Research On Photovoltaic Output Characteristics And Station Group Correlation Based On Mixed Distribution Model

Photovoltaic power generation develops most rapidly in various new energy power generation modes due to policy support and cost reduction.Under the opportunity of high-speed development,it also faces many challenges.Photovoltaic power generation is more vulnerable to environmental impact,which makes it intermittent and random.High permeability photovoltaic grid connection brings adverse impact on the safe and stable operation of the power system,so it has a negative impact on photovoltaic power generation The study of output characteristics,fluctuation characteristics and spatiotemporal correlation of photovoltaic power generation has important practical significance.In this paper,the actual photovoltaic power plant as the research object,the photovoltaic output modeling,volatility extraction,volatility characteristics and photovoltaic output time-space correlation sequence modeling analysis are studied.Based on the measured photovoltaic data of power stations in Jiangsu and Sichuan provinces,the photovoltaic output is divided into measurable part and unmeasurable part by considering the geographical location,altitude,terrain conditions,climate and other factors.For the measurable part,firstly,the theoretical output model of photovoltaic output is established by using the solar radiation intensity,then the daily attenuation coefficient is obtained based on the measured historical output data by using the least square method,and then the theoretical attenuation model of photovoltaic output is established;for the non measurable part,namely the random component under the influence of multiple conditions,the theoretical attenuation output is different from the actual output Here.It can be used as the experimental model of photovoltaic output for in-depth analysis.According to the research on the distribution characteristics of the PV output wave momentum,a method of describing the probability distribution of the PV output fluctuation characteristics based on the mixed distribution model is proposed.First of all,verify and analyze the description degree of the single probability density distribution function to the wave characteristics,and prove that the fitting of the single distribution still can not meet the accuracy requirements.Then,according to the characteristics of each distribution function,build a mixed distribution model,and comprehensively compare a variety of mixed distribution models with different weights.Through simulation analysis and index verification,the TM-TLS distribution function has the best fitting degree and description effect best.Aiming at the correlation analysis of wide area photovoltaic power plants,a modeling method of correlation output series of multi photovoltaic power plants is proposed based on the change of dimensions under the influence of unmeasurable factors.Firstly,based on the idea of spatial dimensionality reduction and the semi variation correction formula,the photovoltaic power plant group is divided into several related areas.On this basis,using the idea of spatial dimensionality increase,the influencing factors affecting the difference of photovoltaic power output are weighed,and based on the probability distribution function,the photovoltaic power generation sequence is constructed by using the fluctuation component of photovoltaic power output Through the simulation of the actual output data of photovoltaic power stations in Jiangsu Province,the accuracy of the correlation of the methods and the accuracy of generating the output sequence of photovoltaic power generation are verified.