Probabilistic Power Flow Simulation Of Photovoltaic Grid-connected System Based On Random Variable State Time Series

The grid-connected photovoltaic system not only alleviates the energy shortage of the country,but also brings new problems to the safe and stable operation of the power grid.In recent years,with the rapid development of solar power generation,along with more and more dense distributed solar power plants and higher penetration of photovoltaic grid-connected,the one-way radial power supply mode of traditional distribution network has been changed.Various uncertainties,such as randomness,fluctuation and periodicity of photovoltaic power generation,make the power quality,transmission efficiency and power supply reliability of the grid occur.Great changes have taken place.In addition,with the progress of science and technology,the high voltage,transmission distance,capacity and the degree of tightness between the power grid,these factors make the current power grid has become a highly complex dynamic system.Due to the emergence of various random variables such as abrupt load change,unit outage and line fault,the operation of power network is full of various uncertainties,which may cause serious consequences at any time.In order to reduce its impact on the power grid,considering the probability of the occurrence of random variables,this paper mainly does the following work:(1)Based on the characteristics of photovoltaic power generation,this paper analyses the relationship between the output power of photovoltaic power generation under several main climatic conditions affecting photovoltaic power generation,and establishes a prediction model of photovoltaic power generation based on BP neural network.The model takes radiation intensity,ambient temperature,relative humidity and wind speed as input,photovoltaic power as output,and historical data as support,and predicts the failure.Photovoltaic power generation over a period of time.(2)Considering the characteristics of distributed photovoltaic power plants with decentralized resources and small installed capacity,the forecasted output power of photovoltaic power generation is merged into the standard system of 30 nodes at different nodes.The power flow distribution of a certain period of time in the case of photovoltaic grid-connected is calculated by Newton Raphson method,and compared with the simulation results in the initial state.Analysis.It is concluded that the influence of PV grid connection on different nodes is different,which enhances the voltage fluctuation of power grid and improves the voltage level of nodes to a certain extent.(3)Modern power grid is a complex dynamic system.The possible fault state of branch at any time will also become an important potential factor affecting the stable operation of power grid.Sequential Monte Carlo method is used to simulate the operation state of power grid with fault probability in time sequence.The probabilistic power flow calculation of power grid considering both grid-connected photovoltaic and grid-operated States will be obtained.The results are compared with the simulation results of the above initial state and the simulation results only considering the photovoltaic grid-connected state.The results are closer to the actual situation,which can effectively reflect the voltage and power flow of the grid operation risk,clarify the state of the power system,find out the weak links in the power grid in time,and formulate accident prevention and power generation in a targeted manner.Network improvement measures provide reliable data for power system planning,and ensure the coordinated and synchronous development of power grid construction and power supply construction as a whole.