with the cumulant method and C-type Gram-Charlier series to calculate probabilistic load flow.The proposed method not only has high computation efficiency compared with Monte Carlo simulation method,but also can reflect the transformation process between variable spaces,and the method also has a good treatment effect on the correlativity of PV generation which does not obey the normal distribution,meanwhile C-type Gram-Charlier series can avoid negative probability density values.The validity of the proposed method is verified by the comparative simulation results in the IEEE 14-bus system and an actual power grid in the south.