Research On Active Power Control Of Grid-Connected Photovoltaic Power Station

At present,the balance between energy and environmental protection is a big problem in the world,so the concept of sustainable development has attracted people’s attention.With the rapid development of science and technology around the world,the value and application of photovoltaic power generation has been gradually explored.As the most important component of photovoltaic power generation,solar panel bears the role of energy conversion.The photovoltaic effect of crystalline silicon can realize the conversion of light energy and electricity.Although solar energy has the advantages of being cheap.clean and efficient:with the increasing number of solar power stations,the proportion of solar power generation system is also increasing.Moreover,the efficiency of solar power generation is greatly affected by the weather,which leads to great changes in grid-connected power,which is not conducive to the smooth operation of the power grid system.Therefore,it is necessary to take corresponding measures at the power station end to control its grid-connected active power reasonably.so as to meet the dispatching requirements and maintain the safe and stable operation of the power grid.After reading alot of relevant domestic and foreign literature,this paper determines that the research is conducted from the following aspects:in the aspect of grid-connected power prediction,the BP neural network prediction method is firstly studied,and wavelet neural network prediction method is introduced on this basis.By comparing the two prediction methods,it is determined that the wavelet neural network has the superiority in the prediction application in this paper.Establishm ent and the weather for the day of the same type similar sample set:using the taonan in jilin province the history of a photovoltaic power station power generation data as input data,sunny,cloudy,rainy day etc of the selected weather types of photovoltaic power station the hour parallel active power to get short-term parallel active power between predicted photovoltaic power station,supply network scheduling plan and photovoltaic power station control.In active power station control strategy.Due to the weather and environmem ental factors,the grid-connected power output of photovoltaic power generation is greatly affected and its output power is unstable.In addition because of the over-use of inverter and the low conversion rate of electric energy in the past grid-connected active power control,this paper proposes an economic optimization control method.Based on the optical power prediction,the active power of the power station is controlled,so that the inverter is started at the same time at the least and the active power of the power station is maximized.